-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
Clocking my X2 3600+ Higher Thread
- Thread starter ps2cho
- Start date
OP
- Joined
- Oct 13, 2004
- Thread Starter
- #42
It's highly unlikely...I mean heck I ran my Athlon XP 2000+ at 2.2GHz 1.95v for a good year 24/7 55c load....
I think its the motherboard due to the fact it looks bent because of the Ultra 120 Extreme...its micro atx sporting the 120X...the fan ports on the board no longer work...and it required a 0.1v increase in the chipset voltage to actually boot into windows
EDIT: just started 3dmark2006 because orthos passed 5 minutes and it instantly crashed. And unfortunately I don't have any other AM2 components to test with...Well I can play all my games at 2.2GHz as the X2 3600+ Isn't a bad chip....
I guess when Phenom comes out I will have an upgrade ready and waiting.
- Joined
- Aug 23, 2006
Well, HTT speeds are associated with FSB which ties to your chipset. Try running stock speed (or a low multi) that you know should work, and up the FSB as high as you can.
- Joined
- Sep 22, 2004
Humm... I basically have the same setup, and I think yours inspired my decision to go with the Biostar 7050, well mine is 7025- same difference.
Your experience here makes me hesitant to ever put my setup in a case, I've been running it on a bench since I got it. I've got an Ultra-120 vanilla on my setup so I'll keep this in mind next time I decided to remount the cooler- don't over tighten.
I'm sure your 3600 is/was a better chip than mine... I have to run +0.200v in bios to run 306x9.5, which ends up reading ~1.45v. The mosfets on these boards get really hot (~70-80C) which I believe is leading to the drop in actual vcore vs. what is set in bios. I just need to order up some Alumina so I can put some sinks I've got on the mosfets. I plan to put a Phenom in this board when they arrive and I think the mosfets will take even more punishment pushing 4 cores vs. just 2.
My setup has been stable at this vcore going on months now running F@H... so I don't think you should worry about pushing the volts to see if you can get past your current wall.
The situation really does suck though... you had a really sweet setup the way it was running. I used to dive in my case and play with settings all the time, but a few experiences like you're having now have turned me away from playing with my setups a lot after they've been dialed in. More times than not I either end up back where I already was (wasted time) or end up in a situation like yourself-
wondering why the machine won't clock like it did before.
Your experience here makes me hesitant to ever put my setup in a case, I've been running it on a bench since I got it. I've got an Ultra-120 vanilla on my setup so I'll keep this in mind next time I decided to remount the cooler- don't over tighten.
I'm sure your 3600 is/was a better chip than mine... I have to run +0.200v in bios to run 306x9.5, which ends up reading ~1.45v. The mosfets on these boards get really hot (~70-80C) which I believe is leading to the drop in actual vcore vs. what is set in bios. I just need to order up some Alumina so I can put some sinks I've got on the mosfets. I plan to put a Phenom in this board when they arrive and I think the mosfets will take even more punishment pushing 4 cores vs. just 2.
My setup has been stable at this vcore going on months now running F@H... so I don't think you should worry about pushing the volts to see if you can get past your current wall.
The situation really does suck though... you had a really sweet setup the way it was running. I used to dive in my case and play with settings all the time, but a few experiences like you're having now have turned me away from playing with my setups a lot after they've been dialed in. More times than not I either end up back where I already was (wasted time) or end up in a situation like yourself-
wondering why the machine won't clock like it did before.
OP
- Joined
- Oct 13, 2004
- Thread Starter
- #45
Hold on a minute guys....
I think I just figured it out. Will edit in a second.
EDIT: yep got it. They moved the damn HT multiplier in the updated BIOS and I completely forgot about it after dealing with my Celeron 420 for the past month.
Alright....300x9.5 loading. Phew...
EDIT2: Ok first thing I am going to do is figure out my max FSB on the board.
I think I just figured it out. Will edit in a second.
EDIT: yep got it. They moved the damn HT multiplier in the updated BIOS and I completely forgot about it after dealing with my Celeron 420 for the past month.
Alright....300x9.5 loading. Phew...
EDIT2: Ok first thing I am going to do is figure out my max FSB on the board.
Last edited:
- Joined
- Jul 8, 2003
- Location
- Dallas, Tx
man, i don't mean to laugh, but, hahahahahahaha
good stuff, it's always something simple
kinda like most accidents happen within 5 miles of your house
glad ur back up and running
good stuff, it's always something simple
kinda like most accidents happen within 5 miles of your house
glad ur back up and running
OP
- Joined
- Oct 13, 2004
- Thread Starter
- #47
man, i don't mean to laugh, but, hahahahahahaha
good stuff, it's always something simple
kinda like most accidents happen within 5 miles of your house
glad ur back up and running
Me too man...me too.
Why the heck would they move the HT multiplier huh? Or I could say...stupid Intel made me forget
Let's get this baby to 3GHz now! To hell with the volts it will require!
2.9GHz 1.35v before, but wouldnt hit 3GHz with 1.4v.... I have my huge CM 690 case with a TON of airflow. I got 5x120mm fans blowing full speed (FOR NOW IF YOU CAN HEAR ME!)
Should I stick with the 9 multiplier or is the 9.5x ok?
