I’ve never used a 2080 or anything like that, I wanted to know:
If I have a game hosted on my machine at home, can I stream it to someone with -a low power machine- right from my house? (I have Fiber…)
But, the REAL question is: Can I do stuff on the machine itself, stuff which maybe uses the graphics card (eg. Blender EEVEE) AT THE SAME TIME WHILE THIS IS GOING ON?? (ie. the 2080 will be being used for 2 things simultaneously!!) Assume that the OS doesn’t crash trying to handle all this or anything…
Do you want him to actually play the game? Because that’s a pretty whole different topic then ‘streaming’ and might involve some serious violation of the terms of using the game in the first place. Let alone the problems you gonna have regarding input lag.
Or is this something you came up with? Then I think it would be easier to setup some web serving capabilities to your game.
Actually I was just using the term “game” generically - substitute “something made in a game engine (like godot or armory)” for that! 
(ie. I will be making it, so no question of any “terms violation” etc.)
Ok, then that isn’t your problem.
Have you tried running that ‘game’ for yourself while working in blender? That should give you a first indication.
It does seem like you’re putting your computer through a lot at once… it all really depend on how demanding the game and your blendering is. Both of these are kind of unknown quantities.
No no, I’m just curious to know whether the 2080 can even handle such things…?
Well we won’t know until you try.
We can’t tell you how demanding that ‘game’ is
It should work just fine. I have been really impressed (my experience is primarily on Windows) with just how sharable a resource the GPU is. You can have two different games running at the same time and things generally work fine. You can start two copies of Blender and have each one do a Cycles GPU render at the same time, and each render runs at about 50% of the speed it would have if it was the only GPU user.
So the GPU resource doesn’t generally seem to behave much differently than the CPU does in terms of running multiple tasks. I find this actually quite impressive because I could totally imagine that swapping between different tasks could be really inefficient in the GPU world.
As long as you have enough total resources for all the tasks (CPU, RAM, GPU cycles, VRAM) it should be good (and a 2080 with 11GB of memory should be easily able to handle two pretty serious tasks and likely more than that).
There could be applications that are poorly coded and eat up more GPU resources than necessary, but I haven’t found too many issues like this.
Ah THAT’S the info I was looking for - thanks so much, Pixelfox
1 Like