X3350 / Q9450 Overclocking Thread

[tenax]

BIOS release V16 for IP35 Pro

Release information:

[ENGLISH]

1. Update AHCI ROM version.
2. Update CPU Micro code.
3. Enhanced 45nm CPU compatibility.
4. Fixed C1E function abnormality.
5. Fixed Quick Power On Self Test function abnormality.
6. BIOS compiled date: 03/18/2008

there ya go..and i just installed it using the windoze abit program. installed fine.

 

[tenax]

well, what do you think it means when they tweak up the heat on the northbridge? (which i would assume means they've played with the voltage somehow) my northbridge wen tup from 43 at a 3500 overclock to 48C, even though i actually lowered it a few notches when i reset my overclock settings post the new bios install. i kid you not..my system temp stayed the same as before (31C) but my pwm went up!

 

[tenax]

Originally posted by: tenax
well, what do you think it means when they tweak up the heat on the northbridge? (which i would assume means they've played with the voltage somehow) my northbridge wen tup from 43 at a 3500 overclock to 48C, even though i actually lowered it a few notches when i reset my overclock settings post the new bios install. i kid you not..my system temp stayed the same as before (31C) but my pwm went up!

edit- it went down to 45 now..i can live with that. and so far in playing with various fbs, multi's etc, did nothing to change my overclock abilities

 

[viol37]

I used to own a Q6600 running all at default speed/voltages using the standard HSF and I was hitting approx 50/55 at idle reading from the BIOS.

I upgrade my HSF to the Zalman 9700NT and my temps dropped down to approx 30/35 so I got a major improvement just be investing in a decent HSF (Zalman).

I am now running a Q9450.

 

[T2k]

Originally posted by: tenax

Originally posted by: themattman
I'm glad that these Q9450's can easily push 3Ghz and many users are approaching (or passing) 3.5 Ghz. I've been planning to get a completely new system, but the infinite delay of E8400's and Q9450's has kept me waiting. (My AMD Athlon XP-M 2500+ @ 2.3 is starting to get a bit long in the tooth)

For the average overclocker with a Zalman or a TRUE, what would be an attainable overclock without having to resort to extreme measures like changing more than just the cpu/ram voltages? I know YMMV, but is around 3.4 out of the question?

Btw, I would love to see some review sites put the Oc'ed E8400 against the Oc'ed Q9450. I want the quad for folding, but I want to make sure that the performance is up there with the E8400.

3400 with only a cpu voltage change is doable and stable for a 9450. it's above 3400 that starts getting a bit dicey. 3500-3600 is easily obtainable, but you start trading off higher voltages. by 3800, you're pressing the voltage up in the 1.4 range on the cpu which is getting a bit deadly and same deal with mch..up in the 1.5 range or higher. i think once you start pushing a penryn past 1.35, you should ask yourself what your goal was in buying..i know mine was cool and quiet with a decent overclock so with that in mind..i want less than 1.3 volts, 3200-3400 stable and a very quiet computer..so i'm happy.

how are you juding performance of an oc'ed quad with an oc'8400? the 8400 will kick the 9450's butt around the block, between having one full multiplier more, able to easily get over 4000 compared to the easily over 3200 of the quad

Why people keep repeating the same, false urban legends over and over again?
There's nothing "deadly" in 1.5V - it simply depends on your cooling. A short visit to the forum of Xtremesystems will tell you people go even higher regularly. My Q6600 was running at 1.5xx stable, all I had to do is upgrade my CPU block to make it stable @3.8xGHz.

 

[fleshconsumed]

Originally posted by: T2k
Why people keep repeating the same, false urban legends over and over again?
There's nothing "deadly" in 1.5V - it simply depends on your cooling. A short visit to the forum of Xtremesystems will tell you people go even higher regularly. My Q6600 was running at 1.5xx stable, all I had to do is upgrade my CPU block to make it stable @3.8xGHz.

Here's link to the same xtremesystems forums, read it carefully

http://www.xtremesystems.org/f...howthread.php?t=175829

There have been reports of wolfdales dying from overvoltage. It's real. Wolfdale/Yorkfield is not Conroe. These are 45nm parts designed to run at a lower voltage than 65nm conroes. If you overvolt too much you risk frying your CPU sooner than later.

