I7 - upgraded DDR3 sticks but no faster ?

[digitaldreams77]

i7 965 CPU
Asus p6T WS Pro Motherboard
Corsair Dominator TR3X6G1600C8D Ram (8-8-8-24)
Basic Bios settings, no overclock.

Memtest 86+ gave 17431 MB/s, great !

Read a recent article here on Toms Hardware where they tested PC3-16000 (2GHz) modules. Noted that the results for most at 1.6GHz were an even more impressive 27240 MB/s....so baught the 2GHz OCZ blade set to get the most from 1.6GHz without overclocking.

Memtest 86+ AGAIN gave 17431 MB/s, what !

Timings had been manually set at 6-6-7-16 as in the article & CPUz confirmed this plus the correct DRAM frequency of around 800MHz (1600).

Any clues ?. I must be missing something stupid here as the timings alone are much better (e.g. CAS 8 down to 6).

Regards

 

[Wayoffbase]

If I recall that article correctly, I think they showed that high performance ram has very little effect on overall system performance. It does give more overclocking headroom, but you are best off sticking with mainstream RAM if you aren't OC'ing your proc.

http://www.tomshardware.com/reviews/memory-scaling-i7,2325.html

Is the the one you read? Because they looked at RAM a few times recently.

 

[digitaldreams77]

No...the article on the front page currently. It clearly shows MB/s when running the modules at 1600MHz and what settings they used.

http://www.tomshardware.co.uk/ddr3-2000-6gb,review-31597-6.html

The point is that the many extra settings in my i7 bios must be throttling/overriding my improved timings - which one(s) ?????

Regards

 

[cjl]

First, what a waste of money. Dominator 1600MHz is more than enough already.

Second, are you sure that the memory is running at the rated speed? Many high performance kits (including the Dominator) will run at reduced speed when first plugged in, and must be manually set to the rated speed.

 

[Ancient_1]

You are comparing Sandra's numbers with those you got from memtest.

Here are my numbers from sandra with a bclck 191 mem @ 1527 7-7-7-19

And a shot of memtest @ 196 bclck mem @ 1567 8-8-8-19

As you can see the numbers aren't comparable.

 

[digitaldreams77]

Ancient_1

If you read my first post you will see I'm comparing Memtest to Memtest and strangely getting no change whatsoever for the reduced Blade timings (esp CAS from 8 to 6). I've never seen this before...must be a bios setting somewhere. Settings/frequency have been confirmed by both Memtest and CPUz.

I'll soon compare performance of both kits using Everest.

I baught the Blade to max out performance at 1600 and then overclock sometime in the future.

Regards

 

[digitaldreams77]

First, what a waste of money. Dominator 1600MHz is more than enough already.

Second, are you sure that the memory is running at the rated speed? Many high performance kits (including the Dominator) will run at reduced speed when first plugged in, and must be manually set to the rated speed.

Cjl.....the Dominator kit was set to use it's onboard 'XMP' settings and ran at the expected frequency & timings (confirmed by both Memtest and the latest CPUz).

Something strange going on here....

Regards

 

[jitpublisher]

I gotta go with the group here and say you are wasting your time and money.
It simply doesn't make much difference at all in real world performance, you will only see or notice any difference at all by running a benchmark.

 

[Ancient_1]

Bandwidth is pretty much just your speed of the ram and since both are running @ 1600 you should expect them to give similar numbers. Changing the cas ratings should help you latency by a fair amount but should have little to no effect on your bandwidth numbers.

Depending on the program some like more bandwidth and some like low latency but I doubt you will find any that you will notice the difference in daily use. Benchmarks like super pi will show the difference you should also see a difference in Everest mainly in the latency.

 

[digitaldreams77]

Well

Tried both sets again @1600, this time with Everest and Sandra benchmarks.

Both Everest and Sandra showed a 'slight' improvement (Everest giving different readings each time it was run, Sandra more stable)

So back to the latest Memtest 86+ & the Blade set. I can't accept that the bandwidth doesn't change with CAS, it has on all my previous rigs. So I pushed it wider, 9-9-9-24 (auto in bios), lo and behold Memtest showed a poorer 15722MB/s. So it DOES change. So why does it seem to hit a limit of 17431 @ CAS 8 & 6 ??

Next step, overclocking above 1600....here the Blade should show it's worth.

Regards