• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

AMD Radeon 7990 reviews thread

Are these dual GPU cards susceptible to the same problems as a Crossfire setup, or does having the two GPUs on the same PCB allow for a superior implementation?
 
Are these dual GPU cards susceptible to the same problems as a Crossfire setup, or does having the two GPUs on the same PCB allow for a superior implementation?

Driver setup is the same. But sources say AMD will be releasing a new driver with the card that addresses many of the CF issues.
 
Are these dual GPU cards susceptible to the same problems as a Crossfire setup, or does having the two GPUs on the same PCB allow for a superior implementation?

The frame timing is typically better on a single card, but it is certainly still an issue.
 
http://www.techpowerup.com/183022/AM...e-Latency.html

Over the past few months, certain tech publications led a borderline-smear campaign against AMD over the way in which its GPUs stream frames to displays. "Frame latency" or "frame time" was purported as a metric of the same importance as frame-rates, in graphics card reviews. Various essentially identical methods were used to show that AMD Radeon GPUs yield higher frame latency (time taken for frames drawn by the GPU to make it to the display) than NVIDIA GeForce ones, even in cases where AMD's chips offer higher frame-rates. AMD has apparently made a significant breakthrough in improving frame latency.

In January, AMD made its first official response to early tests that showed Radeon GPUs to pose higher frame latency. In its defense, AMD stated that frame-latency issues are not a hardware design flaw, and can be ironed out by optimizing drivers to the redesigned memory controllers on GPUs based on its Graphics CoreNext architecture. Sources told us that AMD is ready with its first prototype drivers that fix frame latency issues. These drivers are pre-alpha, and are made available to select industry partners, with an adequate level of competence and expertise. After AMD takes feedback from these partners, the company will begin rolling out the first beta drivers, followed by WHQL-signed ones.
 
Communication could be better on a single card with a pci-e 3.0 spec bridge chip. I also expect AMD to state they are using very particular binned gpu chips for this card. That is not unheard of.
http://news.softpedia.com/news/CES-...PCI-Express-3-0-Launched-by-ASUS-319930.shtml
AMD's chipsets and processors, unlike Intel's, do not support the PCI Express 3.0 interface, which is why motherboard makers didn't bother including the interface in their designs, until now.

It may also be a un enabled feature. I'm not clear if it's available now or not.
 
Weren't the first leaks about this card calling for it to launch in July/Summer? Perhaps their driver fix went smoothly and they accelerated the launch date?
 
They appear to only be paper launching. My guess is the crossfire fix is in some early form where they feel they can launch it and demonstrate it but are still ironing out the wrinkles?
 
They appear to only be paper launching. My guess is the crossfire fix is in some early form where they feel they can launch it and demonstrate it but are still ironing out the wrinkles?

Well, paper launching a card which is already virtually available from third parties and has been for a while. The main thing is whether the drivers will work on all cards, and what might happen to prices of existing already available cards.

Weirdest paper launch ever when comparable cards are already out, assuming the drivers are OK for them.
 
I think you maybe a little surprised Lon.
I expect this new card will be noticeably quicker,run quieter and use less power than the AIB 7990's we've seen so far.
 
If they fix the runt frames issue and it's under $850 I'll be picking one of these up ASAP for my eyefinity setup. I also figure I can pay back around 1/4 or 1/3 of the price mining LTC before the difficulty skyrockets.
 
Well, $1000! Unless there's some special sauce in this card that makes it better than 7970GE Crossfire, it's a very limited niche that will find it better than a crossfire setup.

I suppose it does offer better power consumption, but that's all I can see so far. Maybe it'll O/C real well? Nobody seemed to have the time in the first reviews I've seen to find out.
 
Well, $1000! Unless there's some special sauce in this card that makes it better than 7970GE Crossfire, it's a very limited niche that will find it better than a crossfire setup.

I suppose it does offer better power consumption, but that's all I can see so far. Maybe it'll O/C real well? Nobody seemed to have the time in the first reviews I've seen to find out.

This really doesn't even hold a candle to 7950 CF which would be $540-600, no way you can say this is worth twice the price considering we're not even talking about a large performance gap.
 
The thing I don't understand is that the card that's supposed to be full 7970 ghz crossfire is performing often times much worse than 7970 ghz xfire?

Perhaps they need to turn powertune up to +20%, or the single PCIe slot is the limiting factor? Who knows, but it seems weird to me.
 
How has that ginger Ryan Shrout's hair not turned green yet?

I'm not surprised he didn't use the 13.5 beta driver during the review proper, and I'm also not surprised that he put a his little nvidia jab at AMD's drivers into the conclusion where he showed the drastic improvement in frametime in just one game.
 
The thing I don't understand is that the card that's supposed to be full 7970 ghz crossfire is performing often times much worse than 7970 ghz xfire?

Perhaps they need to turn powertune up to +20%, or the single PCIe slot is the limiting factor? Who knows, but it seems weird to me.

Could be powertune, or whatever they are doing to control power consumption.

Looking at W1zzard at TPU attempt at O/C'ing (Which I didn't see when I looked earlier at the article) it appears that AMD is using some kind of very aggressive throttling to control Temps (and noise?).

power_average.gif


Only using 68w more on avg. than the 7970GHz seems like more than chip binning is going on to accomplish that.
 
Color me dissapoint.

28nm is just an excuse to jack up prices.

Seriously? Almost 1.5 years late to the party, $1k, nothing that hasn't been here other than perhaps binned chips. The crossfire drivers should have been fixed anyway, and will help 7970x2 which is way cheaper.

1.5 years later and you get to pay more for the performance.

8 Games is quite a few, but a $700 card would be a lot better.

What a joke this generation is.

560 ti -> renamed 680, $200 premium
7970 $150 premium
titan $400 premium!
7990 ~$150 premium (over current 7970G prices ~$425)

Maybe that's why AMD and NV keep 'exchanging' workers, they are forming a duopoly to increase prices.
 
Back
Top