View Full Version : The NEXT LAST R600 Rumours & Speculation Thread
By the way, in theory, NV30 has a 25.6GFLOPS when it comes FP32, and a >50GFLOPS when it comes to FP16, while R300 has a 20.8GFLOPS math power.I don't know how you got to those numbers, but errr, congratulations, they're all wrong! :D ;)
NV30 is 500x4x8 = 16GFlops for both FP32 and FP16. R300 is 325x8x12 = 31.2GFlops for FP24. Both IIRC. Of course, NV30 is ridiculously faster than that with FX12, assuming no register pressure. But that's arguably not very useful.
Looking at the detailed specs of NV30/NV35 and R300 with some years gone by and some added perspective, I don't think it's fair to claim that the specs made NV3x seem better than R3xx. The clockrate advantage of NV30 against R300 is pretty much the only advantage NVIDIA had spec-wise, for the entire line-up, imo...
Anyway, back to R600 now, please - and sorry for continuing the OT, but it's not like there's much I/we can say on-topic anyway, except perhaps bitch at everyone linking retarded bullshit! :)
To be honest, to compare R600 with NV30 is an insult to NV30, since after all, NV30 is only 4 months later.:lol:
Oh, nice math there. I guess AMD should have done a "tech preview" in December to get guys like you to stop the clock? Riiiiight.
How could a 50% math advantage fail to produce significant performance gains just because AA and AF are off?
How many GPGPU apps that you know of are CPU limited? That's just one example, of course.
3dilettante
29-Mar-2007, 20:32
How many GPGPU apps that you know of are CPU limited? That's just one example, of course.
That's a good point. I don't know of any GPU isolating test results being discussed in the rumors.
Oh, nice math there. I guess AMD should have done a "tech preview" in December to get guys like you to stop the clock? Riiiiight.
One thing I do have to say: When Nvidia realized that the NV30 was crap, they frantically tried to fix it. AMD/ATI has shown virtually no such sign. Not to say that they aren't frantically trying to fix it behind closed doors, but they're giving the impression that the R600 isn't very good.
3dilettante
29-Mar-2007, 20:46
What signs would we be looking for that would indicate AMD is frantically trying to fix R600?
There are a lot of signs Nvidia was pushing NV30 hard, such as the high clock and heavy shader replacement, but those weren't really known until after the release, and the extent wasn't clear for quite some time afterwards.
What signs would we be looking for that would indicate AMD is frantically trying to fix R600?
There are a lot of signs Nvidia was pushing NV30 hard, such as the high clock and heavy shader replacement, but those weren't really known until after the release, and the extent wasn't clear for quite some time afterwards.
The NV30 was quickly scrapped and replaced by the NV35, with a 256-bit memory bus and somewhat improved shader units. The other things happened relatively shortly after launch, not long after. Not to mention the massive hype machine that nVidia used to try to stem the tide. With the R600, we hear next to nothing. It gives the sense that they don't have anything to show.
The NV30 was quickly scrapped and replaced by the NV35, with a 256-bit memory bus and somewhat improved shader units. The other things happened relatively shortly after launch, not long after. Not to mention the massive hype machine that nVidia used to try to stem the tide. With the R600, we hear next to nothing. It gives the sense that they don't have anything to show.
Or it could mean that the chip is so fantastic there's no need to hype it to all heaven. ATI didn't hype the R300 pre-launch. They just let it detonate over the marketplace.
Wild speculation on marketing, or lack thereof, is pointless at this juncture. We're a few weeks away, *knock on wood*, from a technology demonstration. And a couple of weeks after that from actual benchies.
Let's see what happens.
Deusp,
What, as in there's no R600? :???:
When did we get the impression that R600 isn't any good, let alone crap? And how does one equate R600 to NV30 with contradictory examples (hype machine vs. subdued PR)?
All this waiting is driving me stir-crazy, too, but let's save the funeral for after the birth. There's no sense in AMD releasing specs or benchmarks well ahead of launch, b/c that just gives NV PR something to shoot at.
Bouncing Zabaglione Bros.
29-Mar-2007, 23:48
The NV30 was quickly scrapped and replaced by the NV35, with a 256-bit memory bus and somewhat improved shader units. The other things happened relatively shortly after launch, not long after. Not to mention the massive hype machine that nVidia used to try to stem the tide. With the R600, we hear next to nothing. It gives the sense that they don't have anything to show.
No, it means that like a lot of other companies, they realise there is not a lot of point in wasting your marketing resources hyping up a product without actually having it to sell to people at the same time.
You may remember about NV30's quick replacement, but do you remember the preceding six months where Nvidia hyped "Cinematic Computing" to death? CG? The infamous PS 2/3 comparisons using Far Cry? Cheating on 3D Mark and withdrawal from the Futuremark? Nvidia looked pretty silly after all that resulted in the NV30 instantly cancelled product.
AMD would rather keep it's mouth shut than dig itself into a big hole. Right now all they are is late. They don't want to be late and stupid with it.
At this point, leaks of R600 being quite spectacular is the only thing that will stop people throwing up their hands in the air and buying a G80 out of frustration with AMD not providing anything competitive for half a year.
At this point in time, any leak would help keep me from buying a G80 out of frustration. If they'd say it's atleast similar in speed, then I'd wait for quality comparisons at least.
zgemboandislic
30-Mar-2007, 01:58
At this point in time, any leak would help keep me from buying a G80 out of frustration. If they'd say it's atleast similar in speed, then I'd wait for quality comparisons at least.
No. No, no, no! They can't be six months late AND be similar in speed with somewhat better image quality. From a product being six months late, I expect it to be a) considerably faster, b) cooler and consuming less power, c) equal or better image quality! Especially it being on a smaller process. Come on!
They either fulfill those three major points or their product is burried on launch.
Mcmlxxx IV
30-Mar-2007, 03:04
The NV30 was quickly scrapped and replaced by the NV35, with a 256-bit memory bus and somewhat improved shader units. The other things happened relatively shortly after launch, not long after. Not to mention the massive hype machine that nVidia used to try to stem the tide. With the R600, we hear next to nothing. It gives the sense that they don't have anything to show.
Does ATI itself usually hype up its upcoming cards? Lack of direct hype != a bad card.
No, it means that like a lot of other companies, they realise there is not a lot of point in wasting your marketing resources hyping up a product without actually having it to sell to people at the same time.
You may remember about NV30's quick replacement, but do you remember the preceding six months where Nvidia hyped "Cinematic Computing" to death? CG? The infamous PS 2/3 comparisons using Far Cry? Cheating on 3D Mark and withdrawal from the Futuremark? Nvidia looked pretty silly after all that resulted in the NV30 instantly cancelled product.
AMD would rather keep it's mouth shut than dig itself into a big hole. Right now all they are is late. They don't want to be late and stupid with it.
Well PR is almost always stupid. I rather have a bad product with shit-loads of hype than a mediocre one with no hype. Perhaps it's distasteful, but I think it is the right thing to do from a business standpoint. All this non-noise is only driving people to buy 8800s, which will help dry up demand by the time the R600 comes out unless it's completely awesome. At least with lots of unjustified hype, people would hold out.
I recall that with the R300, while people were indeed impressed, they weren't impressed enough to simply ignore what nVidia was going to show, and only when it was clear the NV30 was crap did most people jump ship. This is not the case with the G80. People are impressed enough that many have already jumped ship.
quest55720
30-Mar-2007, 04:29
Well PR is almost always stupid. I rather have a bad product with shit-loads of hype than a mediocre one with no hype. Perhaps it's distasteful, but I think it is the right thing to do from a business standpoint. All this non-noise is only driving people to buy 8800s, which will help dry up demand by the time the R600 comes out unless it's completely awesome. At least with lots of unjustified hype, people would hold out.
I recall that with the R300, while people were indeed impressed, they weren't impressed enough to simply ignore what nVidia was going to show, and only when it was clear the NV30 was crap did most people jump ship. This is not the case with the G80. People are impressed enough that many have already jumped ship.
No in general people were floored by the r300 it was pretty much the first card that had usable AA and AF on new games. The R300 caught pretty much every one off guard with its 8pipes 256bit bus and incredible performance. It completley laid waste to everything in site at the time. The only people who held out were the most delusional Nvidia fans. I can't remember anyone on this site who did not give mad love to ATI for the R300 when it hit.
Someone's already declared every GPU discussion devolving into R300/NV30 comparisons a B3D Forum Law, right? Me, I prefer car analogies.
I still think R600 won't be cheaper than G80 simply b/c it should have more RAM per comparable SKU. I'm guessing NVIO is a comparatively negligible cost and GPU cost will come out even thanks to 80 vs 90nm. I'm also guessing G80 will continue to have relatively more TMU than shader power than R600, and that ATI most likely won't push more than 4 AA samples/s. That leaves huge bandwidth as the differentiator, and though it wouldn't be out of character for ATI to boost baseline IQ over max performance, I still think performance will be the differentiator at the high end and ATI will have a slight edge with (to borrow an NV term) HRAA scenarios.
Still, the shader scenario continues to intrigue. 320*800MHz vs. 128*1350(++)MHz. Will it matter enough given current benchmark suites? Not to mention that a G80U pushing 128*1800MHz pulls pretty close upon naive inspection, and may not need to clock the rest of its core much higher if it's got more TMUs, which may mean even a G80U consumes significantly less power than a R600XTX.
There, I got all that clueless speculation off my chest. I'm good for another week of waiting.
No in general people were floored by the r300 it was pretty much the first card that had usable AA and AF on new games. The R300 caught pretty much every one off guard with its 8pipes 256bit bus and incredible performance. It completley laid waste to everything in site at the time. The only people who held out were the most delusional Nvidia fans. I can't remember anyone on this site who did not give mad love to ATI for the R300 when it hit.
Kinda like the G80 right now, right? :lol:
I'm pretty sure most people expected Nvidia to respond in kind. Many I recall were quite shocked when it didn't perform as well as the R300.
Ailuros
30-Mar-2007, 07:37
Someone's already declared every GPU discussion devolving into R300/NV30 comparisons a B3D Forum Law, right? Me, I prefer car analogies.
ROFL :D
I still think R600 won't be cheaper than G80 simply b/c it should have more RAM per comparable SKU. I'm guessing NVIO is a comparatively negligible cost and GPU cost will come out even thanks to 80 vs 90nm. I'm also guessing G80 will continue to have relatively more TMU than shader power than R600, and that ATI most likely won't push more than 4 AA samples/s. That leaves huge bandwidth as the differentiator, and though it wouldn't be out of character for ATI to boost baseline IQ over max performance, I still think performance will be the differentiator at the high end and ATI will have a slight edge with (to borrow an NV term) HRAA scenarios.
Still, the shader scenario continues to intrigue. 320*800MHz vs. 128*1350(++)MHz. Will it matter enough given current benchmark suites? Not to mention that a G80U pushing 128*1800MHz pulls pretty close upon naive inspection, and may not need to clock the rest of its core much higher if it's got more TMUs, which may mean even a G80U consumes significantly less power than a R600XTX.
Errr...why only 4xAA or was that just a typo? As for HRAA differences I'd personally prefer to see a few more ROP-interna details before I'll speculate too much about anything. Frankly I would expect to see also here single cycle 4xMSAA and thus at least also 8xMSAA, but a finer description of all the ROP capabilities would help a tad more.
As for the theoretical maximum GFLOP numbers which IMHO will come damn close to the theoretical maximum values (for both) only in singled out synthetic benchmarks, I wouldn't be surprised if NV clocks some sort of "Ultra"-whatever slightly higher and gives it a significant shader domain boost to equal at least those rather senseless numbers.
I'm not saying that I don't expect R600 to not end up any fast than the 8800GTX, rather the contrary; but senseless maximum theoretical numbers don't really mean all that much and that goes for both sides.
Useless without AA/AF information.
I must have missed something. 3dmark-06 uses AA/AF?
Ail, whoops, typo. I meant 4 AA samples/clock, like G80.
IbaneZ, I thought 3DM06 uses no AA and 4x AF by default?
flopper
30-Mar-2007, 09:42
I guess we will see a lot of 8800 cards on ebay after r600 with 1gb ram gets out ;)
For me either ati/amd does not seem worried.
The postpone could been good for several reason,
marketing a whole one for all solution with the spanish baby with a red dragster using a cruisecontrol in all levels of products seems like a nice launch,
its just not a videocard anymore, its a full set of products used in conjunction with eachother.
Spanish baby=barcelona K10 quad genuine cpu.
Red dragster= r600 high end videocard.
cruisecontrol= new chipsets.
Use AMD to bring out the power in your videocards.:grin:
http://www.fudzilla.com/index.php?option=com_content&task=view&id=354&Itemid=1
:wink:
I recall that with the R300, while people were indeed impressed, they weren't impressed enough to simply ignore what nVidia was going to show, and only when it was clear the NV30 was crap did most people jump ship. This is not the case with the G80. People are impressed enough that many have already jumped ship.
As the old saying goes, haste makes waste. The thing about "jumping ship" is that you want to avoid doing it when you can, unless you have a magic wallet and you don't care if you have to jump back...;) If R600 turns out to be appreciably better than G80, then the early "ship jumpers" are going to wish they hadn't been so hasty. OTOH, if G80 turns out to be appreciably better than R600, then the only "penalty" I can see that people will "pay" for waiting to buy it is that they'll probably wind up with much better G80 drivers from the start than the ones that have shipped with the G80s thus far.
IE, since it is well known that the R600 launch is imminent, it really makes no sense to "jump ship" right now, does it? Far better to wait until the product launches to decide on what ship you want to sail. Besides, at the moment, there isn't a single game out there, or which has recently gone gold that my current x1950 Pro AGP won't run with ease...;)
BTW, the only reason people waited to "jump ship" post R300 (I didn't--I went straight to R300 from nv25) is because ATi was a darkhorse 3d gpu maker at the time, and people weren't accustomed to ATi leaping out ahead of the former pack of 3dfx and nVidia. That made it much easier to believe some of the hype nVidia kept spewing about nV30 right up until nVidia canceled it. Since then, Ati has been out front for a long time, and ATi has completely revamped its driver reputation. So I think that today it's much easier to wait and see what ATi is bringing to the table.
