| |||||||
| Extreme Cooling Peltiers,N2,Water...You name it... |
![]() |
| | LinkBack | Thread Tools | Display Modes |
| | #1 (permalink) | |
| Okay, it's my first time posting on this site. I'm not new to overclocking, but I am new to water cooling. I put together my first water system about three months ago. When I first installed it, I was somewhat surprised to find that I couldn't OC ... at all. WIth fan cooling, I had my P4 3.0c running at 3.7GHz without any major voltage increase. After I switched to water, I couldn't increase the FSB at all. Over the last three months, I've managed to OC a little bit, but I can't get the FSB above 220MHz. At first, I thought this was probably just a problem with my particular combination of components, something in the way I had constructed the system, etc. However, 4rm4g3dd0n is a friend of mine. He just finished his first water system earlier today, and now he can't OC. Coincidence? Here are my system specs: Abit IC7-MAX3 P4 3.0c 1GB (2x512) PC4000 Geil Dual Channel DDR ATi Radeon 9700 All-In-Wonder Pro Western Digital 160GB SATA Caviar Drive As for the water system: DangerDen RBX block with the #5 accelerator nozzle Swiftech MCP600 Pump BlackIce Extreme Radiator We're both using the same mobo, so that might be part of the problem. Still, I don't see how switching from one cooling solution to another could adversely impact the OCing abilities of the motherboard to this extent. Any suggestions? Last edited by spyrius; 22-April-04 at 12:10 AM.. Reason: typo | ||
| | | |
| | #5 (permalink) | |
| Icpiper: I spent about 2-3 weeks assembling the water system, disassembling the water system, until I was satisfied with it. As for the CMOS, I've cleared it probably a dozen times in the last few months. The non-OC settings are the same as they've always been. I can't manage to increase the FSB above 220MHz, regardless of how high I crank the V-Core. I can't improve the RAM timings either. I've always kept AGP/PCI Frequency fixed at 66/33MHz. As for 4rm4g3dd0n, I found out he has "fixed" his problem. He just cleared the CMOS a few times, voila, everything's fine. Yay for him. j-dogg: The painting is "Le fils de l'homme" by the Belgian impressionist painter Rene Magritte. You know those damn Belgians, they're just so ... evil. | ||
| | | |
| | #6 (permalink) | |
| interesting, first hr or so i couldnt overclock, but after pulling the cmos jumper 2 times it was back to normal, and im back at my old setting, and much much colder and quieter. im very happy with it. in the meantime, i dont know what happened when i couldnt, like any higher than 215, even after resetting cmos once, it would die for me no matter what voltage, but 2 cmos resets later it worked. i have no clue tomorrow i aim for 3.8 ghz+ but i doubt it will happen | ||
| | | |
| | #7 (permalink) | |
| You know, that is the one trend in Overclocking I hate the most. Peeps are way too fast to raise the VCore. My XP2100+ is OC'd 500mhz at 1.65V, Stock. My P4 2,4C is OC'd 600mhz, Stock VCore 1.525V. On some chips you gota do it, but you don't just listen to what someone else did and start raising that thing, it can ruin a CPU so fast. The next biggest thing you don't do with the VCore is raise it beyond the Core's spec. That means that if you have a 2.4C, and spec is 1.525V Then you check for the highest spec Voltage for that core architecture, which is the Northwood chips, 3.4C at 1.55V. Certainly you can push it up just a little more, but these things die a lot faster then the guys who killed'em want to admit. It's just that if you run it up stack, and can't get it to budge even 200mhz with a Vcore boost, you got a differenct problem and raising the VCore isn't going to fix it. Now I'll tell you this, swaping out Air cooling for water cooling is not going to hurt your OC in any way unless you use a half assed water cooling rig. I doubt you have it set up that wrong. I mean if your temps are fine, then the rig is ok. So you have a different problem. I wish you hadn't cleared the CMOS, but maybe you had to just to get it to boot. But the only thing you should have had to do is swap out the cooling gear, and turn off your fan monitoring for the CPU fan, that way when it doesn't see one it doesn't freak thinking your HSF died. I'll have to look up your motherboard manual and give it a peek, see if something stands out. Chances are there is a default setting that is stopping you, at one time, a common one was that the motherboard wouldn't let you boot if it detected an irregular CPU frequency, so before you could OC, you had to turn this setting off. Can't remember what it was called. I can't get on this site from work these days, Ft. Huachuca has it blocked and I work on post, but I'll investigate and look for an answer and get back with you tommorrow night if someone doesn't beat me too it | ||
