where does the crossfire launch leave r520

[Dave Baumann]

You know, if you read my article I was thinking along the same kind of lines, until it was suggested I look elsewhere.

Realistically there is no point, in monetry / adoption terms to X850 Crossfire - it is there, nw, as a noise maker, a nice little distraction for a few months. However, move down the line a little and it becomes quite a different proposition. Where Crossfire is likely to gain the most traction is not with the X850's, but actually at the X800 XL level.

X800 XL's are already hitting a fairly key price spot for the system builders and its quite probable that you'll many more systems shipping with X800 XL crossfire, or even just the master cards (as, without reliance on same make/model boards makes it easier for the end user just to bang in another and that can be a selling point), as these can be had for a fairly low price, are fairly good performers in the first place.

 

[trinibwoy]

I must say that many people seem to regard the XL as the healing salve for many of ATi's woes. The guy who came up with that SKU must've gotten a nice bonus!

 

[ondaedg]

Onda,

A perfect time to use the Chewbacca defense, but you go with the more original toaster. Nice, but I would have liked the SP reference.

The most salient part of your evidence is VIA, actually. ATI and nVidia both have other reasons to push SLI. What reason does VIA (or actually why did Intel buckle under to nVidia first?) for pushing for SLI/Crossfire capable motherboards?

If there's no market, or it's not significant, why would motherboard chipset manufacturers be either bending over to nVidia (Intel) or trying to leverage ATI's "open" method against nVidia (VIA) to get them to make their multi-gpu method "open" as well?

The only reason I can see is that both Intel and VIA saw their markets shrink because people were purchasing SLI capable motherboards in quantities that caused them concern.

How many people actually put multiple GPUs into those boards doesn't really matter, does it? In fact, I only see it matter to ATI... which is why I think the timing of Crossfire is suspect and rather silly. Crossfire is aimed at the same market as the R520. People who already own 850s and even 800s are the same people who will go out and buy the R520. Maybe they'll buy a "Crossfire capable" mobo as well. I highly doubt they'll buy a "Crossfire capable" mobo, plus a "master card" and then turn around and buy a R520 two months later. (Whether or not they ever buy the second R520, I think is highly unlikely)

The point to my original post is that the capability of SLI is selling motherboard chipsets for Nvidia. Other vendors are rushing to have this same capability. I don't know about you, but if I were picking a motherboard right now, I more than likely would go with the one with SLI as opposed to one that did not no matter who the vendor is. Why? For me, the answer is flexibility and upgradability. Walt and I argued about this way back when SLI was just being released. Based on the current market, I would have to say that SLI has been a success and it looks like it will continue to be well into the future.

I would have to disagree with you in regards to the upgrade path people choose. Especially those who purchased an 800xl. I don't see them suddenly dumping the 800xl and purchasing an r520. If they purchased an 800xl, they probably couldn't justify spending 400+ for a video card, or they simply couldn't afford it. Adding a second 800xl will be alot easier and possibly cheaper then purchasing an r520. Based on preliminary numbers, it is looking like ATI's SLI works very good. Possibly better than Nvidia's even. If it works that good, you may see alot of dual x800 systems.

 

[ondaedg]

You know, if you read my article I was thinking along the same kind of lines, until it was suggested I look elsewhere.

Realistically there is no point, in monetry / adoption terms to X850 Crossfire - it is there, nw, as a noise maker, a nice little distraction for a few months. However, move down the line a little and it becomes quite a different proposition. Where Crossfire is likely to gain the most traction is not with the X850's, but actually at the X800 XL level.

X800 XL's are already hitting a fairly key price spot for the system builders and its quite probable that you'll many more systems shipping with X800 XL crossfire, or even just the master cards (as, without reliance on same make/model boards makes it easier for the end user just to bang in another and that can be a selling point), as these can be had for a fairly low price, are fairly good performers in the first place.

I agree with you about the x800xl. I find a dual x800xl system more attractive than a 6800gt sli system due to the price difference and even more attractive than a 6600gt sli system due to performance. It is a great fit in my opinion.

 

[dizietsma]

X800XL will only take off if 2 of them are faster than 1 r520 or significantly cheaper, or else people will get 1 r520 and then say " well I can always get the other r520 later "

Is 2 x X800XL going to be faster and / or cheaper than 1 r520 ?