Evildeus
30-Mar-2007, 12:17
IE, since it is well known that the R600 launch is imminentDo we? Last time it was imminent also, don't you think? Not that i think AMD will delay once more the launch, but i think it would be wise not to be certain knowing the past of the R600 launch dates ;)
trinibwoy
30-Mar-2007, 12:34
OTOH, if G80 turns out to be appreciably better than R600, then the only "penalty" I can see that people will "pay" for waiting to buy it is that they'll probably wind up with much better G80 drivers from the start than the ones that have shipped with the G80s thus far.
Hmmm what about those people that held on to pretty weak GPU's while waiting for R600 when they could have been having a much better gaming experience on a G80 based card. You know, like those people suffering with X1950Pro's for example :razz:
northfieldz
30-Mar-2007, 12:41
To be honest, it is quite amusing to see fanboys making excuses.
Sorry, but if R600 is anything remote to a monster performance-wise, and if the only problem is yield, ATi won't delay it, it will simply release a degrade version with lower frequence and disabled units, since it will be enough to keep the competition and buy their time and money, thats something anyone whose IQ >50 can figure out.
No,the reason why R600 is so late, is not because the bullsh!t family launch, is not because ATI want wait until there are some DX10 games in the market, nor is because it is too strong so ATI want to hide it to give NV a deathblow (lol, you have to love those die-hard fanboys' funny little logic).
The only reason that why R600 is late, which is quite obviously to anyone whose IQ>50, is because it DO NEED all the fvcking help, and DO HAVE TO clock that high to just keep the pace, and hence the power problem, the yield issue and the delay, delay and delay.
So please, quit day-dreaming, after all, NV30 is only 4 months later (Nov compare to Jul), and it is not hard to image the performance of R600 based on industrial history:lol:
northfieldz
30-Mar-2007, 12:47
btw, when I say NV30 is 4 months later, I mean it is 4 months later than its competitor, R300.
And now R600, is already 4 months later, and is projected to be about 6 months later, judging by the trend, it won't take too much time before ATI is one generation later:lol:
Please stop
Actually the performance of the r600 is very good :wink: and yeilds are probably the only thing holding it back, don't know if lower clocked parts will yeild better, because don't know where the problem is.
Launching a card with poor yeilds will have the same effect if not worse if the product is only competitive in nature or slightly faster. More people will jump ship, because now everyone knows the performance there won't be anything holding these people back.
As you said look at history see what happened in the last 2 generation launchs for ATi, they came out with products with performance slightly better performance (initially for last gen, drivers improved the performance much more later on) but still got pounded on sales. Now after the g80 was released the last gen ATi dropped prices on thier mid top end cards and grabbed marketshare back.
You can't use the nv30 as an example, because the analogy doesn't work out since the nv30 had more issues then being late and hot.
The irony and flame-bait in the above two posts is obvious. :lol:
The irony and flame-bait in the above two posts is obvious. :lol:
Indeed, G80 was supposed to launch in May of 2006 (Computex) and be a direct competitor to R580. instead it was delayed for months and when it finally was released it proved to be a revelation in nV's product portfolio, Delays tell you nothing.
Just.. as the user himself.
INKster
30-Mar-2007, 14:26
Indeed, G80 was supposed to launch in May of 2006 (Computex) and be a direct competitor to R580. instead it was delayed for months and when it finally was released it proved to be a revelation in nV's product portfolio, Delays tell you nothing.
IIRC, G80 was never meant to launch in May of last year.
Why would they do that, just a short two months after the release of both the 7900 GTX and 7900 GT ? ;)
vertex_shader
30-Mar-2007, 14:33
Looks like NV AEG agents wake up for the winter sleep and they are already scared, this i why in the last few days the NV30 hit in this topic number1 spot, very bored....
chavvdarrr
30-Mar-2007, 14:38
some people need to put off their pink , ex-red glasses.
Actually the performance of the r600 is very good and yeilds are probably the only thing holding it back, don't know if lower clocked parts will yeild better, because don't know where the problem is.LOL
So, in order to have "very good performance" (your words!, very good, not "better performance"), ATi needs to increase clock so high that yields are really bad ?
O, wait, we'll hear once again how an error in library was to blame ...
Yes, I think AMD f***ed up buying the wrong company. Better to merge with the winner than to buy the looser.
btw, when I say NV30 is 4 months later, I mean it is 4 months later than its competitor, R300.
Quit repeating that, as it isn't true. NV30 was 7 months later than R300, August 2002 vs March 2003. Trying to use the Nov 2002 "tech preview" as the date for NV30 is a very contentious agenda-ridden device. Even the late January 2003 "previews" were only press. The public could not buy 5800 Ultra until March 2003.
trinibwoy
30-Mar-2007, 14:49
Indeed, G80 was supposed to launch in May of 2006 (Computex) and be a direct competitor to R580.
Well thank the stars that didn't happen. ATI might have been swimming with the fishes by now if R580 went head to head with G80 in any form. Might have had a big impact on AMD's acquisition decision too. I find it hard to believe that Nvidia was aiming so high in May of 2006 though. R580 launched only a few months earlier and was already considered very ambitious for the 90nm process at the time. Heck, G80 launched in November and people were still shocked at what Nvidia squeezed out of 90nm - G80 in May would have been devastating.
I must have missed something. 3dmark-06 uses AA/AF?
Not default, but yes. You can set that.
some people need to put off their pink , ex-red glasses.
LOL
So, in order to have "very good performance" (your words!, very good, not "better performance"), ATi needs to increase clock so high that yields are really bad ?
O, wait, we'll hear once again how an error in library was to blame ...
Yes, I think AMD f***ed up buying the wrong company. Better to merge with the winner than to buy the looser.
And you know they're rising the clock frequencies so high it has heavy impact on yields based... what exactly? Inquirer rumors of clocks? For all we know, the possible yield problems might have nothing to do with the clockrates, if there even is a yield problem.
We do NOT know why it's been really delayed, and those that might know, are under NDA not to tell about it.
I don't know that it was May, but it was in that neighborhood. They talked about it the previous November. That was probably best case. That there was then September rumours and then finally a November release would suggest it took two spins more than the "rosy scenario" that would have allowed them to hit their first target.
Evildeus
30-Mar-2007, 14:56
Till we don't have any datas, we neeed to wait, and i hope that the people waiting to buy R600 won't be displeased :)
On possibly related note, Kingpin, while breaking the freshly made 3DM06 record by Overklokk, said that he was already about to bury the 8800GTX's since he got new toys in already - R600 or 8800 Ultra:?:
trinibwoy
30-Mar-2007, 15:07
I don't know that it was May, but it was in that neighborhood. They talked about it the previous November. That was probably best case. That there was then September rumours and then finally a November release would suggest it took two spins more than the "rosy scenario" that would have allowed them to hit their first target.
Hmm so the 7900GTX that launched in March was just a backup? That's hard to believe but if that's the case then these companies have some pretty damn good contingency plans.
trinibwoy
30-Mar-2007, 15:08
On possibly related note, Kingpin, while breaking the freshly made 3DM06 record by Overklokk, said that he was already about to bury the 8800GTX's since he got new toys in already - R600 or 8800 Ultra:?:
Cool, these guys always get the new toys in advance. Got a link to his post? Although it would be pretty unbelievable that AMD lets R600 out into the wild over a month before its expected announcement.
Cool, these guys always get the new toys in advance. Got a link to his post? Although it would be pretty unbelievable that AMD lets R600 out into the wild over a month before its expected announcement.
http://www.xtremesystems.org/forums/showpost.php?p=2097574&postcount=58
Benching doesn't get any better than this Overklokk
Your doing really great.
I must admit I get pretty lazy sometimes and orb battles are the only thing that light the fire and get me going again.
I was all ready to hang up g80 benching for something(s) else that I cant talk about just yet ..but I will give it one last go.
Ln2 is nice, but takes about 30 litres to get through one 06' session. To be honest, I run my cards at cascade temps 99% of the time, except for when I am pushing the hardest, I wish I had some gpu cascades sometimes to make life alot easier when working the scores up. 3 units all the time can be a headache if your not up to it.
Kingpin is pretty "tight" with NVIDIA right? Wasn't he the guy in that priceless video@G80 launch.
Kingpin is pretty "tight" with NVIDIA right? Wasn't he the guy in that priceless video@G80 launch.
Just did a bit of background check, and indeed, he's pretty tight with nV so I guess it's 8800 Ultra or maybe even GX2 in his hands, then
INKster
30-Mar-2007, 15:27
"Something(s)" ?
Humm...
Hmm so the 7900GTX that launched in March was just a backup? That's hard to believe but if that's the case then these companies have some pretty damn good contingency plans.
Well, some of us think that 7900GTX was a bit later than originally intended as well, so with that in mind the whole roadmap just shuffled a bit to the right. . . :smile: My nomination for "Plan B" would actually be 7800 GTX 512. . .
vertex_shader
30-Mar-2007, 15:50
Kingpin is pretty "tight" with NVIDIA right? Wasn't he the guy in that priceless video@G80 launch.
Yes, like Macci and Sampsa was for ATi.
trinibwoy
30-Mar-2007, 16:15
http://www.xtremesystems.org/forums/showpost.php?p=2097574&postcount=58
Cool thanks, at least it's a good sign that we'll have something new to talk about soon :) I really hope it's not a GX2 though - there are enough SLI benchmarks out there already.
trinibwoy
30-Mar-2007, 16:24
Well, some of us think that 7900GTX was a bit later than originally intended as well, so with that in mind the whole roadmap just shuffled a bit to the right. . . :smile: My nomination for "Plan B" would actually be 7800 GTX 512. . .
Yeah but once they used up their get-out-of-jail card for the 7900GTX delay then they obviously weren't aiming for May for G80 any more. So it's either 7900GTX was late or G80 was late.....can't be both ;)
Yes, like Macci and Sampsa was for ATi.
TBH having followed Sampsa's work since the beginning, even though he was invited to ATI's OC'ing show etc, he hasn't been tied to ATI the same way Kingpin seems to be to nV, not even close, and Sampsa has used both GF's and Radeons always.
Macci on the other hand, well, the guy works for AMD nowadays, I think that tells enough :razz:
Sound_Card
30-Mar-2007, 17:10
Reading thru AMD's GDC slides, I under the the asumption that "Whiteout" was coded on the Frost Bite engine. Or maybe everybody knew that all along?:razz:
Anon Lamer
30-Mar-2007, 17:23
We know what whatever happens, AMD will have their hand forced by the 17th april release of Nvidia lower end parts (I think Fuad got trolled wrt the delay). Its serious now.
I really hope it's not a GX2 though
Me too. I'm tired of GX2's. Double cores just to take the performance crown is weak and boring. :wink: Hopefully Kingpin has something more exciting in his LN2 testbed. :lol:
More OT in the R600 thread. :shock:
http://www.xtremesystems.org/forums/showpost.php?p=2097574&postcount=58
http://www.xtremesystems.org/forums/showpost.php?p=2099141&postcount=61
24k in 06 is just out there. And it looks like a single 8800GTX not 2 as 06 only reports one.
http://www.xtremesystems.org/forums/showpost.php?p=2099141&postcount=61
24k in 06 is just out there. And it looks like a single 8800GTX not 2 as 06 only reports one.
No it's not, it's SLI.
EVGA 680i SLI Dual 8800gtX's SLI CORSAIR Dominator8888 Kentsfield Quad
http://overclockingpin.com/scores.html
http://www.xtremesystems.org/forums/showpost.php?p=2099141&postcount=61
24k in 06 is just out there. And it looks like a single 8800GTX not 2 as 06 only reports one.
Like I said, "while breaking Overklokk's 24k record" [wink]
And it's SLI
Yeah but once they used up their get-out-of-jail card for the 7900GTX delay then they obviously weren't aiming for May for G80 any more. So it's either 7900GTX was late or G80 was late.....can't be both ;)
Yes, I'd agree with that. Presumably they both "moved right" at about the same time in late 2005. Subsequently G80 got one more push from Sep to November 2006. I'd have to go look for Marv's remarks, but I seem to recall it was early November 2005, so probably shortly after that they rejiggered the roadmap.
Macci on the other hand, well, the guy works for AMD nowadays, I think that tells enough :razz:
No wonder these guys get jobs at NV/AMD. Very good PR, and guys like Macci knows how to torture a GPU. :smile:
Power_man
30-Mar-2007, 22:37
. . .Again some Link (http://www.fudzilla.com/index.php?option=com_content&task=view&id=362&Itemid=1) :lol:.
trinibwoy
30-Mar-2007, 22:41
Yaay, it's not loud! That obviously means it shares nothing in common with NV30 :razz:
Benchmarking R600 completely doesn’t make any sense as the driver is super immature and a lot of applications are crushing all the time.
:shock: (lol)
maybe his spell check has the same disease he has? :grin:
Russell
31-Mar-2007, 00:50
I'm glad to see this thread has livened up again.
We know what whatever happens, AMD will have their hand forced by the 17th april release of Nvidia lower end parts (I think Fuad got trolled wrt the delay). Its serious now.
Sounds like it. No other source I've seen has confirmed what he said. Not even the Inq. Unlike all his other R600 news which is semi-common knowledge/rumors he's just reposting, he's alone in this. When the only person saying it is Fuad, I run the other way.
He's probably right about the cooler's loudness though. nVidia did a good job keeping theirs quiet, I don't see why ATI can't too. Plus their cooler looks very elaborate, so I'd expect it to do a better job cooling at similar acoustics.
Sounds like it. No other source I've seen has confirmed what he said. Not even the Inq. Unlike all his other R600 news which is semi-common knowledge/rumors he's just reposting, he's alone in this. When the only person saying it is Fuad, I run the other way.
Actually it wasn't Fuad, it was Charlie at the Inq:
http://www.theinquirer.net/default.aspx?article=38500
Not that it makes it any more reliable..
Sound_Card
31-Mar-2007, 15:08
Some more fuel on the fire beyond guys.:smile:
I have been reading up on RD790 on pretty much every artical I can get my hands on. Has Hyper transport 3.0, HTX support, on board memory in the north bridge to cut down latency, but most of all, has 2 PCIe 2.0 sockets and has a grand total of 52 PCIe lanes.
