On 10/7/2014 10:24 AM, ken wrote:> The calculations John makes are valid as far as they go, valid for a > screen with no applications/windows visible on it. Every time a window ... > In brief, a lot has to happen in addition to the simple rastorization of > the screen that John describes, and so a video subsystem requires many > times the amount of memory he suggests and number of discrete processor > instructions his description would imply. John is right. Laptops with integrated Intel video controllers have all the video processing power necessary for just about anything short of heavy gaming or real-time professional-level image processing. I run a virtual desktop infrastructure. A typical VDI host here will be running around 30 virtual desktops with dual displays at around 30% CPU utilization on 16 cores. Granted, you won't find those Xeons in any laptops, but those virtual desktops (running a full suite of business apps) AND their 60 displays are being emulated entirely in software, and then the display output is compressed and sent to the user over Ethernet, yet the users do not experience the dreadful video performance you describe being the result of not having discrete graphics. Also, each virtual desktop consumes about 3.5GB of memory (including its displays) on its host. The display memory isn't allocated directly as a number, but as a resolution. I seem to recall that it would take up to 128MB to support two 1920x1200 displays, however. -- -Chris