AMD Radeon Pro Duo Performance Numbers Leaked

[Slizzo]

The 295X2 uses 16.1 WHQL Crimson, performance is pretty much the same as current. Many of the games you play are what i consider less demanding. But several of them will make your setup go to it's knees at 2160p if you don't lower IQ. Crysis maxed will for sure, and that game is over 3 years old now. Alot of games won't run on 390 CF at 2160p unless IQ is lowered.

By playable you mean 20-30 fps like consoles or?

Then I understand.

Metro is incredibly demanding, Fallout 4 is not light, but lighter than others. Carmageddon can be kind of crazy (it is on my now modest rig). Borderlands is for sure lighter.

 

[cdawall]

The 295X2 uses 16.1 WHQL Crimson, performance is pretty much the same as current. Many of the games you play are what i consider less demanding. But several of them will make your setup go to it's knees at 2160p if you don't lower IQ. Crysis maxed will for sure, and that game is over 3 years old now. Alot of games won't run on 390 CF at 2160p unless IQ is lowered.

By playable you mean 20-30 fps like consoles or?

Then I understand.

Typical high 40-50's and I cant think of any times I had 20's. That is playable in all those games...

 

[arbiter]

Pcper just released benchmarks using nano in CF, 295x2 and 980TI's in SLI. As i did expect when they released the images here showing a fury pro vs single 980ti to make fury pro look good performance wise, when compared to 2 980ti's in SLI which would be a FAIR test well fury got beat.
http://www.pcper.com/reviews/Graphics-Cards/AMD-Radeon-Pro-Duo-Review

 

[cdawall]

Pcper just released benchmarks using nano in CF, 295x2 and 980TI's in SLI. As i did expect when they released the images here showing a fury pro vs single 980ti to make fury pro look good performance wise, when compared to 2 980ti's in SLI which would be a FAIR test well fury got beat.
http://www.pcper.com/reviews/Graphics-Cards/AMD-Radeon-Pro-Duo-Review

This is the impressive part of all that. I imagine if you got the temps down the fury would lower power consumption, just like the fermi cards (I halfed my power consumption overclocked vs stock coolers).

but on the note of the benchmarks the duo won 3 of the 4 games run for average FPS. Lost 3dmark (go figure) and TW3 (go figure) to me this shows that the pro duo is more powerful at least stock vs stock. Unluckily for AMD we all know how the 980Ti clocks and how the fury does not. Even with the TDP unlocked I don't expect that to change.

 

[arbiter]

This is the impressive part of all that. I imagine if you got the temps down the fury would lower power consumption, just like the fermi cards (I halfed my power consumption overclocked vs stock coolers).

but on the note of the benchmarks the duo won 3 of the 4 games run for average FPS. Lost 3dmark (go figure) and TW3 (go figure) to me this shows that the pro duo is more powerful at least stock vs stock. Unluckily for AMD we all know how the 980Ti clocks and how the fury does not. Even with the TDP unlocked I don't expect that to change.

duo pro won 3 outta 4 games? um are you sure you looked at same page i linked cause according to story it lost 3 outta 4. The one it won was by 2% at 4k but lost at 1440p. 2% could be seem as margin of error so being even where as ones it lost were 10+%.

 

[cdawall]

duo pro won 3 outta 4 games? um are you sure you looked at same page i linked cause according to story it lost 3 outta 4. The one it won was by 2% at 4k but lost at 1440p. 2% could be seem as margin of error so being even where as ones it lost were 10+%.

Shit I read that backwards as hell lol ignore my comment

 

[medi01]

duo pro won 3 outta 4 games? um are you sure you looked at same page i linked cause according to story it lost 3 outta 4. The one it won was by 2% at 4k but lost at 1440p. 2% could be seem as margin of error so being even where as ones it lost were 10+%.

A pity 480p resolutions aren't tested. That's where 980 would really outshine Fury.

 

[Slizzo]

This is the impressive part of all that. I imagine if you got the temps down the fury would lower power consumption, just like the fermi cards (I halfed my power consumption overclocked vs stock coolers).

but on the note of the benchmarks the duo won 3 of the 4 games run for average FPS. Lost 3dmark (go figure) and TW3 (go figure) to me this shows that the pro duo is more powerful at least stock vs stock. Unluckily for AMD we all know how the 980Ti clocks and how the fury does not. Even with the TDP unlocked I don't expect that to change.

When it comes to games, you really need to pay attention to the frametimes that PC Per posts. Those don't lie about actual user experience with the cards. While all the AMD cards did really well in average FPS, their frame times were all over the place.

This will, hopefully, be fixed in a driver. But for now even AMD is not recommending the RPD for gamers.