Now I want to know why is ATi bringing out PCIe 2.0 already? RD790 is slated for a June/july release which is right after R600. PCIe 2.0 cards can work backwards with PCIe 1.0 slots. I'm really starting to think R600 is a PCIe 2.0 card... which makes it a really great investment in the long run.
http://www.dvhardware.net/article15911.html
INKster
31-Mar-2007, 15:31
Some more fuel on the fire beyond guys.:smile:
I have been reading up on RD790 on pretty much every artical I can get my hands on. Has Hyper transport 3.0, HTX support, on board memory in the north bridge to cut down latency, but most of all, has 2 PCIe 2.0 sockets and has a grand total of 52 PCIe lanes.
Now I want to know why is ATi bringing out PCIe 2.0 already? RD790 is slated for a June/july release which is right after R600. PCIe 2.0 cards can work backwards with PCIe 1.0 slots. I'm really starting to think R600 is a PCIe 2.0 card... which makes it a really great investment in the long run.
http://www.dvhardware.net/article15911.html
As i said before, PCIe 2.0 additional bandwidth is of smaller importance the higher you go in the GPU price ladder.
RussSchultz
31-Mar-2007, 17:37
maybe his spell check has the same disease he has? :grin:
We get that from our Asian office and customers all the time (crush rather than crash).
At first I'd correct people, but eventually it became obvious that's just the english word they use for when software breaks.
Silent_Buddha
31-Mar-2007, 18:07
As i said before, PCIe 2.0 additional bandwidth is of smaller importance the higher you go in the GPU price ladder.
Although if rumors are true that R600 can do bridgeless crossfire, than PCIE 2.0's additional bandwidth may be the only thing that allows it.
Regards,
SB
We get that from our Asian office and customers all the time (crush rather than crash).
At first I'd correct people, but eventually it became obvious that's just the english word they use for when software breaks.
ah interesting!
epicstruggle
31-Mar-2007, 18:36
Although if rumors are true that R600 can do bridgeless crossfire, than PCIE 2.0's additional bandwidth may be the only thing that allows it.
Regards,
SB
I was under the impression that PCIE cards werent even using all the bandwidth they currently had.
Sound_Card
31-Mar-2007, 18:59
As i said before, PCIe 2.0 additional bandwidth is of smaller importance the higher you go in the GPU price ladder.
Smaller importance as compared to what? I'm not talking about performance, I'm talking about straight up building a new rig. Say I buy a new mother board and it is all PCIe 2.0. I don't think my "G80" would work would it? I buy R600, put it on my PCIe 1.0 mobo, then buy a new mobo(PCIe2.0), and swap in R600 with no issues.
Anyways, from what I can gather on RD790, it has 4 PCIe slots. 2 of them will be PCIe 1.0, and the other two will be PCIe 2.0. I soppose that is a great move on AMD's part considering a current PCIe 1.0 video card owner may want to upgrade to the mother board.
I'm aware of the bandwith not being all that important. What matters here though is the life in ones purchase.
INKster
31-Mar-2007, 19:31
Smaller importance as compared to what? I'm not talking about performance, I'm talking about straight up building a new rig. Say I buy a new mother board and it is all PCIe 2.0. I don't think my "G80" would work would it?
Actually, yes it would work just fine.
PCIe 2.0 is retro and forward compatible, both in terms of the motherboard and of the add-in cards used.
I buy R600, put it on my PCIe 1.0 mobo, then buy a new mobo(PCIe2.0), and swap in R600 with no issues.
And you can do exactly the same with any other PCI-Express card available now.
Anyways, from what I can gather on RD790, it has 4 PCIe slots. 2 of them will be PCIe 1.0, and the other two will be PCIe 2.0. I suppose that is a great move on AMD's part considering a current PCIe 1.0 video card owner may want to upgrade to the mother board.
I'm aware of the bandwidth not being all that important. What matters here though is the life in ones purchase.
PCIe will only be of substantial use to Turbocache/Hypermemory schemes used in low end GPU's and/or IGP's, since they depend on main RAM access.
Even with a full 32 lane PCI-Express 2.0 path , a GPU with a mere 8GB/s upstream BW and 8GB/s downstream BW is still way crippled compared to the internal bandwidth of top end GPU's (as i previously mentioned).
A standard G80 GTX has ~80GB/s, while the R600 was previously in the ~150GB/s range, until the memory type and speed got downgraded, apparently.
silent_guy
31-Mar-2007, 19:39
Smaller importance as compared to what? I'm not talking about performance, I'm talking about straight up building a new rig. Say I buy a new mother board and it is all PCIe 2.0. I don't think my "G80" would work would it?
Why do you think that? There is this thing called backward compatiblity: AGP x1 still working in AGP x2,x4,x8 slots. USB 1.1 devices still working with USB 2.0 hosts. PCI1.0 still working in early PCI2.0 slots.
etc. etc.etc.
The developers of PCIe2.0 would be insane not to allow this. And, surprise!, there is this tool called Google that allowed me to come up with this (http://www.pcisig.com/news_room/faqs/pcie2.0_faq/). It's a complicated world...
... AGP x1 still working in AGP x2,x4,x8 slots...
offtopic:
AGP x1 (and AGP x2?) 3.3V signaling cards don't work in AGP x8 slot. ;)
INKster
31-Mar-2007, 19:58
offtopic:
AGP x1 (and AGP x2?) 3.3V signaling cards don't work in AGP x8 slot. ;)
Some do.
There is a version of the slot called "AGP Universal", used in a limited number of motherboard models.
I don't know any single AGP8x MB supporitng 3,3V signaling. There are only some graphics cards, which support 8x transfers + AGP1x/2x 3,3V signaling (e.g. RV280)...
Sound_Card
31-Mar-2007, 20:10
The developers of PCIe2.0 would be insane not to allow this. And, surprise!, there is this tool called Google that allowed me to come up with this (http://www.pcisig.com/news_room/faqs/pcie2.0_faq/). It's a complicated world...
this cleared it up for me as well.
http://en.wikipedia.org/wiki/PCI_Express#PCI_Express_2.0
INKster
31-Mar-2007, 20:15
I don't know any single AGP8x MB supporitng 3,3V signaling. There are only some graphics cards, which support 8x transfers + AGP1x/2x 3,3V signaling (e.g. RV280)...
AOpen AK79G Max (Nforce 2), with a simple googling.
Notice the keyless (http://www.bjorn3d.com//Material/Images/252_AK79GnbHSF.jpg) AGP Universal slot.
I was under the impression that PCIE cards werent even using all the bandwidth they currently had.
Last week Tom's Hardware published PCI-E scaling analysis. It shows that x16 is useful at times.
http://www.tomshardware.com/2007/03/27/pci_express_scaling_analysis/
INKster
31-Mar-2007, 20:36
Last week Tom's Hardware published PCI-E scaling analysis. It shows that x16 is useful at times.
http://www.tomshardware.com/2007/03/27/pci_express_scaling_analysis/
Yes, but in actual games the difference between x8 and x16 is already pretty negligible, and the tendency to see any real world discrepancies will only diminish with further increased theoretical bandwidth.
Certainly there isn't going to be any sort of linear increase in performance due to PCIe 2.0.
It's quite curious how GF8800 is so much more bandwidth sensitive than X1950
INKster
31-Mar-2007, 20:44
It's quite curious how GF8800 is so much more bandwidth sensitive than X1950
Indeed.
That's another interesting aspect to compare with G80, once R600 is out.
AOpen AK79G Max (Nforce 2), with a simple googling.
Notice the keyless (http://www.bjorn3d.com//Material/Images/252_AK79GnbHSF.jpg) AGP Universal slot.
Slot is universal, but not 3,3V compatible.... nF2 doesn't support 3,3V signaling. Read last 4 paragraphs:
http://www.pcstats.com/articleview.cfm?articleID=1365
Ailuros
31-Mar-2007, 21:33
Indeed.
That's another interesting aspect to compare with G80, once R600 is out.
Which will have what practical use exactly? Just don't tell me that anyone sane would run any of those boards with less than PCIe x16.
Just don't tell me that anyone sane would run any of those boards with less than PCIe x16.
:lol:
epicstruggle
31-Mar-2007, 21:50
Here is a longer version of a german interview of Vijay sharma ( now in english )
http://www.pcgameshardware.de/?article_id=573129
PCGH: Hi Vijay. AMD decided not to launch R600 officially at Cebit. Can you give us a clue, what was your reason to do or not do so?
Vijay Sharma [Director of Desktop Discrete Product Marketing at AMD]: Sure. Our initial plan as you're probably aware was to do an NDA press media event this week just before Cebit at Amsterdam. And we realized that we had two derivative products of R600 coming. So we had a whole family of R600-products coming within 45, 60 days of each other. So we said rather than have the press come to Amsterdam before Cebit and then another month later do another event lets talk about the whole family at once and include desktop and mobile products as well. So we have a very strong story and we choose to hold on the R600 and have a combined event later in april.
PCGH: So that means the reason for the delay is not R600 not being ready...
Vijay Sharma: No, no. The R600 schedule hasn't changed. So we're building boards, we're on the same hardware-schedule that we've always been. We're just not launching the product until we launch the family.
Ailuros
31-Mar-2007, 21:52
:lol:
Not on purpose....yeah yeah STFU :P
INKster
31-Mar-2007, 21:53
Which will have what practical use exactly? Just don't tell me that anyone sane would run any of those boards with less than PCIe x16.
Actually, i was thinking PCIe 1.0a/1.1 vs PCIe 2.0.
It's April 1st already it seems :lol:
http://www.techarp.com/showarticle.aspx?artno=392&pgno=0
ShaidarHaran
31-Mar-2007, 23:57
It's April 1st already it seems :lol:
http://www.techarp.com/showarticle.aspx?artno=392&pgno=0
This one was easy enough to figure out.
link (http://www.techarp.com/0104.htm)
the answer is the date ;)
ShaidarHaran
01-Apr-2007, 00:11
It's April 1st already it seems :lol:
http://www.techarp.com/showarticle.aspx?artno=392&pgno=0
and the answer is the date (duh): http://www.techarp.com/0104.htm
ShaidarHaran
01-Apr-2007, 00:22
lol, I see there's a bit of a delay in message posting here, sorry about the eleventy million posts
lol, I see there's a bit of a delay in message posting here, sorry about the eleventy million posts
No delay, just your posts went into the mod queue because they contained links. First time posts with links get flagged, since most of that type are from spam bots. Welcome to the mad house :wink:
ShaidarHaran
01-Apr-2007, 00:29
No delay, just your posts went into the mod queue because they contained links. First time posts with links get flagged, since most of that type are from spam bots.
I suspected as much, and even submitted a 3rd post without the http://www part :p
Welcome to the mad house :wink:
It just got a whole lot madder :twisted:
ShaidarHaran
01-Apr-2007, 00:45
Which will have what practical use exactly? Just don't tell me that anyone sane would run any of those boards with less than PCIe x16.
What about all the dual x8 mobos on the market? Let alone the x16 + x4 ones.
trinibwoy
01-Apr-2007, 01:01
Lol, welcome SH :)
No delay, just your posts went into the mod queue because they contained links. First time posts with links get flagged, since most of that type are from spam bots. Welcome to the mad house :wink:
Arn't ya guys using the latest apache and latest bbs? Thought bots don't work with it? sorry for the OT
lol, I see there's a bit of a delay in message posting here, sorry about the eleventy million posts
Oh noooes! :cry: Hide the women and liquor! ;)
Silent_Buddha
01-Apr-2007, 03:16
I was under the impression that PCIE cards werent even using all the bandwidth they currently had.
If that were true then then bridgeless crossfire/SLI wouldn't be limited to slower cards.
I believe Nvidia allows you to use bridgeless SLI on high end cards, however the performance hit from the lack of bandwidth over the PCIE bus cripples the performance of SLI on those cards.
ATI on the other hand limits bridgeless PCIE to X1950 GT (?) and lower only. However while the X1600 series and lower don't see a performance hit to bridgeless PCIE, I believe the x1950 GT does.
[EDIT] - I hadn't realized single cards would be limited at all from x4 and up. Looks like I'm wrong. and considering that was only the 8800 GTS which is significantly less powerful than a GTX. I'm wondering how the GTX would scale in the same situations. It's possible that may be an indicator of a possible performance difference betweeen PCIE 2.0 and 1.0 in certain situations.
Regards,
SB
Mcmlxxx IV
01-Apr-2007, 03:26
If that were true then then bridgeless crossfire/SLI wouldn't be limited to slower cards.
I believe Nvidia allows you to use bridgeless SLI on high end cards, however the performance hit from the lack of bandwidth over the PCIE bus cripples the performance of SLI on those cards.
ATI on the other hand limits bridgeless PCIE to X1950 GT (?) and lower only.
Regards,
SB
I think ATI lets you use bridgeless on all x1000 cards, even x1950XT and XTX. Apparently the hit is not very high, at least on a dual 16x mobo (intel 975). It went in with catalyst 6.11, I believe.
Silent_Buddha
01-Apr-2007, 03:32
My X1800XT crossfire setup won't allow bridgeless SLI. And I haven't heard of X1900XT(X) or X1950XT(X) being able to do crossfire without the bridge.
Regards,
SB
I'm not sure, but I think the bridgeless CF is available on the cards with the connectors on the PCB, and not available on the ones using dongle?
Mcmlxxx IV
01-Apr-2007, 03:50
I'm not sure, but I think the bridgeless CF is available on the cards with the connectors on the PCB, and not available on the ones using dongle?
To my mind, that would still be bridged. Any physical connection, other than the PCI-E slots of course, would be bridged.
My X1800XT crossfire setup won't allow bridgeless SLI. And I haven't heard of X1900XT(X) or X1950XT(X) being able to do crossfire without the bridge.
Regards,
SB
I have not tried this as I have no desire to go dual cards, but here are a couple threads at Rage3d and one review that talk about it. The review used two Asus x1950xtx cards and the picture they have shows no internal (I don't think there is an internal dongle on the xtx cards anyway) and no external dongle. They could be full of it, though. As I said, I have not tried this. The review was linked by someone in the Catalyst 6.11 discussion thread on Rage. Anyway, here are the links
http://www.rage3d.com/board/showthread.php?t=33878673
http://www.rage3d.com/board/showthread.php?t=33874318
Review:
http://it-review.net/index.php?option=com_content&task=view&id=686&Itemid=91&limit=1&limitstart=0
Upon further looking, it may be just x1900 and x1950s, all models, that can do "software crossfire." It may also be only two "slave" cards work with software.
