ASUS P5Q-E (P45) pre-view

[Cloud76]

@ thehinge1
Please submit your impression about bios 1201 compared to 1103 if you are going to flash it.

[siyah]

Siyah forgot to ask but could you post your Q6600 at 3.6 Ghz bios settingsÉ

you can find my bios settings here

Q6600@3,6 on air

[egale]

I have problems with temperature reporting still with 1201. Probe reports a mb temp of 254 and cpu of 111 centigrate! Everest reports the mb temp at either 1 or 25 and CPU at 23.

This along with the fan sensor problems I am having is making be seriously consider switching to another brand of motherboard.

[thehinge1]

Will do, been preoccupied with other things today, but I should have a while to play with it tomorrow... can get just short of good mem scores at 550fsb 1:1 on bios 1103 but got kinda frustrated after hitting DDR1224 on the now-gone P5K with great bandwidth (at only 461fsb though). I may concentrate on 533fsb 1:1 as I'm about to go 4GB 2x2GB Reapers on the P5Q-E and bios 1201 is ready to go

Here is a functional 550fsb setting in the meantime:
ASUS P5Q-E Template :

*Note*: I lost my original 550fsb setting that definitely had some tighter timings and lower voltages in places, but this setup is bomber thus far. I will update again when I find some sweeter spots

Cpu Frequency with this setup is 3.85GHz, initial SuperPI 1M time is 13.235 sec - I think very low 13's or even lower may be possible.

CPU is M0 stepping E7200 retail boxed, air cooled w/TRUE120 & 1500rpm Scythe/Sony FDB SFF21F 120mm
P5Q-E rev. 1.01G on bios 1103
RAM used: OCZ Reaper PC2-9200/DDR1150
No fans over 900rpm in case to keep things nice and quiet

JumperFree Configuration Settings
AI Overclock tuner: MANUAL
CPU Ratio Setting: 7.0
FSB Frequency: 550
PCI-E Frequency: 100
FSB Strap to North Bridge: AUTO
DRAM Frequency: 1000 (BIOS says DDR2-1102)
DRAM CLK Skew on Channel A1: AUTO
DRAM CLK Skew on Channel A2: AUTO
DRAM CLK Skew on Channel B1: AUTO
DRAM CLK Skew on Channel B2: AUTO
DRAM Timing Control: MANUAL

1st Information : 5-5-5-18-11-36-6-11

CAS# Latency: 5
DRAM RAS# to CAS# Delay: 5
DRAM RAS# Precharge: 5
DRAM RAS# Activate to Precharge: 18
RAS# to RAS# Delay : 11
Row Refresh Cycle Time: AUTO (for this RAM is 36)
Write Recovery Time: 6
Read to Precharge Time: 11

2nd Information : 11-5-5-4-5-5-6

READ to WRITE Delay (S/D): 11
Write to Read Delay (S): 5
WRITE to READ Delay (D): AUTO
READ to READ Delay (S): AUTO
READ to READ Delay (D): AUTO
WRITE to WRITE Delay (S): 5
WRITE to WRITE Delay (D): AUTO

3rd Information : 14-13-1-5-5

WRITE to PRE Delay: AUTO
READ to PRE Delay: AUTO
PRE to PRE Delay: AUTO
ALL PRE to ACT Delay: AUTO
ALL PRE to REF Delay: AUTO
DRAM Static Read Control: Disabled
DRAM Read Training: Disabled
MEM. OC Charger: AUTO
AI Clock Twister: AUTO
AI Transaction Booster: AUTO (I think it is PL-13 at this clock freq)

CPU Voltage: 1.33750
CPU GTL Voltage Reference (0/2): AUTO
CPU GTL Voltage Reference (1/3): AUTO
CPU PLL Voltage: 1.66
FSB Termination Voltage: 1.28
DRAM Voltage: 2.04
NB Voltage: 1.34
NB GTL Reference: AUTO
SBridge Voltage: AUTO
PCIE SATA Voltage: AUTO

Load Line Calibration: Enabled
CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled
CPU Clock Skew : Delay 400ps
NB Clock Skew : Delay 200ps
CPU Margin Enhancement: Optimized

Advance CPU Settings
CPU Ratio Setting: 7.0
C1E Suppport: Disabled
Max CPUID Value Limit: Disabled
CPU TM Function: Disabled
Execute Disable Bit: Disabled

Anyone out there please feel free to warn me if you think I'm at 'danger level' on any of these voltages.

