4 ms·
AFAIK, Windows Remote Desktop always renders through virtual display adapter devices, though it can optionally make use of GPU resources to accelerate rendering
by jasomill 2y ago
AFAIK, Windows Remote Desktop always renders through virtual display adapter devices, though it can optionally make use of GPU resources to accelerate rendering to these devices.
I just tested in an ESXi VM[1], and Windows 10 will happily boot and accept an RDP connection without any (virtual) display hardware at all:
https://jasomill.at/HeadlessRDP.png https://jasomill.at/HeadlessRDP.png
Whether a particular computer will boot Windows without display hardware depends on the firmware — in my experience, some workstations will, some won't, and I honestly haven't seen a modern x86 server without some sort of integrated graphics to support out-of-band management, often with a VGA port as the only physical video output.
[1] By hand-editing the VM's .vmx configuration file, setting
svga.present = "FALSE"
and adding the undocumented option
hwm.svga.notAlwaysPresent = "TRUE"
Interesting aside: I figured this out by running "strings" on the hypervisor executable, which turned up this suspiciously Scheme-like bit of configuration code:
https://jasomill.at/vmx.ss.html#L-808 https://jasomill.at/vmx.ss.html#L-808
In fact, two definitions and a couple trivial syntax fixes are enough to get Chez Scheme to evaluate it without warnings:
$ awk -f - vmx.ss <<EOF | scheme -q
BEGIN {
print "(define / fxdiv)"
print "(define (log . args) (for-each display args))"
}
{
gsub(/\(\)/, "'()")
gsub(/ 0x/, " #x")
print
}
EOF
$
So it appears the VMware hypervisor contains an interpreter for a very Scheme-like dialect of Lisp.
- erik_seaberg 2y agoCan a backend app use a single GPU (integrated or not) to support many concurrent users, so that software rendering doesn't become a bottleneck?