AMD: R7xx Speculation

[Gelanin]

Oslo must have been the first point the tracking picked up my last order from! (It was some time since I bought anything from them).

And you may want to add Ireland, Austria, Belgium and France to that list.

I couldnt really be arsed to check where they've expanded to these days

But yeah, with those you basicly have most of northern/western europe covered

 

[Dave Baumann]

It tells you if you go to .com as opposed to any of the local domains!

 

[Silent_Buddha]

1) we weren't discussing the 3850 - everyone knows by now that AMD allocated too many RV670s for HD 3850 production and not enough for HD 3870s.
2) no shit 3870s are finally in stock, they're $30-$50 over MSRP, NO ONE'S BUYING THE DAMN THINGS BECAUSE THE 8800 GT IS A BETTER VALUE!

A bit of exaggeration there? The 3870's are available and in stock starting from 249 USD at Newegg. That's 20 USD over the original price. However, it isn't price gouging from Newegg, as that is the currently recommended list price by ATI due to strong demand outstripping available supply.

The price adjustment was done to cash in on the demand and at the same time to reduce demand in order for supply to catch up. IE - Good business practice when you are in need of cash flow for your company.

So in reality the cards are priced 0 USD over recommended retail MSRP at the current time. Or to put it another way, there's no price gouging going on. If the price is too high for you, that's fine, as even with that price supply is only recently starting to catch up with demand.

Again looking at Newegg the cheapest 8800 GT 512 comes in at 279 USD. So it's still 30 USD over the 3870. While certainly not as compelling as it would be if it was 50 USD more. It's evidently still enough to sway some people to go for it.

Especially if a rumor posted elsewhere is true that ATI has climbed from a 10% DX10 marketshare to a 30% DX10 marketshare. Which would indicate ATI is selling more 38xx series GPUs than Nvidia is selling G92 based GPUs. Again, if the rumors are true.

Personally, at those prices none of them are compelling buys for me right now. I'll be holding out to see what comes along in the springtime. But for someone with an older gen video card, you really can't go wrong with either.

Regards,
SB

 

[Pressure]

Well it sure as hell isn't about tiny little European nations whose GDP is less than that of the state I live in, let alone my country My state (Minnesota) has a GDP of $234 billion, Denmark is only $198 billion. The population of each is similar.

It's actually $257.3 billion with regards to Denmark. GDP per capita may even show a more interesting picture.

But really, there is no point being rude in the process of discussing something as simple-minded as the availability of a graphic card.

 

[nexus_alpha]

It's actually $257.3 billion with regards to Denmark. GDP per capita may even show a more interesting picture.

But really, there is no point being rude in the process of discussing something as simple-minded as the availability of a graphic card.

????????????????????????????????????????????????

This the R7xxx specylation thread

 

[compres]

That american poster is really representative of the negative image the USA has internationally these days. Americans should think a little more before making offensive posts like that.

 

[Acert93]

That american poster is really representative of the negative image the USA has internationally these days. Americans should think a little more before making offensive posts like that.

Can we take the ethno-political jabs to PM? The rest of us Americans minding our manners don't appreciate proper discussion devolving into stereotyped insults that contribute nothing to the topic. Thanks.

 

[Sound_Card]

Im going to go out in the wild and say what if R700 is not a MCM? Perhaps a monolithic 55nm die.

 

[WaltC]

Waiting for WaltC and the boys to come and educate you on the fact that the enthusiast market is irrelevant, it's only the high-volume parts that matter. That would be fun.

Heh... Actually, it seems to me that what's being said here--and of course not by me--is that the high-volume market is irrelevant. Intel seems to have been doing very well there for quite awhile, but anyway...

Someone please explain "enthusiast" to me because when I started playing 3d games back in the '90's with a 3dfx V1 (after playing 2d games for years), the moniker "enthusiast" meant "someone who really enjoys 3d gaming to the extent that he buys dedicated 3d hardware" for the privilege of doing so.

Today, here's what people are saying is *not* "enthusiast":

1) You're not an enthusiast if you play your 3d games of choice in resolutions less than 1600x1200--and a real "enthusiast" turns up his nose at gaming resolutions of 1900x1200 or less.