 

[Charmaka]

What reason does VIA (or actually why did Intel buckle under to nVidia first?) for pushing for SLI/Crossfire capable motherboards?

Again, I think serious consideration needs to be given to mindshare. Currently, VIA frankly aren't generally considered when discussing A64 Mobos, because nVidia solutions do everything you'd want and there's no reason to look further. If VIA can come out with a kick-ass SLI/Crossfire board people will sit up and take notice, and then look at the rest of the VIA range.

 

[RancidLunchmeat]

Onda / DB,

So the market is for X800 users? But, aren't those users existing users? As in consumers who currently have X800s?

Those consumers are unlikely to have dual PCIx16 mobo's, because they're ATI fans. So the market for Crossfire is going to be current X800 users who will be willing to buy a new Crossfire Capable motherboard, plus an X800 Master Card.

And this solution is supposed to be.. what? Compared to the R520? Cheaper? Faster? Both? (Clearly can't be both, I wouldn't think)

I guess the question is why would X800 users (and not X850 users) be willing to purchase a new motherboard and a new master card, rather than simply buying the R520? Is it the hope of a 'future' upgrade path? That the "Crossfire capable" motherboard will be a future investment because later they will want to upgrade to two R520s (or whatever after)?

I still see ATI asking people to upgrade their motherboard (and whatever other system components) at the end of a product's lifecycle in order to make use of past tech when future tech is already on the market or soon will be.

That's not the proposition nVidia proposed to consumers when they launched SLI.

And let's face it.. once the G70 releases, IMO, most of these arguments for SLI/Crossfire in terms of 'future proofing' will go flying out the window.

In fact, I'm beginning to think that Dual GPUs might have a one-two generation life cycle.

 

[Dave Baumann]

So the market is for X800 users? But, aren't those users existing users? As in consumers who currently have X800s?

Personally I wasn't talking about exisitng users, but the system builders that want to attach on to these types of things - the price of the XL is probably a bit of a sweet spot for them.

 

[jb]

Personally I wasn't talking about exisitng users...

Well I have an 800xl so I would be interested. I can sell my current NF4 Ultra board. And get the new xfire 800 version and mombo for only 400 or less depending on what I can sell my nf4 price wise. That gives me the better AA modes and faster perfromace. Granted I could get the R520 but not sure when that comes out and what its AA will be like. As I doubt I can get the same level of AA at the same speed in a single card. Not that I am planning on doing this. But I have this as an option now...and ATM it does not soound THAT bad...as I love good AA...

 

[trinibwoy]

Personally I wasn't talking about exisitng users...

Well I have an 800xl so I would be interested. I can sell my current NF4 Ultra board. And get the new xfire 800 version and mombo for only 400 or less depending on what I can sell my nf4 price wise.

You would really put $400 into a dual SM2.0 setup right now instead of selling the XL and buying a single R520 which would probably come out to be less and have faster performance (in regular AA modes) and full SM3.0 support? That's contingent on next gen engines utilising SM3.0 features to good effect but I would still bet the IQ improvement will be greater than 14x vs 6x. Not to mention an XL will probably suffer badly in next gen games at 6x anyway. This is all speculation but I'm betting the games that you can run on an R520 at 6xAA you won't be able to do better than 10x SuperAA on dual-XL's - you may even have to drop resolution a bit plus you'll be missing out on all of the R520's extra features!!

 

[jb]

Personally I wasn't talking about exisitng users...

Well I have an 800xl so I would be interested. I can sell my current NF4 Ultra board. And get the new xfire 800 version and mombo for only 400 or less depending on what I can sell my nf4 price wise.

You would really put $400 into a dual SM2.0 setup right now instead of selling the XL and buying a single R520 which would probably come out to be less and have faster performance (in regular AA modes) and full SM3.0 support? That's contingent on next gen engines utilising SM3.0 features to good effect but I would still bet the IQ improvement will be greater than 14x vs 6x. Not to mention an XL will probably suffer badly in next gen games at 6x anyway. This is all speculation but I'm betting the games that you can run on an R520 at 6xAA you won't be able to do better than 10x SuperAA on dual-XL's - you may even have to drop resolution a bit plus you'll be missing out on all of the R520's extra features!!

