jaquith :
LGA 1366 -> LGA 2011 ; CPU ONLY ; makes sense - 45% improvement
LGA 1366 -> LGA 1155 ; CPU ONLY ; makes no sense - it's a side grade - 16% improvement
Balance - Gaming typical HD monitor, say I take a GeForce 9800 GT run it on a i7 9xx, SB, or SB-E this is totally unbalanced and the CPU won't do a damned thing to play e.g. Just Cause 2. Conversely, if I take a GTX 570 on Just Cause 2 55.4FPS SB vs 54.1FPS 920 WHERE'S the '16% improvement' poof all gone - 2%.
Take a minute and LOOK with your EYES
😱 at this article ->
http://www.tomshardware.com/reviews/p67-gaming-3-way-sli-three-card-crossfire,2910-6.html ; read some pictures if nothing else.
DUMB CPU 'upgrade' Choice -> 110FPS SB OR 104FPS 920
You are trying to convince the OP to waste their money, in fact IMO for 'Free' RMA the ASUS is bad it has a 3 year warranty.
IF I wanted to improve overall performance: 1. 120GB SSD, 2. new/second GPU, 3. 6GB+ RAM - rendering 12GB.
It's been fun but it's hard to argue with someone who picks and chooses reviews to form opinions while ignoring others and also who just pulls statistics from one benchmark and extrapolates it or worse yet, just makes stuff up off of hearsay without any evidence. Oh by the way, if you want to disprove an article, you do it by disproving methodology not by saying - oh well that doesn't agree with the articles that I agree with so it's wrong.
You want me to read articles, okay let's read the xbit article and let's see if we can get an idea of HOW MUCH better SB is better than X58 okay?
Is the OP even sli'ing? if not, why are you looking at tri sli, tri cf articles?
Source:
http://www.xbitlabs.com/articles/cpu/display/core-i7-2600k-990x_9.html#sect2
Test config: Same testbed, different processor
i960 versus i5 2500k (just remember the 920 performance differential will be even LARGER)
sysmark 2007 - 10% difference
e-learning - 10%
video creation - 4%
productivity - 8%
3D - 11%
Far Cry 2 1680x1060 - 12%
SC2 1680x1050 - 8%
f1 2010 same res - 10%
mafia 2 same res - 5%
metro 2033 same res + physx - 30% difference
civ v same res - 10%
winrar - 8%
true crypt - 0%
adobe photoshop - 30%
adobe lightroom - 0%
itunes 10 encoding - 15%
sonar x1 producer - 16%
x264 video transcoding - -1%
adobe priemere - <1%
cinebench - -1%
3dsmax - <1%
power consumption idle - 54% improvement
power consumption load - 52% improvement
performance in OC: not tested with 960 but differential gap widens between the best i7 and the SB i7
If there is one thing I agree with jaquith about it is that you should just stick with RMA. Don't watse your money on 1366.....
Just from the above figures, a 920 versus a 2500k, you are looking at at least a 25% improvement with a jump to 1155 - especially at resolutions higher than the ones in that article.
Oh btw, from dictionary.com:
http://www.thefreedictionary.com/upgrade - To raise to a higher grade or standard
yes, I would call 1155 an UPGRADE from 1366 and the dictionary would agree with me. Since the numbers are higher, it means it's an upgrade.
anyways, I'm done arguing