Polaris 10 vs R9 390/390X @ GFXBench

[Sweepr]

GFXBench Windows OpenGL Scores

- Manhattan 3.1
P10
Offscreen 9624 Frames (155.2 Fps)
R9 390
Offscreen 10343 Frames (166.8 Fps)
R9 390X
Offscreen 13376 Frames (215.7 Fps)

- Manhattan
P10
Offscreen 7550 Frames (121.8 Fps)
R9 390
Offscreen 8498 Frames (137.1 Fps)
R9 390X
Offscreen 11341 Frames (182.9 Fps)

- T-Rex
P10
Offscreen 29585 Frames (528.3 Fps)
R9 390
Offscreen 39471 Frames (704.9 Fps)
R9 390X
Offscreen 41982 Frames (749.7 Fps)

https://gfxbench.com/compare.jsp?be...wname1=AMD+67DF:C4&D2=AMD+Radeon+(TM)+R9+390X

Polaris 11 (67FF), Polaris 11 (67E8) and Polaris 10 (67DF) results available.

Same ID as the Polaris 10 SiSoftware entry:

 

[thesmokingman]

Authenticity...?

 

[Sweepr]

Authenticity...?

http://videocardz.com/59468/amd-polaris-10-and-11-opengl-benchmarks-spotted

 

[ShintaiDK]

Results does look as expected. 232mm2 and 6Ghz 256bit GDDR5 is what it is. There is a long way up to a 512bit card.

 

[poofyhairguy]

Sooooo, slower than a 390?

That would suck.

 

[thesmokingman]

Sooooo, slower than a 390?

That would suck.

No one knows what it is. This is like beyond early leaks.

 

[jpiniero]

Authenticity...?

Pretty decent. We don't know what the clock speeds are though... it could be at 700 Mhz for all we know.

 

[rainy]

Results does look as expected. 232mm2 and 6Ghz 256bit GDDR5 is what it is. There is a long way up to a 512bit card.

This is reply for you from videocardz.com:

There are few things to consider before you try to analyze those scores. First, we don’t know if those parts are mobile or desktop, secondly we don’t know what platform was used, furthermore it’s unclear if those chips were running at full speed and using full silicons, and finally OpenGL benchmarks are often capped at 60 FPS. So in other words, those values are probably not the best specimens for inter-GPU comparison, if they are useful at all.

 

[thesmokingman]

Pretty decent. We don't know what the clock speeds are though... it could be at 700 Mhz for all we know.

Meaningless numbers. It will probably used as fuel for the opposing side no doubt.

 

[TypoFairy©]

Could even be mobile p10 just to troll everyone

 

[Pinstripe]

http://videocardz.com/59468/amd-polaris-10-and-11-opengl-benchmarks-spotted

Even getting whipped by a GTX 950 lol.

Well, OpenGL was never AMD's strongpoint to begin with.

 

[Olecki]

Sooooo, slower than a 390?

That would suck.

AMD describing Polaris 10 as a mainstream chip. Why so many people expecting similar or better performance than previous generation enthusiast segment? Really powerful card we should get with Vega.
I wouldn't be very surpised if full Polaris 10 will be a little bit slower than 390 but much cheaper.

 

[poofyhairguy]

AMD describing Polaris 10 as a mainstream chip. Why so many people expecting similar or better performance than previous generation enthusiast segment?

Because AMD keeps talking about increasing the number of VR capable systems. That means more computers with a 390 or better performance.

 

[USER8000]

So the first two columns are Polaris 11,the third is Polaris 10 and the last is a GTX950.

Wow,so a GTX950 is faster than Polaris 10.

In fact so is an R9 380:

https://gfxbench.com/device.jsp?benchmark=gfx40&os=Windows&api=gl&cpu-arch=x86&hwtype=dGPU&hwname=AMD%20ASUS%20R9%20380%20Series&did=27959360&D=AMD%20ASUS%20R9%20380%20Series

Yes,OP,totally legit results. No clockspeed listed let alone even the driver revisions used.

Nice,the OP had to make a new thread and not plonk it in the other one.

 

[Insomniator]

Lets say its a desktop part equal to a 390. What is it supposed to cost? I bought a 390 for $230 6 months ago.

I mean... unless its $199 then what is the freakin point. OOO it uses less power go green! Who ****ing cares. Give me performance.

 

[Sweepr]

Pretty decent. We don't know what the clock speeds are though... it could be at 700 Mhz for all we know.

While we cannot go into more details in order to protect the source, we can confirm that AMD Polaris 10 engineering samples are varying in clock between 800 and 1050 MHz, depending on the partner. Our original Polaris 10 story is here.

Assuming 800 MHz it would end up between the two Hawaii cards above at 1050 MHz (Ellesmere XT?). Close to R9 390X at <$300 wouldn't be bad for desktop mainstream parts (to be named R9 480 Series according to Bench-life).

 

[Sweepr]

Wow,so a GTX950 is faster than Polaris 10.

AMD/NVIDIA performance is not directly comparable here.

In fact so is an R9 380:

https://gfxbench.com/device.jsp?ben... Series&did=27959360&D=AMD ASUS R9 380 Series

No it isn't, not the one in your link neither average R9 380 Series scores.

R9 380 Series
Manhattan 3.1: Offscreen 8288 Frames (133.7 Fps)
T-Rex: Offscreen 29107 Frames (519.8 Fps)

So (ES?) Polaris 10 at unknown clocks is beating retail R9 380 Series.

Yes,OP,totally legit results.

Indeed.

 

[selni]

Hasn't AMD OpenGL performance been awful for ages? Wouldn't read too much into it.

 

[Glo.]

Expectations were high, and when the reality does not catch up to the chase, people will downplay the results.

I would not discard those results, nor take them seriously. Take everything with huge grain of salt.

 

[thesmokingman]

Indeed.

Yes,OP,totally legit results. No clockspeed listed let alone even the driver revisions used.

Nice,the OP had to make a new thread and not plonk it in the other one.

Indeed? Ya, I don't think he was giving you props. But go ahead, spin away not like you didn't intend to spin anyways.

 

[DooKey]

Man, the AMD faithful sure hate leaks that don't favor them, but love those that do.

Rinse and repeat for the NV faithful.

 

[Maverick177]

If Polaris 10 is slower than the 390, I'll sell my Fury for $100 then eat my own socks.

 

[poofyhairguy]

Assuming 800 MHz it would end up between the two Hawaii cards above at 1050 MHz (Ellesmere XT?). Close to R9 390X at <$300 wouldn't be bad for desktop mainstream parts (to be named R9 480 Series according to Bench-life).

Glad I haven't sold my 390x yet. I might not want to if your prediction is correct.

 

[maddie]

Expectations were high, and when the reality does not catch up to the chase, people will downplay the results.

I would not discard those results, nor take them seriously. Take everything with huge grain of salt.

What in the world are you trying to say?

If you're not taking them seriously, why keep them?

 

[thesmokingman]

http://videocardz.com/59468/amd-polaris-10-and-11-opengl-benchmarks-spotted#disqus_thread

Many thanks to 2355 for a tip.
Source: GFXBench #1 #2

LMAO, their sources are the unknown benches themselves.