E5200 - Overclocking Thread

[nolonger]

I have seen instabilities appear after the 20th loop. Also, you didn't run IBT for long enough, your system could be unstable. Run it for at least 4 hours.

[nemaca]

Ah, nolonger, I'm not going to do a proper stability until I can actually be able to restart my PC. That 5 loop was just for me to make sure I can operate windows at least. The problem is that I have to restart manually from the button my PC and when I do, I get the "fail overclocking or failed overvoltage" message if I try to tamper with the FSB. Either way, I' wont run anything till this thing is solver, I can't ignore it, since I don't know what's causing it. I disabled the dvd-rw drive, set everything to default and still have the non-reset problem. I unplugged the PSU, gave it 30 seconds and replugged it, but did no good.
What could cause this, I'm really confused. Sound is like not enough power to POST. Isn't my HighPower 500W enough?
I have to shut down the PC from the button, then start it up again from the button. The CPU cooler starts for a second then shuts down for a fraction of second, then restarts then shuts down again, and so on, on a loop.

L.e. This tells me I'm using maximum 300W, with overvoltage and 4GHz CPU, and an additional RAM plate.

[nolonger]

Not really sure on what could cause that, mate. Sorry!

[W4LNUT5]

I had a gigabyte board that used to do something similar to me. It would just fail the post (so it seemed) and keep looping w/o informing me of anything (so i presume). I usually had to clear the cmos (sometimes a few times) and remove power from it for a bit. It would seem to give me 1 good start everytime I did that. After a bios update it was ok.

[nemaca]

I flashed the latest XP BIOS, W4LNUTS, and Wouldn't that clear CMOS? I regret not having my camera home to film it.

So... I'll try to be more specific, and also keep it in this thread since it could be caused by processor and it happened after a BSEL and vmod exactly as in this guy, after checking with intel's specific processor design and heat handling pdf's. I did the modding clean and accurate, no other pins on CPU are touching each other or not being able to connect on socket due to tape or w/e.
The results are correct : 3.44 Vcore, 266 seen FSB => e5200 @ 3333MHz default.
BIOS version is the latest(608 for my mobo), BIOS settings all on defaults, no tweak whatsoever.
I have not changed the OS(Win XP Pro SP3), but I have updated chip drivers(told me was no need), and audio drivers.

So... steps breakdown:

Step 1:
- Computer is stopped. I press Turn Off button from unit case and it powers on nice and smooth, does everything normally, no flaws here. One POST beep in the second it powers up the CPU fan, normal POST, detects 3333MHz, RAM fine, USB controller fine, Video fine, then starts to load Windows, which goes smooth and natural, all the way to me typing here.

Step 2:
- A I am in Windows and I restart the OS from the Start menu/Turn off/Restart. Windows does its job nice and smooth, closes everything, blue normal shutting down screen, then it shuts down the computer as it should do.
- B I am in Windows and I Shut Down the OS from the Start menu/Turn off/Restart. Windows does its job nice and smooth, closes everything, blue normal shutting down screen, then it shuts down the computer as it should do.

Step 3:
- A Computer then begins to restart no matter if I chose Shut Down or Restart: emits one short beep while powering up the CPU fan, video card, system fan, hdd. All this takes half a second.
- B Then the CPU fan stops, like it changed his mind. The rest seem to continue running at same speed: chasis fan, video card fan(replaced by me with a 120mm silent chasis fan)and i believe the hdd too(can't hear it unless I stop all fans). All this takes another half a second in which the CPU fan begins to slow down his rotations.
- C Then it tries to reboot, and CPU fan starts to work full speed again, there is no short beep this time, and this whole thing also lasts about half a second.
- Unfortunately, B and C link together and form a LOOP, not ever going to stop.

The restart button on the case is absolutely useless at this point, the only option I have is to press the Shut Down button on case for 4 seconds and force it to stop. After the forced stop, computer boots like Step 1, but if I tamper in BIOS with the FSB and/or voltages, it gives me the: "Overclocking failed and Overvoltage failed" message on POST and I have 2 options: F1(go back to BIOS and set things to default) and continue booting with defaults BIOS settings. If I choose F1, after I make any changes to BIOS, I press Save and Restart (F10) and this triggers the LOOP also each and every time.
First thing that crossed my mind: PSU or Mobo problem. PSU thermaltake calculator gives me a 288W consumption with overvoltage on CPU @ 4GHz and 2 plates or RAM (I only got one). I checked the motherboard for any shorts and found nothing.

