• 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.

3870 X2 is huge

Who cares if it is huge/small/medium/whatever, the idea is to have a product that will increase performance standards to manage the bad coded titles out for 2008!
 
jaykay said:
SLI seems to be cpu limited showing it ain't that great.
Click to expand...

Its not tho, the bench he posted was purely made up as to say if I overclock 25% i may get these gains...not gonna happen, not in Crysis!

The good thing about Crysis is that the CPU's are not maxed out and thus means you can get a better idea about how the two techs match up!

Race War over.....Crossfires Win! Crossfires Win!
 
Well the WTF are people posting random crap to prove a point.

Ok then, Crossfire scales better than SLI. This should give enough of a boost to make the 3970X2 to perform better than the 8900GX2.

Though the 8900GX2 may have a much better chip on board to help sort out scaling.
 
I don't see how it even could behave as a single card, when amd have been trying for years to do the same thing with cpus and failed.
 
The results I posted were a composite of results from real tests...

What I was trying to point out is the what if - I'm not saying increasing the CPU _would_ change the results... what I'm saying is the results don't have any hard conclusive evidence that changing another system component wouldn't have resulted in a different set of scores for the 8800 - the results do not tell you why the 8800 scored where it did and as its the highest scoring system you have to ask what if...

again if the results were reversed...

8800 GTX Single: 38fps
3870 single: 45fps

8800 GTX SLI: 72fps (+89.5%)
3870 CF: 76.5fps (+70%)

I wouldn't be saying hey SLI scales better than crossfire because I just can't draw a conclusion from those results what the top end of the higher scoring system really is.
 
Energize said:
I don't see how it even could behave as a single card, when amd have been trying for years to do the same thing with cpus and failed.
Click to expand...

Its not people are just assuming that as its on a single PCB it wouldn't use crossfire.

I am nearly 100% positive that i will use crossfire. It will still need software to split up the data on to two GPU's.
It may use internal crossfire on the extra chip but that's probably the only difference compared to two 3870.
 
Rroff - you must have confused yourself or something. here are some simple yet effective points to indicate to you that your argument makes no sense.

1. the GTI's etc are all better than a single 3870,

2. IF and I repeat IF the 3870 had the same performance (and remember it doesnt and as a result doesnt perform as well) in crossfire it would rape SLI EVEN MORE

3. You yourself have posted results PROOVING that crossfire scales better by showing 2 bad cards in crossfire perform better than 2 good cards in SLI and yet you still cant come to terms with the fact that crossfire scales better.

4. your adamant that SLI and crossfire cant be compared when they so blatantly can regardless of what each card specs are on paper/performance.

anything else you dont understand? oh wait Im the one that doesnt understand right?
 
it still would be a ******* to put in antec 900 it would fit as i moded the case to fit 2900 but it would be really close
 
Rroff confuses me. :(

You'll never in a million years be able to compare Crossfire and SLi with the same cards. The only way we can see which one is better if we compared single to dual card scores.

Crossfire as you have shown does a much better job of utilizing the second card.
SLI on the other hand is good but not as good as Crossfire showing it is better.
 
OK lets try this again... using my example results...

3870 single: 38fps
8800 GTX Single: 45fps

3870 CF: 72fps (+89.5%)
8800 GTX SLI: 76.5fps (+70%)

What they haven't shown and what would have provided proof - and I don't see anyone having done this online but if any one can find a benchmark showing it then I would accept it as proof - what would happen if you underclocked the 8800 so that the single card results are equal... would the multi-GPU results then be the same?
 
Rroff said:
The results I posted were a composite of results from real tests...

What I was trying to point out is the what if - I'm not saying increasing the CPU _would_ change the results... what I'm saying is the results don't have any hard conclusive evidence that changing another system component wouldn't have resulted in a different set of scores for the 8800 - the results do not tell you why the 8800 scored where it did and as its the highest scoring system you have to ask what if...

again if the results were reversed...

8800 GTX Single: 38fps
3870 single: 45fps

8800 GTX SLI: 72fps (+89.5%)
3870 CF: 76.5fps (+70%)

I wouldn't be saying hey SLI scales better than crossfire because I just can't draw a conclusion from those results what the top end of the higher scoring system really is.
Click to expand...

Do you have a source for those results or did you just make them up?

That looks like the single GTX is under performing not that SLI is scaling better.

Have a look at this, it compares the GT in sli with the 3870 in xfire
http://en.expreview.com/?p=53&page=1
 
Rroff give up you have lost.

Crossfire gives a better performance increase over single card solution that SLI in general. Hence how it can be said that it scales better.
I see you have GX2, would this be pushing you towards defending your purchase or something? Nobody really cares to be honest, the GTX are beast cards, but when you stick 2 together the gains are just not good enough. Whereas 3870s should have been better, but its good that sticking 2 together does offer acceptable performance boosts
 
peetee said:
Rroff give up you have lost.

Crossfire gives a better performance increase over single card solution that SLI in general. Hence how it can be said that it scales better.
I see you have GX2, would this be pushing you towards defending your purchase or something? Nobody really cares to be honest, the GTX are beast cards, but when you stick 2 together the gains are just not good enough. Whereas 3870s should have been better, but its good that sticking 2 together does offer acceptable performance boosts
Click to expand...

If you bothered to read through I'm not defending SLI

the 7950GX2 is ancient - I've had it what 2 years or something now - why would I be defending my purchase?

If you look at the 3870 v 8800GT link posted above - you will notice that taking COJ out of the equation as its a wildcard in context (you could just as easily find another game where SLI works properly and CF doesn't to balance it) the gains and losses for each side spread over the 5 or so comparable results are about equal...

I'm not looking to win or lose anything here and until someone provides evidence to refute it my point still stands...
 
Rroff said:
My scores were a composite from scores here...

http://it-review.net/index.php?option=com_content&task=view&id=1337&Itemid=91&limit=1&limitstart=1

Which is actually the 2900XT crossfire but the 3870 crossfire results show a similiar differential to the SLI setups they are compared against.
Click to expand...


Have a look at my source it gives much better results, and it compares the 3870 with the GT which is the product it competes with.

It also gives results without AA/AF which we all know is fubar in the latest ati cards.

if you look at the results you will see that xfire gives about a 69% improvement over a single card where as sli can only manage about 52%
 
You must log in or register to reply here.
Back
Top Bottom