2) Along this line of thought, you cannot be an "enthusiast" unless your monitor is at least a 30-incher, or else you are running two or three monitors at the same time.

3) Today's enthusiast isn't worthy of the moniker, I'm convinced, unless he spends 90% of his computer time running benchmarks which spit out nothing but numbers which he is constantly trying to increase in some way, shape, or form. Whether it's MIPs or FPS or MHz, today's "enthusiast" delights in endlessly repeating canned benchmark runs and in spending meaningless hours bragging about the silly numbers his benchmarks produce on various web site forums dedicated to that kind of mindset The remaining 10% of his computer time is spent actually playing the 3d games he claims to be enthusiastic about (I'm guessing it's as high as 10%.)

4) Whether you run SLI, Crossfire, or SLIx3 or CrossfireX x4, you are still not a "true enthusiast" because all true enthusiasts know that these products don't cost nearly enough and aren't anywhere near loud enough or hot enough to be ranked as "enthusiast" products. The "enthusiast" pines in his heart for the sweet melody of the leaf-blower fan and anything less just won't do!

5) The concept of two gpus on a single pcb, originally pioneered in a production product by 3dfx back in the 20th century, does not today qualify for the "enthusiast" label because it's just old hat and the real enthusiast knows that only single gpus on single pcbs are loud, hot, and expensive enough to earn the quality "enthusiast" label (which very often is decided by nothing more substantial than a couple of runs of 3dMark06 which produce numbers high enough to convince the "enthusiast" of his inherent wisdom in spending 4x what he actually had to spend to have just as much fun playing his 3d games as he would have had if he'd spent 4x less.)

6) You can't be an "enthusiast" today unless you subscribe wholesale to the mantra that the only kind of "image quality" worth having in an "enthusiast" product is pixel resolution irrespective of FSAA, AF, and anything else a 3d IHV might wish to build into his product that only improves image quality but otherwise does nothing except *get in the way* of those canned benchmark scores that have come to define the very essence of "the enthusiast." Unless you understand that "there is no difference" in image quality among all 3d products made and sold today--that they differ only in terms of the canned benchmark scores they produce--sorry--you just don't have what it takes to be "an enthusiast."

Let's see--judging by some of the remarks I've read in this thread, it seems to me that I've hit the nail on the head when it comes to the general consensus of who is and who isn't an "enthusiast." Obviously, merely being enthusiastic about playing 3d games no longer qualifies one as an "enthusiast"...

However, getting very pumped up about jerk-wad, brain-dead benchmarks that spit out essentially meaningless numbers, over and over again, in "review" after "review" after "review," seems to define what it takes to be "an enthusiast" today, doesn't it? Yep, for today's enthusiast, intelligent, probing examinations of the image quality produced in 3d games by "enthusiast-grade" 3d products simply isn't done because today's enthusiast doesn't care about piddling crap like that. 'Long time ago we used to care a lot, but now it seems like nobody does at all.

So, now that we've established clearly what it means not to be an enthusiast, what *is* an enthusiast today? I see two possible answers to this question:

1) An idiot
2) A sucker

If someone can think of another possible answer, I'd enjoy hearing it...

I'm hoping that they'll actually have some kind of stab at the enthusiast market with the 3870X2. Because the fact that the GTX and the Ultra have been fairly constant in terms of pricing ever since their release(not truly significant downwards shifts, I think the GTX cheapened somewhat, but that's all) is an indication of stagnation and what the lack of competition brings.

I'm hoping that in 2008 a few pioneering Internet tech journalists will return to the sanity that once defined the "enthusiast." I'm really hoping to see a resurgence in the interest of comparative image quality among competing products as a primary focus in 3d card reviews. That's just ever-so-much-more interesting to me than wading through page after page of repetitive frame-rate bar charts.

In closing, if anyone still isn't clear on what I mean, just consider how many times you see it said in so-called "3d-card reviews" which spend time "comparing" one brand of 3d card with another, these words: "We could see no difference in the image quality produced by either product." That's generally about the average word count given to the topic of image quality in today's "enthusiast-level" 3d-card review. But, after being presented with page after page after page of bar charts denoting often less than a 10% difference in frame-rate performance, have you *ever read* these words about that: "Even though the benchmarks show product X running sometimes higher than product Z in terms of frame rate, our experience in actually playing these games on these products was that there was no subjective difference in performance."....????

