CPU is much faster than GPU: for everyone who has experienced the problem, crashes occur during OpenGL or DirectX rendering. Well, if I may do so, remember that it can’t be the same for everyone. The Question When CPU is faster than GPU? is quite common.
There are some rare conditions where the performance of your graphics card isn’t the problem. A bit long time ago I had such a failure, but I’ve solved it and my question is: if the CPU is so much faster than GPU, why I was not using e.g. +ES_Thread , which gives 20x+ performance to my PC compared to all hardware I have today, and why I didn’t try ECL in the past instead of -kASM , which gives the same speed as -kASM , but less scheduling overhead? By the way, -dynamic-memory limits make it slower, but better supported.
I am using a Thinkpad x200 with i5-3370 CPU and 16 GB RAM, I have GTX 680. After I installed multi-monitor applications I always crash at rendering each frame because some OpenGL or DirectX window managers with horrible interface crash and it takes forever to bring back my application, but all windows look good.
When CPU is faster than GPU?
I saw that render crashes some seconds after the start of the frame so I tried a similar solution to you, kforce2, that I found in your answer: Kill the application on exit and do a restart. It takes less than 3 seconds to load all graphics and renders frames.
Actually, I didn’t like kforce2 because it crashed before I finished rendering a frame (just one). Then I tried the solution recommended by pcgamer.com. I replaced my hardware-accelerated drivers with the vanilla OpenGL and implemented the solution recommended by pcgamer.com.
When I tried it with high resolutions I had some frame stalls at some low resolutions. This has solved it for me. The last screenshot is from a comparison between -KASM (Mac only) and ioshort and ioshort+ouch, which is much better supported and faster. When CPU is faster than GPU?
PC-gaming accounts for a large portion of my time. Therefore I prefer higher performance if possible. You say that some applications might benefit from xrender and this solution, but I can’t find any that really take advantage of the speed that CPU provides (actually, if you consider the fact that most CPU-intensive applications won’t scale well with GPU acceleration but will need a good CPU, this can be a problem).
I’m having some performance problems on my MacBook Pro 2011 with Intel Core i7-3770 CPU and 8 GB RAM (i7-2600k CPU), if you find a solution for this problem, please share it with me.
Are you using emulated hardware or an emulation layer in your 3D software? If so, that’s fine, but if not, are you limiting your frame rate with the game itself, or is there some other source of video memory allocation which seems to have been taxing your GPU? I try to explain to you When CPU is faster than GPU?
Can you take a step back and think about your problem from a different perspective, then you will notice that you can’t slow down the application with OpenGL or DirectX. It’s the CPU that is bottlenecked.
When I played Crysis 2 on my system I got 20-30 fps with an open-source OpenGL Game, but I saw nothing more than 2-5 fps when I played Crysis 2 at 1080p. On the other hand, when I played Battlefield 3 at 1080p I saw much better performance with an emulation layer.
With Battlefield 3 I saw 75-80 fps even at max settings, and I got 60 fps with the open-source game. Now, When CPU is faster than GPU? is not difficult for you.
I think you are using -kASM under Windows, while you should have a -dynamic-memory-limits in a cross-platform application. With ioshort and ioshort+ouch you should be able to set a speed limit of -dynamic-memory-limits and adjust it with the CPU usage. Just be sure to set the default value with kforce2 as described above.Hope You know the answer Now: When CPU is faster than GPU?
Also read: Digital Marketing Services