 

[arbiter]

When it comes to games, you really need to pay attention to the frametimes that PC Per posts. Those don't lie about actual user experience with the cards. While all the AMD cards did really well in average FPS, their frame times were all over the place.

This will, hopefully, be fixed in a driver. But for now even AMD is not recommending the RPD for gamers.

Its fine in say like fallout 4, yes its frame time is 3-4x what 980ti but even then its only 2-3ms. When it looks like it does in GTA5 that is a big problem that needs to be addressed.

 

[Fluffmeister]

Pcper just released benchmarks using nano in CF, 295x2 and 980TI's in SLI. As i did expect when they released the images here showing a fury pro vs single 980ti to make fury pro look good performance wise, when compared to 2 980ti's in SLI which would be a FAIR test well fury got beat.
http://www.pcper.com/reviews/Graphics-Cards/AMD-Radeon-Pro-Duo-Review

Yeah noticed Tech Report did a little article on it, the Duo's frame times are appalling at times.

http://techreport.com/news/30066/pc-perspective-pokes-and-prods-the-radeon-pro-duo

 

[Kanan]

These are child ills that get fixed soon I guess, too soon to disqualify a card for its premature drivers. But clearly, as a gamer, I'd rather get a 980 Ti / 2x 980 Ti or 2x 390 (budget solution), all these solutions have less problems and are very fast too. 4GB (per chip) isn't much for a 1500$ crossfire card, it's maybe good for 1 year maximum (1440p/4K - and it's rather a 4K card...), so you should be really wealthy even considering this card (for gaming).

 

[RealNeil]

I have two GTX-980 Ti cards in SLI, two GTX-980 cards in SLI, and soon I'll have a pair of R9-390X Toxic cards in Crossfire. (I have the first one here already)

I think that I'll just sit back and wait for a few months (maybe longer) because I'm not in dire need of an upgrade. All three gaming PCs are ready.

 

[Fluffmeister]

It's odd, not sure how Crossfire'd Fiji's on a stick qualify for premature drivers, especially when they are flogging the thing for $1500.

A fool and his money as they say...

 

[arbiter]

These are child ills that get fixed soon I guess, too soon to disqualify a card for its premature drivers.

Since card been out for a while and CF pretty much same as this duo gpu i think saying drivers are premature isn't really a valid argument since setting 2 cards up on CF should and would produce same issues.

 

[GhostRyder]

By those standards, I guess we should start judging all cards based on the idea that if 1 game runs bad then all games run bad logic.

I have two GTX-980 Ti cards in SLI, two GTX-980 cards in SLI, and soon I'll have a pair of R9-390X Toxic cards in Crossfire. (I have the first one here already)

I think that I'll just sit back and wait for a few months (maybe longer) because I'm not in dire need of an upgrade. All three gaming PCs are ready.

That's not true, its always a dire need to upgrade .

Pro Duo is overpriced, but not aimed as a gaming card regardless of how illogical that is. Move along, buy two Nanos if you want similar performance and grad a couple of water blocks to gain similar performance all around and pocket the rest of the cash.

 

[RealNeil]

That's not true, it's always a dire need to upgrade .

True, but that's not to say I'll be laying back and not buying anything. I have my eye on a few PCI-E drives for my systems.

 

[Fluffmeister]

Since card been out for a while and CF pretty much same as this duo gpu i think saying drivers are premature isn't really a valid argument since setting 2 cards up on CF should and would produce same issues.

It's DOA, you're spot on.

 

[RealNeil]

Now, with the announcement from NVIDIA about the 1080 and 1070 release, the pressure is really on AMD.
We're going to see a lot of change in the GPU market now. It will be good for consumers too.

Nvidia's prices for these are pretty reasonable.

 

[arbiter]

It's DOA, you're spot on.

fury pro should been out when they announced it months ago. But the delay they put on it pretty much put it in shark tank with the new nvidia cards 2 of them draw same power range as fury but gaming wise should be much faster least on paper cause well SLI/CF is hard to get right.

 

[Kanan]

fury pro should been out when they announced it months ago. But the delay they put on it pretty much put it in shark tank with the new nvidia cards 2 of them draw same power range as fury but gaming wise should be much faster least on paper cause well SLI/CF is hard to get right.

It's not a card for gamers anyway, it's more like a graphics card for pro users (VR programming etc) that maybe game too. It also has FirePro drivers besides normal Radeon Crimson drivers.

Also you can't possibly compare a premium watercooled card, 2 slots, to a normal solution with 2 air cooled graphics cards, 4 slots, without keeping the distinct advantages in mind - that would be highly unwise.