Running with a high HTT and stability.

Associate
Joined
16 Nov 2005
Posts
695
Location
West Midlands, UK.
If you set your HT multiplier correctly, can running with a high HTT in of itself cause stability issues?

I ask as I realised that if I lower my CPU multiplier to x8, increase the HTT to 337 and run on a memory divider of 133, I can run at 2.7GHz(which I'm near to proving is my max stable speed)but increase my memory speed from 192MHz to 225MHz.

Any advice would be welcome as I don't want to start another round of stability tests, only to be baulked by stability issues bought on by having a high HTT.

I don't know whether the CPU and brand of motherboard have an affect on this but I have an Opteron 144 and an EPoX EP-9NDA3I motherboard.

Cheers, dagwoood.

P.S. I realise if I set my HT multiplier to x3 and set the HTT to 337 I'd be running the Hypertransport slightly over spec, but I presume this wouldn't cause any problems.
 
smids said:
You'll need some stability testing guaranteed. Your board uses an nF3 chipset - one which is known to top out at about 280-290mhz - anything beyond is unstable. The nF4 chipsets clock much further than nF3. Run a long prime, I shouldn't imagine it'd last too long.

Thanks for that smids. The nF3 issue would explain my current problem http://forums.overclockers.co.uk/showthread.php?t=17566130

I purchased the EPoX board to replace an Asrock Dual Sata so as I could get my HTT above 274 and it's now looking like I've wasted 50 quid :(
 
smids said:
Sorry to hear that :(. Still, at least you can stop trying to work out the cause of your problems :).

Thanks for sharing my misery and trying to find the bright side :)

It makes more sense now why I have to put 1.50v through my CPU to get it stable with an HTT of 300, but with an HTT of 294 it was stable with 1.45v. It looks like the extra 0.05v is to combat the instability of such a high HTT.
 
smids said:
The Dual SATA2 is a good board and has lots of features above the nF3 so I'd certainly say do that. I'm sure someone can help you out on the settings too if you head into the motherboard section. Good luck :).

And in answer to your HTT question about 300mhz vs 294mhz - I don't think that was the HTT causing that - it sounds more like the memory controller. It would seem that if you managed to get 300mhz stable, then the chipset CAN handle 300mhz (rare but doable on the nF3) but I doubt it'd go much further than that.

I purchased the EPoX board as I'd spent three months of trying to get the most out of the Asrock DS, and I was still stuck at running with an HTT of 274 :( The OCW beta's just wouldn't play ball. Everyone I tried it would never "remember" my voltage settings when I turned my system off then on and so would load the BIOS defaults.

On one occasion after flashing to the OCW beta 2 BIOS, I spent 2 days stress testing to make sure I'd got a stable overlclock of 2.65GHz, only for it to lose my settings when I turned my machine off :mad:

I found the official BIOS's a bit hit and miss as well; you have to be spot on with how you configure your BIOS settings else your machine wont boot. I know this is the case with most other BIOS's, but this seemed worse than any other I've come across. I even got cold boot issues with IDE drives :eek:

On paper, the EPoX board seemed a better proposition. The ability to set the HTT to 400, VDimm setting of 2.8v and the option to increase the VCore over 1.5v. It's a pain in the proverbial I didn't know of the nF3 issues and high HTT values.
 
Back
Top Bottom