still going strong with my dual core e5300 @2.6ghz with 2gb of memory ;) which i built this summer and is fine for what i need (homework and school work mostly). will think to upgrade when it gets a bit slow or breaks but with this in mind when prices go down in a year or 2 will upgrade then :) and give older parts to rest of family
I went with the Blue Ripjaws because they match my MSI GD65 motherboard perfectly, Its silly i know but with the performance difference being so small, they might aswell look good.
Great article, surprised by the performance numbers though I was expecting a bigger spread in results
Rich any chance for an AM3 article like this? I know how boring and mind numbing the testing is but it'd be a good read to see if the AMD chips perform differently with memory latency/frequency changes than the Intel platform does
I'd imagine (from experience, and because AMD CPUs also have integrated dual-channel memory controllers) that the results will be much the same as you see in this article.
Plus, Rich could probably do with a break from memory for a while! :D
Heh yeah I figured he could, Not to mention a break from the AMD platform with all the **** the 965 is giving him heh, but it'd be pretty cool to still cover the memory spectrum that way, I don't think any other site has done it
I know they have the IMC just not sure how much it differs from Intel's TBH, never really seen them compared using the same memory and this style of test, they might prefer different things considering AMD's is much more mature
Originally Posted by Gremlin I know they have the IMC just not sure how much it differs from Intel's TBH, never really seen them compared using the same memory and this style of test, they might prefer different things considering AMD's is much more mature
Ahhh, but the AMD memory controller can handle higher DIMM voltages, so it'd be unfair to AMD for us to test its CPUs with any of this memory anyway.
That's not to say that you can't use Lynnfield memory in an AMD rig and get a lot of overclocking headroom though.
So after reading this article, I'm of the opinion that, when it comes to gaming, GPU > CPU > Mobo > SSD > RAM. Basically just make sure you meet the minimum RAM requirements with a good namebrand and make sure it looks pretty?
Originally Posted by Claave Ahhh, but the AMD memory controller can handle higher DIMM voltages, so it'd be unfair to AMD for us to test its CPUs with any of this memory anyway.
That's not to say that you can't use Lynnfield memory in an AMD rig and get a lot of overclocking headroom though.
I wouldn't call it unfair since there's nothing stopping anyone from just using it at its rated voltage, but yeah AMD handles the higher volts quite happily so it gives you some wiggle room for sure if you were to OC
I'm more asking because I have seen literally hundreds of threads on enthusiast forums since AM3 came out with people spouting the old adage from all the previous AMD evolutions that lower latency over higher frequency is the most valuable asset,
A definitive article would put all that to rest "whats better on AM3, high freq vs lower latency?" because for all we know since nobody tested it AMD's IMC might perform much different than said
Hell it doesn't even need much to test to be honest just some ram ran at different frequencies and latencies, If you guys don't wanna do it send me some ram and ill do it all for you free of charge! :P
Originally Posted by Sifter3000 What bodes better for Intel's new dual channel CPU: high clock, low latency or both? We run the numbers for both so you know what memory kit to pair with your new Lynnfield system.
I'd be interested to see the differences between 4Gb and 8Gb (especially given that you could buy an extra 4Gb instead of springing for high performance memory).
Would love to see you guys do a similar round up for Sandy Bridge. Wonder if its increased performance means that it will be held back by "only" 1600Mhz C9 memory at all?
Comments 1 to 25 of 27
ReplyRich any chance for an AM3 article like this? I know how boring and mind numbing the testing is but it'd be a good read to see if the AMD chips perform differently with memory latency/frequency changes than the Intel platform does
Plus, Rich could probably do with a break from memory for a while! :D
I know they have the IMC just not sure how much it differs from Intel's TBH, never really seen them compared using the same memory and this style of test, they might prefer different things considering AMD's is much more mature
Looks like gaming performance has very little in terms of gains/losses, at least when it comes to frames per second.
When playing online first person shooters, is lower latency more desirable than higher frequencies?
I gotta say, that Hyber Tx1 something looked ridiculous.
Ripjaws look reeeeeal fine.
Ahhh, but the AMD memory controller can handle higher DIMM voltages, so it'd be unfair to AMD for us to test its CPUs with any of this memory anyway.
That's not to say that you can't use Lynnfield memory in an AMD rig and get a lot of overclocking headroom though.
I wouldn't call it unfair since there's nothing stopping anyone from just using it at its rated voltage, but yeah AMD handles the higher volts quite happily so it gives you some wiggle room for sure if you were to OC
I'm more asking because I have seen literally hundreds of threads on enthusiast forums since AM3 came out with people spouting the old adage from all the previous AMD evolutions that lower latency over higher frequency is the most valuable asset,
A definitive article would put all that to rest "whats better on AM3, high freq vs lower latency?" because for all we know since nobody tested it AMD's IMC might perform much different than said
Hell it doesn't even need much to test to be honest just some ram ran at different frequencies and latencies, If you guys don't wanna do it send me some ram and ill do it all for you free of charge! :P
Corsair Dominators 1600Mhz 9-9-9-24 designed for AMD works perfectly at stock speeds 1.55v.
http://img.photobucket.com/albums/v293/wuyanxu/WyXQuad_BuildLog/ee36cb17.jpg
An educated guess would be the two systems are similar, except Bloomfield doesn't OC as far because of triple channel limitations.
Linkeh?
http://www.bit-tech.net/hardware/memory/2009/11/02/intel-lynnfield-memory-performance-analysis/1 - Bindi
http://forums.bit-tech.net/showthread.php?t=171053&page=8
Did they ever pick a winner cause I thought I was the only one with a new idea so I had my hopes up... :(
Yea we did :) It was announced somewhere...
-
« Previous
-
1
-
2
-
Next »
Discuss in the forums