VOGONS


Pentium 3 woes

Topic actions

First post, by stealthjoe

User metadata
Rank Member
Rank
Member

I got two Pentium 3 S370 CPUs a while ago. One was a 600Mhz and the other an 800mhz. The pins were slightly bent on arrival, but with some effort and patience, I managed to straighten the pins. The pins fit into the socket fully albeit not perfectly and with a little resistance (unlike a new processor). However, both failed to post.

So, I got another set of CPUs. This time a 667Mhz and an 800 mhz one. The 800mhz cpu posted fine and it fit like a new processor into the socket with zero resistance. However, the 667 mhz required some effort to straighten the slightly bent pins. I used a credit card (like earlier for a via 733mhz) to straighten it. It finally fit into the socket but with a very little resistance. Alas, the 667mhz processor failed to post. Tried on 2 different motherboards but no luck.

Now it brings me to the question - Do P3 S370 CPUs actually require perfectly straight pins to work or could it be a case of defective CPU? I have managed to straighten pins on a via c3 which fit in the socket with some resistance but it worked fine.

Intel 845GEBV2, Pentium 4 2.4 Ghz, Geforce FX5600 256MB, 512MB RAM, 160GB HDD, Sound Blaster Live! SB0100 - Win 98.

Reply 1 of 29, by dm-

User metadata
Rank Member
Rank
Member

no, it does not require perfect pins, but P3 has exposed die and if the die is cracked, then it will not post.

Reply 2 of 29, by stealthjoe

User metadata
Rank Member
Rank
Member

It has an exposed die but not cracked. I am thinking if straightening the pins using the credit card method could damage the processor.

Intel 845GEBV2, Pentium 4 2.4 Ghz, Geforce FX5600 256MB, 512MB RAM, 160GB HDD, Sound Blaster Live! SB0100 - Win 98.

Reply 3 of 29, by dionb

User metadata
Rank l33t++
Rank
l33t++

All Coppermine CPUs (other than the late Coppermine-T) have exposed dies, so that's normal. As for what killed the CPU... the only time I've killed CPUs straightening pins was when I broke them. One difference though that could be relevant: I use knives, not cards - so conductive metal, not plastic. It's possible the cards generate static electricity and kill through ESD.

Tbh though, if the CPUs were physically abused before you even got them, I'd far sooner expect any damage - phyisical or electrical - to have occurred then too.

Note that some motherboards need specific support for CPU steppings to boot, so if yours doesn't that might be an issue - but I doubt it. Your P3-800 boots as does a C3. Now, it's possible your booting P3 is from the cA2 or cB0 stepping and the non-booting CPUs are cC0 or later and the board doesn't support them - but the chances of a board that does boot a C3 not liking cC0 or cD0 Coppermine is minimal.

To be sure though, please post:
- the S-Spec of the CPUs, both the one that boots and the ones that don't (eg something like "SL3NL")
- which motherboard you are using
- which BIOS version is on the board

Reply 4 of 29, by stealthjoe

User metadata
Rank Member
Rank
Member
dionb wrote on 2025-06-18, 10:58:
To be sure though, please post: - the S-Spec of the CPUs, both the one that boots and the ones that don't (eg something like "S […]
Show full quote

To be sure though, please post:
- the S-Spec of the CPUs, both the one that boots and the ones that don't (eg something like "SL3NL")
- which motherboard you are using
- which BIOS version is on the board

The 667Mhz processor is SL3XW.
The 800Mhz processor is SL3Y2.
The motherboards used were Soltek SL65KV2(via 694 chipset) and iWill VA133Plus(via 693A chipset).
I don't recollect the bios versions but assuming that those are the latest ones as listed on the retroweb site.

Intel 845GEBV2, Pentium 4 2.4 Ghz, Geforce FX5600 256MB, 512MB RAM, 160GB HDD, Sound Blaster Live! SB0100 - Win 98.

Reply 5 of 29, by H3nrik V!

User metadata
Rank l33t
Rank
l33t
stealthjoe wrote on 2025-06-18, 11:18:

The 667Mhz processor is SL3XW.
The 800Mhz processor is SL3Y2.

Both cB0 stepping, it seems.

