Asus P5B (vanilla) Overview

[Thasp]

New Version allows 2.45v vdimm.

New BIOS version or new board? I'd check but the download section at ASUS's site loads at about 20 kbps for me.

If new board, unlucky me, I just bought it today

[ziddey]

where'd you buy from. newegg's been shipping rev1.04g's for quite a while. I got mine back in Feb.

Like I said though, exercise caution with the different vdimm settings, and do note that if you set higher than 2.1v (2.15v in actuality, since it's the same thing), aisuite will crash on load.

[Rockhammer]

So the 1.04g boards have the option for higher vDIMM in the bios? Is there some special trick to access these settings, I don't see it in mine.

[ziddey]

hmm maybe they addressed it in later versions. it probably wasn't meant to be since setting higher than 2.1v would cause aisuite to crash, and higher values were not what was actually being set.

you can vdimm pencil mod for 1.04g. that should be mroe than sufficient

[***Deimos***]

Hey Ziddey,

I noticed you seem to have a lot of experience with P5B.. perhaps you could help out.

Using stock cooling, P4B 1.04G, all volts auto, I got E6420 up to 375 fsb (Corsair at 2.1V, and well below its rated clocks). I tried raising all volts up to 2nd highest and it could only get 380 fsb and no more. Tried again with 6x multi, and same thing.

Any suggestions?

EDIT: yes the P5B 1.04G has higher range voltage for Vdimm, Vnb etc, in the bios.

[The Coolest]

I heard these boards have problems with 333 - 400FSB range. Try 401MHz and see what happens

[***Deimos***]

I heard these boards have problems with 333 - 400FSB range. Try 401MHz and see what happens

same problem. tried 405 and 410 too for good measure.
tried increasing CPU voltage, Vnb, Vfsb... nothing.
used 6x multi. disabled TM, disabled halt state, turned off static read.

when I get back home I'll try upping the south bridge voltage... its just about the only think I havent tried...

[ziddey]

what ram are you using? try using just one stick at first if you're really stuck.

try loose subtimings to rule them out as well. 5-5-5-15-6-42-4-11-7-13. ram 1:1, not 4:5. static read off. try 105 pcie.

You shouldn't need more than 1.4vnb, but try as high as 1.55v for the time being (do note until you verify vnb with dmm, you can't trust what you set)

vfsb should be fine at stock, but bump it one if you must.

how hot is the nb getting?

[***Deimos***]

what ram are you using? try using just one stick at first if you're really stuck.

try loose subtimings to rule them out as well. 5-5-5-15-6-42-4-11-7-13. ram 1:1, not 4:5. static read off. try 105 pcie.

You shouldn't need more than 1.4vnb, but try as high as 1.55v for the time being (do note until you verify vnb with dmm, you can't trust what you set)

vfsb should be fine at stock, but bump it one if you must.

how hot is the nb getting?

thank you for your suggestions.

I have Corsair 6400C4. Have not tested well, but ran games etc at 800Mhz 2.0V 4-4-4-12.. and at 2.1V got at least 890Mhz also to desktop to run programs...

so far all things point to NB/motherboard/CPU causing the fsb wall. The CPU is certainly able to do much higher clocks at 1.25V since it gets stuck at same max fsb with 6x and 8x. Could be CPU cant handle higher fsb. Maybe some poor connection of IC on motherboard... cant tell unless I try different mobo/CPU...

[blind_ripper]

strange , with quad the temp probe will not go below 6°C when putting cascade on the board

[daimaah]

There's a new BIOS, probably beta since it's not been announced on the download pages...
ftp://ftp.asus.com.tw/pub/ASUS/mb/so...B/P5B-1405.zip

1. Update CPU ucode to support new processor
2. Revise SLP 2.0 rule
3. Enable hotplug support under AHCI mode
4. Fix displayed CPU frequency incorrectly after downgrade CPU ratio

[***Deimos***]

There's a new BIOS, probably beta since it's not been announced on the download pages...
ftp://ftp.asus.com.tw/pub/ASUS/mb/so...B/P5B-1405.zip

1. Update CPU ucode to support new processor
2. Revise SLP 2.0 rule
3. Enable hotplug support under AHCI mode
4. Fix displayed CPU frequency incorrectly after downgrade CPU ratio

THANKS!