Things I have tried:

- Reinstall motherboard drivers.
- Check to see if APM in Windows is enabled (however I don't have that option, meaning it's a different thing and not this the problem).
- Disable USB Legacy in BIOS (USB boot control, since I read on some forums it might cause various problems)
- Disable all the CPU features such as: Intel SpeedStep, C1E, etc. ( I did various combinations too)
- Turn power button on PSU to 0 ( I - 0 options), unplug power cord from PSU, let it cool for 30 seconds or more, then replug and power the computer. Result is that It goes through all 3 Steps, exactly as in pressing the Shut Down button on case for 4 seconds.
- While I restart computer (Step 2)in any way (restart button, BIOS restart, Windows restart), unplug the video cord leading from video card to monitor.
- Modify voltages, FSB, RAM ratio, RAM timings(I loosened them) from BIOS.
I hope I got them all.
- Update BIOS through Flash to lates version for my OS.
=== Cleared CMOS, thanks nolonger, and my CPU fan is properly attached.

Things I have not tried:

- Reinstall Windows (somehow I don't think its to blame for this).
- Un-mod the CPU.
- Stop all fans to see if the HDD starts (wouldn't hdd fail be displayed after POST?).
- Thoroughly check for CPU stability. Only did a 5 loop check on IBT, as I do that when I overclock to see if I can at least continue running Windows with browsing internet. This method never failed me.(no comments please, I know its more like a good luck charm thing)
- [please add some more]

I'm not going to continue any OC until this mystery is solved, so I really hope you guys can help me.

Edited on "Things I tried".

[nolonger]

Try a clear CMOS and make sure your CPU fan is properly attached.

[nemaca]

Cleared CMOS, changed PSU to an Antec TruePower 650W, that keeps a friends i7 overclocked, and have same problem.

I put my processor into a friends gigabyte (he was using a e5400) and surprise, worked just fine with both mods (BSEL and vmod).

After making a short brainstorming with some of the most cabable guys in my town, we concluded that the P5KPL-CM simply doesn't like the modded e5200.

So I came up with the idea of trying each mod separately.

Results: gigabyte liked both mods, P5KPL-CM liked vmod separate, didn't like BSEL separate and didn't liked BSEL + vmod.

Solution: I only have now the vmod on my CPU, CPU-Z shows a maximum of 1.368 V on core voltage.

Lol, after nearly 16 hours of hard try-outs, conclusion is that the mobo simply doesn't fancy the mod. Silly.

Anyone else got this problem, or anyone else tried both mods on a P5KPL-CM?

[W4LNUT5]

I was pretty sure it was mobo related and not proc related. Glad you got that sorted out.

If you are interested in pushing this chip further, I would recommend a better motherboard, one more suited towards overclocking. LGA775's should be dirt cheap nowadays considering its an outdated platform compared to whats out there now.

Could try something like these. They should be more OC friendly and have no need for mods. And they are still priced fairly cheap.
http://www.newegg.com/Product/Produc...82E16813128380
http://www.newegg.com/Product/Produc...82E16813135093

It sounds like you chip will go much farther than that. It just seems as if this mobo is limiting you.

[Ahmad Rady]

E5200 @ 4.200 GHz
http://valid.canardpc.com/show_oc.php?id=816243

[nemaca]

@W4LNUT5 I'm sorry but I'm planning an 1366 upgrade and my budget will be tight, I'm not investing in this anymore, except a good thermal paste syringe :P.

I managed to boot ONCE at 3.96GHz, but it crashed when i rushed to start CPU-Z and print it. Tried 2 more times and failed. Booted ok at 3.86 but not stable. I'm keeping it now at 3.51, going to test it thoroughly this week.
As i type I got my eye on prime small fft running in parallel with IBT.

If all goes stable, i figure getting 1GHz out of it with a mobo that is a declared enemy of OC allover the internet, is quite enough, can't possibly ask for more. I kept it at 3.3GHz from the first week I assembled this computer, and for the 58$ I payed for it, it's a steal.