Joined
Sep 6, 2004
Messages
6
Anyone here with an RX Vega and HTC Vive or Oculus want to fire up Nvidia VCAT VR or SteamVR performance tool?

Regards,
Taylor Ratliff
aka TaylorR137
 
I'm interested in this as well, as RTG hasn't been the best choice for a while. It would be nice to see a silver lining to the trash fire that is VEGA.
 
yep, spam searched quite a few times and nothing so far...
 
Here's what I got for my aircooled Vega (balanced profile, ~400w seen on killawatt, 17.8.2 drivers) running with R7 [email protected] and 3200mhz ram.
upload_2017-8-28_8-21-21.png
 
Here's what I got for my aircooled Vega (balanced profile, ~400w seen on killawatt, 17.8.2 drivers) running with R7 [email protected] and 3200mhz ram.
View attachment 34733

I was really hoping that Vega would deliver a bit more in the VR performance dept. Vega is "ok"... trailing a good bit behind Nvidia's last gen offering... and barely managing even half of what Nvidia's top tier cards can offer when it comes to overall VR performance. Granted Vega is pretty good bang for the buck at what it does manage to deliver, but for someone after top shelf VR performance (everything dialed up to max with supersampling) , Vega pales in comparison.

vega.png
SteamVR_test_1.jpg
SteamVR_test_2.jpg
 
I was really hoping that Vega would deliver a bit more in the VR performance dept. Vega is "ok"... trailing a good bit behind Nvidia's last gen offering... and barely managing even half of what Nvidia's top tier cards can offer when it comes to overall VR performance. Granted Vega is pretty good bang for the buck at what it does manage to deliver, but for someone after top shelf VR performance (everything dialed up to max with supersampling) , Vega pales in comparison.

View attachment 34739 View attachment 34735 View attachment 34737

Ahh that's too bad but not unexpected. I need resolution cranked to combat the crappy screen door effect.
 
Same setup, but power target +%50%, max fan profile (super loud) and ram @1000Mhz (core clocks is steady around 1650Mhz where balanced would be around 1400). Killawatt shows around 510w on average.
upload_2017-8-28_11-25-38.png
 
Thx for that but still, this drought of information sux...
 
Here's one from Babletechreviews. It's a VR comparison with the Vega 64 liquid and the GTX 1080 FE and GTX 1080 TI FE.
http://www.babeltechreviews.com/rx-vega-64-liquid-10-vr-games-vs-the-gtx-1080-gtx-1080-ti/

Looks like the GTX 1080 FE is 25% faster in VR then the RX Vega Liquid. I thought they would be about even. Looks like AMD has a lot of work to do on their VR drivers.

They have a lot of driver optimizations needed for Vega still. It's unreasonable to think Vega will be 25% off 1080FE for very long in VR. How does Fury X compare to Vega in VR.
 
They have a lot of driver optimizations needed for Vega still. It's unreasonable to think Vega will be 25% off 1080FE for very long in VR. How does Fury X compare to Vega in VR.

Current performance is pathetic - the $700 liquid cooled 64 can't even come close to a 1070. I was hoping my next card would be AMD, but considering it's barely faster than my 1060 at over 2x the cost, no way. That said the 480/580 can now trade blows with a 1060 in VR where a year ago they were nowhere near, so there could be some hope for driver improvements, though I'm not sure this architecture is different enough that there is a lot of opportunity for optimization left. It'd be completely foolish to get one hoping for driver improvements when a 1070 trounces all Vega versions for the same or less money, though.
 
Last edited:
Thanks for the benchmarks, I purchased a 1080ti a few weeks ago and was worried I'd chosen poorly and hasty. I feel better now, but sad at the same time. Was hoping AMD would have gotten this one right from the start, sadly no...
 
Here is one of my 2x 1080ti on Threadripper 1950x for comparison to Vega when people start posting them here.

VR_Perf_Test.png
 
http://www.tomshardware.com/reviews/radeon-rx-vega-56,5202-15.html This shows some interesting results, with both vega 56 and vega 64 ahead of the 1070 in some games and behind it in others, but below the 1080. What's interesting is that the 56 appears better than the 64 in one of the benchmarks and ties it in another (coincidence that they're both of the games that the vega beats the 1070?)...Not sure how that's possible unless the 64 is heavily throttling itself. It does point to driver issues as the Vega goes from beating the 1070 by a not inconsequential amount in some games to getting it's ass royally kicked in others. Given this, I'd have a tougher time picking between a vega56 and a 1070 at the same price as the vega56 appears to be more powerful but with drivers holding it back. Almost no reason for a vega64, though.
 
I get the same VR ready with my 980ti. I wish they would change it so it has the highest VR ready on a OC Titan, that way we could gauge better.
I would like to upgrade but I want to see what is going to come out.
 
I get the same VR ready with my 980ti. I wish they would change it so it has the highest VR ready on a OC Titan, that way we could gauge better.
I would like to upgrade but I want to see what is going to come out.

Well, you can still get a rough idea as to overall performance potential by the quantity of VR frames that were successfully processed by the test run. More = Better. But I agree, it would be nice if they expanded out the VR Ready gauge line to include zones for "More than Ready" and "Ready for VR Gen 2" :D
 
Back
Top