If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Yeah, I know it's a configuration that's pretty far out there. However, for users like me that need the extra GPU memory there just isn't much of a choice at the moment.
I get what you saying (so incredibly angrily), but for me $4000 for a card that performs similar to an 1080 isn't worth it. Not to sure how much benefit you get over a 1080 with viewport performance either... The extra RAM on board is nice to have.
it is stupid, the 1080 should have had more vram too, atleast 12 gigs, then it would really kick the quadro cards ass so for now only benefit is bigger scenes but it looks like the 1080 are the top dogs, i bought the zotac versions which even more kicks ass
From what I understood, is that with the new architecture two 1080's can share each others memory, thus doubling your VRAM with two cards. Think this is something that Karba would need to implement in the coding using Cuda 8.0... i think.
Totally unrelated with benchmark render performance.
I hope someone can help me here.
I now run one gtx 980 4gb vram.
and it sometimes runs really low in memory, since windows 10 also use up the vram, leaving 2,3gb available memory for rendering.
My plan was like getting titan x 12gb vram for render and still using my gtx 980 for say, windows 10 display etc.
so when i am doing the render, it will have most of the 12gb vram available.
however is this possible?
I did my research around and some says the graphic card driver will have some problem, by having 2 different nvidia cards. (like different architecture etc, pascal, maxwell). some says its possible to run both of the graphic cards but you must use 2 monitor to separate em, so lets say 1 monitor for gtx 980 another for titan, and I gotta use the titan monitor to actually do render. Because to be able to use the graphic card I need to plug it to the monitor? It doesnt really make sense for me. I'm confused now.
anyway when i saw on this thread, some people like rob and karba run gtx 980ti and titan for Fstorm.
when using both, for render,
1. will fstorm use the lowest memory available? say gtx 980 has 4gb, titan has 12 gb. Fstorm available gpu memory 4gb?
2. to be able to use both cards, does the cards needs to be able to run 2 way-SLI?
or is it as easy as plug the titan x in, install the driver and its done?
Really love Fstorm but my GPU memory is holding me back.
Please enlighten me.
Thanks very much.
Best,
k
ps: I use Asus maximus hero 8 alpha for MOBO with 750 watt psu if that affect anything.
1 If using both cards, you are limited by the card with lowest mem.
2 SLI is not relevant, just put both cards in the machine, install drivers and of you go
That's weird I posted my results here a few days ago and it disappeared... I am trying again:
Did some becnhmarks before and after ugrading drivers, and then before and after upgrading Fstorm (0.2.1b to 0.2.4) and then tried some overclocking with my Zotac Amp Extreme 1080. Also compared max 2016 and max 2017.
That's weird I posted my results here a few days ago and it disappeared... I am trying again:
Did some becnhmarks before and after ugrading drivers, and then before and after upgrading Fstorm (0.2.1b to 0.2.4) and then tried some overclocking with my Zotac Amp Extreme 1080. Also compared max 2016 and max 2017.
Comment