AMD Fury X "Fiji" Voltage Scaling

[Sasqui]

Nice work W1zz... as if we needed another reason to love TPU : )

 

[Farmer Boe]

It would be nice to see if there's any advantage going from the standard AMD closed loops to an EK full cover block in a nice loop. Time will tell.

 

[sweet]

Nice work. Could you please share your magic on voltage control

 

[mrwizard200]

Why test with BF3 and not BF4??

 

[Mathragh]

Why test with BF3 and not BF4??

Because that's the game thats being used for all the overclocking tests recently.

 

[Random Murderer]

Congrats, W1zz, on cracking the voltage adjustments and memory overclocking!

 

[the54thvoid]

Great work @W1zzard. Regardless of the results, this is nice work to see being done. Maybe some of the fucknuts that criticise your reviews can come to the front of the class now for spanking.

You went where techies boldly dare to go - into the great unknown of Fury X.

 

[dwade]

Conclusion: Buy a GTX 980 ti.

 

[buildzoid]

AMD cards don't like high core temps (voltage scaling is terrible above 70C core temp) also how hot did the VRM get?

 

[Breit]

how hot did the VRM get?

My version of the review says:

[...]
Once we reach +144 mV, which results in a scorching 1.35 V on the GPU, the maximum stable frequency reaches its peak. At this point the VRMs are running temperatures above 95°C, even though they are cooled by the watercooling loop via a nearby copper pipe. That much heat on the VRMs is definitely not good for long-term use. I would say a safe permanent voltage increase on an unmodded card is around 40 mV or so.
[...]

 

[2big2fail]

Reviews like this is why I keep coming back to TPU. Great job @W1zzard!

 

[el etro]

Great article. Shows AMD IC development needs to be enhanced to be competitive. Both Hawaii and Fiji is having problems with power and heat. Work for the Integrated Circuit Development Team do.


That slides of the 25x20 AMD inittiative shows AMD is working on this.

 

[yeeeeman]

Ok, so this is just the result of the ratio between CUs and ROPs. And I bet that the memory potential isn't even halfway explored. Well, a nice card anyway, but from a gaming standpoint, it can't beat a GTX980Ti. For compute, it is another story altogether.

 

[TheoneandonlyMrK]

Why test with BF3 and not BF4??

Good unanswered question there one game is piddly wizzard what gives time??

Also relative to this do you wizard think there is any chance of improvement with windows ten and newer drivers because this card was designed to make use of
Wddm2 etc? ?

 

[qubit]

Required accessories for overclocking Fury X:

 

[W1zzard]

Why test with BF3 and not BF4??

Because BF3 loads reasonably well, completes in a reasonable time and has very little random variance between runs.

https://www.reddit.com/r/AdvancedMi...ry_x_fiji_voltage_scaling_techpowerup/cti3dx3
It took me 4 hours to get those 3 results. Please understand that my time is not infinite.

 

[ampre]

How can you OC VRAM when Amd officially said it's not possible because it's hardware locked?

One game is also not enough to see if there is a limitation of fury x. Further benchmarks to identify the bottleneck of fury x are needed.

 

[Random Murderer]

How can you OC VRAM when Amd officially said it's not possible because it's hardware locked?

One game is also not enough to see if there is a limitation of fury x.

Because @W1zzard is a damn wizard, that's why.

Also, if you didn't just read his previous post, testing is extremely time-consuming. I recall him stating once that all the testing for a single GPU review can take him weeks.
Don't register just to question the owner of the site, especially when:
1) He is not required to be so thorough in his testing, or even to perform the tests for us
2) He is one of THE names in the GPU industry for he is the writer of GPU-z and Sapphire Trixx
3) He is far more experienced in doing what he does than you are, he has been running this site, doing reviews, and helping to modify video cards for over a decade.
Show some respect.

 

[Deleted member 138597]

Yeah, @W1zzard, do you think Fiji should do better in WDDM 2.0? Any plans to scour this possibility in August?

 

[CounterSpell]

congrats on the research!

 

[ampre]

Because @W1zzard is a damn wizard, that's why.

Also, if you didn't just read his previous post, testing is extremely time-consuming. I recall him stating once that all the testing for a single GPU review can take him weeks.
Don't register just to question the owner of the site, especially when:
1) He is not required to be so thorough in his testing, or even to perform the tests for us
2) He is one of THE names in the GPU industry for he is the writer of GPU-z and Sapphire Trixx
3) He is far more experienced in doing what he does than you are, he has been running this site, doing reviews, and helping to modify video cards for over a decade.
Show some respect.

Never trust a report which you have not faked by yourself!
When a high Manager of Amd said it's not possible to OC the VRAM then the question have to be asked what he have done!

 

[librin.so.1]

Never trust a report which you have not faked by yourself!
When a high Manager of Amd said it's not possible to OC the VRAM then the question have to be asked what he have done!

And if Nvidia says there's nothing with with the GTX 970, that's entirely true, right? RIGHT?

 

[ampre]

It's also a report which I didn't faked!

 

[W1zzard]

Never trust a report which you have not faked by yourself!
When a high Manager of Amd said it's not possible to OC the VRAM then the question have to be asked what he have done!

You can all soon buy a Fiji card and overvoltage/overclock it yourself. Software will be available soon

Yeah, @W1zzard, do you think Fiji should do better in WDDM 2.0? Any plans to scour this possibility in August?

I have no idea, we'll see.

 

[Krieger]

I'm just thankful that W1zzard made this possible. Let all of the other less capable reviewers do the thorough testing.
Some people just haven't been around long enough to know what's what.