Rumor: Radeon R9-290X-X2 to Bring Double the Hawaii?

[EpsylonRad]

Entender y Leer.

Veo mucha envidia aca.

Es Obvio que AMD apuntó con un menor tamaño del chip Hawaii xt a tener la "posibilidad" de hacer un PCB x2.

Donde queda Nvidia con una gráfica de esa potencia???

Google translate:

Understand and Read.

I see a lot of envy here.

It's Obvious that AMD pointed a smaller chip xt of Hawaii have the "opportunity" to make a PCB x2.

Where is Nvidia with a graph of that power??

 

[Novuake]

LOL the name will be appropriate. But I call BS, going from Islands in the pacific to a volcano in the Europe is a stretch.

 

[rostrow416]

I guess amd no longer considers power or heat as issues when designing chips. Pair this up with a fx 9570 whatever and you can replicate Japan's nuclear disaster at home.

 

[Nefos]

and here comes the.. right question
will it run in Crossfire? will there be issues like at 7990?
Will it melt a normal alu case in CF? what radiator should we use for this setup: a quad slot pc, or a smaller car`s?

 

[fatboytyler]

@sbudbud There is no way possible that it would match the price of a GTX 780 Ti. That is not economically pheasable by any stretch of the imagination. Would it be the same as buying two 290x out right? It would likely be cheaper (IMO).

 

[Sakkura]

Quote "Vesuvius, named after the only active volcano in Europe. This is not true, there are many active volcanoes in Europe such as mount Etna for example. .http://www.volcanodiscovery.com/europe.html

Yeah, Etna is in fact erupting right now, which makes it all the more absurd. It is one of the most active volcanoes in the world.

 

[Matous Mojzis]

"While it remains uncertain what the real name will be, it is rumored that the AIBs are going with Vesuvius, named after the only active volcano in Europe."
And what about volcanoes on Iceland?

 

[burmese_dude]

The all new R9 290X2 / heater.

 

[quilciri]

Hey, AMD. Fix your damn frame latency before you try to hock another dual GPU card.

 

[ddpruitt]

Vesuvius, named after the only active volcano in Europe.

I think that's not the name but a description of how it looks when it's running.

 

[Sakkura]

Hey, AMD. Fix your damn frame latency before you try to hock another dual GPU card.

They pretty much have.

 

[epaciga]

AMD: "780ti huh... we'll show them!"

 

[quilciri]

Hey, AMD. Fix your damn frame latency before you try to hock another dual GPU card.

They pretty much have.

Yeah, don't think so. Lack of dx9 or multi-monitor driver support in the bridge version of xfire. Tom's reveiw of the 290x said the frame latency of the pci-e implementation of xfire was better, but not fixed. Even when it's working, the frame pacing still isn't as smooth as Nvidia's.

At this point, AMD has spent the majority of a year trying to correct what Nvidia already had fixed when they started. They'll *probably* catch up sometime in the coming year, but ditched my 7850 xfire because I'm tired of waiting.

 

[Sakkura]

Yeah, don't think so. Lack of dx9 or multi-monitor driver support in the bridge version of xfire. Tom's reveiw of the 290x said the frame latency of the pci-e implementation of xfire was better, but not fixed. Even when it's working, the frame pacing still isn't as smooth as Nvidia's.

At this point, AMD has spent the majority of a year trying to correct what Nvidia already had fixed when they started. They'll *probably* catch up sometime in the coming year, but ditched my 7850 xfire because I'm tired of waiting.

Check the 290X review:

a pair of Radeon R9 290Xes in CrossFire support the company’s frame pacing technology at 3840x2160

Same applies for the 290.

 

[quilciri]

I think you misread. I said the bridge xfire doesn't support multi monitors, which is true. The pci-e version, acc. to tom's review, is just not up to Nvidia's smoothness.

Also from the review you linked:

Here's more evidence that CrossFire likely needs some additional work. This is with frame pacing enabled, and yet the paired-up R9 290Xes demonstrate fairly high worst-case variance, more than doubling what we observe from one card.

With either implementation, it takes some balls to push a multi-GPU card with a straight face right now. Maybe AMD has some heretofore unseen single-board/multi-GPU ace up their sleeve, but I wouldn't hold your breath,

 

[Sakkura]

I think you misread. I said the bridge xfire doesn't support multi monitors, which is true. The pci-e version, acc. to tom's review, is just not up to Nvidia's smoothness.

So you're saying it's bad because of problems that specifically do not affect it?

 

[quilciri]

I'm saying it's not up to par yet because of the other 80% of my post.

The pci-e implementation, which is what the 290 and 290x use, is still not as good as Nvidia's, and Tom's points this out in several places in the very review you linked; I assume you read all of the article. I don't understand what you're trying to refute here.

The sentence that you quoted was in reference to your "correction", which was unecessary in the first place because everything in my original post was factual.