Right now CPU full load is 46c, chipset 46c
- Joined
- Jul 8, 2003
- Location
- Dallas, Tx
i don't care what anyone says, there's nothing wrong with using x.5 multi's
per the book, of course, i would lower your multi and find your best HTT and best memory speeds, then work the clock
per the book, of course, i would lower your multi and find your best HTT and best memory speeds, then work the clock
OP
- Joined
- Oct 13, 2004
- Thread Starter
- #49
Yeh that's what I am doing right now.
So far so good! 330x8 is booting! I never remembered the chipset going past 321FSB!
I'm getting excited
EDIT: This is with the stock chipset voltage too!
340x8 booting right now.
Last edited:
OP
- Joined
- Oct 13, 2004
- Thread Starter
- #51
Ok 335 is the best stable FSB I could get. 340 passed SuperPI, but fails Orthos.
Back to where I was!
Just keep bumping the vcore til it's stable I guess LOL
What you guys reckon a safe vcore on air would be? I upgrade every 2-3 years...so longevity isn't a huge concern to me. As long as it will last a few years then I'm cool with it.
Back to where I was!
Just keep bumping the vcore til it's stable I guess LOL
What you guys reckon a safe vcore on air would be? I upgrade every 2-3 years...so longevity isn't a huge concern to me. As long as it will last a few years then I'm cool with it.
- Joined
- Jul 8, 2003
- Location
- Dallas, Tx
in my experience, errors are from ram, maybe do some tweaking
OP
- Joined
- Oct 13, 2004
- Thread Starter
- #54
Yeh I was just doing some reading and it says that Illegal sumout is related to RAM. I have a feeling the RAM configured by the motherboard is causing the instability.
Here's what I have:
Now someone who knows about RAM, can you tell me the values I need to set in this screen? I tried doing it myself but the board doesn't post when I adjust them.
- Joined
- Jul 8, 2003
- Location
- Dallas, Tx
if you don't get help on the timings here, might wanna head over to xtreme systems and ask a question or 8
OP
- Joined
- Oct 13, 2004
- Thread Starter
- #56
Just bumping the thread. I have an idea that if the BIOS is auto adjusting the RAM timings badly at the lowest divider, then maybe I should try the 533MHz divider? I'm stupid it never crossed my mind...
I mean if the RAM goes up close to 1GHz on the 667MHz divider, then running 533MHz divider should be no problem either!!
Right now I am running memtest 315x9.5 (2997MHz) and so far no errors!!
If the RAM survives a good full memtest86 then I will go and ortho's it and hope that this was my problem all along! Maybe my 3600+ will finally be at 3GHz
I mean if the RAM goes up close to 1GHz on the 667MHz divider, then running 533MHz divider should be no problem either!!
Right now I am running memtest 315x9.5 (2997MHz) and so far no errors!!
If the RAM survives a good full memtest86 then I will go and ortho's it and hope that this was my problem all along! Maybe my 3600+ will finally be at 3GHz
OP
- Joined
- Oct 13, 2004
- Thread Starter
- #57
Ok still failed so I am going back to the 667MHz divider and trying to increase the volts.
I first tried 2.1v and I get a rounding error.
Tried 2.2v still rounding error
Right now testing 2.3v
I still can't believe this ram is memtest86 stable at 996MHz
EDIT: Yep 2.3v failed.
Let me try raising the vcore now....I'm at 1.41v right now, but my max load is 45c which is very cool for this voltage.
EDIT2: This whole problem is so weird...I wish I could just determine what the heck is causing this!!! Motherboard has been tested up to 335FSB stable...RAM seems to pass memtest86, but I am running Ortho's memory test and it gives me a rounding error.
Right now I am running the stress CPU with Gromacs Core and I'm 5minutes stable at 3GHz. Using the memory stress, it fails at 3minutes every time. I just don't understand
EDIT: I think I've determined its my 2nd core that is holding me back. Prime95 on core0 runs fine, but core1 never gets past the first iterations test. Orthos fails on the 2nd core every time.
Looks like I'm stuck huh!
I first tried 2.1v and I get a rounding error.
Tried 2.2v still rounding error
Right now testing 2.3v
I still can't believe this ram is memtest86 stable at 996MHz
EDIT: Yep 2.3v failed.
Let me try raising the vcore now....I'm at 1.41v right now, but my max load is 45c which is very cool for this voltage.
EDIT2: This whole problem is so weird...I wish I could just determine what the heck is causing this!!! Motherboard has been tested up to 335FSB stable...RAM seems to pass memtest86, but I am running Ortho's memory test and it gives me a rounding error.
Right now I am running the stress CPU with Gromacs Core and I'm 5minutes stable at 3GHz. Using the memory stress, it fails at 3minutes every time. I just don't understand
EDIT: I think I've determined its my 2nd core that is holding me back. Prime95 on core0 runs fine, but core1 never gets past the first iterations test. Orthos fails on the 2nd core every time.
Looks like I'm stuck huh!
Last edited:
- Joined
- May 25, 2007
- Location
- With Yahweh in Colorado
you could try to burn it in more then what it is. some people swear by it. and hell what u got makes me want to go buy one being that new egg has them for like 55 bucks.
- Joined
- May 25, 2007
- Location
- With Yahweh in Colorado
no dude i think this might be the end for him. im running into the same thing with my 5600+. i can come so close to having it stable at 3.35 to 3.4Ghz core 0 is sweet but core 1 is crappy. i put some volts into mine to. i went up to 1.575 . he could try to up the voltage a little more.
Similar threads
