Recent content by barcaman196835

You should test UE4 with Radeon GPU Profiler, and you will see that the engine will do nearly 500 cache flush before the first draw. With this bug you will loose 10-15% performance in general, but earlier there was no other profiler that allow them to see this problematic behavior. Rolando from...

I'm not familiar with UE4 nowadays. I needed a much faster engine, so don't use Unity/UE-like solutions. These are just too slow. Nvidia happily tell you how much you win. ;) I optimized my approach with these informations in mind. Sure it is much worse for a dev to always ask the IHV, but we...

I don't understand what are you talking about... DSBR is already implemented and working for a lot of titles. There is a driver mode for the developers to force the different rasterization forms, so we can compare the results. My engine saves a lot of bandwith (~10-15%) with DSBR, compared to...

Pay attention that Doom is also using the GCN shader extensions, so every GCN-based Radeon use the vendor specific fast shaders, while the other GPUs use the standard SPIR-V shaders.

Tessellation on it's own is not an issue. Sure it is purely overtessellated, but even with an x16 driver limit the effect is still slower than a TressFX 3 implementation, which is really sad if we count that TressFX calculate OIT. Without OIT, TressFX is still 10 times faster than Hairworks with...

With that license it is hard to modify it. You can't change the performance critical parts of the code. With this kind of restriction, open source don't give you a real advantage. But sure, we can see now why is this effect so slow. nVidia was much more open earlier. Something happened in the...

No it isn't. Most Gameworks effects are using standard functions, but the whole middleware is totally wrong for the users. The basic concept of Gameworks is to control the performance, so nVidia can deicide how fast an effect can run. They don't really care about harming AMD or the devs. The...

Guys, the problem is that a lot of devs are crying for these features for years. AMD just serve the needs, if Microsoft and Khronos didn't do it. In my opinion it is shame when an IHV solves the biggest programming problems before the API providers.

This is a little bit different. I didn't have much time to analyze these extensions, but there are only four functions that are standardized in SPIR-V: readFirstInvocationARB, readInvocationARB, gl_SubGroupInvocationARB, and ballotARB. All others are proprietary to AMDs SPIR-V extensions...

DX12 don't need a lot of invest in the driver side. The layer is very thin, and a lot of management done in the application itself or some universal OS layer/driver comes with Windows 10 (VMM/DXKG). So doing an "app ready" driver is not necessary. The IHV can change the shaders, optimize for...

Kyle! I don't see the problem here. Ashes of the Singularity is a game with an integrated benchmark, just like some other games. You can now download the upgrade on Steam, and play the actual game with the upgraded renderer. So while AMD mention benchmark it the driver notes, these basic...