Mcmlxxx IV
01-Apr-2007, 03:56
Sorry for the double post, but I cannot edit my above post. I misread what you meant Kaotik. I thought you meant using the internal dongle instead of the external. I see what you mean now, but the review link I showed, and what some people have tried, it seems to work on the cards with no internal dongle connector like the x1950xtxs shown in the mini review. One person on Rage even claims to have an x1900xtx and x1950xtx in software crossfire.
Twinkie
01-Apr-2007, 04:30
It's April 1st already it seems :lol:
http://www.techarp.com/showarticle.aspx?artno=392&pgno=0
renderX.. two R600 packed into the same die?
:roll:
Ailuros
01-Apr-2007, 06:00
What about all the dual x8 mobos on the market? Let alone the x16 + x4 ones.
Our dear administrator was laughing because when I got the 8800GTX I was dumb enough to not consult my mobo's manual. The thing has two PCIe slots (20 PCIe lanes in total) of which if you insert a single GPU into the secondary slot it operates only at PCIe x1. Since I only play in high resolutions with AA I had ~30% less performance in 1600+4xAA and ~20% less in 2048+4xAA.
Through that stupidity crisis I had already ordered another mobo and CPU, until I found out what an idiot I am, broke more plastic parts out of the case to fit the monster in, inserted the card in the primary slot and cancelled said order.
I had spend 650 Euros back then for the Gainward Bliss (~$825) and I would have actually shied away for spending another 70-80 bucks for another mobo if it would hold back the GPU?
Classic :lol:
at least on a dual 16x mobo (intel 975)i975 != dual 16*.
Runs 1*16 or 2*8.
Well, my DFI (http://us.dfi.com.tw/Product/xx_product_spec_details_r_us.jsp?PRODUCT_ID=4696&CATEGORY_TYPE=LP&SITE=US) does anyway.
My X1800XT crossfire setup won't allow bridgeless SLI. And I haven't heard of X1900XT(X) or X1950XT(X) being able to do crossfire without the bridge.
Regards,
SB
Bridge/cable-less CF is enabled for X1300 series, X1600 series, X1900 series, X1800GTO/XL (not XT), X1950GT/XT/XTX (not PRO).
http://ati.amd.com/technology/crossfire/charts.html
vertex_shader
01-Apr-2007, 10:21
First Crysis Beta DX10 benchmarks with pictures.
1920x1440x,4x aa 16x af, XQ6700, 2x1gb ram.
X2900XTX 52.7 FPS
8800ULTRA 29.2 FPS
Link (http://img458.imageshack.us/img458/3503/dx10benchsc8.jpg)
To my mind, that would still be bridged. Any physical connection, other than the PCI-E slots of course, would be bridged.
I meant that the cards that have the connector on the PCB can do it with or without the actual bridge between them, while the ones that use dongle can't.
Twinkie
01-Apr-2007, 10:45
First Crysis Beta DX10 benchmarks with pictures.
1920x1440x,4x aa 16x af, XQ6700, 2x1gb ram.
X2900XTX 52.7 FPS
8800ULTRA 29.2 FPS
Link (http://img458.imageshack.us/img458/3503/dx10benchsc8.jpg)
Thanks for wasting 5 seconds of my life.
Rebel44
01-Apr-2007, 10:48
First Crysis Beta DX10 benchmarks with pictures.
1920x1440x,4x aa 16x af, XQ6700, 2x1gb ram.
X2900XTX 52.7 FPS
8800ULTRA 29.2 FPS
Link (http://img458.imageshack.us/img458/3503/dx10benchsc8.jpg)
Nice benchmark :lol: thats 7th benchmark I saw in just few hours - which means R600 was released yesterday - I love 1.April:twisted:
Thanks for wasting 5 seconds of my life.
Honey, you've wasted more than just 5 seconds on these boards and you've done it entirely yourself. :lol:
Twinkie
01-Apr-2007, 23:17
Honey, you've wasted more than just 5 seconds on these boards and you've done it entirely yourself. :lol:
Ok you got me there :lol:
Mcmlxxx IV
01-Apr-2007, 23:21
Bridge/cable-less CF is enabled for X1300 series, X1600 series, X1900 series, X1800GTO/XL (not XT), X1950GT/XT/XTX (not PRO).
http://ati.amd.com/technology/crossfire/charts.html
Doesn't quite match their release notes. Check the release notes for the 6.11s. Here is what it said:
"This release of Catalyst® introduces software CrossFire™ support for any combination of Radeon® X1900 and X1950 installed in a system running either the Windows XP or Windows XP Professional x64 Edition operating system.
Note: Software CrossFire™ is not supported on the Radeon® X1950 Pro product line."
I take that to mean all x1900/1950s except the pro. I'm surprised the x1800xt is not supported while the GTO/XL and faster x1900 line is. Wierd.
I meant that the cards that have the connector on the PCB can do it with or without the actual bridge between them, while the ones that use dongle can't.
Yeah, I got your meaning right after I hit the reply on my first message. People have gotten, or so they claim, cards to software crossfire that were supposed to use the external dongle. One on Rage3D says he has an x1900xtx and an x1950xtx, both slave cards with no internal connector slot, in software crossfire.
Classic :lol:
i975 != dual 16*.
Runs 1*16 or 2*8.
Well, my DFI (http://us.dfi.com.tw/Product/xx_product_spec_details_r_us.jsp?PRODUCT_ID=4696&CATEGORY_TYPE=LP&SITE=US) does anyway.
Yeah, I meant dual 8. I was in a hurry and mistyped.
ChrisRay
02-Apr-2007, 07:23
Has this interview been posted here yet?
http://www.pcgameshardware.de/?article_id=573129
Has this interview been posted here yet?
http://www.pcgameshardware.de/?article_id=573129
Interesting uh?
PCGH: You felt it was worth the hassle compared to the proven process of 90nm?
Vijay Sharma: Oh the R600 wasn't the first 80nm part we've done.
PCGH: But I'd bet definitely by far the most complex...
Vijay Sharma: O certainly it was the most complex. But from a process perspective we have not had 80nm process issues on the R600.
PCGH: Rick told us on the press conference that you had HD-Audio and HDMI integrated into the RV6x. Is that true for all R6x-products from top to bottom?
Vijay Sharma: Yes.
Is it fair to say that Vijay has very little technical background? some of his comments makes you go "Hmm.."
Indeed interesting, some real actual rumors :smile: Or arguably just more fud :???:
In the 5x-architecture we have the ringbus memory controller - there have been some changes in the 600-family. some changes or some changes?
nicolasb
02-Apr-2007, 09:51
Is it fair to say that Vijay has very little technical background? some of his comments makes you go "Hmm.."You need to allow for English/German/English translation errors.
epicstruggle
02-Apr-2007, 10:10
Has this interview been posted here yet?
http://www.pcgameshardware.de/?article_id=573129
Yep: http://forum.beyond3d.com/showpost.php?p=959971&postcount=1337
:razz:
You need to allow for English/German/English translation errors.
It read very much like it wasn't translated at all, and that the conversation took place in English. Pretty conversational in places, a nuance which is hard to maintain during translation. Plus I don't think Vijay speaks German :razz:
CarstenS
02-Apr-2007, 12:27
It read very much like it wasn't translated at all, and that the conversation took place in English. Pretty conversational in places, a nuance which is hard to maintain during translation. Plus I don't think Vijay speaks German :razz:
Jep. This is the original english version of the conversation i had with Vijay. Plus, I do not believe that he lacks technical background - he just had to be very careful what he was telling me, since this was a no-NDA talk. :)
Geeforcer
02-Apr-2007, 16:40
Interesting uh?
Vijay Sharma: Oh the R600 wasn't the first 80nm part we've done.
PCGH: But I'd bet definitely by far the most complex...
Vijay Sharma: O certainly it was the most complex. But from a process perspective we have not had 80nm process issues on the R600
Doesn't Vijay know that the whole R600 has been due to move to 65nm? Also, where is the mention of the yield for the 32-linepipes™ part?
Doesn't Vijay know that the whole R600 has been due to move to 65nm? Also, where is the mention of the yield for the 32-linepipes™ part?
Quasar should've just asked if they had problem with the 65nm production process ;)
Vijay Sharma: Oh the R600 wasn't the first 80nm part we've done.
PCGH: But I'd bet definitely by far the most complex...
Vijay Sharma: O certainly it was the most complex. But from a process perspective we have not had 80nm process issues on the R600.
Seems odd that he specifically said that "from a process perspective we have not had 80nm process issues" instead of just saying "we have not had any issues". Almost like it was a hidden joke. Because you wouldn't have 80nm issues if you were actually on 65nm, now would you?
*sigh* Speculating is fun for only so long. Then you want the curtain drawn to see where you were right, where you were wrong and what you hadn't even considered.
Geeforcer
02-Apr-2007, 20:48
I think one would be REALLY hard-pressed to read that interview and conclude that R600 is based in anything BUT 80nm. To me, it's as black on white as it gets.
The "from the process perspective" qualifier is interesting though. To me it leaves open a possibility that R600 might have had non-process related problems. Like, an overly ambitious or simply buggy design.
I think one would be REALLY hard-pressed to read that interview and conclude that R600 is based in anything BUT 80nm. To me, it's as black on white as it gets.
The "from the process perspective" qualifier is interesting though. To me it leaves open a possibility that R600 might have had non-process related problems. Like, an overly ambitious or simply buggy design.
7-5-14
7-5-14
This (http://www.fx57.net/?p=544)?
This (http://www.fx57.net/?p=544)?
ocworkbench actually..
silent_guy
02-Apr-2007, 23:20
I think one would be REALLY hard-pressed to read that interview and conclude that R600 is based in anything BUT 80nm. To me, it's as black on white as it gets.
The "from the process perspective" qualifier is interesting though. To me it leaves open a possibility that R600 might have had non-process related problems. Like, an overly ambitious or simply buggy design.
I agree completely. It's possible that they had bugs in, say, their 80nm PLL, IO pads or any other part that requires circuit design. That doesn't mean that 80nm by itself is hard, complex, or "sucks" as some enlightened individual suggested. :wink: It just means it's a bug that could have happened in any other process.
My general feeling is that, on this forum and on enthusiast websites in general, way too much importance is given to the process problems and too little to simple human failure. In real life, the ratio between those 2 is something like 50 to 1 in 'favor' of latter (if not higher).
That only time I remember where switching from one process to another was considered to be a real problem was going into 130nm, because suddenly stuff like SI and IR drop were certain to bite you if you chose to ignore them.
Power_man
03-Apr-2007, 00:04
. . .Again some Link (http://www.fudzilla.com/index.php?option=com_content&task=view&id=362&Itemid=1) :lol:.
Wazzzaaaa (http://www.fudzilla.com/index.php?option=com_content&task=view&id=380&Itemid=1) ??? :?: Maybe they fu*k with us . . .:evil:
Russell
03-Apr-2007, 06:27
Wazzzaaaa (http://www.fudzilla.com/index.php?option=com_content&task=view&id=380&Itemid=1) ??? :?: Maybe they fu*k with us . . .:evil:
That does fit in with both expectations and (memory) products available on the market. That doesn't make it true, of course, but at least it's possible that it is.
vertex_shader
03-Apr-2007, 11:02
AMD on the Rebound: A Chat With Henri Richard
HWZ: With the merger, it is understandable that certain products have been realigned, but could we have an update on the R600 delays?
Henri Richard: The R600 will be out in the second quarter. The reason we decided to delay the launch was that we wanted to have a complete DX10-enabled solution top to bottom. A lot of people wrote that the reason it is delayed is because of a problem with the silicon, but there is no problem with the silicon. We are demonstrating it. We can ship it today. But if you think about it, looking at where the market is at, the volumes are going to be in the R610 and R630, so it makes sense for us to do a one time launch of the entire family of DX10 enabled products. That meant delaying the R600 for a few weeks, but frankly it doesn't make a difference in the life cycle of the product and talking with our customers and partners, they felt that it would make a bigger impact with one full launch. So we decided to do that.
Link (http://www.hardwarezone.com/articles/view.php?cid=30&id=2229)
icecold1983
03-Apr-2007, 11:11
that entire explanation makes absolutely no sense, delaying r600 so u can launch it alongside low end cards can do nothing but potentially hurt sales. i dont buy it for a second
Silent_Buddha
03-Apr-2007, 11:48
How in the world could it "hurt" sales by launching an entire family.
The low end and midrange parts are certainly not going to take away from sales of the high end part.
And the low end and midrange parts can only gain sales from trickle down effects of reviews of the high end product. At the worst, they would still only sell as well as they would if they didn't launch with a high end product.
There is the argument that they have lost sales on the High End by delaying it another few weeks, but considering they already missed the Christmas season, a few weeks sales is a bit of a drop in the bucket when you consider the entire lineup.
I wouldn't be shocked to hear that high end products only made up 1-5% of the total PROFIT (not revenue) from the entire graphics card lineup top to bottom. In that case, a loss of a few weeks worth of sales if it has the potential to boost the sales of low end and midrange parts makes sense from a buisness standpoint. And considering that's a few weeks in a part of the year NOT particularly well known for high hardware sales...well, it's not exactly end of the world stuff.
Granted as an enthusiast in the SMALLEST mainstream graphics niche market, it certainly doesn't sit well with me. Well, HTPC might be a smaller market than the enthusiast market, but I wouldn't be so sure about that.
AMD is a business. They are doing what they think will get them the most money. In this case, boosting sales of low end and midrange parts due to the delay of the R600. The card has been delayed months already, if delaying a few weeks will boost sales of the entire line then that might more than make up for the loss of revenue from lost R600 sales over the past few months. It may or may not work out as they planned, but there "is" a reason for it.
Then again without any of us having any insider knowledge, who's to say they aren't blowing wind up our arses. But considering the situation AMD is in. It wouldn't exactly benefit them to outright lie to the public and shareholders, eh?
