Got the chip and payment today Bones. Thanks.:thumbsup:

Set it up already and let's see some numbers! :ohcrap:

Did you get things fixed NY?

Having some troubles myself since the Ultra D had condensation drip onto it (Again) and letting it dry out for now. Maybe it will come back one more, maybe not. I did though get some great WPrime runs off with the 146 while things were working.

Guess you can say I've gone a bit 939 crazy myself....

Condensation comes from being well into the sub-ambient:ohcrap: What have you been up to? Sounds like it's time for some kneaded art eraser and paper towels.

3532MHz on the 4000+ so far with NO TIM. Maybe improve that today. :cool3:

lol, busted!

Man this briongs back memories, Make we want to grab a set up, been a while since I played in the AMD arena ....:blink:\

Wonder if Bulldozer will be a contender ....

NY, TIM+ Eraser = WIN

Yeah, I think he'll get the win.

NY, you should be able to take a few Golds with ease once it's setup properly.

Golds I doubt, but maybe a few more points with some LN2 are possible. Today the chip acts funny though......won't go up past 3.4ish. I think maybe my gpu swapping(ati to nvidia) without proper cleaning could have corrupted something. I cleaned and swapped back, but it acts the same.

I have little time to do anything more this weekend. I plan to reformat and try it again when I have time.

I saw that CPU-Z run you posted and dude - 1.8v's+ ? :Dizzy:

I hope you didn't damage it somehow because it is possible to do even with better cooling used.
You can ask Mr. Scott about that one sometime.

Basing the following on my HVAC experience - I could be wrong or at least not spot on with the following but here it is:

You might be running into a "Thermal Latency" problem with it.

Not the chip's fault and remember whatever materials your pot, heatsink or whatever is made from will only remove so much heat/BTU's within a given timeframe based on cooling used and what you've posted, the chip still has it's lid on that could be the thermal bottleneck causing it.
With the materals of the block or pot and lid plus the TIM, it all adds up to your thermal efficiency.
Good cooling temps are great but core temps themselves could be getting too high if the heat isn't passing through all the materials fast enough = Heat buildup in the core itself. That's a big factor in your results, esp in the longer benchies such as 32M Super PI or 1024M WPrime since those runs make your CPU run for a much longer time than the short runs will, giving more time for heat to buildup.

I know the guys here have experience with extreme cooling so maybe they can weigh in on this.

^ right on

Excellent info Bones, I have often wondered about that myself even in relation to cold Air cooling. No one delids anymore...

It is certainly possible and I am good at destroying things. ;) but I kept the temps. in check.....no higher than 15C, and that was when I wasn't using any TIM. I wasn't trying to run any benches with more than 1.7v, so hopefully everything will be good this afternoon.

Gotta go get my taxes done first, maybe I'll be able to afford some juice this weekend......:bench:

EDIT: for some reason this 4000+ does seem to run a tad bit warmer than my other 939 singles.

Only bad thing though is it only takes it getting too hot just once to do the damage. I really hope it's OK and you'll continue to have fun with it.

Glad you guys found what I posted helpful.

Oh yeah, finally had to buy a new monitor since the old CRT finally kicked the bucket but good. I was waiting on that to happen and now I have a 23' Asus on the way here.
http://www.newegg.com/Product/Produc...82E16824236117

Not bad for the price I paid with the specs it has.
Now that the deal is done, I can finally quit worrying about that problem.

New stuff will not last as long as old stuff.

then again, a new LCD will not weigh 100 pounds or get wonky looking in hte corners like a 24" CRT.

Colors are not as good, detail is indetermined... might have been sharper on new CRT, but over time, LCD is better. (well I havent had a LCD last 5+ years yet)

Heat, depends on the LCD/CRT. Not a big differnce so I do not beleive the whole Wattage difference. My 28" 1920x1200 monitor heats up my room better than my PC.

Well I have some good news and some bad news about the 4000+ chip :blink:

Yesterday it didn't act right.......I had the chiller fully cold, so the core was ~-10C.........would not boot windows set to 265x12 with any voltage and it boots that with 1.392v normally.

Today the chiller had not been run and the chip boots right up to 265x12, and with slight vcore increase 270x12. Also was able to reach the same 3532MHz with the same 1.856v as before. Increase to 1.888v and only clocks up to 3515MHz....just like before.:laughing:

Wait a minute.......CBB between 0 and -10C???....with an AMD chip?? It would boot up and run, just not at quite as high settings. And it definitely would not clock nearly as high even with vcore.

Very strange if you ask me.:Dizzy:

We do.;)

Lol. Which time?

:ohcrap:Evidence of my theory on the behavior of the 4000+................

"A64 4000+ @ 286x12=3442Mhz, Radeon X1950XT @ 625/1800, 2x512mb Kingston Hyper-X "BH-5" @ 264Mhz 2.0-2-2-5 1T, DFI LP UT nf4 Ultra-D, CPU cooled by asetek Vapochill, despite Coldbug." -Turrican

4000/4400 /4800 were beasty chips. Not surprised they have thermal issues

High or Low.

4200 /4600 had a much smaller cache IIRC, and so was easier to OC.

Back off on the voltage a little - You're beating it to death and it shows. :eek:

I'd only volt it that high for suicide runs for only short periods of time but not for trying actual bench runs even if the MHz isn't as high. It's not wanting to clock up as high simply because it's getting too hot at the core and -10c isn't enough to keep it from overheating at 1.8v's+. It takes a temp of -18c to hit 0 degrees in farenhight (Misspelled that one for sure :laughing:) so your temps need to be even lower to safely run that kind of voltage.

I've been benching at 1.65-1.7v......I only turn the vcore up that high for cpuz runs.

I was just making the point that this chip acts like most other 939 single cores in the respect that the max vcore is ~1.856v and if you go over that it won't clock any higher or even as high.

I ran wp32m last night @ 3369MHz/1.648v.(still couldn't quite get your score Bones:blink:)anyways......But it still won't get past ~72% of the way through 1024 no matter if I set it to stock or what.

Going to give the other Expert a try with the 4000+ today (before I punish the 148), as Barton has made some good suggestions about possible corruption of my system.

I was thinking that this crashing was caused by something I disabled in BIOS;), but it crashes 1024 regardless. I'll get to the bottom of it.....you can count on it.

Make sure you've done everything you can for stability. If it's not stable, it won't complete the run no matter what you do.

I will say that if the CPU gets hot, it will slow down it's time between iterations. Try it with even less volts and see what happens, even if your MHz isn't as high for the attempt and go from there.