• 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.

Core i7 Multi-GPU SLI/Crossfire Game performance review

Fortunately the cheapest i7 doesn't look too expensive. Its the platform around it that costs a bomb.

But that is still a major factor though surely? You can't have the cheap(ish) cpus without the expensive platform around it to run it.;)

And although nice seeing it benefits with dual and tri SLI I still think its only going to be for the rich people out there who don't care about value for money but just to have the fastest system possible.

A comparative i7 system is £300 more than a 775 setup. For that £300 you can have a 2nd graphics card or watercooling on everything.

So for the same money you can have an xfire or sli 775 rig which will outperform the i7 rig with a single graphics card.

Of course, I have noticed some lucky people on here already run GTX280 in SLI so there's Intel's market but its not a big one really?

And it amazes me that people like Strife are excited about the gains in not been cpu limited yet are running 3.2 core 2 duo which would have already seen a massive gain in going to a q95xx running at 4Ghz.
 
what the hell is going on in this thread.

This is ALL there is to it, they tested a brand new quad core, against an older DUAL CORE. The test is utter tripe, there is no if's or buts here, every other review puts gaming performance essentially equal, no matter what level of sli/crossfire you're using between the X48 and the older chips to the X58 and newer chips.

There will never be a large difference, on die memory, triple channel memory and a new platform means nothing, games are still gpu limited, always have and always will be.

The review, is bogus, absolutely. There will be no large performance gap in the future, in a few years when the new quads are twice as fast and games actually need a little more cpu power maybe, but the architechture will play no part in that.
 
I think the point is more to highlight that Core 2 Duos aren't the be-all and end-all of all new computer games when it comes to multi-card setups which do actually scale well with extra CPU crunching power than anything else.
 
I think the point is more to highlight that Core 2 Duos aren't the be-all and end-all of all new computer games when it comes to multi-card setups which do actually scale well with extra CPU crunching power than anything else.

But sadly due to the costs it is more likely that people with q9xxx clocked at 4Ghz + will be the ones wanting to buy i7 and they will be disappointed with the gain (if any) after they have spent all that money on a i7 except they will have something new to play with.

And with that core2 duo maybe running at 4Ghz on a decent mobo then even against an i7 at 4Ghz the differences in framerates in dual and tri sli may have still been small.
 
I think the point is more to highlight that Core 2 Duos aren't the be-all and end-all of all new computer games when it comes to multi-card setups which do actually scale well with extra CPU crunching power than anything else.

for the love of god, seriously, i7 quad cores are NO FASTER< AT ALL than their previous quad cores, like the Q6600, for gaming.

My point was people are still discussing and still saying the difference is big. THere is NO DIFFERENCE. they did not compare a quad to a quad, they compared a dual to a quad in a cpu limited situation. Every other review site compared a old quad to a new quad and found no difference, a couple of games are 1-2% faster on the new setup and a couple other games are 1-2% faster on the old setup.

multicard setups do scale better when cpu limited with more cpu power, the point is this review is attributing a massive gain purely to the i7, its actually the going from dual to quad core, if they'd used a quad core on both platforms there would have been a 1-2% difference either way on all the games. i7 is utterly pointless as a gaming upgrade, the whole point is this review is awful, its not comparing apples to apples, at all.

You would have gotten the same numbers entirely if the i7 rig was simply replaced with a X48 rig with a Q6600 in at the same speed.
 
You would have gotten the same numbers entirely if the i7 rig was simply replaced with a X48 rig with a Q6600 in at the same speed.

I agree except for this. I think it would take a q9xxx which is about 10% faster clock per clock over the q6600 to give the same results. :p

But yes, major money for little gain and now we know the none top end cards max out in their overclock at 3.6ghz cause Intel has fixed them to that so they will be slower in games than a q9xxx at higher than 3.6Ghz then a 965 at around £700-£800 for any gain at all is a very expensive gain.
 
Can't really see the appeal for most cost-conscious gamers at the moment. With the cheapest I7 and mobo weighing in at nearly $300 a piece, you're talking an outlay of £350+, and that's before you even take into consideration the fact that most people in that category will likely have DDR2 RAM at present.

Sure, if you've got the wonga for Tri-SLI / Quadfire then there are gains to be had, but the vast majority of people will be GPU limited at typical settings (say 1920x1200 2xAA 8xAF or similar) in most games.
 
If you can afford an i7 you probably have a 30" monitor, SLI 280/4870x2 and a Q9xxx at 4Ghz (sounds familiar;)).

From the tests the performance gain of a i7 will be marginal, I think this years CPU upgrade budget is going on a better HDTV.
 

except its sli limited, motherboard limited, nothing to do with the quad. Why didn't they show a single benchmark with X48 vs x58 both with a quad core? why because every single other site that has done so shows they are identical. ALl that shows is that Nvidia's own implementation of sli on their own older motherboards is still massively limited, the fact we know its cpu limited and over doubled the available cpu power barely made a dent vs the 680i dual core setup points to the board being the limit not the cpu.

THey are also using , 2gigs of mem vs 3 on the i7 platform, with VISTA, which tends to give a chunk of system memory for each graphics card.

Its simply crap testing not on equal footing at all, they are both motherboard and memory tested on a platform where just about no one on earth would recommend less than 4gigs.


Again EVERY OTHER REVIEW IN THE ENTIRE DAMN WORLD, shows its no better for gaming, and more to the point, when you look at the high res, theres still very little difference. Hands up if you have triple sli and won't be using minimum 1920x1200 and almost certainly some AA and AF, at which point even the 2gig 790i limited platform performs the same as the less limited 3gig platform with a "better" cpu.
 
So from what I can gather, i7 is only useful at the ultra high end of the market. If you've got less than 2 current gen GPUs you won't see a bottleneck with an overclocked Core 2 Duo.
 
THey are also using , 2gigs of mem vs 3 on the i7 platform, with VISTA, which tends to give a chunk of system memory for each graphics card.

No, no it doesn't. You're either talking about a 32bit OS (it's still not 'giving' memory, but I guess you could scrape by with that idea) - or the virtual memory solution should the VGA card(s) run out of memory. The full 2GB would still be available to apps in Vista, unless they ran the 32bit version with 2 280s, in which case it would indeed be slightly lower than 2GB.
 
Back
Top Bottom