I can think of maybe seeing something like that written once or twice in the last 30 or so "reviews" I've read, but most of the time the "enthusiast" reviewer doesn't point out the essential lameness of those pages and pages of bar charts it's taken him several hours to compile in an effort to appear to know what he's talking about...

Let me end here by wishing everyone the best of new years, and by assuring all reading that I mean to pick on no one in particular. It's just frustrating for me as someone who does--despite the common current definition of the word--consider himself to be "an enthusiast" about 3d gaming. It's been a long time since I've been surprised, satisfied, or intrigued by what I've read in a 3d-card review. I think that's pretty sad, and I'm hoping that in '08 this trend will begin to reverse itself. Always the optimist...

 

[Sound_Card]

So, now that we've established clearly what it means not to be an enthusiast, what *is* an enthusiast today? I see two possible answers to this question:

1) An idiot
2) A sucker
3) A person with lots of money
4) All of the above

If someone can think of another possible answer, I'd enjoy hearing it...

Fixed.

 

[INKster]

What about developers ?
Being able to test future game engines on powerful but cheaper commodity hardware surely must be an advantage, and even Nvidia mentioned "seeding" many of them with 3-way SLI rigs in one of their slides very recently...
AMD probably does the same.

The alternative is the slow software emulation method, surely not a productive way to work in modern times, with the proper hardware just peaking outside.

 

[ECH]

Heh... Actually, it seems to me that what's being said here--and of course not by me--is that the high-volume market is irrelevant. Intel seems to have been doing very well there for quite awhile, but anyway...

Someone please explain "enthusiast" to me because when I started playing 3d games back in the '90's with a 3dfx V1 (after playing 2d games for years), the moniker "enthusiast" meant "someone who really enjoys 3d gaming to the extent that he buys dedicated 3d hardware" for the privilege of doing so.

Today, here's what people are saying is *not* "enthusiast":

1) You're not an enthusiast if you play your 3d games of choice in resolutions less than 1600x1200--and a real "enthusiast" turns up his nose at gaming resolutions of 1900x1200 or less.

2) Along this line of thought, you cannot be an "enthusiast" unless your monitor is at least a 30-incher, or else you are running two or three monitors at the same time.

3) Today's enthusiast isn't worthy of the moniker, I'm convinced, unless he spends 90% of his computer time running benchmarks which spit out nothing but numbers which he is constantly trying to increase in some way, shape, or form. Whether it's MIPs or FPS or MHz, today's "enthusiast" delights in endlessly repeating canned benchmark runs and in spending meaningless hours bragging about the silly numbers his benchmarks produce on various web site forums dedicated to that kind of mindset The remaining 10% of his computer time is spent actually playing the 3d games he claims to be enthusiastic about (I'm guessing it's as high as 10%.)

4) Whether you run SLI, Crossfire, or SLIx3 or CrossfireX x4, you are still not a "true enthusiast" because all true enthusiasts know that these products don't cost nearly enough and aren't anywhere near loud enough or hot enough to be ranked as "enthusiast" products. The "enthusiast" pines in his heart for the sweet melody of the leaf-blower fan and anything less just won't do!

5) The concept of two gpus on a single pcb, originally pioneered in a production product by 3dfx back in the 20th century, does not today qualify for the "enthusiast" label because it's just old hat and the real enthusiast knows that only single gpus on single pcbs are loud, hot, and expensive enough to earn the quality "enthusiast" label (which very often is decided by nothing more substantial than a couple of runs of 3dMark06 which produce numbers high enough to convince the "enthusiast" of his inherent wisdom in spending 4x what he actually had to spend to have just as much fun playing his 3d games as he would have had if he'd spent 4x less.)

6) You can't be an "enthusiast" today unless you subscribe wholesale to the mantra that the only kind of "image quality" worth having in an "enthusiast" product is pixel resolution irrespective of FSAA, AF, and anything else a 3d IHV might wish to build into his product that only improves image quality but otherwise does nothing except *get in the way* of those canned benchmark scores that have come to define the very essence of "the enthusiast." Unless you understand that "there is no difference" in image quality among all 3d products made and sold today--that they differ only in terms of the canned benchmark scores they produce--sorry--you just don't have what it takes to be "an enthusiast."

