Seems That if you have your Nforce 2 motherboard in Dual channel mode you should set your Tras at 10 or 11 for a nice increase in bandwidth.
Give this site a read and let me know what you think
http://www.geekextreme.com/modules.p...howpage&pid=30
Seems That if you have your Nforce 2 motherboard in Dual channel mode you should set your Tras at 10 or 11 for a nice increase in bandwidth.
Give this site a read and let me know what you think
http://www.geekextreme.com/modules.p...howpage&pid=30
I have heard of this before, but i have never seen any hard evidence, im sure that it allows me to run a higher fsb too.
Dont suppose you have anything on setting the divider to 6/6 rather than 3/3?
Good find
Will
| XP1600-m | ASUS AN78X Deluxe | r9700 pro | 2x512mb pc37000 |
Thanks will, thought it was worth sharing.
Dont have a clue about 6/6 over 3/3 sorry
tras?
AMD Athlon 4400X2 @ 2.565PenisextentionMhzOriginally Posted by Knoxville
Dual Layer, Gold Plated, LED Power,Dual Golden OMG IT MAKES MY CodPiece BIGGER 1-1-1-1 DDR62.3 @ 1222.3433Mhz
5 X 400GB Porn Array
X1800XT Dildo enchanged 3D Version, 512MegaLongJohn
Oh, did I mention.....I like sheep.....
WWW.MrsBurley.CO.UK now updated
Here's a link to more hard evidence
http://forums.amdmb.com/showthread.p...hreadid=237991
Yup I switched to TRAS of 11 about 4 months ago after discovering on other forums and benching wise it certainly yields better performace.
Also the jist of it is that this is not confoined to just nf2 chipsets or platforms.
errrr lol, what does "confoined" mean?
The reasoning behind this comes down to understanding what the memory latency stands for and how a computer works at the lowest possible levels,
The issue I have with both sets of tests on the pages mentioned is that they are on the Synthetic side of tests, i.e not performing operations likely to be seen in real world applications.
Its all to do with operations that the system is going to be performing. If you considure a processor a list of road instructions, turn left, give way, move to the right lane,
The instructions that the synthetic benchmarks will be doing is say accelerate, an easy operation to do, and to maximise its efficency from 0-60mph you do this in first gear with full revs to the red line.
Doing this for 0-60 works a treat but for normal every day operation you need to use breaks which doesn't work great with this operation.
It may appear like you are getting better performance and sure you get better numbers come out of things shown, but guys this does not equate to better real world performance.
Otherwise don't you think that memory manufacturers would be suggesting to putting your memory to that setting as standard.
Its just a way to tweak your figures to make your computer appear faster than it really is. Its for those sad people that just want their computer as a thing to boast about.
TiG
-- Hexus Meets Rock! --
"Confined" an obvious typo, not used to the keyboard on this lappy - or for your benifit I should say Laptop ?Originally posted by KeZZeR
errrr lol, what does "confoined" mean?
Last edited by Pita^Norf; 21-12-2003 at 04:03 PM.
I use Crucial ddr400, not the most high performance of RAM, but using these settings, will it be okay? It wont blow them or anything?
[XP2500-M @ 200x11] [2X512MB Crucial PC2700] [MSI K7N2 Delta ISLR] [160GB 8MB Samsung SP1614N] [160GB 8MB Maxtor DiamondMax9] [Megaworks 250Ds] [GeforceFX 5900] [Viewsonic VP171s]
no it will be fine. If you put the timings too tight the computer may not boot, just reset the CMOS, or hold insert on nf2's. Only voltage can cause you problems with RAM.
Will
| XP1600-m | ASUS AN78X Deluxe | r9700 pro | 2x512mb pc37000 |
Very interestiung results there, and not only did it increase benchmarks (tbh not particularly important) it helped system stability, as 2,2,2,11 was the only setting stable for him @ 219, maybe worth a shot for an extra couple of mhz on the fsb
Wibble
Currently wanted:
DVB- satellite pci card
TiG - i think we ned some results with usefull software - games, encoding software etc.. to find out if it helps..
saying that though id like to point out that it makes very little difference acctually runnign in dual chan in the first place, and it doesnt affect benchmarks much for me... though things that usually give any increase in synthetic benches will usually make a good, if unnoticable and very small improvements anyway in normal apps so im going to change it, then leave it, and not particually care what it is
Row Address Strobe Timing.Originally posted by Swafeman
tras?
t = ns
RAS is the same a CAS but is rows instead of columns.
I dont like sig pics so i turn off sigs Which doesnt help when i dont know what ive written here! DOH!
Been on XS.org for quite a while. Seems to yield good results witn 11-2-2-2.5 as well which I was surprised about!!! 11 is a really loose timing compared to the others (I could run 4-2-2-2 all day long on my NF7-S when I had it but didn't know about this!)
.: Predator :.
- Shuttle SN25P - A64 3700+ San Diego @ 2.7GHz - 1GB PQI Ultra DDR - X850XT - Asus DVD-ROM - 200GB Maxtor + 2*80GB SATAII -
what is better for gaming.. cause i can run 2-2-3-6 or 2-2-2-11 both stabily and was just wondering which one would be better... or should i use something like pifast to determine which is better???
And what is a good program thatll tell you your FPS in any game??
There are currently 1 users browsing this thread. (0 members and 1 guests)