Regards,
SB
Subtlesnake
03-Apr-2007, 11:52
If we take that quote at face value then it's good news because there should be R600 supply at launch (rather than a month or a few weeks later as sites have suggested).
epicstruggle
03-Apr-2007, 12:30
Link (http://www.hardwarezone.com/articles/view.php?cid=30&id=2229)
Thanks for the link mate. Im just note getting a good vibe from the people AMD is sending out to do press. I also dont think that their stock is going to be doing very well this month.
epic
And it took AMD to make them understand that launching the entire line is faaaaaaaar more profitable?After years of being in the business?Riiiiiiggghhhhhht, I can see the light now, before they simply didn`t give a shit about the low-margin high volume parts that make the bulk of sales because...umm...they simply didn't, duuh. Anyway, since it`s ATi, they be telling the truth, no doubt about it. So on launch day, everyone and their dog will be able to get an R600. Cool.
trinibwoy
03-Apr-2007, 13:01
Anyway, since it`s ATi, they be telling the truth, no doubt about it.
So I see ATI's halo still fits after all this time :lol: So if they really had problems you think they would come and pour their hearts and souls out to us begging for an apology for the delay? I don't see how any logical person could think that a "family launch" could possibly be beneficial to R6xx sales. The market for high-end R600 products just doesn't care whether it's launched alongside its middling siblings or not. Even if they are trying to redefine the brand or reinvent themselves or some other such marketing move I don't buy it. It's not like it's a AMD thing - they launch their CPU's haphazardly all the time.
So on launch day, everyone and their dog will be able to get an R600. Cool.
If anyone should even want one by then, that is. Otherwise, fully agree. They must have thought "nah, all this high-end mindshare just sucks, let's miss a few months of revenues and lose more of our image so we cann sell more low-end chips afterwards". Yeah right.
AMD on the Rebound: A Chat With Henri Richard
HWZ: With the merger, it is understandable that certain products have been realigned, but could we have an update on the R600 delays?
Henri Richard: The R600 will be out in the second quarter. The reason we decided to delay the launch was that we wanted to have a complete DX10-enabled solution top to bottom. A lot of people wrote that the reason it is delayed is because of a problem with the silicon, but there is no problem with the silicon. We are demonstrating it. We can ship it today. But if you think about it, looking at where the market is at, the volumes are going to be in the R610 and R630, so it makes sense for us to do a one time launch of the entire family of DX10 enabled products. That meant delaying the R600 for a few weeks, but frankly it doesn't make a difference in the life cycle of the product and talking with our customers and partners, they felt that it would make a bigger impact with one full launch. So we decided to do that.
Link (http://www.hardwarezone.com/articles/view.php?cid=30&id=2229)
I think you might've bothered with quoting the second paragraph of Henri's answer there, too....;)
Also increasingly in particular with Vista, as we've seen with the competition, it doesn't matter if you're shipping the silicon if the drivers are not stable. There is nothing more frustrating than having bought a new graphics card and having your system crash repeatedly because the drivers are not ready. Although we today, even by Microsoft standards, have the best and most stable drivers in the entire industry, the few weeks will give us even more time to continue improving the drivers. Again, the decision lies in the fact that we will have a top to bottom DX10 offering with drivers that will have a very very high level of stability and the only difference is a few weeks. So it seems to make a lot of sense to do it that way.
I know from previous discussions on the matter that some people think that ATi is in such a hurry to "get something out" to compete with G80 that it's going to rush out an incomplete product line with unpolished drivers, but this quote would seem to dispel such notions. Being realistic about the matter, should the R600 family meet market expectations at its release then the fact that it shipped post G80 will be among the first facts to be forgotten. IE, if R600 meets expectations then who cares if it shipped later than G80? About the only group I can imagine that would care about it is the group which jumped the gun for G80 and couldn't wait...;) Time will tell it all, one way or the other, and by all appearances we don't have long to wait.
Evildeus
03-Apr-2007, 13:39
IE, if R600 meets expectations then who cares if it shipped later than G80? About the only group I can imagine that would care about it is the group which jumped the gun for G80 and couldn't wait...;) Time will tell it all, one way or the other, and by all appearances we don't have long to wait.AMD should care. G80 has a great impact on brand recognition. By the time the mid/low range get out, we do know that it will benefit of the high end situation. And right now, G80 is quite welcomed it seems, whereas R600...
trinibwoy
03-Apr-2007, 14:01
IE, if R600 meets expectations then who cares if it shipped later than G80?
I'm sure AMD cares a lot more than you do Walt :) I'm amused that you're trying to imply that no product is better than a product with relatively unstable drivers. Which one do you think has greater benefit to the manufacturer?
trinibwoy
03-Apr-2007, 14:16
If anyone should even want one by then, that is. Otherwise, fully agree. They must have thought "nah, all this high-end mindshare just sucks, let's miss a few months of revenues and lose more of our image so we cann sell more low-end chips afterwards". Yeah right.
Of course there will be high demand for R600. But what about all those guys who would've stretched a little further to go for the big boy. Now that the market will be flooded with attractive and cheap DX10 parts from both IHV's it's a bit silly to think that high-end R600 sales would not be impacted.
I'm really amazed that some people are trying to argue that having no product is a good strategy. I must have missed that class during my marketing course :lol:
vertex_shader
03-Apr-2007, 15:03
I think you might've bothered with quoting the second paragraph of Henri's answer there, too....;)
I forget to quote the second part :wink:
Of course there will be high demand for R600.
But only if the competition doesn't offer a faster and/or cheaper product in the same ballpark. Which is not the case now, and probably also won't be for long after R600 appears (I ass-ume that nV will bring the successor of the G80 either simultaneously or imediately after the R600 release, be it G81, G90, GX2 or whatever).
In order for R600 to become a success, it HAS to crush the competition speed-wise, price-wise or both. I don't believe for a second that it will do either.
Though it's all just my personal opinion and unfounded predictions, but that's what my common sense tells me.
In order for R600 to become a success, it HAS to crush the competition speed-wise, price-wise or both. I don't believe for a second that it will do either.
Surely either or both would signficantly help. :lol: Don't know if I agree tho that it can't be a reasonable success if it's a competitive part with whatever its going up against at a competitive price. And our front page seems to reflect that the GPGPU wars are continuing to heat up, and at the volumes we're talking about for tippy-top high-end parts, that could be a signficant factor in ultimate commercial success as well.
With success I mean back to the 1st place in sales charts, no less :)
GPGPU is nice and all, but I think that market is quite a bit smaller than the "sure" buyers (the gaming crowd).
What I find amusing is, My top selling card for the last 3 months has been the 1950Pro, and sales have increased as each month goes, with march being the best so far.
I fail to understand where some people get the idea that the 8800 has helped sales when NV simply do not have any cards to compete.
Sound_Card
03-Apr-2007, 15:35
In order for R600 to become a success, it HAS to crush the competition speed-wise, price-wise or both. I don't believe for a second that it will do either.
I believe it will crush speed wise. Price wise maybe not, but value wise??? Maybe so. Then add in feature wise...
What I find amusing is, My top selling card for the last 3 months has been the 1950Pro, and sales have increased as each month goes, with march being the best so far.
I fail to understand where some people get the idea that the 8800 has helped sales when NV simply do not have any cards to compete.
very simple, nV's over all sales for number of cards for the g80 last quarter was ~3.5% which is 1.5% higher then normal, so AMD's not being in the high end is showing up there, this is a nominal figure when looking at the entire market that nV and ATi prior to the take over. But nV is about to release thier mid range before AMD's entire line up shows up as well, possibly a month before hand, where will those 1950 pro sales go after that, not to mention if there is 1 or possibly 2 top end SKU's coming out its going to drive prices of exsisting SKU's lower, I don't believe in the rumors of $1k or $800 cards, its not feesible now unless they are for only a select few crazy enough to buy, unless the performance of these new SKU's truely are a very good deal faster then the r600 (like 2 times the performance), which doesn't seem likely.
I believe it will crush speed wise. Price wise maybe not, but value wise??? Maybe so. Then add in feature wise...
not going to happen, IMO marginally faster in games yes 15%-20% at most in some settings over current cards, features, there are a couple features that seem interesting enough, but nothing that gamers will go googleye over.
trinibwoy
03-Apr-2007, 15:52
What I find amusing is, My top selling card for the last 3 months has been the 1950Pro, and sales have increased as each month goes, with march being the best so far.
I fail to understand where some people get the idea that the 8800 has helped sales when NV simply do not have any cards to compete.
Well besides the fact that sales in the sub $200 range are irrelevant to a discussion on G80 I'd be interested in seeing a more detailed breakdown of your sales numbers if you don't mind :) e.g I'd like to see how the 8800GTS is faring against the X1950XT/X.
Would it be possible to provide some hard numbers on your top sellers? Also, do you sell online or are you a B&M shop?
bigtabs
03-Apr-2007, 16:31
not going to happen, IMO marginally faster in games yes 15%-20% at most in some settings over current cards, features, there are a couple features that seem interesting enough, but nothing that gamers will go googleye over.
I think it's going to be quite hard to come up with many situations at launch where the 8800GTX would be GPU limited enough that a hypothetically 40% faster R600 would actually show it's muscle. Perhaps in 25x15 or whatever those crazy resolutions are, but that doesn't equate to 'many situations' in my book. At least not in the sense that most reviewers will probably care about, when they are benching Quake 4 and Oblivion at 16x12.
icecold1983
03-Apr-2007, 16:42
How in the world could it "hurt" sales by launching an entire family.
The low end and midrange parts are certainly not going to take away from sales of the high end part.
And the low end and midrange parts can only gain sales from trickle down effects of reviews of the high end product. At the worst, they would still only sell as well as they would if they didn't launch with a high end product.
There is the argument that they have lost sales on the High End by delaying it another few weeks, but considering they already missed the Christmas season, a few weeks sales is a bit of a drop in the bucket when you consider the entire lineup.
I wouldn't be shocked to hear that high end products only made up 1-5% of the total PROFIT (not revenue) from the entire graphics card lineup top to bottom. In that case, a loss of a few weeks worth of sales if it has the potential to boost the sales of low end and midrange parts makes sense from a buisness standpoint. And considering that's a few weeks in a part of the year NOT particularly well known for high hardware sales...well, it's not exactly end of the world stuff.
Granted as an enthusiast in the SMALLEST mainstream graphics niche market, it certainly doesn't sit well with me. Well, HTPC might be a smaller market than the enthusiast market, but I wouldn't be so sure about that.
AMD is a business. They are doing what they think will get them the most money. In this case, boosting sales of low end and midrange parts due to the delay of the R600. The card has been delayed months already, if delaying a few weeks will boost sales of the entire line then that might more than make up for the loss of revenue from lost R600 sales over the past few months. It may or may not work out as they planned, but there "is" a reason for it.
Then again without any of us having any insider knowledge, who's to say they aren't blowing wind up our arses. But considering the situation AMD is in. It wouldn't exactly benefit them to outright lie to the public and shareholders, eh?
Regards,
SB
how in the world could delaying the launch of a high end part possibly help sales of any part
seahawk
03-Apr-2007, 16:47
how in the world could delaying the launch of a high end part possibly help sales of any part
Simple answer. Most reviews will headline the top-end-part, so if you slightly win there, but clearly loose in the mid-range, the later fact will get much less publicity. I said that long ago.
I think it's going to be quite hard to come up with many situations at launch where the 8800GTX would be GPU limited enough that a hypothetically 40% faster R600 would actually show it's muscle. Perhaps in 25x15 or whatever those crazy resolutions are, but that doesn't equate to 'many situations' in my book. At least not in the sense that most reviewers will probably care about, when they are benching Quake 4 and Oblivion at 16x12.
even in games that will be GPU limited it depends on what is limiting the GPU.
Trini, I`m sure you`ve noted the humorous note of my post ;).
If this bloody industry has certainties, one of them is that high-end dominance is crucial, or at least having a competitive, comparable part at the high-end. The ammount of exposure one gets from that if extremely important. The 5200 sucked donkey balls, but it still sold to a considerable extent, because nV had a high-end thing on the market. Did it suck compared to its competitor?Well, yes, but it was there, it got mentioned in games reviews, CPU reviews, all kinds of reviews.
You don`t want to be yesterday`s news at the high-end, unless you want to throw away heaps of freeish publicity and mindshare. So irrespective of the ammount of gibberish being spouted across forums about how cool it is to leave your direct competitor dominate the market for 6 months(at least), because you want to have super polished drivers and a full product line to boot, I don`t believe that it was am unforced decision to delay launching the r600. IMHO.
icecold1983
03-Apr-2007, 17:01
Simple answer. Most reviews will headline the top-end-part, so if you slightly win there, but clearly loose in the mid-range, the later fact will get much less publicity. I said that long ago.
this is the case regardless.
Sound_Card
03-Apr-2007, 17:02
not going to happen, IMO marginally faster in games yes 15%-20% at most in some settings over current cards, features, there are a couple features that seem interesting enough, but nothing that gamers will go googleye over.
We will have to wait and see. 15% to 20% faster than G80 on today's current dx9 games is a warm welcome to me actully considering that with G80(and soon to be R600), we are very much CPU bound on popular resolutions such as 1280x1020, 1400x900, 1020x760, etc. As time goes by, I really look forward to the performance delta between G80 and R600 in future DX10 titles because I get a feeling that 20% gap may strecth just a tad or two.
For features, I really look forward to what ATi has to bring to the table. I hope for a possible DX10.1 ready logo on a box, and it would bee cooler if it was PCIe 2.0. But I won't be sad if that does not happen. I'm already impressed with what has been leaked(HDMI). Can't wait till NDA lifts and see what other goodies thay have to offer. SSAA, and 32xAF are two IQ features I hope to see in R600.
trinibwoy
03-Apr-2007, 17:04
Trini, I`m sure you`ve noted the humorous note of my post ;).
Lol, yeah I did. But it was a good spring board for my rant :D
this is the case regardless.
Exactly, I don't see how launching the high-end part alongside the mid-range is more beneficial than launching it before.....:?:
icecold1983
03-Apr-2007, 17:15
We will have to wait and see. 15% to 20% faster than G80 on today's current dx9 games is a warm welcome to me actully considering that with G80(and soon to be R600), we are very much CPU bound on popular resolutions such as 1280x1020, 1400x900, 1020x760, etc. As time goes by, I really look forward to the performance delta between G80 and R600 in future DX10 titles because I get a feeling that 20% gap may strecth just a tad or two.