@ thehinge1
Please submit your impression about bios 1201 compared to 1103 if you are going to flash it.

[Cloud76]

CUT

Anyone out there please feel free to warn me if you think I'm at 'danger level' on any of these voltages.

I think your 1.66 CPU PLL voltage is high for a daily use, expecially because you are running a 45nm CPU.
Real voltage, considering mobo overvoltage, must be 1.70-1.72 or even a bit higher.
Are you sure you need so much? I think that around 1.54 (bios setting) can be enough, I would no go much more high with a 45nm CPU for daily.

can get just short of good mem scores at 550fsb 1:1 on bios 1103 but got kinda frustrated after hitting DDR1224 on the now-gone P5K with great bandwidth (at only 461fsb though)

to improve bandwidth first try to enable "Static read control", it improve performance but decrease stability, test it and if all is good try to lower performance level. I don't know if you read P5Q DL discussion, however, the 1:1 FSB/RAM is the worse divider for PL (13... max 10...), for better PL (max 8... or 7 if you are fortunate) you must use other divider.

[wongnog]

New bios official 1201 released 11-Aug-2008 on Asus website. Pretty much same changelog as previous vers, "Enhance the memory compatibility" according to Asus.

[thehinge1]

I think your 1.66 CPU PLL voltage is high for a daily use, expecially because you are running a 45nm CPU.
Real voltage, considering mobo overvoltage, must be 1.70-1.72 or even a bit higher.
Are you sure you need so much? I think that around 1.54 (bios setting) can be enough, I would no go much more high with a 45nm CPU for daily.

to improve bandwidth first try to enable "Static read control", it improve performance but decrease stability, test it and if all is good try to lower performance level. I don't know if you read P5Q DL discussion, however, the 1:1 FSB/RAM is the worse divider for PL (13... max 10...), for better PL (max 8... or 7 if you are fortunate) you must use other divider.

Thank you! I think I can get it to 1.58 stable by increasing CPU GTL's about .02 each, 1.54 & 1.56 were no-go. I even killed my BIOS and had to recover then flash to v1201 when I started playing with NB GTL (it is back on auto for now).

How high is a safe CPU GTL with 45nm and Vcore set at 1.22750 in BIOS?

Yes, I am aware of the 1;1 issues but I want to find the 'fairly reasonable' topend of the chipset and am still very early in testing this board.

I may just back down a bit on some voltages and run a lower fsb with divider to boost the RAM... but I am actually getting close to my old P5K's stats when it was at 461/1224 and maxed out.

[mojado]

hello people, I have the same specs that thehinge1 (E7200 + P5Q-E + Reaper 1150).
I followed step by step your BIOS configuration but I can´t pass 525 to FSB.

What´s something wrong? Any idea? Please, any suggestion??

Thanxs in advance.

[thehinge1]

Are you sure your RAM timings are correct for Reaper 1150? Your screenshot shows 5-5-5-15 and OCZ designed them to be set at 5-5-5-18, I also have Reaper 1066 and they are rated 5-5-5-15. It does make a difference with my kit.

Are the rest of your RAM settings the same as mine? I have gone manual with every single setting in the RAM area at this point and can post them if you want.

Additional unsolicited info: I have gotten my rig to 533fsb/DDR1066 with much some pretty decent (lower) voltage settings, and can also do 450fsb/DDR1200 or 500fsb/DDR1200 on the 1150's after some more experimentation. Ultimately I'm looking for the fastest configuration for my cpu regardless of indicated fsb or RAM speed.

hello people, I have the same specs that thehinge1 (E7200 + P5Q-E + Reaper 1150).
I followed step by step your BIOS configuration but I can´t pass 525 to FSB.

What´s something wrong? Any idea? Please, any suggestion??

Thanxs in advance.

[Cloud76]

New bios official 1201 released 11-Aug-2008 on Asus website. Pretty much same changelog as previous vers, "Enhance the memory compatibility" according to Asus.

At the moment I'm writing Asus removed 1201 and also 1103.

Thank you! I think I can get it to 1.58 stable by increasing CPU GTL's about .02 each, 1.54 & 1.56 were no-go. I even killed my BIOS and had to recover then flash to v1201 when I started playing with NB GTL (it is back on auto for now).