Let's see--judging by some of the remarks I've read in this thread, it seems to me that I've hit the nail on the head when it comes to the general consensus of who is and who isn't an "enthusiast." Obviously, merely being enthusiastic about playing 3d games no longer qualifies one as an "enthusiast"...

However, getting very pumped up about jerk-wad, brain-dead benchmarks that spit out essentially meaningless numbers, over and over again, in "review" after "review" after "review," seems to define what it takes to be "an enthusiast" today, doesn't it? Yep, for today's enthusiast, intelligent, probing examinations of the image quality produced in 3d games by "enthusiast-grade" 3d products simply isn't done because today's enthusiast doesn't care about piddling crap like that. 'Long time ago we used to care a lot, but now it seems like nobody does at all.

So, now that we've established clearly what it means not to be an enthusiast, what *is* an enthusiast today? I see two possible answers to this question:

1) An idiot
2) A sucker

If someone can think of another possible answer, I'd enjoy hearing it...




I'm hoping that in 2008 a few pioneering Internet tech journalists will return to the sanity that once defined the "enthusiast." I'm really hoping to see a resurgence in the interest of comparative image quality among competing products as a primary focus in 3d card reviews. That's just ever-so-much-more interesting to me than wading through page after page of repetitive frame-rate bar charts.

In closing, if anyone still isn't clear on what I mean, just consider how many times you see it said in so-called "3d-card reviews" which spend time "comparing" one brand of 3d card with another, these words: "We could see no difference in the image quality produced by either product." That's generally about the average word count given to the topic of image quality in today's "enthusiast-level" 3d-card review. But, after being presented with page after page after page of bar charts denoting often less than a 10% difference in frame-rate performance, have you *ever read* these words about that: "Even though the benchmarks show product X running sometimes higher than product Z in terms of frame rate, our experience in actually playing these games on these products was that there was no subjective difference in performance."....????

I can think of maybe seeing something like that written once or twice in the last 30 or so "reviews" I've read, but most of the time the "enthusiast" reviewer doesn't point out the essential lameness of those pages and pages of bar charts it's taken him several hours to compile in an effort to appear to know what he's talking about...

Let me end here by wishing everyone the best of new years, and by assuring all reading that I mean to pick on no one in particular. It's just frustrating for me as someone who does--despite the common current definition of the word--consider himself to be "an enthusiast" about 3d gaming. It's been a long time since I've been surprised, satisfied, or intrigued by what I've read in a 3d-card review. I think that's pretty sad, and I'm hoping that in '08 this trend will begin to reverse itself. Always the optimist...

Something that should have been in video card reviews from the very start! What good is it to say that card A, runs 4, 5, 6, 7, 8, 9, etc FPS better with game A1 using max settings at certain resolution(s) then Card B, when game A1 needs 55 FPS in order to play (for example) the game as intended and both cards fall short? Calling card A the better card isn't being subjective at all when the reviewer doesn't provide any data regarding the frame rate needed to maintain the responsiveness and smooth immersion the developer intended for that game (be it 55 FPS or otherwise)! There should be data regarding each game benchmark to show the frame rate required to play the game as intended. It can be through subjective opinion or from the developer(s). Even if both Card A and Card B obtains the frame rate needed to play said game, regardless if one is faster still doesn't change the immersion of the game. It doesn't get any faster then whats required to play the game. So where does that leave you? It leaves you with:
-cost of each video card
-IQ of each video card
-features/technical prowess offered with each video card
-availability
-etc

 

[Arun]

I thought the saying was 'If you think you understand quantum mechanics, you don't understand quantum mechanics'. I didn't know it applied to statistics too - oh wait, it doesn't.

Hint: Performance with certain settings in a certain game doesn't tend to be completely unrelated to every other benchmark out there, especially but not exclusively in the same game. It's not because 10% doesn't make any difference in one specific case that it never makes any difference; furthermore, subjective improvements are gradual, not instant. It's not like anyone wouldn't notice anything at 16% and would at 17%...