I also chime in with "I've never messed up a P3 cpu by straigthening pins" and I'm a bit lazy, so I don't straigthen them to a fit, where they just slide right in. "If it CAN get in the socket, it DOES" 🤣 I have had fears though, that pins might snap off during installation or removal 😁

If it's dual it's kind of cool ... 😎

--- GA586DX --- P2B-DS --- BP6 ---

Please use the "quote" option if asking questions to what I write - it will really up the chances of me noticing 😀

Reply 6 of 29, by stealthjoe

User metadata
Rank Member
Rank
Member
H3nrik V! wrote on 2025-06-18, 12:43:

"If it CAN get in the socket, it DOES" 🤣 I have had fears though, that pins might snap off during installation or removal 😁

You mean you force the cpu pins into the socket even if it doesn't align quite well? If so that sounds scary! 😱

Intel 845GEBV2, Pentium 4 2.4 Ghz, Geforce FX5600 256MB, 512MB RAM, 160GB HDD, Sound Blaster Live! SB0100 - Win 98.

Reply 7 of 29, by dionb

User metadata
Rank l33t++
Rank
l33t++
stealthjoe wrote on 2025-06-18, 11:18:

[...]
The 667Mhz processor is SL3XW.
The 800Mhz processor is SL3Y2.

Hang on, you first mentioned four P3 CPUs (600, 800 and 667MHz that didn't work, and another 800MHz that did), these are two. Is this the P3-800 that did work or the one that didn't?

The motherboards used were Soltek SL65KV2(via 694 chipset) and iWill VA133Plus(via 693A chipset).
I don't recollect the bios versions but assuming that those are the latest ones as listed on the retroweb site.

Assumptions are the mother of all... but that said, these are late, well-supported and when it comes to supported CPUs downright slutty boards. The iWill board even specifically mentions supporting Coppermine to 1GHz, and the 1GHz Coppermine is at least a cC0 CPU.

So unless it turns out that the working P3-800 isn't the SL3Y2 and it's a cB0 or older CPU, I'd say you just lucked out on the CPUs.

Reply 8 of 29, by AlexZ

User metadata
Rank Oldbie
Rank
Oldbie

If they fail to post on two motherboards then most likely they are dead. To be sure you would have to have a 1Ghz version working on one of the boards to rule out BIOS incompatibility issue.

Having so many dead CPUs looks suspicious. I would suspect you perhaps touched the pins and killed the CPU. Or you bought CPUs with cracked dies. I never saw so many dead CPUs. The dead rate is about 5%. Some people touch electronics everywhere and kill it. I bought a GeForce 9800 GTX and the guy brought it in hands touching pcb. Miraculously the card survived undamaged. One person shipped me an AM2+ board that they somehow damaged but the CPU survived. Many people who sell electronics do not know how to handle it.

Pentium III 900E,ECS P6BXT-A+,384MB,GeForce FX 5600, Voodoo 2,Yamaha SM718
Athlon 64 3400+,Gigabyte GA-K8NE,2GB,GeForce GTX 275,Audigy 2 ZS
Phenom II X4 955,Gigabyte GA-MA770-UD3,8GB,GeForce GTX 780
Vishera FX-8370,Asus 990FX,32GB,GeForce GTX 980 Ti

Reply 9 of 29, by stealthjoe

User metadata
Rank Member
Rank
Member
dionb wrote on 2025-06-18, 16:05:

Hang on, you first mentioned four P3 CPUs (600, 800 and 667MHz that didn't work, and another 800MHz that did), these are two. Is this the P3-800 that did work or the one that didn't?

The first 2 processors were 600MHz - SL3XT and 800MHz - SL3Y2. Both didn't work on the initial tries. The 800MHz die later cracked possibly as one of the board's heatsink was heavy. Didn't realise that.

The other 800 mhz SL3Y2(again) in the second set worked.

Assumptions are the mother of all... but that said, these are late, well-supported and when it comes to supported CPUs downright slutty boards.

Interesting term 😀 😀. So would it matter if the bios versions are the later ones in this case?

So unless it turns out that the working P3-800 isn't the SL3Y2 and it's a cB0 or older CPU, I'd say you just lucked out on the CPUs.

The working one is indeed 800mhz SL3Y2 (2nd set).

Intel 845GEBV2, Pentium 4 2.4 Ghz, Geforce FX5600 256MB, 512MB RAM, 160GB HDD, Sound Blaster Live! SB0100 - Win 98.

Reply 10 of 29, by stealthjoe

User metadata
Rank Member
Rank
Member
AlexZ wrote on 2025-06-18, 16:34:

If they fail to post on two motherboards then most likely they are dead. To be sure you would have to have a 1Ghz version working on one of the boards to rule out BIOS incompatibility issue.

Having so many dead CPUs looks suspicious. I would suspect you perhaps touched the pins and killed the CPU. Or you bought CPUs with cracked dies. I never saw so many dead CPUs. The dead rate is about 5%. Some people touch electronics everywhere and kill it. I bought a GeForce 9800 GTX and the guy brought it in hands touching pcb. Miraculously the card survived undamaged. One person shipped me an AM2+ board that they somehow damaged but the CPU survived. Many people who sell electronics do not know how to handle it.