How high is a safe CPU GTL with 45nm and Vcore set at 1.22750 in BIOS?

Yes, I am aware of the 1;1 issues but I want to find the 'fairly reasonable' topend of the chipset and am still very early in testing this board.

I may just back down a bit on some voltages and run a lower fsb with divider to boost the RAM... but I am actually getting close to my old P5K's stats when it was at 461/1224 and maxed out.

I am a little 'concerned, I saw dying too many bios in the forum

About GTL, you can play with them to stabilize your OC, I don't know if there is a safe area, I made a little research, you can find some information about GTL at the following link:

http://www.thetechrepository.com/showthread.php?t=87

http://www.edgeofstability.com/artic.../gtl/gtl1.html

http://edgeofstability.com/articles/...fault_gtl.html

http://www.overclock.net/intel-mothe...s-formula.html

If you want to read them... hope it could help.

[mojado]

Are you sure your RAM timings are correct for Reaper 1150? Your screenshot shows 5-5-5-15 and OCZ designed them to be set at 5-5-5-18

But the OCZ EPP profile say 5-5-5-15 @2,1v.

[thehinge1]

@Cloud76: I went back 2 BIOS to 1003, 1201 worked (before it blew up, lol) but the temps read 10c high... ASUS is REALLY starting to get me cheesed off between the defective P5K Pro and now continued bios issues on P5Q-E. I have already read those GTL articles, but thanks nonetheless and the small increase I tried was probably fine - I was just looking for yet more info!

@mojado: EPP is really optimized for Nfarce chipsets from what I have read, they co-developed the standard. OCZ even has the 5-5-5-18 settings on their product page for our RAM as preferred for Intel chipsets and it is plenty speedy. I knew you are on Intel chipset so I did not bother with the EPP stuff.

Question: Do your reaper stix have a white sticker without a max voltage (old style) or a 'chrome/holographic' sticker that says '2.1v'?

OCZ changed the chips back in January(maybe December) from the old 2.35v max voltage rating to the new 2.1v one as the new chips need much less voltage. Mine are new style and running at DDR1216MHz@2.03v as I type this. If you are new style I'll send my whole settings to you or post them here, I'm running bios1003 at 507fsb/DDR1216 quite stably on the 5:6 divider now with very reasonable voltages.

[Dee]

OK, straight up replies if you will -- I can get the P5Q Deluxe at a great price. Is it worth it over the PQ5-E which I can get cheaper?

[thehinge1]

If I was going to use a quad I would have gone Deluxe for its 16-phase PWM, I chose P5Q-E's 8-phase for my dualie. The 'E' even seems to clock ever-so-slightly higher on dual-cores. I was looking at a $40 US price difference, and am happy. Straight up enough?

OK, straight up replies if you will -- I can get the P5Q Deluxe at a great price. Is it worth it over the PQ5-E which I can get cheaper?

[Dee]

Yep! Thanks very much. I'm going for an E8600 in a couple of days so I'll keep your advice in mind. I saw a couple of people mention that they think the P5Q-E is the best choice for duals. Thanks again bro.

[wongnog]

Odd, ASUS has removed bios 1103 and 1201 from their website, latest one up is 1003. All are still available on their ftp site.

Does anyone else find they get very slow boot times when they plug something in this board on IDE? I have on DVD+RW drive, and my Vista x64 start times are soooo slow when I enable IDE on the board. The drive seems to make audible clicking noises too when Vista is starting up (during that loading screen with the green progress bar). As soon as I disable IDE, Vista starts up in seconds. Time for me to get a new drive I guess?

[Loser777]

I just flashed 1201... XD.

[thehinge1]

Noooooooooooooooooooo!

ASUS pulled it and 1103 off of their site... It worked great for me till it corrupted violently yesterday North of 550fsb when I got the NB GTL too low. Kaboom!

Anyhow - 1003 is working really well for me thus far and is quite stable fwiw.

I am researching CPU GTLs right now and they ARE your friend - looking like a great way to get more FSB without actually putting more voltage through your core! More testing to see if the theory holds water, but I definitely have shaved some Vcore off thus far without any obvious ill effects. I imagine the same could be done on the NB when I get brave enough to return there.

I just flashed 1201... XD.

[Leeghoofd]

Noooooooooooooooooooo!