So either you test EVERY SINGLE setting for EVERY SINGLE game that has ever existed and will ever exist (where's your time machine now?) and rate your subjective experience on EVERY SINGLE graphics card you need to test... Or you just do things the current way and try to extrapolate the differences to other possible settings or workloads.

And no, I don't think it's fair to say saubjective differences in IQ are 'negligible' either; ideally you should look into what causes them and if there are cases where they may indeed be more problematic thn in the one you're currently looking at. A good example of that is transparency antialiasing: it's not because two algorithms are decent in one case that they will be in all, and the final IQ difference can be huge in some games/scenarios.

 

[no-X]

Holland and Denmark do not exactly fulfill the statement "throughout Europe". Inkster states they are not available where he lives.

…

2) no shit 3870s are finally in stock, they're $30-$50 over MSRP, NO ONE'S BUYING THE DAMN THINGS BECAUSE THE 8800 GT IS A BETTER VALUE!

For Czech Republic and Slovakia: GT's are not available, too. They arrive from time to time in a very limited quantity, but they are overpriced (~$375-$390, expected in the second half of January). HD3870 is not in stock, but you can get them in a week (~$285-$299). HD3850 is usually available ($225 in stock, $199-210 in 14 days)

 

[AlexV]

1) An idiot
2) A sucker


I'm hoping that in 2008 a few pioneering Internet tech journalists will return to the sanity that once defined the "enthusiast." I'm really hoping to see a resurgence in the interest of comparative image quality among competing products as a primary focus in 3d card reviews. That's just ever-so-much-more interesting to me than wading through page after page of repetitive frame-rate bar charts.

My god, it's full of words...."2008:A verbose Oddysey". I'd really love people to get off their budget, pseudo-communist horse that seemingly allows them to rag on enthusiasts, be it in any market. Yes, a Veyron does the same thing a Mitsubishi Colt does, so why get the Veyron?Because some can, and that's their option, and by God it's POS reasoning to rag on them for it. Yes, a butt-ugly woman performs approximately the same functions as Halle Berry(or whoever you fancy), so why get Halle for a ride, when she'd be significantly more expensive and the inherent differences wouldn't be earth-shattering.

It's funny that the decision that the enthusiast segment is irrelevant has come up just as ATi failed to deliver anything targetting that segment. Quite the coincidence, eh?Just like in the CPU business it's suddenly OK to have piss-ass clocked parts that are outperformed across the board, but are a bit cheaper so inherently they're better value. And they provide a...solid(???) experience. What you ppl don't really process is that both the enthusiast AND the high-volume market matter, and are intrinsically linked. I've had this discussion before here and elsewhere and I know that the argument is mostly wasted, but I'll ask you something I asked some dude on R3D who had a similar line of reasoning:do you consider AMD/ATi to be completely retarded?Were they completely retarded when they WERE targetting enthusiasts with FX CPUs and X1900XTXs or whatever high-end expensive board?Is this a business revolution, finally figuring out that enthusiasts don't mean jack?I would ask you if you consider Intel and nV retarded as well, but you probably do as they're the spawn of Satan or somesuch.

Ragging on reviewing practices simply because it doesn't really paint the picture you'd like is also not really useful. What would reviewing be if all we'd get is pages upon pages of subjective experiences and nothing even vaguely comparable to something posted on another site/by another reviewer?Hell, you have fansite forums for that, to read pages after pages of dribble about something that's hot actually isn't hot, something that doesn't perform really performs, something that looks like crap actually looks good, and about how each and every driver revision magically improves IQ and so on. A review has to use a measuring stick that's directly comparable to what other sites use and what John Doe can do at home/will see at home. The bar-charts you so much hate. There has to be a standardized(more or less) system of measurement, and having some wanker harp about whatever he perceived isn't close to being useful, simply because with us humans, opinions and interpretations are abundent, and individual, and quite different. You know that old Sumerian sayingpinions are like assholes-everybody has one. Is the "bar-chart" approach perfect?No, but ATM it's probably the best compromise that can be achieved.

