They do 2.5,3,3,8 too . But having some boot issues with the latest BIOS.
Thanks for the info Gautam, Ill install a win2K SP4 partion for some 3Dmark2001 scores.
I'm stable at 2800mhz too now :)
Printable View
They do 2.5,3,3,8 too . But having some boot issues with the latest BIOS.
Thanks for the info Gautam, Ill install a win2K SP4 partion for some 3Dmark2001 scores.
I'm stable at 2800mhz too now :)
Make sure you press F5 as soon as installation begins and install the 'MPS Uniprocessor PC'.
28000 sounds nice. I've been through 3 Dothans so far and none of them even come close to 2.8GHz. :( One of them can't do 2.5, the other falls below 2.6, this one used to be able to do 2.7 before I chipped its core. :-/ Perhaps the fourth time around....
Thanks for that Gautam.
http://img175.imageshack.us/img175/1...lity6bl.th.jpg
Just cranking her up. Hitting 2832 Stable ATM.
Won't do 2850 stable though, but my P4C800ed is dropping bad to around 1.54v under load(1.6 set in BIOS). So need a droop mod and maybe a Vcore too.
I also have some PC3500 BH5 corsair sticks that I can use for a fast run.
As for the ballistix. Just ran 5 full loops of memtest(All tests) @ cas2,3,3,8 PAT on 235mhz :)
Pentium M 740
2x256MB OCZ BH-5
P4P800SE (droop mod + 3.3 = vdimm)
Prom mach2 R507
Going to do a vcore mod tommorrow to get higher. The proc doesnt seem to be a very good one but its nice for now.
The first screen is a bit f*cked but srry about that :p:
http://home.kabelfoon.nl/~vnrswk/spi24750dothan.jpg
http://home.kabelfoon.nl/~vnrswk/dothan3020pi.jpg
2 pics of the setup/mounting while running a test to see if the mounting was any good:
http://home.kabelfoon.nl/~vnrswk/rig1.jpg
http://home.kabelfoon.nl/~vnrswk/rig2.jpg
sweeeeet
Very nice,
3Ghz is still not to be sniffed at
@ The Doctor : I see you have a digital uniMeter; did you maybe measured real CPUs supplied Vcore (Vcc) & how? (to compare it with Bios/monitorProgram readout)
--
Regards, Spajky
It matches exactly :) If you selected 1,6V it gave 1,6V (I was at The Doctor yesterday ;) )Quote:
Originally Posted by Spajky
From WHERE was made the measurement? on adapter somewhere (where?) or on MoBo´s MosFets?
The doctor is that not likely to get condensation around the socket on the upside of the board?
ANy chance of a step by step for us idiots? :)
2 pics of the spot to measure vcore on (same spot):Quote:
Originally Posted by Spajky
http://home.kabelfoon.nl/~vnrswk/vcore1.jpg
http://home.kabelfoon.nl/~vnrswk/vcore2.jpg
The pics ive posted before were from the mounting test run (no issolation at all) when i was confident i worked ok i issolated it all so condensation is not a problem :)Quote:
Originally Posted by marvt74
may you post a picture about how you realize to put the VAPO LS on the dothan adpater?Quote:
Originally Posted by Mrki
Just to see haw to mount it and how to insulate between the mobo and the dothan adpater.
Thanks
how do i get my scores in the team. :D heres a link http://service.futuremark.com/compare?2k1=8622086
__________________
and one more .this is a24/7 set up :D
A random though here on the 400MHz and 533MHz FSB chips and their overclocking abilities... Has anyone checked cache latency between the 2? (At the same FSB and overall clockspeed, same RAM, ratio and timings, etc)... I'm wondering if there's any difference between the two and that's why the 533MHz chips can clock to higher FSBs?
Edit: Ignore that - just realised that it would really show up in the performance if the 533MHz parts / C0 cores had a higher cache latency (don't know what I've been smoking ;))
Here's an issue I just found happening with my 730.
I seem to have hit a FSB wall.... regardless of multiplier I cannot boot (not even a post) at >247 fsb.
I can boot at 247 X 11 with 1.55V in bios. Thats not totally stable, but I can atleast pi 1M at that speed. But at 248.... no post. Strange thing is, I can and have been booting at 237 X 12 as seen in my sig.
So I figure ok, lets find the max FSB. Now this is on a P4C800E that did 275 fsb on my 14 multi pressy just a few weeks back. 10 X 250 = no post, 9 X 250 = no post, 8 X 250 = no post, 8 X 245 = no post...... Have not tested lower.
So WHY can I boot at 11 X 247, but 8 X 245 will not post at the same vcore :confused: My ram can handle up to 250mhz 1:1, but even with a divider I still could not post at 250 (ram at 200).
I have not tried clockgen yet, well not fully that is. I booted at 237 X 12 (max stable so far) then clockgened up to 240 X 12 .... seemed ok until I minimized clockgen .... locked up.
Questions:
Most important: I have the 4pin P4 connector plugged into the mobo, and I am feeding the adapter with its 4 pin power connector as well. Does the mobo power connecter need to be there? or could it be hindering anything ?
Is there any particular mod that should be done to the adapter itself? (have droop modded the mobo a while back).... strange thing is, that while there does not seem to be any droop, I set 1.5 in the bios for instance, but yet the sensor reads 1.47ish.... It used to overvolt at idle.... 1.5 would be 1.53ish, so that I find strange, but I figure the sensor is just reading wrong. I have not done the U wire mod (using the new beta bios for vcore / multi adjustment)
Is 1.6v safe for a 24/7 system ? I have been hesitant to go over 1.55v, but I have tried 1.6 for a few mins... still no post (cooling in sig)
I just don't understand why I cannot boot at higher FSB at low multipliers..... I guess the cpu just doesn't like higher fsb, but why does 11 X 247 work, but 8 X 245 not :confused: :confused: :confused:
Thanks for any input :)
@ The Doctor :
According to the pics, measuring was done on Mosfets on MoBo, you should have done it on the adapter with temporary very small HS provisorly mounted-no fan (to have access to adapter onBoard Vcore regulator) @ stock clock speed, I am sure the values would be different than from MoBo ...