The only s370 working P3 is the 800MHz I have currently. Getting these processors locally is a real PITA if not impossible. The ones I got were all imported and I am thinking if the Xray machines at the customs might have wrecked it (every parcel goes through x-ray whether big or small here).

I am very careful when it comes to handling electronics. I always touch grounded metal surfaces prior to touching any electronic boards and also ensure to touch only the edges wherever possible. Also in my place the humidity is very high.

But again as dionb stated, thinking if there might have been a static generated by the plastic card when straightening the pins. I didn't do any intervention on the 800Mhz SL3Y2 as it fit perfectly in the socket and it worked perfectly as well.

I was also thinking that by straightening the pins, the electrical connection between the pins and the base might have been broken. Is this even a possibility?

Intel 845GEBV2, Pentium 4 2.4 Ghz, Geforce FX5600 256MB, 512MB RAM, 160GB HDD, Sound Blaster Live! SB0100 - Win 98.

Reply 11 of 29, by Archer57

User metadata
Rank Oldbie
Rank
Oldbie

Honestly my opinion - in case of 3 dead CPUs - they were already dead when shipped. Some freak accident could have happened with one, be it static electricity, broken solder joint, whatever (forget about xray at customs, to put it mildly - not likely). But all this things are not very likely and i've rarely seen dead CPUs at all, apart from physically broken die in cases where it is exposed. So 3 of them happening at the same time would be so improbable it does not make sense to even consider.

Or may be they are not really dead and you are dealing with either compatibility issue or some motherboard issue. Though with 2 boards it is not very likely either...

AthlonXP 2200+,ECS K7VTA3 V8.0,1GB,GF FX5900XT 128MB,Audigy 2 ZS
AthlonXP 3200+,Epox EP-8RDA3I,2GB,GF 7600GT 256MB,Audigy 4
Athlon64 x2 4800+,Asus A8N32-SLI Deluxe,4GB,GF 8800GT 1GB,Audigy 4
Core2Duo E8600,ECS G31T-M3,4GB,GF GTX660 2GB,Realtek ALC662

Reply 12 of 29, by pete8475

User metadata
Rank Oldbie
Rank
Oldbie
stealthjoe wrote on 2025-06-18, 10:28:

It has an exposed die but not cracked. I am thinking if straightening the pins using the credit card method could damage the processor.

That would surprise me, I've had to straighten the pins of dozens of P3 and Celeron 370 processors over the years.

Reply 13 of 29, by H3nrik V!

User metadata
Rank l33t
Rank
l33t
stealthjoe wrote on 2025-06-18, 14:40:
H3nrik V! wrote on 2025-06-18, 12:43:

"If it CAN get in the socket, it DOES" 🤣 I have had fears though, that pins might snap off during installation or removal 😁

You mean you force the cpu pins into the socket even if it doesn't align quite well? If so that sounds scary! 😱

Force is such a strong word here ... 🤣 but well, sometimes, they do have to be a bit coerced to pop in - under the therory that the socket will straigthen them a bit too ...

If it's dual it's kind of cool ... 😎

--- GA586DX --- P2B-DS --- BP6 ---

Please use the "quote" option if asking questions to what I write - it will really up the chances of me noticing 😀

Reply 14 of 29, by AlexZ

User metadata
Rank Oldbie
Rank
Oldbie

One way to avoid such issues is to ask seller to test them before sending. I sell CPUs and always check they still boot to BIOS.

I straightened pins using credit card and it didn't result in the CPU being killed.

Pentium III 900E,ECS P6BXT-A+,384MB,GeForce FX 5600, Voodoo 2,Yamaha SM718
Athlon 64 3400+,Gigabyte GA-K8NE,2GB,GeForce GTX 275,Audigy 2 ZS
Phenom II X4 955,Gigabyte GA-MA770-UD3,8GB,GeForce GTX 780
Vishera FX-8370,Asus 990FX,32GB,GeForce GTX 980 Ti

Reply 15 of 29, by stealthjoe

User metadata
Rank Member
Rank
Member
AlexZ wrote on 2025-06-19, 07:29:

One way to avoid such issues is to ask seller to test them before sending. I sell CPUs and always check they still boot to BIOS.

I straightened pins using credit card and it didn't result in the CPU being killed.

I did ask the seller to test them prior to shipping for which he affirmed. Not sure why it turned out otherwise.

