Answer the question
In order to leave comments, you need to log in
I wonder how two CPUs on the mother share the calculations during games?
Here are vidyuhi in a pair and in a trio in a specific way.
Separately physics separately texture? How about two CPUs on the mother then?
Well, when working with a database, you can simulate two PCs with two databases there. Divide and glue the results of the sample in half.
And how to divide the flow of instructions and data from different channels and ddr bars for different CPUs and cores?
It's not easy and wastes time and resources, doesn't it?
Is it different for amd and blue?
Answer the question
In order to leave comments, you need to log in
They don't share at all. Each application (game) thread is executed by one core. The operating system is responsible for distributing threads among the cores and switching them.
If the game is not able to work in multi-threaded mode, then in general only one processor core will be loaded.
Multiprocessor systems are necessary for optimizing operations of the same type, namely their parallel computation. This is necessary for servers, not all games support multi-core, let alone multi-processor ... Its development is not worth the blood and sweat of programmers.
It's easier to hang everything on a video card, now there are more people with paralleled video cards.
Multi-core, for databases.
No way. If the game is not designed for multiprocessing - it uses one processor. The second will be stupidly idle.
Everything works a little differently. During the creation of the program, the programmer determines which parts of the program should be executed in parallel. The level of this programming can be different, it depends on many factors. During the execution of the program, the parallelization of calculations occurs according to the code.
Didn't find what you were looking for?
Ask your questionAsk a Question
731 491 924 answers to any question