No, we haven’t gone crazy and that’s it it would be possible to make graphics cards without their own memory. After all, they are the most used on the market and we do not just say it, since the one integrated into the processor of your PC complies with said specification. However, we are talking about releasing them as expansion cards. Well, in this article you will learn why nobody makes them and they are not for sale.
The reason why different components have their own memory is to avoid conflicts and fights to access them with the CPU and because they work much better with their own memory. This is the case with graphics cards. However, today we bring you a concept that may seem strange to many, but it’s not that crazy, given that it has happened in many systems in history, why not on PC? Well, we explain it to you.
Graphics cards without memory?
Well, even if it seems impossible to you, it would be possible to make them, however, they would not have the same performance as units designed for gaming because they would use system RAM for this, which they would access through the PCI-Express port. That is to say, they couldn’t be better than an integrated graphics card, but the good thing is that we could see them as an M.2 expansion card thanks to it. Why don’t they see each other? Well, due to the fact that by not achieving more power than an iGPU, well, they’re still redundant and stupid.
However, it is possible to do it and all this is due to a peculiarity and that is that GPUs are used to working with two different memories:
- One is local memory accessed through the GPU’s built-in memory controller and is usually found in video memory, ie what we call VRAM.
- The other is system RAM, which the GPU accesses through its DMA units.
Now, many of you will wonder if this is not counterproductive due to the latency due to the distance between the two parties. Well, we have to think that GPUs have the ability to hide latency to some degree. Unsurprisingly, even with the video memory itself, we can find access times between 5 and 10 times compared to the usual RAM and which are not higher than when the card itself accesses its own memory. .
We’ll see it in the future, but not the way you think
With the advent of disaggregated chips, we’ll see how integrated GPUs in processors will separate from the main chip. We get a glimpse of this with the Ryzen 7000 desktop, where we have a small on-chip graphics card which is the IOD, but we’ll see the separation for the first time in Meteor Lake, where although it’s not possible to speak of a graphics card, extension, it will be the first time that we will have a separate GPU from the main unit without its own memory.
However, we will not see these units in M.2 units and that is that we have left one item for last to complete our argument and that is that the power delivered to said expansion cards is not greater than 10 W, while in the frame of the processor can receive much higher power. So in the end, we would not only be limited in the capabilities of an iGPU by memory bandwidth, but also in terms of clock speed due to low power consumption.
And that’s why there are no graphics cards without memory, because the performance would be low enough to be useful in any case and would therefore be a waste of time and resources. Moreover, the input range is already absorbed by the integrated graphics and gradually disappears.
