- Joined
- May 18, 1997
- Messages
- 54,860
Spitballing Vega GPU Performance
This is an interesting article over at Tech Report that is "speculating" on AMD's new Vega GPU and its performance. Now keep in mind that Scott Wasson, the owner of Tech Report, works for AMD. So if I am going to jump on the rumor train with any site, it is probably going to be Tech Report.
Either way, none of these dart throws suggest the eventual RX Vega will have what it takes to unseat the GeForce GTX 1080 Ti atop the consumer graphics-performance race, as some wild rumors have postulated recently. I'm willing to be surprised, though. We also can't account for the potential performance improvements from Vega's new primitive shader support or its tile-based Draw Stream Binning Rasterizer, both of which could mitigate some of these theoretical shortcomings somewhat.
My sources in the industry have indicated a couple of different positions on how they think Vega will stack up against NVIDIA. My most trusted sources seem to think that it will edge out GTX 1080 by a slight margin, and I have a couple that think Vega will fall just slightly short of GTX 1080. What I am surely not getting from any sources is that Vega is a "1080 Ti killer." Wherever Vega lands, we certainly need AMD's Radeon Technology Group back in the mix at the high end even if it is not the ultra-high end.
What we do know from AMD's past, is that its cards have done great in some applications like bitcoin mining, and even trounced NVIDIA when it came to comparative products. And while I was watching Raja demo Vega running DeepBench and being successful against NVIDIA in that demo, it did not give me huge warm and fuzzies about Vega and gaming. As always though, the proof will be in the pudding.
This is an interesting article over at Tech Report that is "speculating" on AMD's new Vega GPU and its performance. Now keep in mind that Scott Wasson, the owner of Tech Report, works for AMD. So if I am going to jump on the rumor train with any site, it is probably going to be Tech Report.
Either way, none of these dart throws suggest the eventual RX Vega will have what it takes to unseat the GeForce GTX 1080 Ti atop the consumer graphics-performance race, as some wild rumors have postulated recently. I'm willing to be surprised, though. We also can't account for the potential performance improvements from Vega's new primitive shader support or its tile-based Draw Stream Binning Rasterizer, both of which could mitigate some of these theoretical shortcomings somewhat.
My sources in the industry have indicated a couple of different positions on how they think Vega will stack up against NVIDIA. My most trusted sources seem to think that it will edge out GTX 1080 by a slight margin, and I have a couple that think Vega will fall just slightly short of GTX 1080. What I am surely not getting from any sources is that Vega is a "1080 Ti killer." Wherever Vega lands, we certainly need AMD's Radeon Technology Group back in the mix at the high end even if it is not the ultra-high end.
What we do know from AMD's past, is that its cards have done great in some applications like bitcoin mining, and even trounced NVIDIA when it came to comparative products. And while I was watching Raja demo Vega running DeepBench and being successful against NVIDIA in that demo, it did not give me huge warm and fuzzies about Vega and gaming. As always though, the proof will be in the pudding.