In this article we are going to talk about which model is the best right now to move video games, if not how said processor should be. This does not mean that we know more than those who design the processors. All this so that you understand that there are a series of decisions that lead to the design of PC processors, which do not make them perfect for all tasks. Although we can’t deny that it’s unclear what each of those leaving the factory will be used for and it would be a hassle to have a chip for a specific design.
What would the ideal processor for gaming look like?
First of all, we must clarify that we are not talking about a processor that does not work if we edit a Word document, but one that can work in all cases, but takes a different approach to designs conventional. We must therefore start by ruling out a series of elements on the entry table that will not be in our ideal processor for games.
Full cores and no multithreading
Multithreading does not double the power of the processor, but rather doubles the parts to fetch and decode the instructions so that in a period of stoppage, which is called an execution bubble, said core can approach 100% performance. However, this has a problem, it limits the maximum clock speed that can be achieved on Boost by enlarging the cores. The reason for this is that the voltage a CPU can achieve is inversely proportional to its area.
The other issue is communication, although they are not twice as comprehensive, they have internal communication like this. At some levels rings are used to communicate that data is rotating in one segment per clock cycle, the problem is that the more clients there is, the more latency there is between them. Which is to say, if today’s games need 16 execution threads in 8 cores, what better way to give them a full 16 cores with the ability to hit as high GHz as possible.
No integrated graphics card in our ideal processor for gaming
Nobody plays with an integrated graphics card in the PC and now there is an obsession to give them support tasks like video encoding in streams. For those who want to play a game with the highest fidelity, this part takes up space on the chip and generates heat, it is a useless element that adds nothing but inconvenience. Our brand new NVIDIA GeForce, Intel ARC or AMD Radeon are already good enough for all kinds of work. In addition, we do not want manufacturers to tie us in a draconian and monopolistic way to hardware ecosystems that they will abandon in a few years and that limit the way in which we are going to configure our PC.
DMA units directly connected to the graphics card
DMA units allow the graphics card’s GPU to access the main RAM to get the list of commands for the next frame. A more advanced version is the ReBAR which performs the reverse step. That is, it is the processor that writes to the graphics card memory using the direct memory access mechanism of the graphics card.
The idea we’re talking about goes much further, since the PCI Express controller today is inside the CPU, the goal would be to build the lists for the GPU into the CPU’s last level cache and send them directly via a dedicated channel to GPU VRAM. Obviously, we are not talking about graphic assets, just the list of instructions to do.
What would we gain from it? Well, several key milliseconds, because we would save the whole process of going to the cache to send the data to the graph looking for the memory address we want to write. Obviously, said DMA unit will have the ability to convert memory addresses into the GPU’s own space.
video game manual
In a general-purpose processor, able to move any type of instruction, these are built from other simpler ones. But what happens when a complex operation is used continuously? Well, that takes a lot more cycles than an implementation of this instruction from scratch.
The idea is not to add new support units, but instructions that are not important in other areas, here they are treated as first class citizens. Either by requiring fewer instruction cycles, avoiding resource access conflicts between them, or failing that, by creating new mechanisms for the same command, but making it work up to 10 times faster. What would that translate into? Well, obviously, since a video game is a program and is made up of instructions, it should run much faster.
Table of Contents