ASUS RX 480 STRIX OC 8 GB

[mrthanhnguyen]

In all fairness, in the reviews the OC numbers are all much lower what you can pull off at home. On the 390/390X cards they usually listed 50-80MHZ OC, when in reality you could easily OC them a good 150MHZ.

Why dont you buy the same card when its available and OC it the way you have claimed and show the result here? My gtx 670 TOp and the msi gtx 970 OC capacity are very close to Wizard review.

 

[W1zzard]

Oh boy this is bad .. looks like perf summary has been wrong all along since we introduced the new graphs

 

[dj-electric]

oh... shit.
This is serious. so many people count on these as a reference

Have fun fixing it w1z. :X

 

[W1zzard]

oh... shit.
This is serious. so many people count on these as a reference

Have fun fixing it w1z. :X

Internet will explode

 

[ShurikN]

Oh boy this is bad .. looks like perf summary has been wrong all along since we introduced the new graphs

When was that?

 

[W1zzard]

New graphs are up in this review

 

[the54thvoid]

Oh boy this is bad .. looks like perf summary has been wrong all along since we introduced the new graphs

How far back?

And whoopsy!

 

[CAT-THE-FIFTH]

I got 5.65% faster for the FE GTX1060(when I manually added up the FPS) at 1080P and the new graphs say 8% or are there new numbers for some of the games??

 

[W1zzard]

Come here to check and discuss my math: https://www.techpowerup.com/forums/threads/tpu-math-check.224326/

 

[sutyi]

When was that?

If look careful enough you can find some on ancient cave walls.

 

[Mtom]

Oh boy this is bad .. looks like perf summary has been wrong all along since we introduced the new graphs

Oh boy, there goes some overtime

 

[tajoh111]

I added up the FPS at 1080P for both cards and came up with the following:
1.)Asus RX480 1110.6 FPS
2.)FE GTX1060 1173.3 FPS

That would make the FE GTX1060 5.65% faster.

Also if you look at the TPU and computerbase.de images of the cooler,it looks the same as the GTX1080 one Asus use and it only really two heatpipes make decent contact out of the 5. The other two make periphery contact,so it looks a bit of a bodge job by Asus!

I re added the numbers twice and although your 1060 numbers are correct. Using the rx480 numbers from the 1060 review, the total FPS is 1074.8. I think you might have added the 900p number during your calculation somewhere.

This makes the FE edition 9.16% faster than a rx480 and this is without WOW. With wow, its 12.5% faster.

 

[W1zzard]

I re added the numbers twice and although your 1060 numbers are correct. Using the rx480 numbers from the 1060 review, the total FPS is 1074.8. I think you might have added the 900p number during your calculation somewhere.

This makes the FE edition 9.16% faster than a rx480 and this is without WOW. With wow, its 12.5% faster.

Don't add FPS numbers, you'll weigh higher performing benchmarks more than low fps ones. See my math thread here for how I do it: https://www.techpowerup.com/forums/threads/tpu-math-check.224326/

 

[Shatun_Bear]

W1zzard you really, really need to add DOOM to your game testing suite. And you really need to consider swapping out some games that are now either very old or quite irrelevant these days that you are currently using.

There seems to be little point in having BF4 AND BF3, which is even older than an old game. Surely BF3 can be laid to rest in your testing suite in favour of DOOM, and preferably tested in Vulkan. This would give your benches more legitimacy.

 

[W1zzard]

W1zzard you really, really need to add DOOM to your game testing suite. And you really need to consider swapping out some games that are now either very old or quite irrelevant these days that you are currently using.

There seems to be little point in having BF4 AND BF3, which is even older than an old game. Surely BF3 can be laid to rest in your testing suite in favour of DOOM, and preferably tested in Vulkan. This would give your benches more legitimacy.

Yes, I'll add Doom on next full rebench.

I agree BF3 is old. BF3 is included for OC stability testing, best and most reliable stability test for me. I have to get numbers for all cards anyway so I can show them in the OC gains chart, so I thought including them in the main review too could be useful for some people.

 

[rtwjunkie]

BF3, which is even older than an old game

I understand your point, but only someone extremely young would label BF3 "old"

Try hitting middle age, and you'll see that quite a few years need to pass before something is actually old. BF3 is still quite a technically good and impressive looking game. So not actually old.

But I understand your point about wanting a new API game in the testing.

 

[CAT-THE-FIFTH]

I re added the numbers twice and although your 1060 numbers are correct. Using the rx480 numbers from the 1060 review, the total FPS is 1074.8. I think you might have added the 900p number during your calculation somewhere.

This makes the FE edition 9.16% faster than a rx480 and this is without WOW. With wow, its 12.5% faster.

Nope,I added the numbers for both correctly.

They are both at 1080P.

Edit!!

I am looking at this review,and you are looking at the reference model,so that is why we have difference scores.

 

[Shatun_Bear]

Yes, I'll add Doom on next full rebench.

BF3 is included for OC stability testing, best and most reliable stability test for me. I have to get numbers for all cards anyway so I can show them in the OC gains chart, so I thought including them in the main review too could be useful for some people.

Thanks for the response I appreciate your hard work with the benches. Looking forward to seeing Doom added in there. I also appreciate you only have the time and resources to bench only so many games.

I understand your point, but only someone extremely young would label BF3 "old"

Try hitting middle age, and you'll see that quite a few years need to pass before something is actually old. BF3 is still quite a technically good and impressive looking game. So not actually old.

But I understand your point about wanting a new API game in the testing.

Lol ok I'm nearly 32.

BF3 is an old game compared to the rest of the titles in the line-up. But sure, if you want to play semantics of what 'old' is and place it in a line of games that were released in the last 30 years, it would be newer than it would be old, but that would be slightly ridiculous. By normal standards, most gamers today would call BF3 'old' in the context of benchmarks when BF4, BF:HL and soon BF1 have succeeded it.

 

[Moofachuka]

Yo we need an actuary here =P

 

[G33k2Fr34k]

A couple of problems with Polaris 10 and AMD chips in general:
1- Too much FP64 resources which adds to the power ratings of the cards and wastes die area.
2- Their DX11 driver overhead is quite bad.

On a side note, it seems that Polaris 10 has much better Tessellation throughput than the previous AMD cards:

 

[arnold_al_qadr]

definitely will save some money and buy gtx 1060..

 

[LAN_deRf_HA]

So it looks like aftermarket cards aren't making the huge overclocking difference people were hoping for. Temps are good but that noise is oddly high for a 2016 Strix card.

 

[Fluffmeister]

Yeah solid enough card, but not quite the silver bullet many had hoped for.

@W1zzard are you likely to have the equivalent 1060 Strix in the labs at some stage?

 

[las]

Uhm, very noisy and the OC potential seems terrible on these chips so far? 3% GPU OC?!?!

 

[W1zzard]

Yeah solid enough card, but not quite the silver bullet many had hoped for.

@W1zzard are you likely to have the equivalent 1060 Strix in the labs at some stage?

Yes I have the 1060 STRIX here