Intel 845GEBV2, Pentium 4 2.4 Ghz, Geforce FX5600 256MB, 512MB RAM, 160GB HDD, Sound Blaster Live! SB0100 - Win 98.

Reply 16 of 29, by stealthjoe

User metadata
Rank Member
Rank
Member
H3nrik V! wrote on 2025-06-19, 06:12:

Force is such a strong word here ... 🤣 but well, sometimes, they do have to be a bit coerced to pop in - under the therory that the socket will straigthen them a bit too ...

I get it 😊. I am a little paranoid about using even the slightest of force for fear of breaking these things. These stuff are not easily available where I live and it makes me even more cautious about handling them. But darn, the odds always work against me!! 😞.

Intel 845GEBV2, Pentium 4 2.4 Ghz, Geforce FX5600 256MB, 512MB RAM, 160GB HDD, Sound Blaster Live! SB0100 - Win 98.

Reply 17 of 29, by Spark

User metadata
Rank Member
Rank
Member

I've bought five S370 cpu's off ebay over the last few years.
None of them would go into the socket on arrival, all needed many legs bending back with a pen-knife blade, and all worked.
Even then I always have to give them a wiggle and a push when swapping in.
I think you've been unlucky with the cpu's. I doubt you've broken them by bending back the legs.

One thing I did do was loosen off the heatsink clip a bit for fear of cracking the die. That thing was on there like a vice when I first got the PC.

Reply 18 of 29, by momaka

User metadata
Rank Oldbie
Rank
Oldbie

My take with s370 P3's - they tend to snap pins very easily. I've had AMD and P4's come with pins literally bashed at 90 degrees and still managed to straighten them without breaking anything. With P3, if the pins are even slightly more bent, there's always the risk of them snapping. This leads me to believe P3 CPUs don't have very thick connections going to the pins... so I think it's possible that perhaps some of your straightened pins are not making a good connection to the CPU substrate anymore and thus the reason for not working. That said, try re-inserting and re-testing each CPU several times into the socket. Also, does the behavior change if you apply pressure to the CPU (without the heatsink) and power on the board?

And other thoughts to consider: do your boards consistently work with that one working 800 MHz P3? As in, can you do several boots in a row? What about if you leave the board turned off for a few hours and then try? The idea here is to rule out any motherboard shenanigans (be it due to possibly having bad caps or whatever else may be wonky with that hardware.)
Also, with the working CPU, can your RAM always and reliably pass tests like Memtest?
PSU? Known good brand and model and checked for bad caps?

I know I keep mentioning the caps like I have some kind of obsession with those (well, I won't lie that I do to some extent 🤣 )... but I've been finding too much old hardware lately that's affected by them (half the time without any showing any signs.) So always worth investigating on that front too.

As for damaging the CPU due to using a CC - I highly doubt it. I've used CC for straightening the pins on a dozen of CPUs and never had one go bad due to that.
Also, since you say you live in a place with high humidity, that really lowers the chance of damage from static discharge pretty much down to nothing.

AlexZ wrote on 2025-06-18, 16:34:

Some people touch electronics everywhere and kill it.

I do that all the time (handle electronic boards by touching wherever) and have never killed a single one by static discharge - even when I'm often walking at home with my wool socks (in the winter) on the carpet.
It really takes bad luck to kill modern stuff by way of ESD.
I do agree that 3 dead CPUs in a row is too much, though - especially P3's. I've also not seen P3's with cracked/chipped dies as much as Athlon/XP's. So all in all, I don't think this is the issue either.
Probably like you said - either really bad luck or maybe O/P or O/P's hardware is doing something "funny" to make the CPUs not work.

Reply 19 of 29, by stealthjoe

User metadata
Rank Member
Rank
Member
momaka wrote on 2025-06-19, 20:51:

And other thoughts to consider: do your boards consistently work with that one working 800 MHz P3? As in, can you do several boots in a row? What about if you leave the board turned off for a few hours and then try? The idea here is to rule out any motherboard shenanigans (be it due to possibly having bad caps or whatever else may be wonky with that hardware.)
Also, with the working CPU, can your RAM always and reliably pass tests like Memtest?
PSU? Known good brand and model and checked for bad caps?

I checked the 800mhz processor after switching off, dismantling and then reconnecting the board after a day or two. It posted fine with no issues. The PSU I am using is Gigabyte P450B which is supposed to be a decent and safe one.

I had earlier checked the first set of processors (600 and 800mhz) on 3 different boards with no luck. So it looks like the issue is with the processors.

Intel 845GEBV2, Pentium 4 2.4 Ghz, Geforce FX5600 256MB, 512MB RAM, 160GB HDD, Sound Blaster Live! SB0100 - Win 98.