darn, UPS just delayed my cpu/mobo delivery until Monday :(
Printable View
darn, UPS just delayed my cpu/mobo delivery until Monday :(
same here with my much needed thermal pads ( last piece of system = $1.50) ugh
All this "magical" thermal diode is is a simple resistor, when the temperature changes, so does the resistance. The cpu measures the resistance of this resistor and just "assumes" that is the correct temperature...Quote:
Originally Posted by Fred_Pohl
This is why this system is flawed :)
I built one yesterday, it could boot OS X at about 3.8 GHz but it was not stable under full dual core load. The system would not post at 3.9 GHz with good air cooling. OS X appear to be perfectly stable at 3.69 MHz (FSB 410). The vCore require for perfect stability at 3.6 MHz appeared to be no more than 1.525V.Quote:
Originally Posted by theonlybabyface
The situation might be different with better cooling, but so far BadAxe2 is not much better than BadAxe1 here. I have another BadAxe2 board that I will probably test on Monday, so we will see if my results thus far are typical.
8 x 462 @ 1.475 stable for 10 hours orthos blend...
figured i was too greedy at first... slowly moving back up to 8 x 475.
I think this analysis is flawed for two reason:Quote:
Originally Posted by fhpchris
First, I do not think that even the analog thermal diodes are just "simple resistors", rather I would expect them to be transistor-like devices based on the semiconductors Intel uses in the rest of the chip. However, Core 2 Duo's goes well beyond this now, they use digital sensors which are most certainly not just "simple resistors".
Second, even if Core 2 Duo's used a "simple resistor" the assumptions being made have nothing to do with the sensor being at the "correct temperature". Clearly that would be absurd. Rather, what is assumed (and probably flawed) is that the sensor voltage (or resistance) varies linearly (like a straight line) with temperature, when in fact it is non-linear function (complex).
um, is that with an AW9D by any chance? :stick:Quote:
Originally Posted by shimmishim
:)
yeah i wish!!! still on the asus p5b.-deluxe.Quote:
Originally Posted by ErikaeanLogic
the abit board can do 9x423 small fft stable but can't run blend worth crap.
then i tried running memtest (dos version) and i got errors like crazy).
so i switched over to the asus to make sure my mem wasn't bad.
i memtested my memory at 475 and it ran with no errors. so i determined it was the board... can't do anything about it... so i'm waiting for the new bios with the 1333 strap then i'll switch BACK to the abit... so the asus is just a temporary fix :)
shimmishim,
What are you settings man VFSB etc? Trying to get to grips with P5B Deluxe bios here.
Not being funny, but why only temporary?
Cheers
RLM
for 8 x 462 @ 1.475, i did in order after the vcore voltage...Quote:
Originally Posted by RLM
1.4
1.25
1.6
1.057
so i think the vfsb was 1.25.
as i said before, i think my current problem has to do with my NB overheating.
I am moving my desk and computer over from my bedroom into my second bedroom which has no heat running in it so it'll be nice and cold in there :)
Cheers man, much appreciated.
I followed Freecableguy's preparing P5W DH guide and cleaned up and applied AS5 on the NB and SB. Also stuck a couple of fans on them, one off an extreme spirit 2 and one off a blue ice. Should be okay??? MB 37c
Still testing at the mo. Currently on 405x9 @ 1.51v. Volts read 1.45/6 in bios hardware monitor and probe, so figure that one out:confused: New terminology to me but I think I'm correct in saying I have a vdroop of about .03v at load aswell. Idle temp reporting 58c in probe 54/55c in Coretemp WTF. Load goes up to about 70c, which isn't so unusual.
You mentioned in one of your posts about lowering the voltages for better results, and I seem to be experiencing the same. No benefit currently from upping voltages too much to VFSB etc.
Anyway fun'n'games:)
Cheers again.
RLM
yes.. it's all about finding that balance between voltage and temps...
since my aw9d is sitting idle for now, i'm about to remove the mx159cu from it and put it on the p5b deluxe :) hopefully that'll help it cool it some more and run orthos blend longer...
its amazing how long i've spent trying to find a stable 24/7 overclock speed for both of the chips i own...
okay... well after a week of testing with this chip, i've determined that the limiting factor is HEAT. yes, this chip is generating a lot of heat... almost 10C's higher than my e6700es.
and i have a feeling it has to do with the IHS again...
i was adding a thin layer of AS5 to the chip, too thin a layer and when i tried to run blend, my computer would freeze up.
so i took off the waterblock and looked at the imprint and noticed that only two edges had made contact with the cpu.
so i applied a slightly thicker layer and it made a better imprint but temps were still running a bit high.
took the xeon out and popped in my 6700 and i've been able to blend test at 8 x475 for over 4 hours now...
i think this chip might need a bit of sanding as well :)
anyway... wonder if this chip could do more if i could just deal with this heat issue
What kind of temps are you looking at? I find it hard to believe my H2O setup is any better than yours and have zero problems running blend as I've already proven.Quote:
Originally Posted by shimmishim
upper 70's, lower 80'sQuote:
Originally Posted by RacerX
on the 6700, mid to high 60's.
i might have hit a wall also with the chip... i didn't want to try more than 1.565 volts in the bios..
might have to crank it up some more to get it stable at 3.8. guess this chip isn't as good as i thought it'd be... oh well!
I hate to pollute the xeon thread, but until I can buy a nice 3070 xeon, this is all I can play with :(
4ghz dual 32m pass @ 1.63v
http://www.sspmustang.com/ot/x6800/4...al32m-run1.jpg
http://www.sspmustang.com/ot/x6800/4...al32m-run2.jpg
Installed new BI GTX 240 rad and noticed quite improvements in temps and OC. New SS is 4.25 which I will put in my sig later but to compare to fhpchris I came up with this. I finally notice my P5W64 WS Pro is becoming a limiting factor. Waiting for new Asus Commando 965 board.:banana:
http://home.comcast.net/~phippy1/32superPI.JPG
Nice Xeon! what is the stepping?Quote:
Originally Posted by RacerX
I think my windows installation is really screwed up after trying to get this x800 to work, and later corrupting it a few times :(
Decent results Racerx. Decent temps aswell. What's your ambient out of interest?
RLM
so guy's what board are you saying this runs better on the abit or the asus? i have the chip hear in 2 days but unsure abotu what motherboard to get
Quote:
Originally Posted by RLM
Room temp between 70-72c. I think also I might have better contact with block/IHS this time around, only 2nd time mounting the block on the Xeon. Also 80mm fan blowing over ram slots making the system temp lower than what it really is. Vcore set at 1.51v via bios but board overvolts quite nicely +.05v but actually drops to 1.54v underload which I measured on the board itself.
fhpchris, stepping is a L628B127 I believe you have one as well. Can't wait to see your Xeon 3070 results.
very nice.. :) hehehe.. i think fhpchris did mention that sooner or later we will seeing someone priming at 4GHz.. :)
:woot: I hope my xeon will be here soon.. :) hehehe will be using DFI RD600 :rocker: (I hope :D)
Glad the radiator helped you RacerX. I thought that you said you had a batch L628B167 earlier in the thread. Is this the same cpu?
Yeah, the rad help tremendously thanks for the push towards a new one. And thats the same and only Xeon I own at the moment.Quote:
Originally Posted by Ron 61
Still pushing for more.....but will need vmch mod for 4.3ghz.
http://home.comcast.net/~phippy1/4.2ghz.JPG
Congrats, looks like you have the strongest xeon so far. Your getting to the point where everything has to be dead on in order to go any further.