@ Frozendeath :
The mobo power connector does not need to be there! (adapter has its own Vcore regulator feed with electricity by that "floppy" plug ... To rise the Vcore, IMHO you have to do a "VID pins wire trick mod" on the socket of the adapter & do not go over 1,6V ... If this does not help, means that adapter onBoard Vcore regulator can not go higher much ... find Intel datasheet for example [C-M] 30311004.pdf (p#16&36) for Dothan pins & Vcore definition ...
I'm VERY VERY upset now. My 730 hits a wall at 210-215 just like my 715 did at ~166. There is absolutely NO WAY to stabilize above 210-215 on either p4p800 or p4c800 with any of the 2-3 bios I've tried on each. This would be an easy 2700mhz chip if not for this issue.
Bah!
Frozendeath and Lithan
Please read my and highlanders observations about multipliers and FSB issues with boot .
http://www.xtremesystems.org/forums/...490#post940490
onwards from that post.
not true... I get up to 1.8Vcore using a motherboard Vcore mod...Quote:
Originally Posted by Spajky
jup... just did the p4c800 vcore mod (50k pin8) on mij p4p800 wich uses the same chip for vcore and works great.Quote:
Originally Posted by charlie
Quote:
Originally Posted by Frozendeath
hm, originally i wanted to quote only the first part about 8x2hundredsomething
now i just quoted the rest because i may have some input for you there.
anyway. i have somewhat the same issues. multi 8x is a no go, but then
again a bunch of guys around me have the same problem. you can select
the 8x multiplier with almost anything (i didn't try anything below 200), but
rest assured, that it *should* not work. it seems this is some kind of bios
issue. but a general one, as it affects p4c800-type boards as well as p4p800
type board - i am running a p4p800-se.
anyway, i found 9x to be the best multi available for some aircooled high
fsb-action. so far i had it up to 311 via clockgen. 8x doesn't work, 7x can be
selected, but magically switches over to 6x. so the multi-settings left are
6x, 9x, 10x, 11x, 12x and whatever multis you have available via your cpu.
i prefer the higher ones 755 and upwards...
you may try to boot at 8x133 or 8x166 - shouldn't work... so don't bother
with 8x - even selecting it via eist in windows makes my rig crash before
i can say wow...
with my p4p800se i can boot at 250 - 9x 10x, whatever.. maybe a single
mhz more, but 252 is a no go. doesn't matter what kind of mem-divider or
pat-setting i'm running. so live with it. whatever your max bootable fsb, be
happy with it - it's not the end of the road, as i found out via a couple of
days of testing and some good hints around here.
btw another thing regarding your overclocking trials from 237x12 towards
240x12 via clockgen. it might be wise to drop the multi a notch - either via
setting 11x in the bios options or by dropping it via eist in windows. trust
me on that one. unless you run 1:1 and your memory craps out on you, you
should be able to rise the fsb a good deal above your max boot. sure there
has been the occasional "bad" board that won't do much more anyway, but
in most cases you can go higher using clockgen.
a word of notice regarding clockgen. use small steps to up the clocks if you
want to shoot for max-fsb. i usually boot at 10x250 or even 9x250 (on going
for >300 fsbs) and then - knowing i can go beyond 300 - i up the fsb in steps
of about 10 mhz til i am at ~290. like 260, set, 270, set, 280, set, ... when
i reach 290 i use smaller increments, until i'm in the killzone where i only up
it by clicking once, max twice on the arrows to up. then set -> bench spi1m,
next, ... on first trying for max fsb, i went from 250 to 30x in increments of
1mhz fsb... took 30mins to one hour to finally get there ;). but sometimes if
you make to big a step, the sys locks up even if it *could* run with the
selected fsb-clocks (& overall clocks).
regarding the p4 plug. leave it plugged into the board. it's there for a reason,
so keep it that way, it's better and all...
regarding 1.6v 24/7. well writing this reply i don't see your sig ;), but anyway,
i would make that choice depending on what your temps are loaded/unloaded
- i try to keep my rig below 50°C loaded.. but then again i'm still only using
the default asus-heatsink. my 24/7 clocks & voltages are this:
266x10 @ 1.52v (or whatever that exact setting in the bios is). it works with
1.51x, but i usually up it one notch for better stability. that way i can run
it all day and all night. on benching i go up to 1.6v and ~2.8ghz (barely
stable enough) resulting in about 54°C loaded, but i wouldn't run it at 1.6v
24/7 with that stockcooler. using a half decent waterblock, like that zalman
zm-wb2 (that i still need to find over here for a decent price) or some kind
of phasechange, i wouldn't think about it. with phase i would probably run
upwards of 1.7v if it would yield significant gains...
true!!! 8x doesn't work on my P4C800... 9x++ work great!! Sky, your 311fsb, is that 1:1?? The best I can do 1:1 is 295fsb,,, with some G.Skill.Quote:
Originally Posted by sky
C
Should i go with the 7x0 dothans instead of the 7x5? Ive heard rumors of 7x0 doing better.
Im thinking of buying dothan rig and might use it for 24/7 action with ghetto mounted watercooling. It just has to do better than my 2,8Ghz Venice :D