 

[nexus_alpha]

My god, it's full of words...."2008:A verbose Oddysey". I'd really love people to get off their budget, pseudo-communist horse that seemingly allows them to rag on enthusiasts, be it in any market. Yes, a Veyron does the same thing a Mitsubishi Colt does, so why get the Veyron?Because some can, and that's their option, and by God it's POS reasoning to rag on them for it. Yes, a butt-ugly woman performs approximately the same functions as Halle Berry(or whoever you fancy), so why get Halle for a ride, when she'd be significantly more expensive and the inherent differences wouldn't be earth-shattering.

It's funny that the decision that the enthusiast segment is irrelevant has come up just as ATi failed to deliver anything targetting that segment. Quite the coincidence, eh?Just like in the CPU business it's suddenly OK to have piss-ass clocked parts that are outperformed across the board, but are a bit cheaper so inherently they're better value. And they provide a...solid(???) experience. What you ppl don't really process is that both the enthusiast AND the high-volume market matter, and are intrinsically linked. I've had this discussion before here and elsewhere and I know that the argument is mostly wasted, but I'll ask you something I asked some dude on R3D who had a similar line of reasoning:do you consider AMD/ATi to be completely retarded?Were they completely retarded when they WERE targetting enthusiasts with FX CPUs and X1900XTXs or whatever high-end expensive board?Is this a business revolution, finally figuring out that enthusiasts don't mean jack?I would ask you if you consider Intel and nV retarded as well, but you probably do as they're the spawn of Satan or somesuch.

Ragging on reviewing practices simply because it doesn't really paint the picture you'd like is also not really useful. What would reviewing be if all we'd get is pages upon pages of subjective experiences and nothing even vaguely comparable to something posted on another site/by another reviewer?Hell, you have fansite forums for that, to read pages after pages of dribble about something that's hot actually isn't hot, something that doesn't perform really performs, something that looks like crap actually looks good, and about how each and every driver revision magically improves IQ and so on. A review has to use a measuring stick that's directly comparable to what other sites use and what John Doe can do at home/will see at home. The bar-charts you so much hate. There has to be a standardized(more or less) system of measurement, and having some wanker harp about whatever he perceived isn't close to being useful, simply because with us humans, opinions and interpretations are abundent, and individual, and quite different. You know that old Sumerian sayingpinions are like assholes-everybody has one. Is the "bar-chart" approach perfect?No, but ATM it's probably the best compromise that can be achieved.

No one said the enthuiast market is irrelevant some of us felt it was getting to much relavance. ATI is doing quite fine without hitting this segment data (not sure if it is concrete) shows ATI gained 20% more market share in the dx 10 segment. Sorry must of us cannot afford $700 gpus and $1500 cpus therefore mainstream products matter to us. But everytime we want to discuss mainstream to highend (not ultra) gpus some "enthuiast" tries to drown us out with that gpus suckerz pay $100 more for 10-20% more fps you will feel better. Ok no. We all understand the trickle down effect of technology which is why enthusiast segment is important. But some persons hype it out of proportion.

Bar charts are irrelevant when when fps are greater than 60-70 fps because the human eye cannot tell the difference.

Yes opinions are like assholes not a gripe against you but some persons keep forcing us to beleive the enthusiast segment is the end all be all of gpus.

 

[AlexV]

No one said the enthuiast market is irrelevant some of us felt it was getting to much relavance. ATI is doing quite fine without hitting this segment data (not sure if it is concrete) shows ATI gained 20% more market share in the dx 10 segment. Sorry must of us cannot afford $700 gpus and $1500 cpus therefore mainstream products matter to us. But everytime we want to discuss mainstream to highend (not ultra) gpus some "enthuiast" tries to drown us out with that gpus suckerz pay $100 more for 10-20% more fps you will feel better. Ok no. We all understand the trickle down effect of technology which is why enthusiast segment is important. But some persons hype it out of proportion.

Bar charts are irrelevant when when fps are greater than 60-70 fps because the human eye cannot tell the difference.

Yes opinions are like assholes not a gripe against you but some persons keep forcing us to beleive the enthusiast segment is the end all be all of gpus.

Reading comprehension is quite important these days...failed to go through the part where I said that the enthusiast segment and the lower ones are intrinsically linked?You can't cut out a part because your favourite company doesn't perform there and say that it's suddenly not that important, after it's been important for so long. What you don't understand is that those whom you so lovingly refer to as budget-conscious most often don't care about the qualities you harp, and will gladly get some IGP/POS cheap thing/whatever Dell puts in, without second thoughts. They'll hardly care about AA/AF or enabling that uber-nice shader in the game's menu.

