GTA 5 beta 0.271

[Portal2x2]

Version updated, download again
Increased performance for those users who see huge fps drop (AMD videocards and/or high drawing distance in advanced graphics options).


Portal2x2
If you don't see the difference, then just don't need it. Personally i can't look at ugly flickering shadows and areas where antialiasing not applied at all.

Boris tell what mode MSAA should stand to see it?

[JF049]

Increased performance for those users who see huge fps drop (AMD videocards and/or high drawing distance in advanced graphics options).

Still the same for me. I don't feel uncomfortable with lower fps, so I personally don't really need this issue to be fixed (right now, at least).

[ENBSeries]

JF049
Sorry, i decreased buffers updates from 3000 in the worse scenario (highest draw distance and view of entire city) to around 20. There is no single reason to have performance loss which i know. When UseEffect=false, it do not modify only per object data and keep executing various other things like reading hotkeys, updating global data, timings, etc. So, your rig or just AMD videocard/driver heavily stuck with some of few drawed and modified objects. This can be detected only by making several tests where code is partially disabled.

Portal2x2
I use MSAAx2 for screenshots, but any enabled should work.

[robi29]

I agree. Still have performance drop. And that's OK with me, because I don't feel fps drop.

Of course the problem doesn't occur everywhere, only in some places.

[ENBSeries]

Okay, here is the test attached. Modify F1 temporary variable to disable some processing. Values 2, 3, 4 are triggers. Post results if any of these values have fps impact and how much.

EDIT: file deleted

[JF049]

Okay, here is the test attached. Modify F1 temporary variable to disable some processing. Values 2, 3, 4 are triggers. Post results if any of these values have fps impact and how much.

Downloaded, will post the results in 30 minutes.

[Pfuscher]

Is this compatible with reshade/sweetfx? Your awesome msaa would be really nice with smaa.

[JF049]

Tested in this area where Michael lives:




APPLY CHANGES was pressed with every variable change.
Var1 Var2 Var3 Var4 MaxFPS
0 0 0 0 49
0 0 0 1 49
0 0 1 0 49
0 0 1 1 49
0 1 0 0 49
0 1 0 1 49
0 1 1 0 49
0 1 1 1 49
1 0 0 0 49
1 0 0 1 49
1 0 1 0 49
1 0 1 1 49
1 1 0 0 49
1 1 0 1 49
1 1 1 0 49
1 1 1 1 49
No effects - 61

I'll drive around the city later with the same test if it is really needed.

[robi29]

Still the same. I just tested and any values didn't affect fps. Sixteen combinations and any of combinations doesn't have impact to fps.

[ENBSeries]

It's wrong. Variable 1 must be set to 1, 2, 3, 4 for testing, not variables 1-4.