| | | |
| | #10 (permalink) | |
| I think that it might be some sort of saftey feature in your BIOS. Like mine you have to set it to Aggresive or you cant O/C. Maybe its something like that. If not then I would suggest resetting your CMOS, Unplug the power cord from your PSU, the CMOS battery, and the jumper. Let it sit for 10mins or so, and then plug it all back in, and try again. | ||
| | | |
| | #11 (permalink) | |
| Icpiper: Don't worry about the v-core. I understand that raising v-core on the northwoods helps little if at all for overclocking. When I said "regardless of how high", I mean that I raised it from the stock 1.525V to about 1.6V, and saw absolutely no improvement. As for the obligatory overclocking setting, I've looked at pretty much every inch of the BIOS, and there's nothing I can find. I have all the alarm, auto-shutdown, CPU-fan check settings turned off at this point, so I doubt that's it. :-/ | ||
| | | |
| | #13 (permalink) | |
| Here something that I had to do with my A64 rig. I was having a very similar prob as you, I had just installed a new waterblock and I couldn't overclock as far a before. (even though I had replaced one block with a better) Nothing else had changed on the rig. My problem was that the water block (prior to tightning down) had a little slop for side to side alignment. Now with the RBX, the center of the water block needs to be sitting as close to the center of the cpu die as possible to work at its highest levels. By resetting the block and ensuring that the center of the RBX and the CPU were as close to concentric as possible I was able to obtain even higher overclocks. This may not be what is holding you back, but it wouldn't hurt to look at it as a possible cause. You don't have to pull the block, just loosen it and position it then tighten back up. | ||
| | | |
| | #15 (permalink) | |
| I tend to agree with CPU Killer and Anonymous. A slightly misaligned block mount can give you seemingly good temps, but an uneven cooling of the core is gonna knock the crap outta your OC. If you're not used to the temps you can get watercooling you might never notice just looking at temp readings. It's very easy for tubing to pull a block out of true. A block should hang roughly parallel to the cpu with the hoses attached BEFORE you bolt on the block. Something to check. | ||
| | | |
| | #16 (permalink) | |
| errr iseem to see everyones forgoten somthing - unless its just me on a very short fallen whim. Water coolers dont use the fan connector on the motherboard right? so what if the bios has somesort of overclocking locker that reduces the OC level down to a minimum? tried reseting the bios? or simply the CPU might like it nice snug and warm for a decent overclock! | ||
| | | |
| | #17 (permalink) | ||||||||||||||||||||||||
sorry about that ic lemme clarify what i meant was its not a bad chip (some dont overclock as well as others, etc) and it wasnt a new bios setting, he did the exact same thing as me. i took out all my old hardware, installed the water, and didnt change a thing in the bios. before the water i could run 270fsb or 3.515 ghz cpu constantly, or even 3.661 for benchmarking, but after the water it wouldnt go any higher than 215 even though i didnt change anything in the bios. first cmos reset didnt help, but somehow second one did. i dont get it, but mines working now, and when i get home ill hopefully be posting screenies of 3.8+ ![]() | |||||||||||||||||||||||||
| | | ||||||||||||||||||||||||
| | #18 (permalink) | |
| Nev, that is just the angle I was heading for. At a minimum, in almost everyones case, they must turn off the warnings, alarms, and safeties for CPU fan monitoring or else the missing fan will look like a dead fan. Now if you were running with these features off before, well fine, but if not, then here is the deal. If you don't remember doing this, then what else don't you remember. But it's moot as you have cleared the CMOS so we have to look at this from scratch. I have blown an OC before and not even been able to boot till the next morning. I cleared the CMOS several times and it just didn't want to accept it. Remember, with P4's you have CPU throttleing, so if your running hot, or the BIOS just want's to kick this into play, then your going to run slow and cool no matter if the HSF is dead, your on water, or have a Prommie running. Just things to concider, but in the short run, anyone who says they swapped air for water and didn't do a thing to the BIOS has me wondering simply because you don't have that CPU fan connected anymore. And if you take the typical mid to high powered 80mm and connect it to a Mobo fan header, your asking to burn that header, and maybe the whole Mobo out. | ||
| | | |
![]() |
| Bookmarks |
| Thread Tools | |
| Display Modes | |
| |