Overclocking a 1.8A GHz P4

[jagojago12]

I'm going to buy a P4 1.8A GHz processor from newegg or so pretty soon, but I was just wondering what were the chances of this thing hitting at least 3GHz using either: watercooling, peltier, and water/peltier combination. I'm hoping to get a 100% overclock, but if I end up with 3GHz I guess I will be satisfied "ok."

I'm going to order RAM that does 200MHz (in case we can OC by 100% 18x200MHz = 3.6GHz). I'm not too sure on which motherboard I should get. Do the SiS655 motherboards overclock well? How about the i845PE chipset? I'm not too interested in the E7205 since I'm going to limit my motherboard budget to 100-ish (flexible though).

However, if the E7205 is that much better, or would significantly increase my chances at 200MHz FSB, then I'm all for it. I normally research this stuff myself, but I've been real busy lately (Haven't posted much in here for the past month or so) so I would greatly appreciate it if someone can suggest a motherboard that overclocks well.

[number1sixerfan]

With a peltier, you might get 3ghz. 3.6 is really pushing it and I don't think you could possibly reach that high without prometia, dry ice, or LN2 cooling. Even if you do reach 3.6ghz, which again is highly unlikely, it wouldn't last very long. But heh, prove me wrong and reach that massive overclock.

I don't know much about P4 boards and don't know which will overclock best. But good luck, you'll definately need it.

[jagojago12]

Thanks for the insight, however I was more interested in a suggestion for some hardware, rather than to be doubted. There's no doubt that nothing in overclocking is guaranteed, we all face this hard truth. Another reason why you are doubting this so much is because almost all P4 1.8A overclocking results you've seen have been on the B0 stepping (I'm assuming here, forgive me if I'm wrong) which generally won't overclock as well as the C1 stepping. The C1 stepping is the same process that the 3.06GHz Pentium4 is made on, so there is definetly a good chance of it hitting 3GHz since I suspect Intel's production line has been tweaked and refined so it has become hard for them to actually make a CPU that won't overclock well. It's similiar to why the A3 stepping on GeForce4 Ti4200's were almost all Ti4600 class or better in overclocking potential. It also explains why the XP1600+ Palomino had the same overclocking potential as the XP2100/2200 (Back when AMD's best steppings were the AGOIA/AROIA). Get what I'm saying here? While I may seem overly optimistic to you, I do have a reason.

I don't need anyone reminding me that I might not make it. Why state the obvious? I didn't think you meant any harm, and neither am I - so don't take what I'm saying seriously. To give you a hint, this isn't going to be for personal use, but will rather be an experiment. Whether or not it will be stable at 3.6GHz is totally different from actually achieveing those marks. I just need some data .

Anyway, what are people's opinions on the Epox i845PE chipset? I just need something that is able to get to 200MHz CPU FSB easily, so I know for a fact my CPU isn't limiting me. I want to rule out as many variables as possible. What I basicaly need is a motherboard that can easily overclock to 200MHz FSB, have at least 1.8V core (actual, no undervolted stuff) and preferably at least 2.8V on the RAM. The Epox i845PE chipset has all of these features I want, but the 200MHz FSB limitation scares me (Based on anandtech's Epox 845 review). I know that a 200MHz limitation doesn't necessarily mean it won't get to 200MHz, but more often than not, a motherboard will never overclock to it's maximum specified FSB.

[number1sixerfan]

Yes I know about the B0 and C1 steppings. Most of the overclocks I've seen are from C1 steppings. Yes, they do overclock higher than the B0 steppings. I was just saying that it would be hard to reach 3.6ghz. Anyways, you might want to look at the ABIT IT7-Max2v2 board. It's a very popular board for overclocking.

[fobugee]

hey buddy, i have a 2.53 C1 Stepping, and the most i can get on air is 3.15 GHz....if i had water, maybe i could get 3.2 - 3.3 GHz. If i get those with a 2.53, i dont know how that's gonna happen with a 1.8...you would have to get like a prometia, and if you're gonna do that, you're better off just getting a 3.06! then again, like you said, you're doing this for "experimental purposes" and i respect that.

[CrawlingEye]

Originally posted by fobugee
hey buddy, i have a 2.53 C1 Stepping, and the most i can get on air is 3.15 GHz....if i had water, maybe i could get 3.2 - 3.3 GHz. If i get those with a 2.53, i dont know how that's gonna happen with a 1.8...you would have to get like a prometia, and if you're gonna do that, you're better off just getting a 3.06! then again, like you said, you're doing this for "experimental purposes" and i respect that.

Not neccessarily, if he intends to vmod, he could get significantly higher. With that kind of cooling, I'd definitely vmod either 'board I got, or just crank the voltage.

[jagojago12]

Originally posted by fobugee
hey buddy, i have a 2.53 C1 Stepping, and the most i can get on air is 3.15 GHz....if i had water, maybe i could get 3.2 - 3.3 GHz. If i get those with a 2.53, i dont know how that's gonna happen with a 1.8...you would have to get like a prometia, and if you're gonna do that, you're better off just getting a 3.06! then again, like you said, you're doing this for "experimental purposes" and i respect that.

I'm not dogging your overclock, but you are running at a relatively low V core on your CPU. 1.55 volts isn't very high, and I suspect that with the vcore that I intend to use (At least 1.8V) that I will have a better shot at with the overclocking results than what you've experienced. Of course it'll be a long shot to reach my goal, but I don't see it being impossible .

So the ABIT IT7-Max2v2 board can get to 200MHz FSB (800MHz QDR) easily? Any other suggestions?

[Extol]

Well if your going to attempt to get there you need a C1 1.8a with the SL6LA product code. Some guys have hit 3.6ghz on air and a minor vcore bump to. But your chances of getting a gem chip like that are like winning the lottery. But good luck with your overclock.



edit- if you run your vcore at 1.8, then you risk losing it fairly quickly or you may not.

[jagojago12]

After checking out the IT-7 Max 2 Ver.2, I am impressed.

Upto 1.85 vcore, upto 3.2 vmem, 4-phase power supply and high quality MOSFETS. I think this motherboard can have a shot at achieving 200MHz/800MHz QDR FSB.

I think I will get this motherboard. If anyone still has any suggestions, feel free to post them.

I do have one question though. The 3.2 vmem on the IT-7 Max2 Ver.2 is only acessible through a modified BIOS. There are BIOSes out there that already come modified, but the older BIOS only has a vcore of upto 1.7V which is far less than the 1.85V the current BIOS gives. How would I go about editing the BIOS to get 3.2 vmem?

[FaTs]

Originally posted by jagojago12
After checking out the IT-7 Max 2 Ver.2, I am impressed.

Upto 1.85 vcore, upto 3.2 vmem, 4-phase power supply and high quality MOSFETS. I think this motherboard can have a shot at achieving 200MHz/800MHz QDR FSB.

I think I will get this motherboard. If anyone still has any suggestions, feel free to post them.

I do have one question though. The 3.2 vmem on the IT-7 Max2 Ver.2 is only acessible through a modified BIOS. There are BIOSes out there that already come modified, but the older BIOS only has a vcore of upto 1.7V which is far less than the 1.85V the current BIOS gives. How would I go about editing the BIOS to get 3.2 vmem?

My older B-1 ? stepping 1.8a does 3ghz @ 1.95v.. which is far to high of a vcore to run day in day out. I'll throw some water and ice at it and see what i can get it to on 1.75v and let you know.