MikeyLikesItSI

stan, in terms of performance, how do you like dual core over hyper threading? big difference? or is the additional core negligable over HT?

I don't know, i think i'm going to get out of AGP while there is still somewhat of a market for it.... This board blew the one stick of memory, so thats money down the tube.... I had a bad experience with my asus p4c800e-dlx, so i'm hesitant to go back to ASUS, but i've been hearing great things about their new boards!

Thats what i've been trying to weigh out, whether i'm going dothan or dual core. I figured dual core might be better to get some money for this 6800 GT while there is still somewhat of a market for it, but i cooked that one stick of DDR600 i have anyway, so i'd have to by memory regardless, Dual core sounds like fun, but the numbers dothan is putting out is insane..... hmmm Ace: the clocks started going downhill way back in the spring. Unlike ford, dividers didn't help my case. Also, after several months of having it backed down, I was pushing it up with some mild volts and about 280 FSB and it pooped out this stick of ram here.

I just wouldn't have expected it with a fully sinked chipset, watercooled northbridge, and a pretty heavy vortex of air blowing through the case... well, i suppose its time to start looking at newer chipset

I know the old northwood CPU suffered from degeneration after lots of volts and high clocks, but is it possible a motherboard chipset could do the same? My 630 used to clock to 4.1 with no problems on relatively low voltage. Well, now i can't get it stable over 3.8ghz. I bought a 551 CPU, and i can't get it stable much over 3.8 either. i've tried different memory configurations and different sticks of memory with no luck. Unfortunatly i don't have any other platform to try these CPU's out on to see how they perform. So, back to the question at hand, is it possible for components on a motherboard to suffer degeneration?

Whenever it shows default CPU speed at post while overclocking, its because the memory timings are too tight for the speed they are clocked too. Sometimes that will happen if CPC isn't disabled while upclocking. If you put the other sticks back in and loosen up the timings, or use a mem divider, it should take care of that problem.

Well, my setup has been very content all summer long, i backed everything down a wee bit, 260fsb, 1.450 vcore, 2.7 vdimm, 1.6 vagp. The weather was a touch cooler the other day and i was in the mood to play. I only pushed voltage to 1.550 vcore, 2.8vdimm, and 1.8vagp (CPU and NB underwater) which is not uncommon for 24/7 winter use. So using clockgen for a personal best superpi score, i was pushing FSB to about 280. After a few re-boots, and some minor tweaking, i pushed it up to 284 FSB. Well, it re-booted half way through the super pi sequence, and didn't come back up. I got the never ending LOOOOONG beeep, followed by a long pause. So, pulled the jumper, the battery, and the power cord...... after a few minutes, i put it all back together and it was still beeping. Popped the processor and everything looked ok. Then i started to change memory slots. Beeping went away, but still no post. I eventually got it down so it would boot with the sticks in a certian order, but it only reports 764megs of mem out of a gig. So i located the good stick and just left that in for the time being. I tried the bad stick in another computer and it showed 512 at post. I tried it back in this board by itself, and it won't post. Anybody ever experience anything like that before? this is using the new 6/27 bios as well.

http://www.dfi.com.tw/Upload/BIOS/75TLD627.zip

Here's a vote for BigStan. He always was very generious with his knowledge, and helped me on a 1 to 1 basis almost everytime i ran into a problem.

What type of PSU are you using? Have you tried some other memory in dual channel on your 875P-T? Did you try it dual channel in the orange slots?

i think i've needed less volts then you cause i'm on water. although your cooling setup is pretty impressive for air. But i used to be rock solid stable at 270 fsb with 1.4250 vcore. now even with 1.550 its not, but only 5 fsb less, i'm ok at 1.400, maybe less, i haven't really tested yet.

LOL i dunno. I haven't had a problem at 265 yet. I'm runnin 1.400 vcore. But yet i can't get 270 stable at 1.550 vcore? strange, especially since it was stable previously. I could see the little dot where the pin makes contact with the pad, i could see it on all of them. (until i went cross eyed :-) )

Passively cooled northbridge is definitly a week spot on this board. Especially with a water cooling setup. When cpu is air cooled, the air from the CPU cooler blows over the NB as well. But once you take that air supply away, you'll need to either put a better heatsink with fan on it, or adapt a fan as you are going to try to do. A NB waterblock is really difficult to keep it stabilized, but through the use of some thin neoprene, that is achievable as well. just my .02

It was hard to get a good shot with the lighting conditions i was using. All of the contacts are actually just subsurface. the insulator that goes around all the contacts has just a super small amount of raised effect to it. Those particular pads, i couldn't find anything more then just a slight bit of discoloration to the bad. I didn't know if it was a form of oxidation, or arcing. I figured if it were arcing, it would leave some blackened marks, or 1 spot that was a different color, not the whole pad. 270 proved to be unstable no matter what i did. using the 5/4 divider seemed to be ok, but then that failed too. I'm testing for stability at 265 now.

2.5-3-3-7 always seemed to be TCCD's favorite for me. i used that at 275 mhz, and at 280, any combination of timings i tried didn't make it any better.