Good morning, you yourself are an enthusiast. Because you're here, you're talking about IQ, the relevance or lack of relevance of FPS, differing architectures. It has nothing to do with getting expensive stuff(burn the heathens who can actually afford it, they're idiots, aren't they?Power to the ppl, equality to all, stop the wasting of resources...sigh), but it has everything to do with awareness. You're an enthusiast once you start having a darned idea about what's going on under the hood and start caring about getting AA and so on. And you'll look at high-end stuff and you'll evaluate and you'll see how things trickle down to what you can afford/can justify buying. And there are a billion reviews upon release of top-end GPUs, and everyone is drooling over them-everyone in your category, the category that actually knows a thing or two. And what you see at the top end kindof trickles down to lower market segments(this may change with future multi-GPU approaches, but time will tell)-or do you imagine that ATi designed the R6xx family starting with the low end 2400 and scaled it upwards?

You don't understand the importance of having a universal measuring stick in terms of comparing products that are eterogenous in nature, but are meant to perform the same task. How am I to know that pen A is better then pen B?If I listen to dudes saying:whoa, the grip is the awesome, I can scratch my nose exceedingly well with it(the nose and hand shapes are fairly individual, in case you wondered), how useful would that be to me, with my different hand and nose?But if you see a paper that says:A can write underwater, B cannot, A can write at below 0 temps, B cannot, A cannot write on rice paper, B can, I have a more objective way of comparing the products and seeing how they would be of use to me. If in the future a job involving scuba-diving and writing or going to the arctic regions is expected to come up, i'll get a supply of A, if I'm going to write love-letters on rice paper at some point I'll get B, even though at present none of those apply. And I bet you're preparing the argumental-nuke just about now because GPUs aren't pens. Wait a tad, all will be made clear.

Benchmarks are useful and relevant even when they're above 60-70 fps due to a number of reasons. First, an average of 100 over an average of 80 might mean that the lows were higher in the first case(it matters a helluva lot). Second, it might mean that you'll be able to get 4x AA and 16X AF instead of 2X AA and 8X AF. Third, it might mean that when an config raping game like Crysis comes out, 1280x1024 would be a feasible resolution instead of 1024x768 or 800x600. Fourth, it is something that can be neutrally appreciated, unlike comments like:"the image on architecture A was far more vibrant, the IQ(another fairly subjective element, tastes are non homogenous) was outstanding, we felt"-WTF does that tell me?Maybe I feel the opposite upon seeing them in action. Of course considerations WRT image-quality and so on are necessary, but you can't build a review solely on them. Ever wondered why Jeremy Clarkson does mention BHP numbers and times from 0-60, instead of only saying:that car is fast and powerful?

 

[Geo]

Errm, didn't there used to be a Radeon R7 generation thread around here somewhere?

Anyrate, I see that Inq is claiming it taped out recently, and is boldly predicting an easy time of it and a 2Q release. I think that's a little too bold for my stomach at this point, tho I think it likely that R7xx is not nearly so ambitious a change in arch as, say, R5 --> R6. If it really has taped out (and that's the kind of basic fact that Inq is usually more likely to be right about) then certainly a summer release is not an unreasonable target.

 

[Geeforcer]

Especially if a rumor posted elsewhere is true that ATI has climbed from a 10% DX10 marketshare to a 30% DX10 marketshare. Which would indicate ATI is selling more 38xx series GPUs than Nvidia is selling G92 based GPUs. Again, if the rumors are true.

ATI does not have anywhere close to total 30% DX10 market share. There must be more context to CJ's numbers (such as time period) because I can tell you that right now #ATI DX10 GPUs sold/# total DX10 GPUs sold <<< 0.3

We can continue this discussion in the industry forum if anyone is so inclined.

 

[Dave Baumann]

ATI does not have anywhere close to total 30% DX10 market share. There must be more context to CJ's numbers

Q3 Mercury numbers were 36% DX10 marketshare. (And NVIDIA's Q2 CC revealed they had 75% in Q2, where we had just begun the ramp of mainstream parts).