 

[bka4u2c]

I thought there was someone here on this forum that killed his 45nm chip using 1.5 voltage also. I wouldn't approach 1.5 without water cooling or something better and even then I would only do it for benchies. Not 24/7.

 

[Markfw]

Originally posted by: fleshconsumed

Originally posted by: T2k
Why people keep repeating the same, false urban legends over and over again?
There's nothing "deadly" in 1.5V - it simply depends on your cooling. A short visit to the forum of Xtremesystems will tell you people go even higher regularly. My Q6600 was running at 1.5xx stable, all I had to do is upgrade my CPU block to make it stable @3.8xGHz.

Here's link to the same xtremesystems forums, read it carefully

http://www.xtremesystems.org/f...howthread.php?t=175829

There have been reports of wolfdales dying from overvoltage. It's real. Wolfdale/Yorkfield is not Conroe. These are 45nm parts designed to run at a lower voltage than 65nm conroes. If you overvolt too much you risk frying your CPU sooner than later.

Yes, its not false urban legends. You have an opinion, thats fine, just don't try to convince others to fry their CPU.

 

[Foxery]

Originally posted by: T2k
Why people keep repeating the same, false urban legends over and over again?
There's nothing "deadly" in 1.5V - it simply depends on your cooling. A short visit to the forum of Xtremesystems will tell you people go even higher regularly. My Q6600 was running at 1.5xx stable, all I had to do is upgrade my CPU block to make it stable @3.8xGHz.

There's more going on inside a CPU than just heat. Extreme voltage causes electromigration. Detailed writeup in the Highly Technical forum here: (very long)
See posts by CTho

In a nutshell, no amount of cooling will stop the transistors from breaking apart when you push too much current through them.

 

[T2k]

Originally posted by: fleshconsumed

Originally posted by: T2k
Why people keep repeating the same, false urban legends over and over again?
There's nothing "deadly" in 1.5V - it simply depends on your cooling. A short visit to the forum of Xtremesystems will tell you people go even higher regularly. My Q6600 was running at 1.5xx stable, all I had to do is upgrade my CPU block to make it stable @3.8xGHz.

Here's link to the same xtremesystems forums, read it carefully

http://www.xtremesystems.org/f...howthread.php?t=175829

There have been reports of wolfdales dying from overvoltage. It's real. Wolfdale/Yorkfield is not Conroe. These are 45nm parts designed to run at a lower voltage than 65nm conroes. If you overvolt too much you risk frying your CPU sooner than later.

Thanks for the link though I knew that.

1. All that single affected CPU - an 8400, not a quad-core - has max Vcore recommendation of 1.45V - 0.05 away from 1.5V, which by itself cannot warrant for such sudden death unless Intel grossly overestimated their tresholds.

2. When you OC the rule of thumb is always try to achieve stability with maximum FSB with minimum voltage or you don't raise voltage for starter but rather when it's necessary and even then do it little by little - in other words you slowly ramp up your voltage and only when it becomes essential to achieve stability.

3. As they pointed out there it's really about "FSB termination voltage" settings as it supposed to be kept pretty low (some cases lower than some mobos minimum!), not Vcore,

4. Never trust a single temp source indicator. Almost all are off by a little and worst offenders are on-board diodes of motherboards.

5. Similarly I never trust BIOS nominal values when dealing with voltages - almost every X38 mobo I recently came across - tried, read about etc - overvolts.

6. Know your cooling solution. People kept telling me that my water-based cooling - for details see my config link in my sigline - is nothing better than a high-end air cooling when it comes to OC. Well I disagree, I'd never approach 1.5V with any air colling. Maybe it's just me but that radiator w/ 3x 12cm fans on it and pumps circulate a lot (dual 5.25" bay-sized) of water give me lot more confidence.

7. It's been two months now and I haven't seen en masse 8400 death from overvolting. This doesn't mean it cannot happen with the just-arriving QC Penryns easier but I find it unlikely. Anyway we just have to wait and see.

