It's usually the graphics processing unit that limits performance, causing noticeable drops in frame rate.
It's usually the graphics processing unit that limits performance, causing noticeable drops in frame rate.
The screenshot shows Shadow Warrior 2. Many of my PC games struggle with low frames per second at 1% difficulty. For my current setup, I chose an affordable i5 CPU with low power consumption and a small cache size. If I had picked a more powerful i7-8700, offering higher cores and threads along with faster turbo speeds, would the experience be smoother? While CPU speed is important, several factors influence performance—core count, thread count, turbo capabilities, cache size, and even memory bus width all play a role.
Bubbles and stalls. Some piece of code waiting on another thread that is waiting on another thread that is waiting on another thread. Its a fundamental issue with parallization and multitasking. Some background task demanded attention, CPU gave it some it stalled a thread, but a different thread who needed the first thread to synch up with it before going back off doing its own thing is now stalled out to. Sometimes its a misspredited branch. Your code path has an IF (x){ do this } else do this other thing It does "this" at the same time it was evaluating X betting on X being true. X was not infact true and now it has to go and do other thing. and throw out all the data. It could be it needed data that it just threw out of cache so now its stalled waiting for going to the next cache or ram, or storage. What exactly is causing your specific stutter is different almost every time. Most of the time its due to how the game is written. sometimes its not. Often times using a more powerful CPU is just brute forcing it and ignoring the underlying reason because people just be like, lol moores law, its the computer engineer's problem now. Yes, a 4.6ghz CPU vs 3.3ghz with all the same microarchitecture will be significant smoother, when threads need to synch up, there is less latency just due to pure clock speed.
The optimal settings vary based on the game engine and the scene being rendered. To maintain a steady 60 fps, each frame must finish within about 16.6 ms. This covers CPU processing time and data transfer to the GPU, as well as the GPU's rendering workload. Sudden events like new enemies appearing can temporarily raise CPU demands, especially for loading textures and decompressing data. The quicker the CPU handles these tasks, the less likely you'll notice frame rate drops. Multi-threaded engines benefit from extra cores since texture loading and decompression can run in parallel. If the engine isn't multi-threaded or resources are fully used, prioritizing faster cores becomes more important. Caching also helps by reducing reliance on slower system memory. Ultimately, it depends heavily on the specific engine and game behavior.
The importance of 1% lows varies widely depending on the situation. Different titles prioritize various aspects—some favor many cores/threads, others focus on speed, cache size, memory latency, or bandwidth. While every game has its needs, no single factor dominates universally. For example, a CPU with strong frequency might not always improve 1% low performance in certain games like CS:GO. The 7800X3D benefits from large L3 cache in most cases but doesn’t always boost averages in others. Ultimately, what matters most is balancing these elements according to the specific game and conditions.