For features, I really look forward to what ATi has to bring to the table. I hope for a possible DX10.1 ready logo on a box, and it would bee cooler if it was PCIe 2.0. But I won't be sad if that does not happen. I'm already impressed with what has been leaked(HDMI). Can't wait till NDA lifts and see what other goodies thay have to offer. SSAA, and 32xAF are two IQ features I hope to see in R600.
do people rly buy a high end card to play at such low resolutions?? no wonder people think g80 is cpu limited. for those of us with bigger lcds the r600, if its all its talked up to be, cant come fast enough
Evildeus
03-Apr-2007, 17:19
this is the case regardless.I would say more. It's better to have much publicity over time meaning review of high end parts, then mid then low, then mobile parts etc than everything in the same period of time. Moreover the publicity by users implify the impact of the reviews of mid/low/mobile parts
icecold1983
03-Apr-2007, 17:21
anyone who's going to be aware of the positive publicity is surly going to realize your low end card sucks
Sound_Card
03-Apr-2007, 17:27
do people rly buy a high end card to play at such low resolutions?? no wonder people think g80 is cpu limited. for those of us with bigger lcds the r600, if its all its talked up to be, cant come fast enough
Back when I bought my x1800gto2(moded now to x1800xtx), I gamed on 1280x1020(still do), my friend thought I was crazy and that a 7600 or x1600 should suffice my needs. Boy, is he eating his words...
I know what your saying, but their are plenty of people that buy high end for a resolution like 1280x1020. And now a day's, most gamers have LCD's. Your locked on your native resolution unless you want like what LCD's look like outside that resolution. I just don't feel like going out and buying a 1980x1080 monitor and having to refresh my graphics card every 6 months or less just to use it. I would be a broke ass.:razz: Buying R600 for 1400x900 or 12080x1020 gives me life while still looking good to the eye. At least I don't have to upgrade my card every so often and maybe upgrade with in a year in a half. By then R600 whould be in the bargin bin for 200.00 flat and I may just crossfire for more life. I don't wan't to buy a el cheapo graphics card just because I have a lowly resolution of 1280x1020, because then I would be cought refreshing my graphics card every 6 months or less to enjoy that resolution. Buy R600, crank up the AA/AF, and game on and rock out on 1280x1020.:razz:
We will have to wait and see. 15% to 20% faster than G80 on today's current dx9 games is a warm welcome to me actully considering that with G80(and soon to be R600), we are very much CPU bound on popular resolutions such as 1280x1020, 1400x900, 1020x760, etc. As time goes by, I really look forward to the performance delta between G80 and R600 in future DX10 titles because I get a feeling that 20% gap may strecth just a tad or two.
15-20% in a GPU limited situation depending on what is limiting the GPU sorry for being ambiguious and keeping your hopes high ;), when do you look at reses like that when buying a high end card? yeah that 20% will stretch a tad or two precentage points thats about it.
The performance of the low and mid parts will be what it is.
Regardless of their performance, it seems the best thing for their sales might be to have them available at the peak mind-share of the high-end part. The question is, when will that be? After dominating the high end performance charts for a couple months and getting a "reputation," or right after a flood of reviews and publicity. Despite that most of us here are sticklers for value analysis among the competing products, there exists in the greater market a strong brand (and family) association for many people. As has been mentioned above, winning the high-end battle is very important for mid/low sales; and timing of those sales wrt the high-end battle can be optimized business-wise. IMO, of course. I'm not in the business, but this just makes sense.
ERK
icecold1983
03-Apr-2007, 17:44
Back when I bought my x1800gto2(moded now to x1800xtx), I gamed on 1280x1020(still do), my friend thought I was crazy and that a 7600 or x1600 should suffice my needs. Boy, is he eating his words...
I know what your saying, but their are plenty of people that buy high end for a resolution like 1280x1020. And now a day's, most gamers have LCD's. Your locked on your native resolution unless you want like what LCD's look like outside that resolution. I just don't feel like going out and buying a 1980x1080 monitor and having to refresh my graphics card every 6 months or less just to use it. I would be a broke ass.:razz: Buying R600 for 1400x900 or 12080x1020 gives me life while still looking good to the eye. At least I don't have to upgrade my card every so often and maybe upgrade with in a year in a half. By then R600 whould be in the bargin bin for 200.00 flat and I may just crossfire for more life. I don't wan't to buy a el cheapo graphics card just because I have a lowly resolution of 1280x1020, because then I would be cought refreshing my graphics card every 6 months or less to enjoy that resolution. Buy R600, crank up the AA/AF, and game on and rock out on 1280x1020.:razz:
im already thinking my 24 inch isnt big enough, couldnt imagine playing on something smaller.
SugarCoat
03-Apr-2007, 17:55
do people rly buy a high end card to play at such low resolutions?? no wonder people think g80 is cpu limited. for those of us with bigger lcds the r600, if its all its talked up to be, cant come fast enough
thats a silly comment. Look what happens after a year let alone 2. The X800XTs and 6800Us of old CHOKE on games these days and can barely push medium settings in many games at 1280x1024. If you replace your high end card literally as soon as the next best thing then you could reasonably question it. However those that keep their cards 12-24 months before replacement are doing it smart, i include myself in this. If i had some ultra high resolution to feed i'd be handicapped much faster and be forced to purchase an upgrade to remain at the same settings as i would at 1280x1024.
Before my x1800xt cooked it was just starting to show its age too in a few of the latest titles and almost certainly would have when the many games using the UE3 engine come out. Not to mention games with things like HDR which can potentially hammer the GPU. Simply put buying a high end card for a 'low resolution', as you put it, of 1280x1024, or even 1600x1200 which isnt seen as an overly taxing resolution these days, is not a dumb move in any way. I play at 1280x1024, i have an 8800GTX, i'll enjoy my games at maxed settings for awhile to come and thats the way i like it. I simply dont have the budget to feed some ultra high resolution monitor and keep the same settings that i like for a decent amount of time, nor do i see any value in the limited amount of time you can squeeze out of a $600 card before its obsolete under such situations.
caffeinated
03-Apr-2007, 18:28
I tend to keep my hardware for quite a while. I broke down and upgraded my 6800 GT when it stopped displaying the green part of the spectrum :grin: . That, more than performance issues, was what finally pushed me over the edge. I would imagine that I'll be using my 8800 GTS for a long time.
icecold1983
03-Apr-2007, 18:40
well u can usually just sell your high end card every 6 to 8 months for not much of a loss, also assumed people buying these $500 cards had money to spare
bigtabs
03-Apr-2007, 18:42
I tend to keep my hardware for quite a while. I broke down and upgraded my 6800 GT when it stopped displaying the green part of the spectrum :grin: . That, more than performance issues, was what finally pushed me over the edge. I would imagine that I'll be using my 8800 GTS for a long time.
I agree. I bought an X1900XT for playing at 12x10 with high settings over a good amount of time. The next card I buy will be high end, and for the same reasons... unless high end becomes something relatively a lot higher than what it has been. That kind of thing is hard to know though isn't it. Who knows what must have titles are going to come and kick your card's ass?
I'm glad of my choice. There are some games that you can see struggle slightly at 12x10 now with this card. I pay for the privilege of not having to scrimp on settings for a good while. A higher resolution or lower spec'ed card would just increase the frequency of my purchases.
icecold1983
03-Apr-2007, 18:45
wow i wasnt trying to attack anyone just to clarify.
Bouncing Zabaglione Bros.
03-Apr-2007, 18:50
Remember that this is no longer just about R600 - this is also about AMD launching itself as a graphics and chipset vendor, ie a total platform supplier. With R600 so late, the midrange cards almost here, it no surprise that AMD, (like most of today's marketing led companies) has decided to go for a "big bang" bells and whistles launch where they can sell themselves, their new graphics business, their new products, new chipsets, maybe even new processors in there too.
Even if you believe that R600 is done and piling up in warehouses as of a month ago, someone high up in AMD has decided that the benefit of selling them a few weeks earlier is not preferable to doing this big razamataz launch for everything they want to be seen as, both now and in the future.
caffeinated
03-Apr-2007, 18:53
well u can usually just sell your high end card every 6 to 8 months for not much of a loss, also assumed people buying these $500 cards had money to spare
That's just it, though. Who pays $300 or $400 for a used video card? I want a warranty (even if I do plan on voiding it immediately :lol:). Even if I could sell it at $200, then I would have to turn right around and spend another $250-$300? That's a bit wasteful to me. I can dispose of my income just fine without upgrading my video card every 6 months, though granted there are some people that would go this route.
Robin B
03-Apr-2007, 19:57
Dont know if this is posted, dont kill me if it is ........:lol:
Here you'll find the unabridged and uncensored interview transscript of PCGH with Vijay Sharma at Cebit 2007. It was a freely held conversation - no scripted questions and answers. So please bear with us the casual manner of speech and grammar. The Interview was conducted by Carsten Spille on behalf of PC Games Hardware, all rights reserved. (2007)PCGH: Hi Vijay. AMD decided not to launch R600 officially at Cebit. Can you give us a clue, what was your reason to do or not do so?
http://www.pcgameshardware.de/?article_id=573129
trinibwoy
03-Apr-2007, 20:05
this is also about AMD launching itself as a graphics and chipset vendor, ie a total platform supplier.
Yeah but the burning question is - are they doing it on purpose? :lol:
Sound_Card
03-Apr-2007, 20:07
15-20% in a GPU limited situation depending on what is limiting the GPU sorry for being ambiguious and keeping your hopes high ;), when do you look at reses like that when buying a high end card? yeah that 20% will stretch a tad or two precentage points thats about it.
I'll crank up the AA/AF along with any in game or control panal goodies and make it as close to gpu bound I can get. As mentioned earlier, plenty of people buy high end cards for resolutions like that. That performance delta could stretch alot more than that when talking about Crysis, UT3, and bioshock with max detail and AA/AF.
icecold1983
03-Apr-2007, 20:11
has anyone figured out what kombatant was talking about in your sig soundcard? because only the staunchest ati fans are happy about the constant r600 delays, and i dont think that is many people
Bouncing Zabaglione Bros.
03-Apr-2007, 20:16
Yeah but the burning question is - are they doing it on purpose? :lol:
They are now!
I really think it's just the case that with previous delays, the AMD buyout of ATI, etc everything has just arrived within these couple of months, so they elected to do it all together instead of one-by-one. They decided to build up stock, do a harder launch, get the marketing done right.
AMD have already missed competing with G80, Nvidia's midrange isn't here yet, so that's a non-issue, AMD have missed the Christmas period and Vista launch, so a few more weeks makes little difference to them if it means they can supply decent availabilty and do a big marketing song and dance that benefits the whole DX10 range and the new AMD as a whole.
After all this though, the products had better be stellar.
Geeforcer
03-Apr-2007, 20:25
I don't know, 2 months (Mid-March to Mid-May) seems like quite a bit of time.
Sound_Card
03-Apr-2007, 20:44
has anyone figured out what kombatant was talking about in your sig soundcard? because only the staunchest ati fans are happy about the constant r600 delays, and i dont think that is many people
I keep reading hints on a full product line up(x2200, x2400, x2800, Radeon mobil, Firestream, Fire GL) and platform launch(RD790 and below, K10).
I don't think ATi has ever mentioned a hard release date, but only one time(cebit). So technically they have only delayed once. So ATi fans are not up set that they keep delaying... their upset because they are late to the party sort of speak.:smile:
icecold1983
03-Apr-2007, 20:58
I keep reading hints on a fully product line up(x2200, x2400, x2800, Radeon mobil, Firestream, Fire GL) and platform launch(RD790 and below, K10).
I don't think ATi has ever mentioned a hard release date, but only one time(cebit). So technically they have only delayed once. So ATi fans are not up set that they keep delaying... their upset because they are late to the party sort of speak.:smile:
i fail to see how delaying r600 to launch it alongside a bunch of crappy low end cards and a mb that wont be the best in its class will please many people
caffeinated
03-Apr-2007, 21:19
I still find it extremely difficult to believe - despite the vociferous denials - that the delay isn't due to a problem. If they had the hardware ready right now to ship, wouldn't they?
Maybe it is a driver issue? It wouldn't surprise me if they delayed it because of a driver issue. AMD in the SS7 days and the VIA chipset days had some pretty bad drivers (though that isn't AMD's fault) and ATI had in the past some driver issues. Maybe they are more sensitive to wanting good drivers because of the flack they took in the past?
I wonder if we will ever really know...
Something must have caused the delay. The only other thing that it could be is that the mainstream and value products are really great, while the R600 is not quite as great as they would like for it to be, so they are launching all of them at once to try to ride on the good reviews of those products. Certainly if R600 is a disappointment, they don't want hardware reviews pointing out lackluster performance for a month or two before the great midrange products ship. If R600 gets so-so reviews, but tremendous bang for the buck with the mainstream parts, they could potentially mitigate the damage by spreading reviews around..? Just a theory..is it a plausible one?
There is more to conducting business profitably than rushing out products as soon as possible.
I agree with BZB.
Dont know if this is posted, dont kill me if it is ........:lol:
No, you're the first (http://forum.beyond3d.com/showpost.php?p=960663&postcount=1372). :razz:
Russell
03-Apr-2007, 21:46
I still find it extremely difficult to believe - despite the vociferous denials - that the delay isn't due to a problem. If they had the hardware ready right now to ship, wouldn't they?
No doubt they have some hardware to ship, but likely not enough for the kind of widespread availability they want. I'd say the most likely theory is that their yields simply haven't been great and they need the extra time to ensure that they have more parts for launch.
Sound_Card
03-Apr-2007, 21:46
i fail to see how delaying r600 to launch it alongside a bunch of crappy low end cards and a mb that wont be the best in its class will please many people
sheesh hard to please?
What do you mean a mother board that wont be best in class? :roll: How do you know that?
The delay for R600 could be for several tactical reasons, some we may never know. It could be a mixture of things. Full product and platform line up along with lot's of availibilty from top to bottom. Performance drivers from day one, etc. I mean we won't know untill NDA lifts. It's not like they decided to delay out of the blue, they must of saw a oppertunity.
nutball
03-Apr-2007, 22:04
Performance drivers from day one, etc.
According to boasts at other places their drivers have been tickety-boo and rock solid for months now. Drivers can't be a problem.
It's not like they decided to delay out of the blue, they must of saw a oppertunity.Sorry, what sort of revisionist history is this? Out of the blue for whom? Out of the blue for people who had tickets booked to Amsterdam, or out of the blue for ...?
caffeinated
03-Apr-2007, 22:07
sheesh hard to please?
What do you mean a mother board that wont be best in class? :roll: How do you know that?