ASUS pulled it and 1103 off of their site... It worked great for me till it corrupted violently yesterday North of 550fsb when I got the NB GTL too low. Kaboom!

Anyhow - 1003 is working really well for me thus far and is quite stable fwiw.

I am researching CPU GTLs right now and they ARE your friend - looking like a great way to get more FSB without actually putting more voltage through your core! More testing to see if the theory holds water, but I definitely have shaved some Vcore off thus far without any obvious ill effects. I imagine the same could be done on the NB when I get brave enough to return there.

GTL's are more important then you would think. For the bios dissapearing and popping up again, Asus got server issues ( what's new lol ) and sometimes the mirroring and co goes wrong and users see the old list... just go via the ftp and it should be fine... there's nothing wrong with the bios, as some suspect it has been pulled or such... ofcourse for some users it's fab while for others it might give issues... too much hardware setups floating around...

I kindda like 1201 on the E and Dlx...

[TI66ER]

1201 is fine on my p5q-e,my temps seem to report ok and i have zero issues.

[roubagalinhas]

i have a p5q-e bios 1103 + E7200 + 4GB DDR2 1000MHz+

i cant boot windows over 440FSB, it stops always @ windows loading bar
i've tried raising pll, nb, vsb, and other values without success, as well as straps
tried other bioses as well

the previous owner of this cpu guarantees it makes at least 550FSB(on P35), and this board made 500FSB @ minimum voltages with my previous E6600

any clues?
my current config is cpu @ 400*9.5 1.3v, mems 500MHz 1.8v, all board voltages @ "minimum"(first value above minimum)


thanks

[thehinge1]

Turn up the 'suck'! (bios 1201)

My temps are fine now, 'cept OCCT has its knickers in a twist because AIProbe, CoreTemp, and RealTemp are all a-ok. Weird because all were 10c high except CoreTemp in my earlier bios exploits.

I was finally getting some good speed with lower voltage, decided to flash back up to 1201 since it worked well initially. No stability, possible issues between 'desired' and actual voltages.

So now I am back @bios1103 and she seems to run fine again. What a total what of time trying to test my friend's Platinum 1066 while my board was borked

My P5W-DH may not run high fsb over low 400's but at least it is dead-nuts reliable.

[Leeghoofd]

Turn up the 'suck'! (bios 1201)

My P5W-DH may not run high fsb over low 400's but at least it is dead-nuts reliable.

Think this is more end user related then board related... your best friend is a digital multimeter... some voltages have been adjusted apparently and need to be redialed to get stable again...

1201 is rockstable on 2 E's and on the Dlx as well... all CPU's at 4000Mhz : 2 wolfies at (8 x 500) and one Qx9650 (at 10 x 400) on the dlx mobo. All rigs folding 24/7 at 100% and the quadcore is used as game platform too...

Really no hickups here

[wongnog]

FWIW, I seem to be stable at 9x400 on my Q6600 (G0) with bios 1201. Wasn't even able to boot into Vista before with other bioses. Will post some screens when I get a chance to fully stress test it

[thehinge1]

Part of my frustration is exactly that - I just had a shoulder reconstructed so I cannot get the darned thing apart to properly bench it. Although here's the weird thing: ever since I put 1201 on it, if I do crash I can't just flip the power switch on the PSU & 'F1' to get back into BIOS upon powering back up - now it makes me clear the jumper and nothing else has changed in my setup. I've never had this type of issue in dozens of builds so I think I had/have some type of BIOS anomaly??? I THOROUGHLY cleared it, then total erased/reflashed to 'known good' settings in bios 1103 at 507fsb/DDR1216 and all seems well again.

I wish I could just yank the board out, plop it down on the bench & then use both digital and analog MM's on it. The odd 'continuous reboot' cycle was exhibited in bios 1003, 1103 and 1201, and started with 1201.

It just doesn't seem to be the same after the big kaboom

Got any ideas that may help me while temporarily one-armed?

Tx

Think this is more end user related then board related... your best friend is a digital multimeter... some voltages have been adjusted apparently and need to be redialed to get stable again...

1201 is rockstable on 2 E's and on the Dlx as well... all CPU's at 4000Mhz : 2 wolfies at (8 x 500) and one Qx9650 (at 10 x 400) on the dlx mobo. All rigs folding 24/7 at 100% and the quadcore is used as game platform too...

Really no hickups here