Sli > Crossfire?

[BubbySoup]

Yep, I realised that after I posted it.

I was going to do a sneaky ninja edit but got distracted by one of those new kit-kat bars, yum!

 

[Final8y]

thats just 1 review which is fairly old. more recent reviews at firingsquad, hardocp, anandtech, and guru3d show sli having much greater efficiency than that of crossfire.

it must also be a platform issue since that review you posted shows the crossfire setup using intel X38 chipset and the SLI setup using the nforce 680i chipset.

newer reviews use nvidias newer chipset for sli.

Indeed it is old but you post as if SLI has always been more sufficient than crossfire.
And only recently has sli caught up crossfire & beat it with cards that are faster in the first place.

 

[Cyber-Mav]

Indeed it is old but you post as if SLI has always been more sufficient than crossfire.
And only recently has sli caught up crossfire & beat it with cards that are faster in the first place.

well looking at the reviews sli has very good scaling, more so than crossfire. if just that the platform for sli is limited to nvidia chipsets and when using nvidias latest and greatest chipset as seen in the first post i made in this thread even 2 x 9600gt's using 780i nvidia mobo end up hammering crossfire 3870xt setup.

it would be really good if nvidia would open up to allow sli on intel chipsets, but then who would buy nvidia mobos then?

 

[Final8y]

well looking at the reviews sli has very good scaling, more so than crossfire. if just that the platform for sli is limited to nvidia chipsets and when using nvidias latest and greatest chipset as seen in the first post i made in this thread even 2 x 9600gt's using 780i nvidia mobo end up hammering crossfire 3870xt setup.

it would be really good if nvidia would open up to allow sli on intel chipsets, but then who would buy nvidia mobos then?

I will repeat That now sli looks like it has better scaling with cards that are all ready faster.
Where were your comments when the link i posted was new? & most other reviews at the time were giving the same results, how would have been able to make the statements your making now back then.

not most places, but most people on this forum say crossfire scales better but there is no proof out there to show that it does.

People say this because of the overall history of sli v crossfire & have shown proof has been presented for when it is the case that crossfire scales better than sli & it is the reverse of your statement with the Nvidia cards being faster than the ATI in single form at the time as well, but as i have said you must have been off the net for all the time that crossfire was scaling better in percentage terms than SLI & links were being posted at the time as proof or some one else was using your account.

 

[J.D]

Also Cyber-Mav you have to remember that 2x 3870XT's are just faster than the G80 GTX in some cases, drawing level in a few and ahead in a few also. So in that scenario you would have to take the performance gains separately and then match the percentage gained. If a single GTX is hitting 63FPS at 1600x1200 4xAA/16xAF/HDR and one 3870XT is hitting 40FPS and 2x are hitting 70FPS then obviously adding another GTX into the equation is surely going to make Crossfire look not as impressive.

I'd have to see percentage gains of SLI vs Crossfire from a recent review to even think about accepting your view. Get me some results that show me then I'll believe you. Until then I know of only what I've read and Crossfire did scale better than SLI. I'm in no way saying that Crossfire is better than SLI, only that an ATI card would get more FPS for the second card than Nvidia would. Obviously 2x 8800GT/GTS/GTX easily beats the crossfired cards but when you remember that a single G80 GTX is just a little bit weaker than 2x 3870XT's then that's why SLI looks so much better than Crossfire to some people I think.