 

[T2k]

Originally posted by: Foxery

Originally posted by: T2k
Why people keep repeating the same, false urban legends over and over again?
There's nothing "deadly" in 1.5V - it simply depends on your cooling. A short visit to the forum of Xtremesystems will tell you people go even higher regularly. My Q6600 was running at 1.5xx stable, all I had to do is upgrade my CPU block to make it stable @3.8xGHz.

There's more going on inside a CPU than just heat. Extreme voltage causes electromigration. Detailed writeup in the Highly Technical forum here: (very long)
See posts by CTho

In a nutshell, no amount of cooling will stop the transistors from breaking apart when you push too much current through them.

Of course. Too high voltage will kill it, insane one even immediately, that's not a question.

However unless Intel didn't keep a 'safe margin' I don't see how 1.5V could be deadly.

 

[T2k]

Originally posted by: Markfw900

Originally posted by: fleshconsumed

Originally posted by: T2k
Why people keep repeating the same, false urban legends over and over again?
There's nothing "deadly" in 1.5V - it simply depends on your cooling. A short visit to the forum of Xtremesystems will tell you people go even higher regularly. My Q6600 was running at 1.5xx stable, all I had to do is upgrade my CPU block to make it stable @3.8xGHz.

Here's link to the same xtremesystems forums, read it carefully

http://www.xtremesystems.org/f...howthread.php?t=175829

There have been reports of wolfdales dying from overvoltage. It's real. Wolfdale/Yorkfield is not Conroe. These are 45nm parts designed to run at a lower voltage than 65nm conroes. If you overvolt too much you risk frying your CPU sooner than later.

Yes, its not false urban legends. You have an opinion, thats fine, just don't try to convince others to fry their CPU.

Maybe I wasn't clear enough, sorry - I took issue with his wording: 1.5V is NOT 'deadly', let's stop these urban legends here.
Dangerous? Of course, could be. But if you really want to play safe then do not OC or only whatever the safe voltage zone allows you.

 

[T2k]

PS: I'll have to see more of these problems before I get concerned about it, to tell you the truth. Maybe I'm foolish and these 45nm ones have such an extremely low overvoltage tolerancy that +0.05V over Intel's guidelines could kill them but I think we have to see more data before we declare this +0.05V 'deadly'...

 

[Markfw]

BTW, I got my new TRUE 120extreme today, and I bumped the cpu vcore up to 1.35, and all the others to max, and I still can;t even boot at 451. 447 is it, and it will do that at 1.312. I really think these chips don't like the high FSB (over 440). So I backed everything down, and my temps are much nicer, but NO OC improvement.

3576 is it !

 

[themattman]

I just ordered a Q9450 from microcenter online, a TRUE from tankguys, and two Scythe fans from newegg. I will be having some fun on Friday night

And if anyone else is interested, I also got:

Antec 900, Corsair HX520, Gigabyte DS3L, 2x2gb Patriot 1066, Evga 8800GTS, WD Caviar 640GB, Asus Xonar DX (boo Creative), DVD Burner, Vista Home Premium 64 bit, Logitech G15 (old style), Dell 2407WFP-HC, logitech MX 310

I think that is it This will be a *slight* upgrade to my current:

AMD XP-M 2500+ @ 2.3, 1GB Corsair DDR PC 3200, ABIT NF7-S2, ATI All-In-Wonder 9800 Pro 8x AGP, 120 GB hd, Antec Sonata II

Hopefully I will be able to get the 3.4-3.6 overclock everyone else has been gaming.

 

[tenax]

Originally posted by: T2k

Originally posted by: Markfw900

Originally posted by: fleshconsumed

Originally posted by: T2k
Why people keep repeating the same, false urban legends over and over again?
There's nothing "deadly" in 1.5V - it simply depends on your cooling. A short visit to the forum of Xtremesystems will tell you people go even higher regularly. My Q6600 was running at 1.5xx stable, all I had to do is upgrade my CPU block to make it stable @3.8xGHz.

Here's link to the same xtremesystems forums, read it carefully

http://www.xtremesystems.org/f...howthread.php?t=175829

There have been reports of wolfdales dying from overvoltage. It's real. Wolfdale/Yorkfield is not Conroe. These are 45nm parts designed to run at a lower voltage than 65nm conroes. If you overvolt too much you risk frying your CPU sooner than later.

