Cant boot at 400fsb = P5W DH.

[Jimbo]

I'm trying to run 7x400 fsb on my E6600 and I was hoping it would work at that lower multi so I could reduce vcore and therefore bring down temps at load on the cpu.

Dosent even get to the post screen and I have to power off to get back in the bios.

I'm running at the mo - 9x333fsb mem @ 833mhz vcore and mem volt auto, and nb/sb 1.50v/1.05v

Any ideas?

[Dani]

Try 401MHz

[taemun]

I can boot at 8x400 fine here ... no raised vMCH or anything ...

t

[Jimbo]

Well I treid 400mhz and ram at 800mhz but it wont work, cant see 401mhz working mate.

Ram is Gskill 6400 NR. @ 5/5/5/15.

[Mykou]

Well I treid 400mhz and ram at 800mhz but it wont work, cant see 401mhz working mate.

Ram is Gskill 6400 NR. @ 5/5/5/15.

Seems like your mobo not from a good batch as mine

can boot@3.954 ghz but not at 400 fsb

[caater]

vnb 1.65v and mem volt to.. 2.1-2.2v maybe. dunno what chips NR kits are using..

[Icer]

Sooo, your currently @ 9x333 = 3GHz but you want to run @ 7x400 = 2.8GHz? If you want to run at a lower GHz, lower Vcore, and lower volts, why not just try 9x300 = 2.7GHz.

[Jimbo]

If that would let me lower vcore I suppose I could try it, thanks for that. Just until I get a Tunic Tower 120 lol.

[Durzel]

I'm trying to run 7x400 fsb on my E6600 and I was hoping it would work at that lower multi so I could reduce vcore and therefore bring down temps at load on the cpu.

Dosent even get to the post screen and I have to power off to get back in the bios.

I'm running at the mo - 9x333fsb mem @ 833mhz vcore and mem volt auto, and nb/sb 1.50v/1.05v

Any ideas?

See this post:
http://www.xtremesystems.org/forums/...5&postcount=23

Long and short of it is - I need 1.3250v vCore to dual Prime stable at 10x340 (3400Mhz), but I need 1.4250v vCore just to POST at 6x410 (2460Mhz).

[Icer]

If that would let me lower vcore I suppose I could try it, thanks for that. Just until I get a Tunic Tower 120 lol.

You say you already have Vcore set to auto? How much lower are you trying to get?

[CopperCAT]

See this post:
http://www.xtremesystems.org/forums/...5&postcount=23

Long and short of it is - I need 1.3250v vCore to dual Prime stable at 10x340 (3400Mhz), but I need 1.4250v vCore just to POST at 6x410 (2460Mhz).

Does this mean that the P5WDH boots up with a 9x multi and then switches back to the user-setting?

[caater]

Does this mean that the P5WDH boots up with a 9x multi and then switches back to the user-setting?

yup.
p5w and p5b change multiplier by modifying exactly the same cpu register as, for example, crystalcpuid does.
first, post at default values and then bios applies new multiplier and boots up.
ES and XE chips are different - they are true unlocked cpus.
i totally missed that one from first post
indeed, to post at 7x400 you have to give cpu enough vcore like post at 9x400.

[mach82]

I'm trying to run 7x400 fsb on my E6600 and I was hoping it would work at that lower multi so I could reduce vcore and therefore bring down temps at load on the cpu.

Dosent even get to the post screen and I have to power off to get back in the bios.

I'm running at the mo - 9x333fsb mem @ 833mhz vcore and mem volt auto, and nb/sb 1.50v/1.05v

Any ideas?

My P5W DH couldn't boot at FSB 400 MHz, and the max value was 390 MHz. That's quite normal, that some can get past 400 MHz and some not.

[mach82]

yup.
indeed, to post at 7x400 you have to give cpu enough vcore like post at 9x400.

that's weird, 'cause I tried 7x450 1.3V with an E6400 and 8x400 1.3V with the same CPU. Both ran fine. Yet it couldn't boot at 450x8 with 1.3V. So what you've written doesn't add up, as far as my experience with P5B DLX and P5W DH is concerned.

[Durzel]

yup.
p5w and p5b change multiplier by modifying exactly the same cpu register as, for example, crystalcpuid does.
first, post at default values and then bios applies new multiplier and boots up.
ES and XE chips are different - they are true unlocked cpus.
i totally missed that one from first post
indeed, to post at 7x400 you have to give cpu enough vcore like post at 9x400.

