i never ran nb below 1.63v lol
and my nb temps never hit over 40c on air i had a katana3 heatsink on the nb:D
water a bit better
personaly i dont like low voltage ram :p:
be patient they will pop up
Printable View
Still running 1.52 daily here stock heatsink with an antec spot cool cooling it... temps under load are 51°C still running strong with 2 x 2Gb 1600C8... QX9650@4ghz quad at 400FSB...
I'm getting a 12GB Sniper 1600C7 kit for review John... keep you posted how they work on the old hog...
@rotamonkey
I guess you have that swiftech block on the original heatsink arrangement, therefore the high temperature.
That original heatsink arrangement is no good, if you want to get that temperatur down I suggest you switch it for something better.
Back in the days when i ran my REX I used a koolance CHC-122 for the MCH, it was in the same loop as the CPU (Q9650) and during full load the MCH rarely went over 40C even though I used 1,60v +.
do you guys think a swiftech GT cpu block could be modded to fit the NB on the rex?
http://i52.tinypic.com/14sfc4g.jpg
yeah definately goin to have to do something about this, its gettin colder here now and the temps arent as bad anymore but under load definately around 50 cant believe you guys arent even cracking 40... holidays next week so Im gonna pull the board and see what I can do...
Do you think perhaps we should some how make a QVL for modern DDR3 RAM (if possible).?
I am just thinking as a lot of manufacturers tend to list the old and not very easy if at all possible to get hold of RAM.
Would be nice if they did test the newer RAM on this board. :(
John
don't think it will be interestign anymore John, hardly anyone uses this board for 24/7 ( I still do at my parents and you : that makes two :p )
You have a very good point there Leeghoofd :(
I would like to try and think of a valid counter argument, but facts are facts and the amount of people who are still on 775, yet along Rampage Extreme is diminishing everyday.
I am clinging on until Skt 2011
I eagerly await your test of the 12GB Kit on this board. If possible please could you test with 8GB as well as the 12GB :)
ASUS gave me a BIOS which fixes the VID Ratio error with the QX9650, I am happy with it, they also updated the RAID ROM too. BUT took away TREF :|
John
Anymore 4GB module test reports guys?
John
I was using a REX 24/7 in gf's pc up until a month or so ago. Not many people left who would be though that's true.
16GB for $135, no matter how crappy the speed/timings are thats good value for a cheap desktop. SB just loves crap ram too so thats where the market is now.
Is anyone still benching with his REX or am I the only one left? :D
I've been testing my Dominator GT 2000C8 on my REX for the first time yesterday and today, trying to get a fast 32M at 2400MHz.
This is the result:
http://img41.imageshack.us/img41/815...dr31600cl6.jpg
My goal was to hit 17m15s ( or lower of course ), as you can see I'm not quite there, but hardware-wise this is pretty much the limit. This was done at 1.55V Vnb and 1.70V Vdimm, I tried to get tRP 4 at up to 1.75V Vdimm, or tCL 5 and/or tRCD 5 at up to 1.60V Vnb and 1.80V Vdimm, but this didn't work out.
If you have some ideas on how to improve the timings I'm all ears! :D
Software-wise this is also almost maxed out. A fresh OS might give me small gains, but I don't think it'll gain me 2 seconds.
In case anyone's interested in the bandwidth, here's an Everest shot right after reboot ( same settings as the 32M above ):
http://img202.imageshack.us/img202/7...1t2400mhze.jpg
That is good going Don Dan :)
Has anyone had any luck with Ubuntu on this here board?
I cannot seem to BOOT Ubuntu unless I disable ACPI via a command line?
(and then I am stuck with crap performance and 1 core)
This is the LIVE BUILD DISK I am using
and I have tried both the x86 and x64 distros from here
I still cannot get Ubuntu to get to the desktop after the splash screen unless I boot with the disable ACPI option.
HOWEVER the distro I am using works fine on a few other PCs :(
John
Its just you ;)
I have some results here: Hyper testing on REX
Maybe your ram just doesn't like tRP 4?
Thanks John! :)
I'm sorry I can't help you with your Linux problem, I've never used that OS.
Haha, I was afraid that is the case. Not having any money for new toys sucks... :rolleyes: :D
Your scores were actually the base line for my testing, since only few people have tested these 2GB sticks on the REX.
I guess that's the truth, or it's the board. :shrug:
I already have an idea on how to improve the score, I'll give it a try again some day. :yepp:
I've still got 5 boards......4 of them working ....one died on the front lines after 6 hours @ -135. :D
One day soon will bash them some more. :up:
I still have mine too and a nice dual core. Just haven't used it since it was in my GF's PC running the Q9650, before that it was all insulated up for fun times. Hopefully one day I'll fire it up again!
Tried another kit of ram Don?
Try BIOS 0403, and get another kit on there see if you can tighten it up. You using 266 strap? Pull-ins on or no boot at tRD 5 with them enabled?
At 6 x 400 and DDR3-1600 its very limiting, can only tighten so much. Its probably down to OS tweaks if you want much more speed. Why the 17m15s target?
That score was done on the 0403 BIOS! ;)
I'm on the 200 strap, but now I think the 266 strap might have been faster. I tried all the Pull-Ins at tRD5 and enabled them one after the other, but they all resulted in no boot, even at higher Vmch and Vdimm.
I'll try the 266 strap and also another kit when I have my semester break.
The score was done with all the OS tweaks I know, no magic left. :D
My friend Hyperhorn is at 17m13.985s and I'd like to beat his score, but he believes sub 17m10s is possible, so that might be the new goal. :)
Damn you man, you are making me want to haul out my mobo lol. I don't have a PSU or a HDD I can use either so I'd have to make a quick order. Must resist.
I have no idea how SPI32M is so much fun. To a casual observer anyone running that bench must look completely insane, staring into space, mentally extrapolating from each loop time what the final time will be, willing it to run that bit faster each loop. We're all nuts.