The delay for R600 could be for several tactical reasons, some we may never know. It could be a mixture of things. Full product and platform line up along with lot's of availibilty from top to bottom. Performance drivers from day one, etc. I mean we won't know untill NDA lifts. It's not like they decided to delay out of the blue, they must of saw a oppertunity.
Is anyone under NDA yet?
No doubt they have some hardware to ship, but likely not enough for the kind of widespread availability they want. I'd say the most likely theory is that their yields simply haven't been great and they need the extra time to ensure that they have more parts for launch.
Well, if they didn't have *enough* hardware, yes that would be bad, but it would be more than what is out now...:). Wouldn't it be better to release it and have reviews declaring it the latest and greatest until they can get more out? I'm just trying to consider what tactical reasons they would have to put off selling the product, and to be perfectly honest, it just doesn't seem plausible if they have it ready to be sold.
And if no one is under NDA yet, either (I don't know whether they are or not, that's why I asked), that just makes it seem all the more likely that there is some reason why they can't ship it. I have to say that my guess is starting to become that it isn't as high performance of a part as they would like.
I just can't see any company passing up the opportunity to get a lot of good excitement going about their product, especially since it has to be clear to AMD that many people are tired of waiting. I would imagine the AIB makers are twice as tired of the wait. That's why I came up with the "maybe their mainstream parts are great, and R600 is so-so" theory.
trinibwoy
03-Apr-2007, 22:16
It's not like they decided to delay out of the blue, they must of saw a opportunity.
Or maybe the delay just wasn't up to them? ;)
I'll crank up the AA/AF along with any in game or control panal goodies and make it as close to gpu bound I can get. As mentioned earlier, plenty of people buy high end cards for resolutions like that. That performance delta could stretch alot more than that when talking about Crysis, UT3, and bioshock with max detail and AA/AF.
Keep dreaming, Crysis not going to happen, UT3 highly unlikely, bioshock, don't have info on that game.
Or maybe the delay just wasn't up to them? ;)
heh, and God bistowed on this world three great fruits of processors, the g80, core duo, and the missing r600 ;)
Razor, you know something we dont? ;)
Sound_Card
03-Apr-2007, 22:30
Keep dreaming, Crysis not going to happen, UT3 highly unlikely, bioshock, don't have info on that game.
not going to happen what? Highly un likely what?
icecold1983
03-Apr-2007, 22:35
sheesh hard to please?
What do you mean a mother board that wont be best in class? :roll: How do you know that?
The delay for R600 could be for several tactical reasons, some we may never know. It could be a mixture of things. Full product and platform line up along with lot's of availibilty from top to bottom. Performance drivers from day one, etc. I mean we won't know untill NDA lifts. It's not like they decided to delay out of the blue, they must of saw a oppertunity.
when ati releases a class leading motherboard ill eat my words. waiting to launch an entire lineup at once and claiming your doing it because its the best thing to do makes no sense. the only people who follow these events for the most part dont rly care about anything except the r600. its delayed due to technical or performance issues. dont see how anyone can think otherwise.
Razor, you know something we dont? ;)
kinda, there should be a few people here that already know, actually talked with my friend last week, working with a certain engine mentioned above, the difference isn't that pronoucned, also talked with Crytek about this a couple of weeks back same thing came from them, they say performance will still increase because they haven't worked much with the card at least not as much as they did with the g80, but then again the drivers for the g80 still aren't up to spec either.
Icecold, I think I'm spotting a trend in your posts. You want it all, right now, don't you? Or do you like the opportunity to bitch about it even more? ;)
icecold1983
04-Apr-2007, 01:00
of course i want r600 right now......i dont think im the only one
Well, anyone who says they weren't wanting it last December is either fibbing or an Nvidiot. :lol:
Russell
04-Apr-2007, 02:47
Well, if they didn't have *enough* hardware, yes that would be bad, but it would be more than what is out now...:). Wouldn't it be better to release it and have reviews declaring it the latest and greatest until they can get more out?
While we're at it, why don't they send out 100 review samples to all of the important web sites and paper launch the thing. That way they can declare how it's the latest and greatest while we wait for availability.
We'd all like to know what R600 will look like (performance wise, we've seen the pics), but if they're going to release with inadequate supply then they may as well go back to paper launching their products.
icecold1983
04-Apr-2007, 02:52
a launch with low supply is better than just not launching.
SirPauly
04-Apr-2007, 03:18
Is it?
If they launched with a very low supply to actually buy --- then its not worth launching. Consumers would be screaming left-and-right about not finding any and so would the press. It would taint the reviews and AMD over-all as a company.
If released with no volume; all it would do is feed the very few that want to get their hands on the hardware but doesn't help AMD over-all. Where are the real positives?
This product needs to help bring a positive halo effect on DirectX 10 and Vista; to help fuel them all --everthing that is AMD. If marketed correctly -- AMD could make some serious noise with Vista and DirectX 10 - using these two important bullet points to fuel the entire family at once.
Delay is a not a good thing -- but rushing a good product too early doesn't help either. If the product is as good as many think it may be -- what's a few weeks to make sure there is better drivers, more volume and springboard their entire family. If there ever was a launch that is important, well it is this one; you have a new operating system; a new DirectX; a new generation; a new direction for AMD to show leadership. Let's hope it's more substance, data and delivering the actual goods instead of paper hype, PR marketing and excuses.
Look, I want to know what the product is as well but what can you do? We will find out if ATI/AMD made the right decision.
Sound_Card
04-Apr-2007, 03:43
when ati releases a class leading motherboard ill eat my words.
? I honestly don't understand this. Have you ever owned a ATi chipset? Let alone read what they have out right now? Their chipsets are top notch.
Silent_Buddha
04-Apr-2007, 03:45
It's a business it's all about making the most money with the opportunities that you have.
R600 already missed the big selling season, Christmas.
R600 was and still is going to launch during a time of the year the year when computer sales are slow.
Thus, by delaying the launch for a few weeks so that an entire line can be released at the same time they are hoping to...
1. Mitigate lost revenue from R600 missing the holiday buying season by hopefully pushing up sales of mid-range and low-end parts which are far more profitable anyways.
2. They may have seen an opportunity to also launch other products that aren't graphics related.
3. Flesh out GPGPU programming on the R600 family. If this works out, they stand to make millions more than selling R600 as a graphics card. The OIL industry just by itself I'm sure would pay millions for the computing power this would bring them. After all they are already spending millions on hardware/software that is significantly slower than what this brings to the table.
4. Showcase an entire lineup and possible platform for big OEM customers like Dell, HP, etc.
Compared to the lost revenue from missing the holiday season, missing out on a few weeks at this time of the year is basically nothing. They already missed their big opportunity, now it's all about how to make the most of their current situation.
ATI in the past would have launched the R600 as soon as it was possible to launch because that's basically all they have.
AMD on the other hand is looking at what they can do with ATI to bring in the most money. And I'm sure if they can piggy back other launches on R600 they will do it.
Does it mean I like it? Not really.
Is it going ot help R600 be successful? Probably not. But what it will probably do is make AMD more money than if they had launched R600 by itself only a few weeks earlier than they could have launched the other stuff.
At the point they missed the holiday season, everything had to focus on how best to make back lost revenue from a lost opportunity.
Myself, I'm still waiting to see which one releases something for me faster. Either R600 is launched with good Vista 64 drivers or Nvidia finally get's off it's arse and releases good Vista 64 drivers. Whichever comes first gets my dollars.
Regards,
SB
caffeinated
04-Apr-2007, 04:16
SB you make some good points, but then it also sounds as though you are suggesting that they don't care about the performance market because they missed their big window of opportunity for it. Are you thinking that they are going to go after the large business market and stop worrying about the consumer market so much?
Or did I misinterpret what you were saying? I realize that there is a profit in making stream processors, but I would think there is a profit in also making OEM products, too...otherwise NVidia and possibly Intel wouldn't be doing it, right?
To be honest, that doesn't really sound like a healthy business choice. In particular, since, as we were reminded very recently in this very thread, they were all set to go in Amsterdam and poof, pulled the plug on that idea. That truly does not seem like a choice, it seems like a move that they were forced to make.
So is anyone under NDA yet? Is that what you were hinting Razor1?
? I honestly don't understand this. Have you ever owned a ATi chipset? Let alone read what they have out right now? Their chipsets are top notch.
A couple years ago their chipset was in every box sold at a brick and mortar shop and I have to say they left a lot to be desired. I guess they were $400-$600 desktops, but they were cheap chipsets. I don't remember ATi having as many enthusiast level chipsets as other companies.
I just want ATi/AMD to release some information so my computer fund doesn't get sucked into the dining room fund by my wife. The longer they hold out, the less likely I'll upgrade the 'ole computer. In addition, I fear if I don't upgrade now, I'll just go to consoles. Therefore, hurry up and give us something! <rant over>
:roll:
as I said 15-20% I'm not expecting any more then that at GPU limiting settings over the 8800 GTX, fillrates are going to play an important roll in benchmarking these new cards more then raw bandwidth. And the demo that is to be released is an UE3 based demo, should be out very soon. Its a physics demonstration on the UE3 engine. One of the designers of this demo used to work for me, so I do trust his judgement. And in many cases from what he has heard, it edges out the 8800 GTX, I'm not sure if that is real or not but the person he got the information from is an ex ATi employee now working in the game industry but this could be due to drivers which has more creedance.
So is anyone under NDA yet? Is that what you were hinting Razor1?
I'm not sure if press is under NDA as of yet, I know some developers are. I did get more info, I hinted at some new features, which are pretty cool, but can't really talk about them, I myself don't have full details on these features just some things that were muttered to me :), and probably won't be getting it soon.
caffeinated
04-Apr-2007, 04:30
I'm not sure if press is under NDA as of yet, I know some developers are. I did get more info, I hinted at some new features, which are pretty cool, but can't really talk about them, I myself don't have full details just some things that were muttered to me :), and probably won't be getting it soon.
Meaning that you won't be getting an R600 anytime soon, or that the press won't be getting an R600 anytime soon, or were you saying you won't get details any time soon? I'm not trying to pressure you, and certainly do not want you to say more than you are comfortable with saying - I just didn't understand the last part :D. As always, you have good input.
kemosabe
04-Apr-2007, 04:31
Longshot guess, but maybe R600's bride-to-be RD790 is closer (http://www.pcisig.com/developers/compliance_program/integrators_list/pcie/) to release than expected, and they didn't want the groom walking down the aisle alone? :smile:
/ducks Geo's flying frying pans.
icecold1983
04-Apr-2007, 04:31
i dont recall ati ever releasing an mb as good as nvidias.
INKster
04-Apr-2007, 04:34
i dont recall ati ever releasing an mb as good as nvidias.
ATI and Nvidia sell motherboards now ?
Meaning that you won't be getting an R600 anytime soon, or that the press won't be getting an R600 anytime soon, or were you saying you won't get details any time soon? I'm not trying to pressure you, and certainly do not want you to say more than you are comfortable with saying - I just didn't understand the last part :D. As always, you have good input.
nitty gritty details about the features, but my old sig, I do believe I was correct, as for ALU structure I'm not going to get that any time soon either, finer details are very hard to come by :wink:
icecold1983
04-Apr-2007, 04:38
ATI and Nvidia sell motherboards now ?
they design them.
caffeinated
04-Apr-2007, 04:44
nitty gritty details about the features, but my old sig, I do believe I was correct, as for ALU structure I'm not going to get that any time soon either, finer details are very hard to come by :wink:
Now I understand what you were saying. Fine details do seem to be in short supply...or rather, fine details that weren't pulled from the nether regions of rumor-mongers :lol:
icecold1983
04-Apr-2007, 04:47
razor what was your old sig?
INKster
04-Apr-2007, 04:49
they design them.
They just design optional layouts of motherboards, they don't sell or manufacture them.
razor what was your old sig?
750-850 mhz 64 vec 4 units (I'm lead to believe they are vector units, I'm not sure about the number of them or type of vector units, but 320 madd units, doesn't leave too much to the imagination)
The bus was already pretty much confirmed.
edit:
Added to this I do believe the pixel granularity to be less then the g80, which would give insight to why DICE mentioned the geometry shader perforamace to be more then that of the g80, but this is my speculation.
Added to this I do believe the pixel granularity to be less then the g80, which would give insight to why DICE mentioned the geometry shader perforamace to be more then that of the g80, but this is my speculation.
What's the connection between pixel granularity and GS performance?
What's the connection between pixel granularity and GS performance?
That, like the previous generation, ATI will do nothing about the ROP count and keep focussed on mathematic performance inside their VPU?
That, like the previous generation, ATI will do nothing about the ROP count and keep focussed on mathematic performance inside their VPU?
I'm not sure I follow you..
I'm not sure I follow you..
you're not a sheep, you're not supposed to!
nicolasb
04-Apr-2007, 10:54
when ati releases a class leading motherboard ill eat my words.Well, of course, the rumours about the RD600 chipset originally made it out to be an amazing piece of kit. Among other things it was supposed to support x16/x16 Crossfire, with all 32 of those lanes coming off the north bridge (unlike, say, the Nvidia 680i where half of them come off the south bridge which is much less cool).
According to the conspiracy theorists, the reason why RD600 was so late was that, after the AMD/ATI merger, AMD insisted that RD600 be redesigned and deliberately crippled, because they wouldn't stand for an AMD-badged chipset that promoted the sale of Intel CPUs by offering optimum performance on an Intel platform.
See, for example: http://www.theinquirer.net/default.aspx?article=37712
ATI's (not yet DAAMIT then) answer at Computex was clear: 2 x PCI-E 16 lanes in those two graphics slots, plus one PCI-E x 4 slot (x 16 connector) for the physics accelerator
[...]
Then DAAMIT came to be, and so did many other things, including uncertainty over the fate of this chipset, coupled with substantial success and performance-crown-snaffling by the Nforce 680i.
[...]
So when, after all the delays and other shenanigans, DAAMIT's RD600 finally appeared, we were hoping to finally see the long-promised, pole-position competitor to the 680i. But what a disappointment!
The number of PCI-E lanes, at 2 x 8 plus 1 x 4, remains the same as on the two-year old Intel D975XBX 'Bad Axe' mainboard! This definitely won't compete with the Nforce, and frankly it won't stand a chance against Intel's upcoming chipsets a few months down the line.
