4 ms·
Wonder if it make sense to architect computers with a small sidecar CPU that is not as powerful but it also runs at ultra low power ... so tasks like these can
by albert_e 1y ago
Wonder if it make sense to architect computers with a small sidecar CPU that is not as powerful but it also runs at ultra low power ... so tasks like these can be delegated to it while allowing the main CPU to enter low power state when nothing else is placing demand on it.
- cwillu 1y agohttps://en.m.wikipedia.org/wiki/ARM_big.LITTLE https://en.m.wikipedia.org/wiki/ARM_big.LITTLE
- pbhjpbhj 1y agoThis link is to Arm's big.LITTLE, but other companies have efficiency cores (E-cores) too and may run fewer cores at lower clock to save power.
- phire 1y agoThe extra CPU usage isn't that bad. Updating a single number isn't that hard. The bigger problem is waking up the GPU and all the communication between components, which is why the computer with integrated graphics takes a smaller hit than the one with dedicated graphics. And why the ARM laptop did even better, because they were optimised for this usecase.
- userbinator 1y agoI'm not sure if the GPU is even needed for an operation like this. Basically modifying a few dozen bytes in the framebuffer every second. It would be interesting to disable all graphical acceleration, causing the CPU to do all the work, and see what difference that makes to the test.
- o11c 1y agoWindowing systems these days tend to spam the GPU in the name of performance, unfortunately.
- jpc0 1y agoI would be much more interested in having the ability to inform both the gpu and cpu to stay in low power mode while performing this operation. I does not need to be scaling up to high performance to: - Read a piece of memory - Increment with one and modulo - Display a section of a texture on a section of the screen. You will very likely spend more time passing memory around than otherwise, and to be honest if it happens every second I would hope it stays in cache so you wouldn’t ever even bother the memory.
- phire 1y agoYeah, some mobile GPUs actually have hardware compositors that can do. They can even support moving windows around with pretty low overhead. But the software support to take advantage of it isn't really there. There isn't a standard API to access such functionality, and so the hardware compositors end up unused, so the don't really put much effort into improving them. But with proper software support and a hardware compositor with enough flexibility, you could easily put the clock in it's own texture and update it with very low power consumption. Actually, Desktop GPUs already have a single hardware sprite that gets used for moving the mouse cursor around with very little overhead (and lower latency).
- jeroenhd 1y agoIntel already has that in the form of E-cores, and ARM has had big.LITTLE for quite a while now. I don't think any Microsoft employee is going to spend a day writing CPU scheduler code to make sure the seconds display in the task bar is being sent to the right low-power cores. I'm surprised they even bothered to port that from Windows 10 to Windows 11 to be honest.