EDIT: new BIOS didn't fix any of my issues... except that the frequency is reported correctly now when using 7x multiplier on E6420 (default 8x).

[ryandinan]

So, will this board now support quad core cpu's? Did it before...?
-Ryan

[Rockhammer]

Yes.

http://event.asus.com/mb/quad_core/

[***Deimos***]

P5B vanilla can do 440fsb at least
different E6420

[Volkswagen]

what do u guys use to cool the northbridge and the heat pipe joint ? I just added a 40mm fan just in case even tough I am doing only like 460fsb? Also does the nb Voltage need to be raised for that type of fsb?

[ziddey]

which board are you referring to? the p5b vanilla doesn't have any heatpipes. what vnb are you using? it definitely wouldn't hurt to glue a 40mm on the heatsink

[jacky89]

Has anybody tested the new bios released 06/27/07?

[***Deimos***]

Has anybody tested the new bios released 06/27/07?

I have tried 1405 for P5B. No real advantages.

BTW: Sorry to repeat, but this is P5B thread NOT P5B Deluxe, which has different layout and different BIOS, and different cooling.

P5B - small regular NB heatsink. No RAID. No firewire. newest BIOS 1405. Many component for power circuitry are smaller or missing.
P5B Deluxe - heatpipe cooling solution. All the goodies. Better power circuitry. Better overclocking

[elec999]

Is there any way to unlock more options in the bios.
Thanks

[ziddey]

like what
it'd help to say what you want to accomplish

[Timbosan]

Hey guys i recently bought some new gear and this thread has been really helpful to me. My setup consists of:

e6300 week 27, stock intel HSF
asus P5B (vanilla) rev. 1.03G (no mods)
2 x 512MB Corsair XMS2 DDR2-800
POS video card (8800GTS coming soon)
Thermaltake Toughpower 750w PSU

OC so far (100% stable):
cpu @ 3.2 GHz (458 x 7)
vcore = 1.29v (real, idle)
FSB = 1832 MHz
vMCH = 1.20v
vdimm = 2.1v
Ram = 916 MHz
Ram Timings (read down the screen) 5-5-5-15-6-42-6-12-10-18

This board is great. It has no trouble reaching 470+ mhz at only 1.20v (ram not stable tho)

[Timbosan]

I have the data sheet for RT9214.

http://www.richtek.com/www_en/Docs/DS9214-06.pdf

I may have worked out the pencil mod. Pin 6 is FB and pin 3 is GROUND - if i drew line in no#2 pencil across IC from pin 6 to pin 3 it would decrease pin 6 FB signal and increase output voltage. I have done similar mod to my 6800 vanilla to incease memory voltage. I'll try it later tonight with some crap ram. Anyone wanna try it first on a P5B they dont need?

[ziddey]

that's exactly right, but since the ic has 8 pins, 3 and 6 are on different sides. by drawing from one side to the other, you won't get much resistance drop at all.

if you can't find a resistor in the area that actually has continuity between leg6 and ground to shade, get conductive ink and draw from one leg to just about the other leg. then use pencil to try to finish the loop. that'd allow more resistance drop.

[Timbosan]

that's exactly right, but since the ic has 8 pins, 3 and 6 are on different sides. by drawing from one side to the other, you won't get much resistance drop at all.

if you can't find a resistor in the area that actually has continuity between leg6 and ground to shade, get conductive ink and draw from one leg to just about the other leg. then use pencil to try to finish the loop. that'd allow more resistance drop.

thats what i did on my video card - made the middle 80% of the line with conductive silver & joined the pins to it with pencil. I'll try to find a resistor with continuity to pin 6 at one end & other end to ground and pencil that, but if i cant i'll try connecting the IC leg 6 to IC leg 3.

There can only be 3 possible resistors, R2, R3 or R4 in RT9124 data sheet.