Yup, you're right.. except ES CPUs are the same. Don't know about XEs though - Im guessing from that fact they've always been unlocked that there's something special about them internally.

From my experience (I've got an E6700 ES) I can't boot at 6x410 unless I give it 1.4250v vCore - which is way more than is required for the clockspeed (2460Mhz). I can dual prime stable for 18+ hours at 3.6Ghz (10x360) at that vCore.

My hunch, like you said, is that Asus multiplier unlocking works the same as CrystalCPUID - but that when you first POST it will try and boot at whatever your default multiplier is, which in my case is 10x... this would explain why I need so much voltage to boot if its trying to boot at 10x410 (4100Mhz).

that's weird, 'cause I tried 7x450 1.3V with an E6400 and 8x400 1.3V with the same CPU. Both ran fine. Yet it couldn't boot at 450x8 with 1.3V. So what you've written doesn't add up, as far as my experience with P5B DLX and P5W DH is concerned.

What CPU have you got? E6400? If you're giving the CPU 1.3v vCore, which you say is fine for either 7x450 or 8x400, then maybe your PC is capable of booting at 8x450 (3.6Ghz), even if it wouldn't get into Windows at that speed (which wouldnt matter anyway because Im guessing the multiplier gets changed as soon as it properly POSTs).

Jimbo on the other hand has an E6600, and if hes trying to boot at 7x400 then I think its initially POST'ing at 9x400 (3600Mhz), which is clearly too much for stock voltage.

[itznfb]

7x400 = 514MHz NBCC, maybe you need to up the volts on your NorthBridge?

[Jimbo]

You say you already have Vcore set to auto? How much lower are you trying to get?

I have only so far been succesful botting at 1.3v anything below that and my current oc doesnt work.

[caater]

7x400 = 514MHz NBCC, maybe you need to up the volts on your NorthBridge?

i'm afraid that multiplier has nothing to do with nbcc..
but that's possible that his nb voltage is too low for 400 fsb.

jimbo, have you tried to bootup at 333, mem at 1:1 or 4:3, lowering mult with crystalcpuid and using setfsb/clockgen to determine your max fsb with your current settings?

[itznfb]

i'm afraid that multiplier has nothing to do with nbcc..

NBCC = DefaultMulti/SetMulti * FSB
http://www.xtremesystems.org/forums/...d.php?t=114998

[caater]

NBCC = DefaultMulti/SetMulti * FSB
http://www.xtremesystems.org/forums/...d.php?t=114998

you can easily disregard that nbcc information, it's just sandra bug
also control panel->system and various older programs have trouble detecting real clockspeed.
they just don't know how to read fsb
just use clockgen/setfsb and see if sandra detects new fsb..
but what sandra is capable of, is reading multiplier.
default multiplier is readable from cpu registers (msr 2Ah, bits 26:22), bootup frequency is also supposedly readable from msr (i think from E7h or/and E8h).
if multiplier is changed, sandra detects it, assumes that clock speed is the same (because it lacks support for reading from clockgen) and displays that weird clock speed. and that weird clock speed is indeed DefaultMulti/SetMulti * FSB but that's not nbcc the real nbcc still is FSB * multi_Defined_by_strap.. or smth like that. as explained by Tony.

[itznfb]

yea, so you're saying all the information that Tony and FCG provided is false?

[caater]

yea, so you're saying all the information that Tony and FCG provided is false?

well, it was FCG's post, so unfortunately he made an error this time
Tony's posts about straps and so on are very true and quite helpful.

besides, if you think carefully, this statement also defies logic.. A laptop with T7600 in low-power mode (6x mult) would have 387MHz nb instead of 166MHz
not exactly power-saving..

[itznfb]

well i can't find any evidence of NBCC=xDefault/xSet * FSB being incorrect. the only posts i can find prove it to be true. i don't have time however to read through all the posts right now. do you have evidence pointing to anything that disproves the NBCC or bandwidth results?

[caater]

do you have evidence pointing to anything that disproves the NBCC or bandwidth results?

bandwidth tests (tonys findings) are correct for p5b and there's no need to disprove these. and they are totally unrelated to fcg's theory about nbcc.
and nbcc - read my earlier post about that.

but let's concentrate here on Jimbo now

[itznfb]

well, this is directly related to the ontopic issue. if there is an increase in NBCC related to the multiplier defaul/set ratio, whether it resets with straps or not could be causing his problems. which would explain the cases were some can boot at 401, but not 400.