Comment on Huang declares Israel Nvidia’s “second home” with record-breaking campus investment
WhyJiffie@sh.itjust.works 1 week agojust wait until you know what your graphics card could actually do. live OCR on screen contents, face detection and training on anything ever displayed on the screen.
unless I miss something then also direct access to all system memory (when SR-IOV is disabled, or has been set up improperly), write access to onboard firmware, probably access to your drives and network too.
same applies to any pci express device you put into your machine.
hendrik@palaver.p3x.de 1 week ago
Sure. I’m not entirely sure how PCIE works these days. But in it good old days we had methods to read pretty much arbitrary memory regions via PCIE or early Thunderbolt(?).
I just figured it’d be massively complicated to wait for the user to pull something on the screen, do computationally expensive OCR, some AI image detection to puzzle documents back together, and then you’d only get a fraction of what’s really stored on the computer and you’d still need a way to send that information home… When you could just pick a plethora of easy options like read all the files from the harddisk and send just them somewhere. I think it’s far more likely they do some easy and straightforward solution.
WhyJiffie@sh.itjust.works 1 week ago
I think that’s easier to do for GPUs, and they could do it nonstop with say a 5 fps rate.
but also now that I think of it more, the graphics card doesn’t really need to do all this, the manufacturer given binary blob kernel driver has easier access to disk and network