When and if the R520 is out and can do all that then I will relook at what I have. I will weigh what we know at the time about the R520 and its price vrs availbility vrs what I can get out of cross fire. I also have 4 older systems that I use at my place (son has one, keep an extra one to play games with in case a buddy comes over). If and when the R520 was the better buy then I can pass down and/or sell the 2 800xl. Its been almost 2 years before we have seen what PS2.0 engines can do. I think it will take more time out before we see any real use of PS3.0. Please keep in mind that I am an AA nut and would rather have better AA and give up on other IQ features...

 

[trinibwoy]

When and if the R520 is out and can do all that then I will relook at what I have. I will weigh what we know at the time about the R520 and its price vrs availbility vrs what I can get out of cross fire. I also have 4 older systems that I use at my place (son has one, keep an extra one to play games with in case a buddy comes over). If and when the R520 was the better buy then I can pass down and/or sell the 2 800xl. Its been almost 2 years before we have seen what PS2.0 engines can do. I think it will take more time out before we see any real use of PS3.0. Please keep in mind that I am an AA nut and would rather have better AA and give up on other IQ features...

Ok, thought you were looking at it as some sort of long-term solution. If you are willing upgrade at any time then you can go for anything

 

[karlotta]

I wonder if the 110nm Fab for the ATI XL and the NVDA G70 cards will put a crimp in eithers ability to get product to market. They are both on TSMCs 110nm fab?It looks like the first direct competition for the same Fab in years.

 

[trinibwoy]

I wonder if the 110nm Fab for the ATI XL and the NVDA G70 cards will put a crimp in eithers ability to get product to market. They are both on TSMCs 110nm fab?It looks like the first direct competition for the same Fab in years.

I don't think the competition will be that great on 110nm. There will most likely be much higher demand for XL's than top-end G70's anyway. The problem will be at 90nm when both companies entire lines are using the same process.

 

[ondaedg]

I think the one thing alot of us are failing to realize is that the average consumer for the most part do not visit Beyond3d's forums nor have the same level of knowledge that we do in terms of market value for each video card. Their buying cycle may also be different. There are alot of people in the world who would give a shop 2000.00 and tell them to build them a kick ass gaming machine right now. The shop is not going to tell the customer: "I would wait until the R520 comes out in the late summer."

 

[RancidLunchmeat]

I think the one thing alot of us are failing to realize is that the average consumer for the most part do not visit Beyond3d's forums nor have the same level of knowledge that we do in terms of market value for each video card.

Well, that's true, and I mentioned it in one of my long-winded babbles above.

How many people go to a "system builder" with a fist-full of cash and say 'build me the best system'?

Perhaps that market is larger than I think it is, because I normally segment the market into three groups: Those who buy off the shelf, those who build themselves, and those who buy off the shelf but "upgrade".

Point being, I don't see how Crossfire appeals to people outside of those who visit Beyond3d's forums (or something less technical) because those are likely to be the only ones who really understand/are interested/have even heard of Crossfire.

Maybe I'm jumping to incorrect conclusions because there's a segment of the market (those who spend lots of money with system builders) that exists at a level that would be stunning to me.

 

[Diplo]

I have to agree that both SLI and CrossFire are of zero interest to me at the moment and I don't actually know of anybody who is interested. Not only will it mean that most people will have to buy a new motherboard, but it often entails buying a new PSU and more case fans to cope with the extra power drain. Apart from making all the lights in your house dim the two cards will be generating mucho heat and noise. Then you are bound to get many situations were games are plain buggy, IQ is diminished or you have to wait for driver fixes to come out. Sorry, forget it.

 

[nutball]

How many people go to a "system builder" with a fist-full of cash and say 'build me the best system'?

How many people buy a system because it's got 1024 megathingmes, or 16 kilowhatsits with "HT" technology (Hormone Therapy? I dunno), or because it's 3.2 gigaflaps fast?

 

[Geo]

It seems to me that another factor that has to be considered is that many (most, is my impression) of the folks who have pcie X8xx-class cards already (and thus are the potential "upgrade" market with the best price leverage) have them in Dell (or similar) systems with 2 or 3 yr warranties. . .and are the kind of folks who would buy a Dell (instead of building) in the first place.

 

[Geeforcer]

SLI has been out for many months now. Does HP or Gateway offer an SLI system? I don't expect that Crossfire will fare any better in that market.