 

[quilciri]

The inclusion of bridge-based xfire is for perspective. AMD spent at least 10 months on frame-pacing for their bridge implementation. It doesn't support multiple monitors or DX9 games.

They have now switched to an entirely new topology for xfire, which leaves anyone who bought a 7k or older AMD card out in the cold (they have to hope for a complete in-driver fix - which, given the new hardware xfire seems less likely to happen)...and that new topology is still not as good as Nvidia's solution.

 

[Sakkura]

I'm saying it's not up to par yet because of the other 80% of my post.

The pci-e implementation, which is what the 290 and 290x use, is still not as good as Nvidia's, and Tom's points this out in several places in the very review you linked; I assume you read all of the article. I don't understand what you're trying to refute here.

The sentence that you quoted was in reference to your "correction", which was unecessary in the first place because everything in my original post was factual.

The frame time variance with Crossfire 290X is fine in most of the games tested.

 

[quilciri]

The frame time variance with Crossfire 290X is fine in most of the games tested.

There is 5 millisecond 95th percentile variance in Tomb raider, almost 9ms in Metro, 10ms in Crysis 3 and 11ms in Skyrim, the last of which is DX9 and unsupported by xfire (still!). 11ms is heavy microstutter. 5 is noticeable. that's a problem in 4 of the 7 the games tested.

Tomshardware called the 2.64ms variance in Bioshock infinite "not problematic, but notable". I can agree with that (though Nvidia's variance was less than 1ms there)

When you're talking 2, 3, and 4 times that much difference in frame rendering the 95th, that is not "fine." That is noticeable microstutter.

 

[Sakkura]

It's not all about the 95th percentile. It's better to look at a graph of frame variance (or raw frame time) vs. percentile. But even at the 95th percentile, there are several cases where Crossfire performs better than SLI. Which is why I said Crossfire is pretty much fixed. It's not exactly as good as SLI yet, but it's close enough to no longer be a serious issue, at least with the 290 and 290X.

 

[quilciri]

It's not all about the 95th percentile. It's better to look at a graph of frame variance (or raw frame time) vs. percentile.

Respectfully, yes it is. If you have a 10ms 95th percentile variance in a 30fps game, as is often the case in that reveiw we've been referring to, that means that 1 1/2 times every second you're going to see a ripple go down your screen as a large part of one frame overlaps with the next.

But even at the 95th percentile, there are several cases where Crossfire performs better than SLI. Which is why I said Crossfire is pretty much fixed.

Please, Link where crossfire had better frame latency than SLI. I'm eager to see that.

It's not exactly as good as SLI yet, but it's close enough to no longer be a serious issue, at least with the 290 and 290X.

To you maybe. Tom's hardware disagrees.

Once again, the latency issues in the 290x xfire are in addition to the big middle finger AMD just gave to people who bought a pair of their previous cards. Switching hardware to improve crossfire implies to everyone else that this the improvements to in-driver frame pacing aren't long for this earth, better buy a new card. The people who've faithfully bought a pair of 7x cards while waiting and hoping for AMD's drivers to catch up seem to be screwed (ditto for those that bought a 6x card, thinking they'd xfire later)

bit disgruntled? You bet I am....P.S. there's a pair of used Asus 7850's up for sale on craigslist in the north chicagoland area.

 

[spookyman]

Will it be the same fiasco as the 7990 was?

 

[Novuake]

Will it be the same fiasco as the 7990 was?

Fiasco?

 

[Sakkura]

Respectfully, yes it is. If you have a 10ms 95th percentile variance in a 30fps game, as is often the case in that reveiw we've been referring to, that means that 1 1/2 times every second you're going to see a ripple go down your screen as a large part of one frame overlaps with the next.

Please, Link where crossfire had better frame latency than SLI. I'm eager to see that.

To you maybe. Tom's hardware disagrees.

Once again, the latency issues in the 290x xfire are in addition to the big middle finger AMD just gave to people who bought a pair of their previous cards. Switching hardware to improve crossfire implies to everyone else that this the improvements to in-driver frame pacing aren't long for this earth, better buy a new card. The people who've faithfully bought a pair of 7x cards while waiting and hoping for AMD's drivers to catch up seem to be screwed (ditto for those that bought a 6x card, thinking they'd xfire later)

bit disgruntled? You bet I am....P.S. there's a pair of used Asus 7850's up for sale on craigslist in the north chicagoland area.

Crossfire has better frame times than SLI in Bioshock Infinite:

Source: PC Perspective (and that's just the variance graph, the raw frame times are even more in favor of Crossfire).

The 95th percentile is not all that matters. The whole spectrum of percentiles matter.

Tom's Hardware disagrees with your conclusions. In Bioshock Infinite, for example, they point out that there's a difference, but not a noticeable one.

As for the serious issues Crossfire had in the past, or currently has on other cards, that's irrelevant to the 290, 290X, and the hypothetical 2x290X.