Yes, its not false urban legends. You have an opinion, thats fine, just don't try to convince others to fry their CPU.

Maybe I wasn't clear enough, sorry - I took issue with his wording: 1.5V is NOT 'deadly', let's stop these urban legends here.
Dangerous? Of course, could be. But if you really want to play safe then do not OC or only whatever the safe voltage zone allows you.

dangerous, deadly..i'm not going to argue symantics with you..the point is...better safe than sorry..remember how many newbs come to sites like this looking for overclocking advice and then their processor or board goes left on them and the poor sucka is going what the hell? and quite often those newbs have invested what money they have for one new cpu, one new board etc that will last them a long time versus, "kill a processor? heck i'll just buy a new one"

 

[bryanW1995]

Originally posted by: tenax
ip35 pro owners, attention!

here's the new official 16 bios..see the text notes when you unzip the file..exciting!

<a target=_blank class=ftalternatingbarlinklarge href="ftp://ftp.abit.com.tw/pub/download/b...ro/m629b16.zip">ftp://ftp.abit.com.tw/pub/d......b...ro/m629b16.zip</a>

good thing I went through all that crap to get the beta bios...grrrrr...

 

[tenax]

aren't you happy though that you now have the latest bios?

 

[bryanW1995]

I know that it's probably exactly the same and that I would have had the exact same problems installing it...but still...

re the overvoltage issue that has become the hot topic:

1.5v on a Q6600 is much different than 1.5v on a yorkie. 1.5 is less than .2v over the default iirc. in fact, my e6750 has a default of 1.35v. A Q9450, however has what, 1.15 or 1.2 as default? 1.5 on one of those is like running 1.65-1.7 on a Q6600. THAT is dangerous, no matter what cooling you're using.

 

[Markfw]

Speaking of default... My motherboard says my default vcore is 1.175, but coretemp says it is 1.0675, and I am running it at 1.312. And even with a true 120 extreme, lapped, I am getting 65c full load.(on the highest core)

 

[runawayprisoner]

Regarding the voltage thing, we conducted some... researches on the degradation myth over at XSF. Here's the thread, so please read up, guys. (And maybe change your PSU afterwards)
http://www.xtremesystems.org/f...howthread.php?t=179965

Mmm... and Mark, may I ask for your voltage settings... at the moment? Maybe I can help with something?

 

[Obsoleet]

The stuff in my sig arrives on Friday. I'll be joining you all soon in this worthy quest. :thumbsup:

 

[themattman]

Originally posted by: Obsoleet
The stuff in my sig arrives on Friday. I'll be joining you all soon in this worthy quest. :thumbsup:

I have almost the same setup as you coming in . We should see what the difference is in 3d marks when you have the 9450 paired with a 8800gt/8800gts.

 

[bryanW1995]

Originally posted by: Markfw900
BTW, I got my new TRUE 120extreme today, and I bumped the cpu vcore up to 1.35, and all the others to max, and I still can;t even boot at 451. 447 is it, and it will do that at 1.312. I really think these chips don't like the high FSB (over 440). So I backed everything down, and my temps are much nicer, but NO OC improvement.

3576 is it !

did you try jumping over 451? I have that problem with my e6750, it won't do 450-460, but 449 or less and 461+ are gtg.

 

[Markfw]

Originally posted by: bryanW1995

Originally posted by: Markfw900
BTW, I got my new TRUE 120extreme today, and I bumped the cpu vcore up to 1.35, and all the others to max, and I still can;t even boot at 451. 447 is it, and it will do that at 1.312. I really think these chips don't like the high FSB (over 440). So I backed everything down, and my temps are much nicer, but NO OC improvement.

3576 is it !

did you try jumping over 451? I have that problem with my e6750, it won't do 450-460, but 449 or less and 461+ are gtg.

wel,, when I first tried, I stated at the know good 509, and went down by 5, it wouldn;t even post until I got down to ~460. Wont boot into windows over 455 and isn;t stabe over 447. 451 is close, but not F@H stable 100% load.