So why such a long release delay, since the RD600 chipset demos seemed to work well at Computex and the guys in charge were saying it was about launch-ready? Maybe, after the merger, the new boss re-evaluated the positioning and asked for a revision to, say, halve the number of PCI-E lanes and do a "down differential" as compared to the offerings for its own chips - and that would need some time.
RussSchultz
04-Apr-2007, 11:07
I
1. Mitigate lost revenue from R600 missing the holiday buying season by hopefully pushing up sales of mid-range and low-end parts which are far more profitable anyways.
Depends on what you mean by 'profitable'.
The high end stuff always makes more margin. (I.e. more money back per dollar put in)
The lower end stuff makes more volume.
And I'm really not buying any reasons trying to say "delaying will make them more money". It just doesn't make sense.
icecold1983
04-Apr-2007, 11:31
Well, of course, the rumours about the RD600 chipset originally made it out to be an amazing piece of kit. Among other things it was supposed to support x16/x16 Crossfire, with all 32 of those lanes coming off the north bridge (unlike, say, the Nvidia 680i where half of them come off the south bridge which is much less cool).
According to the conspiracy theorists, the reason why RD600 was so late was that, after the AMD/ATI merger, AMD insisted that RD600 be redesigned and deliberately crippled, because they wouldn't stand for an AMD-badged chipset that promoted the sale of Intel CPUs by offering optimum performance on an Intel platform.
See, for example: http://www.theinquirer.net/default.aspx?article=37712
it still comes down to ati chipsets sucking compared to the competition.
trinibwoy
04-Apr-2007, 11:59
And I'm really not buying any reasons trying to say "delaying will make them more money". It just doesn't make sense.
No it doesn't and that's why noone has been able to put forward a viable way in which they could accomplish this. You don't forego revenue today in order to make it up tomorrow. That's not how it works - time value of money and all that. So not only does R600 have to spur post launch sales sufficiently to make up x months of lost revenue it also has to make up the interest lost on cash receipts in that period!! And it's worse if R600 is impressive relative to G80 because that means sales would have been high should it have been on the market.
It's also highly likely that the delay incurred additional expenses, further reducing income from R600. But there are those who continue to believe ..... :)
What's the connection between pixel granularity and GS performance?
sorry put more things together, the batching of the ALU's if the granularity is smaller, less ALU's per cluster, so when doing things like GS, it might be possible to have more free units to do other things. The comments that DICE made sounded like there might be something that is holding back the g80 when doing geometry shaders compared to the r600, I'm just throwing out ideas.
Entropy
04-Apr-2007, 12:31
Depends on what you mean by 'profitable'.
The high end stuff always makes more margin. (I.e. more money back per dollar put in)
The lower end stuff makes more volume.
This only works out if the volume of the high-end part is sufficient to amortize development cost. Engineering design time, testing + redesign/re-tape out costs, PCB design and testing costs et cetera, are specific to the R600, and I'm not at all convinced that the part will see sufficient volumes at sufficient margins to recoup those given the size of the chip and the cost of surrounding silicon. It would still make sense to release it given the hit already taken, and even at a net total loss it could still be worthwhile for AMD in terms of brand recognition. But as a consumer I'm not thrilled with the money poured into the high-end parts and poster child technologies such as Crossfire and stream computing. These technologies definitely don't seem to be paying for themselves, which only means that the associated costs are carried by the rest of the product line, making it less competitive. And that is not even taking into account the human resource aspect of tying up talented people in endeavors that ultimately will affect very few, rather than engaging them in making the high-volume projects more successful.
This only works out if the volume of the high-end part is sufficient to amortize development cost. Engineering design time, testing + redesign/re-tape out costs, PCB design and testing costs et cetera, are specific to the R600, and I'm not at all convinced that the part will see sufficient volumes at sufficient margins to recoup those given the size of the chip and the cost of surrounding silicon.
It would be really, really stupid if the basic development wasn't common/shared for the whole lineup of the R6xx gen. So I don't think there are any huge extra costs involved there. The initial costs went into the high-end and the mid-/low-end gets the cut-down version if they planned with any sense for reality.
nicolasb
04-Apr-2007, 13:29
it still comes down to ati chipsets sucking compared to the competition.Well, if the conspiracy theories are correct, RD600 would have been entirely non-sucky, but was actively crippled at the request of AMD. That means ATI is capable of producing a chipset that doesn't suck, it merely hasn't yet been allowed to do so.
nicolasb
04-Apr-2007, 13:35
The Inquirer sort of confirms that audio output via HDMI will be found on R600 as well as RV6xx:
http://www.theinquirer.net/default.aspx?article=38714
The ATI sound implementation is not GPGPU code. It is dedicated silicon, probably brought on by the Vista DRM infection and MS twisting arms to force it on people.:shock:
Sound_Card
04-Apr-2007, 13:43
Well, if the conspiracy theories are correct, RD600 would have been entirely non-sucky, but was actively crippled at the request of AMD. That means ATI is capable of producing a chipset that doesn't suck, it merely hasn't yet been allowed to do so.
It's still not a sucky board.
Entropy
04-Apr-2007, 13:52
It would be really, really stupid if the basic development wasn't common/shared for the whole lineup of the R6xx gen. So I don't think there are any huge extra costs involved there. The initial costs went into the high-end and the mid-/low-end gets the cut-down version if they planned with any sense for reality.
Obviously this is the way it works.
I'd say that even so, it is questionable if the R600 will pay for itself.
A hint may be where the rest of the line-up is in terms of price and silicon costs. Look at the additional cost of the R600 die (yield?) and surrounding silicon, factor in development cost and low volume...
My guess is that the R600 is made for marketing purposes more than anything.
Twinkie
04-Apr-2007, 14:40
No rumours on its performance except the not so good ones, but lots of hype/rumours based on GPGPU, sound/audio and video engine etc.
Im starting to get a feeling that AMD or ATi got way too out of focus.. like theyre covering up something with other not so important stuff for what a GPU should be. Performance and quality for 3d apps should be the foremost highest priority.. but im not hearing much about that other that teraflops, sound card, etc etc.
trinibwoy
04-Apr-2007, 14:45
My guess is that the R600 is made for marketing purposes more than anything.
The high-end is a source of revenue just like any other market segment. Yes, of course cheaper, high volume parts bring in the most total revenue. But to imply that the high-end is a negligible source of profit is a bit of a stretch. And I would think they contribute considerably to overall gross margin as well. Think of how important the X19xx series was given the mediocrity of X16xx.
Also, did you consider how market dynamics would change if you aimed low and your competitor aimed high? You would get muscled out of lucrative market share faster than you could say uncle and be relegated to the high volume, low margin segment.
Doomtrooper
04-Apr-2007, 14:46
I wouldn't come to that conclusion, but it is fun reading these threads before launches.
trinibwoy
04-Apr-2007, 14:50
No rumours on its performance except the not so good ones, but lots of hype/rumours based on GPGPU, sound/audio and video engine etc.
Well we did get wind of devs claiming GS performance was more promising on R600. But yeah I agree - a lot of fluff and promises and no substance so far to the rumours. We have ATI claiming that everything is peachy and drivers are top notch - but these are the same people that claimed Crossfire would work perfectly with every title and never require profiles.....
The Inquirer sort of confirms that audio output via HDMI will be found on R600 as well as RV6xx:There is actually a direct quote from an AMD guy (http://www.pcgameshardware.de/?article_id=573129) :idea:
PCGH: Rick told us on the press conference that you had HD-Audio and HDMI integrated into the RV6x. Is that true for all R6x-products from top to bottom?
Vijay Sharma: Yes.
CarstenS
04-Apr-2007, 15:09
Would it be a great surprise, if AMD was going common sense and realized that with GDDR4 and immense clockrates those cards would be to expensive for "normal" buyers and cancelled the high-end models, reworked the top-class to accomodate the more affordable GDDR3 RAM and a silent dual-slot-cooling solution? :)
So they can sell enough cards to the segments that really matter. After all, Athlon FX and Pentium/C2D Extreme Editions are for a very very small market also.
edit:
Reason for this: Look at Kombatants citation floating around in one of the signatures here: A pleasant surprise to many people (with pleasant and many in bold). Additionally, Vijay mentioned in our interview not something about a performance hammer or G80-Killer but rather stressed the point, that R6x-Familiy will deliver a very good price-performance-ratio - which is usually not found in the high-end.
SirPauly
04-Apr-2007, 15:58
Would it be a great surprise, if AMD was going common sense and realized that with GDDR4 and immense clockrates those cards would be to expensive for "normal" buyers and cancelled the high-end models, reworked the top-class to accomodate the more affordable GDDR3 RAM and a silent dual-slot-cooling solution? :)
So they can sell enough cards to the segments that really matter. After all, Athlon FX and Pentium/C2D Extreme Editions are for a very very small market also.
edit:
Reason for this: Look at Kombatants citation floating around in one of the signatures here: A pleasant surprise to many people (with pleasant and many in bold). Additionally, Vijay mentioned in our interview not something about a performance hammer or G80-Killer but rather stressed the point, that R6x-Familiy will deliver a very good price-performance-ratio - which is usually not found in the high-end.
Also did get out of that nice interview and other quotes that the R-600 family may offer some nice flexible price points and really target performance/value offerings. Only time-will-tell for most of us to truly know.
However, didn't get out of it no flag-ship product at all. Let's say this theory is right and there is no flag-ship product offerings at all.........you leave a sector and price-point alone for your competition. It may give the wrong message to potential customers and gamers over-all. The product can't help build the name-brand. The product can't help build mind-share. The product can't show more technology leadership for important bullet points like DirectX 10 and Vista. The product can't help sell more product in other sectors that matter more to consumers. You leave potential revenues and margins on the table over-all for others to take.
I don't see why ATI/AMD can't offer great value and performance by creating a flexible R-600 family to target more potential customers and offer a Flag-ship as well.
Don't see why great value and performance translates into no Flag-ship at all.
trinibwoy
04-Apr-2007, 16:10
Don't see why great value and performance translates into no Flag-ship at all.
I don't think the concern is about having a flagship - the concern is that ATi is stressing everything but dominance at the high end which is what most of us care about the most. Midrange price/performance is great but that's not what we've been waiting for :)
caffeinated
04-Apr-2007, 16:12
No rumours on its performance except the not so good ones, but lots of hype/rumours based on GPGPU, sound/audio and video engine etc.
Im starting to get a feeling that AMD or ATi got way too out of focus.. like theyre covering up something with other not so important stuff for what a GPU should be. Performance and quality for 3d apps should be the foremost highest priority.. but im not hearing much about that other that teraflops, sound card, etc etc.
That is also my feeling. Another way of looking at it is not that they got too out of focus, but that they are having to focus on other things besides what is typically important for a GPU.
Teraflop performance is interesting, and certainly useful for some applications, as is an HD sound solution. Those are nice to haves; but generally, most people care about what features as far as being a graphics card it brings to the table. Unfortunately, everything that has been rumored so far about it's GPU performance has not exactly been favorable.
As others have pointed out, there just doesn't seem to be a good reason for the delay besides problems. First, we have the rumors of massive power consumption. That wouldn't be enough to put off some buyers, but it would definitely put off a lot of people if it required an exorbitant amount of power (relative to the competition) without delivering an equally exorbitant amount of performance.
If the performance is there, I would think that rumors would abound by now. Is it possible that they are delaying to sell off the older parts, in order to make as much profit off of those as possible? Every single reason that I have heard for the delay being "a good business decision" just doesn't seem to hold water.
Unless of course, there is something wrong. Then delaying is the only business decision that could be made.
I'm willing to entertain any other theories, though :D. One thing that we haven't heard yet, is from "the press" on this particular forum ;) Is anyone here under NDA yet, or have any idea when that will happen? Or is that something that those under NDA are unable to talk about - i.e. are you unable to say you are under NDA if you are?
(Side Note: Wow, that was a tough question to phrase...before too long I'm going to be comparing great products and what to call other great products that are as much greater as other great products...! :D)
trinibwoy
04-Apr-2007, 16:14
Since we're in another one of those transitionary phases I thought this article would be a cool read for those that have never seen it. It basically tells the 3D story from NV1 up to R300 and outlines the pitfalls and triumphs that the companies have encountered along the way. Who knows, maybe ATI's acquisition, the emergence of DX10 and the R600 launch might have a significant impact on the future of 3D hardware and the fate of the companies involved.
http://www.aceshardware.com/read.jsp?id=60000292
karlotta
04-Apr-2007, 16:35
it still comes down to ati chipsets sucking compared to the competition. suck at what? There is no sucking ATI chipsets. There are performance wins and losses, but none suck?
chavvdarrr
04-Apr-2007, 16:41
I don't think the concern is about having a flagship - the concern is that ATi is stressing everything but dominance at the high end which is what most of us care about the most. Midrange price/performance is great but that's not what we've been waiting for :)
huh. I don't care if high-end card has dual-triple-quadro slot cooler. I won't buy it, so WE (people like me who are not proffesional gamers) care about price/performance.
trinibwoy
04-Apr-2007, 16:52
I don't follow chav. What we've been waiting on is a new architecture, new features and better performance. Why exactly are you waiting on R6xx in particular for price/performance? The currently available solutions don't do it for you?
Don't get me wrong - I'm not saying that the mid-range market isn't important. But are you really waiting for better price/performance in general or are you waiting to see what R6xx based cards are capable of?
jimmyjames123
04-Apr-2007, 17:17
I think it is clear why we have heard so little about R6xx performance in games vs G8x. In most games on the market today, R600 really doesn't have a significant performance advantage, if any, over G80. AMD/ATI believes that the R6xx cards will have more of an advantage in DirectX10 games. That is why their employees keep talking about having the best DirectX10 cards. However, it's pretty risky to bank on the fact that DirectX10 performance will be superior.
CarstenS
04-Apr-2007, 17:17
I don't think the concern is about having a flagship - the concern is that ATi is stressing everything but dominance at the high end which is what most of us care about the most. Midrange price/performance is great but that's not what we've been waiting for :)
The man catches my drift... One other thing to consider would be the omnious answer (and geos - i believe - signature of the nes 6800 GT wearing red) Vijay gave regarding the gap between G80-GTX and -GTS.
vBulletin® v3.8.4, Copyright ©2000-2010, Jelsoft Enterprises Ltd.