Congratulations, you made it to the last page, the most important page of every review. In case you skipped right to this page from whatever page in between, don’t worry, so does 90% of the readers ;-) The last page is where it’s at, the page that every manufacturer gets to, to see if they look good in the end, and go home with an “award”. If you looked at the previous pace you'll probably have seen that some kits are cheaper than others, but to be honest: the price differences in the midrange segment aren't that huge, and it boils down to your own preference about design and availability for the most part.
To make a long story short: every memory kit reviewed today deserves its place in a core i7 setup. You want more proof: just look at the thumbnails below and much will be clear.
Average increases 1333 ->1600 (click for larger version)
Best vs. worst results @ 1333Mhz (click for larger version)
Best vs. worst results @ 1600Mhz (click for larger version)
Best vs. worst results 1333/1600 combined (click for larger version)
The first thumb shows the average increases when you look at the memory at 1333Mhz, versus the numbers at 1600Mhz. It is also the difference between a low latency setting vs a higher latency one with increased speeds. The result : the higher speed operation is consistently faster, but only 1,33%, on average.
The second thumb is more interesting, it compares the worst performing module in each test with the best performing one. The bottom line also shows the average difference: 2,33%. The third thumbnail does the same for 1600Mhz operation : result : 1,98% difference.
The last thumb is the most interesting: suppose you got an average performing 1333 kit (almost all do cas 7 nowadays), and compare it to the best performing midrange kit. The average performance increase you can expect from your system is 3,59%.
I’ll let it up to you to decide whether this is a worthwhile performance increase, it is certainly something you will not be noticing in your day to day pc tasks (you need about 10% speed difference to really note a difference). The conclusion of today must therefore be that any of the tested kits will do fine in your i7 rig, of course, if you do want the last bit of performance, and if you’re an avid benchmark enthusiast (like myself), just go on and pick up that upper midrange kit you’ve been drooling on…
(click for larger version)I'll decide, if I must
So which kits do run away with our recommended tags this time ? As performance differences were so small, and were actually very different from test to test, I decided to let mathematics decide. It's nice to hide behind maths once in a while, don't you agree ;-)
The conclusion : OCZ gets the highest average placement, followed by Mushkin, Crucial and Geil. I think it is fair to say once more, however, that none of the kits was standing out in any clear way. One could say though that the highest rated (cas 7) kits did end in front of the group, and the lowest rated (cas 9) kits at the back. In this way, logic is respected in the end.Time out : two more coming in
As I mentioned on the previous pages, there were two manufacturers that I contacted because of - in my opinion - strange behavior from their memory kits when we tried to overclock them. Those two are Corsair and Mushkin, and both have been looking into the issues, and have sent us a replacement kit to test. However, I decided not to include the results in this review yet, as there wasn't actually anything wrong
with the kits at their rated speeds. Both did the rated speeds and timings without any issues, they just didn't overclock to any satisfying level. I have been assured by both manufacturers that the headroom presented is atypical for both kits, and more can be expected - on average.
The follow up review of these kits is already in the works, but I just don't want to keep you waiting too long, so here are three screenshots from the first testing of the new arrivals:
Corsair Dominator PC12800C8 rev. 2.2 ; Mushkin Redline
Both these kits are dramatically different from their predecessors included in the review. So radically different actually, that we can talk about a total different part. Corsair only upped the rev. from 2.1 to 2.2, but it's a world of difference. Cas 8-1T at DDR1866 (PC15000) is simply stunning for a midrange kit.
The upper midrange (and more expensive) Mushkin’s do even better : they reach 960Mhz (DDR1920) without any problem, manage 7-8-7-1T at DDR1866 (PC15000) and even booted 8-9-8-1T at 1Ghz (but weren't completely prime stable - working on that) For a full look at these modules ni comparison with the OCZ Blades, have a look at our followup article
Time to gather all the boxes and wrap this review up...
(click for larger version)The good, and the bad
For each kit I'd like to present you with the some major good and "less good" items that crossed my mind when reviewing these parts:
Great headroom for a value-rated kit
Value is not an empty word : very good price/performance ratio.
Nice looks for modules and packaging.
Availability in Europe could be better.
High cas ratio does not tell the whole story (better performance at equal cas settings)
Overclocks very well when used at the appropriate cas setting (cas 10)
A little slower than the competition
We couldn't find this part : price and availability ?
Kit easily beat its own specifications, with quite some overclocking headroom
Amazing price for 6Gb of 1600Mhz memory : absolute top value
Great looking kit
Difficult to find in Europe, mostly replaced by cas 8 part at slightly higher prices
Great stock performance.
Very good looking kit, sure to stand out
Not overclocking like it should - we were probably unlucky.
Expensive compared to the (midrange) competition (Europe)
Nice looking heatspreader design with good quality feel.
Excellent overclocking potential.
The kit could use name tag
Great value in the US, but (too) expensive and hard to find in Europe
Top performing kit in this roundup
Amazing price/performance ratio
Very good overclocking headroom
Heatspreader design is starting to look a little dated
Probably nearing End of Life
Top notch look and feel, heatsink quality.
Very good performance at lower latencies
Good value for money considering the quality, with excellent availability
Kit does not scale to higher latencies/overclocking speeds
Rev. 2.2 is on the way : make sure you get the right revision.
Very good performing kit, easily beating its own specifications
Very good overclocking potential, scales with voltage.
Rather plain looking heatspreader design
Quite expensive compared to the midrange competition
As for our recommended tags, after some debate we decided to hand over two of them for now. Here we go... Patriot Viper PC12800C9
OCZ platinum PC12800C7
At this point a well deserved and deeply meant "thank you" goes out to all manufacturers that have put trust in our site, and who continue their support for our future memory reviews. The list is long this time, but I'd like to give it to you anyway:Cathy from Geil
Glen Haley from Qimonda-Aeneon
Mary Phuong from Patriot
Lili Chen from G.Skill
Michael "Michel" Klein from Mushkin
Tobias Brinkmann from OCZ
Gareth Ogden from Corsair
Joan Lunny and Gill Kelso from Crucial
Thank you all, I hope to welcome you to another review in the near future.
Last but not least, I would especially like to thank our sponsors from the Tones shop, which have come through for us more than once lately. Thank you Bram and Emanuel, we hope you continue to support us in the future. If you live in Belgium and are looking for some great value PC products, just go and have a look at the shop if you don't buy there already.
Well, that's it, I'm signing off : thank you all for reading and hope to see you soon.