The vast majority of CPUs today are heterogeneous chips that contain not only multiple CPU cores, but also what we call an iGPU or integrated graphics chip. We have always interpreted this component as a piece less easy to move a little more than the Windows desktop, a kind of replacement that as soon as it is time to do the heavy lifting, that is to say to generate the graphics in games, it passes over to the main graphics card to take a vacation. However, this is a huge waste of resources.
Why will the integrated GPU become important in games?
The main job of a graphics card is to generate images, however, for some time they have been used for other different tasks, but in different markets than the PC. We don’t have our Radeon or our GeForce to help synthesize a protein or to create a military simulation. Of course, there are a number of things that excess GPU power, caused by downtime, is used to solve common problems in games, especially those related to collision detection or calculating the game physics.
However, to perform these tasks, many games don’t use the graphics card and let the extra CPU cores do the work, leading to one of the things that allows the flexibility of contemporary GPUs to be obsolete. The reason for this is that an NVIDIA graphics card is not programmed the same way as an AMD card, and instead Intel and AMD processors use the same code.
The concept of preframe
One of the most important calculations in games is that of visibility, that is, knowing which objects are actually seen in each frame and which are not. This is done by the main GPU when generating the frame, however, for some time now the concept of pre-rendering has been used, which is not to generate a frame with all the complete information, but to generate the image as simply as possible as long as we know the objects visible in the next frame.
This is easy to implement simply because the driver only has to run the next frame in a cropped fashion to get the corresponding information and the graphics card acts accordingly when it then renders the final frame. Think of it as a sketch that a draftsman makes with few details before making the final drawing.
What is its use?
Well, there are several of them and we will list them below:
- This allows us to completely remove all geometry that is in the display list, but is superfluous for the following reasons:
- off camera
- behind a larger object
- Too far and therefore the details would not be differentiated.
- It allows the creation of a tree-like data structure that tells us where each object is, which is essential for Ray Tracing.
- It allows us to order the geometry according to its position on the screen, which facilitates the possibility of rendering by tiles and thus reduces the impact on the video memory.
How is the preframe constructed?
This is to create the next frame with less information, since it will not be seen on the screen, some parts of the 3D pipeline are omitted in order to speed up the construction of the pre-frame.
- No graphic shader is applied, in any of the steps.
- No texture or color information is applied either.
- No image post-processing effects of any kind are applied either.
The idea is to be able to have it ready in a few milliseconds, with the aim of having the information ready for the GPU to use in its favor to generate the final frame. Remember that the purpose is to indicate the visibility and positioning of the geometry in the scene, essential information to generate an image. And this is where the trap behind this whole mechanism comes in.
It is in the steps after the rasterization step that the most data ends up moving and therefore the ones that require the most power, by removing them from the equation it means that we can move the execution of the preframe to much simpler graphics hardware, like as it’s graphics integrated into the processor, so it has to be active in order to use it.
Complications related to its implementation
Alright, everything we’ve discussed so far looks great on paper, but now it’s time to talk about the raw part, and we’ll only see this solution when the integrated GPU architecture and graphics of the card will be identical or at least compatible. This, unfortunately, is already a problem for NVIDIA, as it’s the only one of the three brands that doesn’t have an x86 processor. Moreover, AMD would be the only company that could take advantage of this situation to force the use of its Ryzen processors and Radeon graphics cards in the future.
On the other hand, the idea of the preframe is great for GPUs with a large number of cores, however, with the gradual increase in chip build costs, any outside help is welcome, even if it goes through the board integrated graphics. in the processor. After all, prices cannot go up continuously, there will come a time when the market says enough is enough.