• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Should NVidia drop SLI, has it had it's day ?

running a single nvidia card with a 4k screen, every setting is as high as possible, can't do SLI as would struggle with AIO water coolers for the card

actually never considered SLI for the pc :)
I should've added that i mean if one wants high consistent fps they would need to go SLI at 4K, of course you can fiddle settings to get acceptable performance with a single card but it's not for me at least. Really though it depends on the game, demanding titles don't run well at all.
 
Last edited:
SLI support has already been dropped, by me. And also by plenty of other people who got sick of buying 2 cards and getting the performance of one card in that latest game.

In actual fact, I had 2 copies of that latest game because the "SLI ready" card manufacturer gifted me a game that wasn't SLI ready with each card.

Nvidia can continue to "support" SLI for as long as they like, but I won't. If a game can't run at 60fps @ my current resolution on the fastest single card available then there can only be two solutions:

1 Drop down a resolution.
2 Don't buy the game.

As I game on a ROG Swift paired with a 980Ti... My choice will be option 2. I certainly won't be buying more 980Ti's to achieve the same frame rate, or even worse... Ghosting and flickering and juddering and whatnot.

I can't agree more.

I've been a long time user of SLI, (god knows how much money I've spent on graphics cards over the years...)

Historically it did seem to work quite well, but my take on it is that games have gotten so complex, are often built to run on multiple platforms, meaning PC can't get the dedicated focus it once did, as a result things like SLI just don't get the attention of the devs.

The end result is that right now - not a single game I play regularly, properly supports my 2x 980GTX's, I can sort of get SLI to work with varying degrees of success, by playing with inspector and forcing some version of AFR, but the reality is that all it does is cause both cards to be used, it doesn't improve the frame rate that much, I might gain 10fps - certainly not a whole second cards worth (£300-400...)


I think it should be dropped, unless you're benchmarking it seems to now be a massive waste of money, and I say that as a long time user of SLI.
 
I think those hoping that DX12 will fix all of their SLi problems are going to be disappointed. From what I've read the API exposes multiple GPUs in a much more sensible and usable way but it also completely shifts the development work from Nvidia and AMD to the game devs. So then the question becomes: who will put in the effort to support it? And can engines like Unreal implement it in a way that can be used generically across games that use it? Given the history of dev support for what is really a niche market I'm not optimistic.

And VRAM stacking is a pipe dream as far as I can see. Whilst theoretically possible the amount of work and potential for bugs is so massive I'd be very suprised if anyone ever released a game that supports it.
 
You've got Gregster :)

LMAO... sorry, but that was pretty funny :D (I think even Gregster would laught at this!)

I don't think SLI should be dropped, but I think there does indeed need to be a more common standard so that dev's are not having to split work. It should be almost native to all companies for multi-cards.
 
on the upside kaap, if sli and crossfire stopped you would save a furtune only having to buy just the one amd and nvidia card every year
 
What I'd like to see is a discount when buying multiple cards or a discount on a card if you can prove that you've already purchased one the same.

It's too expensive to be viable for most people with the straight multiplier on cost.
 
Back
Top Bottom