• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Do you think the 9x285 3xhtt on Neo2 plat has worth while performance increase over 10x255 4xhtt?

N

nealh

Diamond Member
I am trying to decide whether I should try and increase the fsb to 284/285 3xhtt vs running at 10x255 4x htt

what you guys think..also do you think a 285 fsb is too much stress on the mobo and system...I realize if pci/agp is set t o67 they are locked
 
I don't think there will be much difference between the two (I'd run with the higher multiplier myself--easier), though the Neo2 can probably handle even 300MHz HTT well.
 
Higher hypertransport or higher FSB is always better than increasing the clock multiplier as FSB/ HT increase bandwidth and decrease latencies as well as increase clock speed, whereas multipliers simply increase clockspeed alone.

-Kevin
 
I'd have just left it that way because it's a bit easier and there's not a whole lot to be gained from higher HTT and a low multiplier versus a higher multiplier, at least on this platform, from what I've seen. As it is, you don't gain much over asynchronous operation by running memory synchronously with the HTT when overclocking. As for these latencies you mention, I doubt they would be noticeable either, though I'm not quite sure exactly what these are in reference to.
 
how about your memory at each setting?

Do you have winrar? test it's internal benchmark on

284/285 3xhtt wih memsetting 166 2.5-3-3 1T

vs.

running at 10x255 4x htt memsetting 200 whatever timings ram can handle

vs.

running at 10x255 4x htt memsetting 166 2-2-2 1T if your ram can handle


IMO the last one will be fastest by far and winrar seems to be best test for analysis.
 
To be honest with my cpu i just tested my 2.6 on my 3200 and i am running my ram at 260 1:1 and im running the 3x htt and i am running fairly lax timings... but i tested it under sandra and it just does as good or not better than the 4x... it says i actually have like 10pts higher in a few of the categories.. i may of had it at 255 in the first one i tested but thats alot of htt that is gone with the 3x changed so i say it has no affect and plenty of people will also agree.. if ur that worried about it.. get sandra run ur two settings make a ss of them both and compare than you will know the truth for urself
 
more curosity..if the fsb at 255 vs 285 is essentially the same as everyone suggests..I will stay at 255..seems like it would be less stress on components, ie the mobo..I realize the pci/agp are locked

Zebo ..I should have mentioned but my ram is best 133...seems to do fine at 2.5-3-3-10 2t at 2.7v

I am pretty sure the ram 133 speeds at 285 fsb may truly be an issue for my ram in DC...so I would use ram100
 
On the K8N Neo2 Platinum, what exactly does the "HT multiplier" affect? And what relationship does it have with the RAM and CPU? I understand bumping up your FSB essentially overclocks the whole system, but what does the HT multiplier really affect?

Thanks
 
Hey Neal, "Dad" here; based on my recent experience (and thanks to you) I doubt the additional speed is that great; I've been able to successfully OC to 2.4 Ghz on a Newcastle 3500+ (my signature gives more details) and to tell you the truth that's probably pretty conservative but I've very happy with the results. Runs rock stable under Prime. So when you're pushing the envelope for those last few megaflops my thinking is that there is a small risk of burning out the RAM, particularly if it's value RAM.

My guess the HT (Hypertransport?) multiplier has something to do with raw transfer rate between the CPU and the memory subsystem. However, as you know the docs are clear as mud on this point.
 
I decided not to try 285 fsb again..I am worried this may stress the mobo too much?? or my ram..and when I ran 9x285 vs 10x255 I saw no difference in performance..I may test again as I changed my bios and Prime95 is now again stable at 2550

in fact..I have even adjust my ram settings..ram 133 (170) 2.5-3-3-10 1t @ 2.7v
 
After playing with my FX-55 for awhile I can tell you this. When running at 11x273Mhz 3XHT I can brake 8000 in Sandra's memtest. At 11x266Mhz 4xHT my score drops down to 7400-7500.. Other then that I don't notice a differenece.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top