- Oct 13, 1999
- 22,377
- 2
- 81
I know this is old news to some here, but last night I was REALLY jazzed about this...
I've been wanting to try this for months. The system used to be one that I used daily, so I didn't want to mess with it. However, it has been replaced. The old system is a P3-700@933 on an Abit BX133-RAID plus the usual assortment of other stuff. The replacement system was a Celeron 1.0A@1.33 on an Abit VH6-T (plus the rest of the system). I gave away the 1.0A to a friend and got a 1.2, which would not hit 133FSB, only working stable at 130FSB and a somewhat high temperature (wanted to keep the system fairly quiet, so had a slow speed CPU fan). So, I backed down a step to a Celeron 1.1A@1.46 (does go higher, but I want the easy 133FSB with everything else running at spec). Now I had a replacement system, what should I do with the old system that I no longer used daily? Put the Celeron 1.2 in it, of course! Since it wasn't my "main" machine, if I happened to break it, oh well. So, I go to my old bookmarked pages and re-study all the information. Alright, voltage tweak, mask those three pins, connect those two pins... flash the BX133 BIOS to the newest version (I was still running the original release BIOS, hehe, but the system had been stable and overclocked for 25 months using Windows Me without any reinstalls, used daily for almost 20 of those 25 months), ream out three holes in the CPU socket so the masked pins can fit... Didn't work out too well because I had problems masking the pins and trying to get the CPU to sit properly in the socket. Finally I got fed up and just got some needle-nose pliers and broke those three offending pins off by rocking them back and forth. Okay, after straightening the pins around those three that I bent, I inserted the CPU (had to rock it a bit since everthing had been bent up and restraightened) and hit the power switch. Nothing. Took the CPU out and noticed that the wire connecting the two pins (one of which I broke off only half, so the wire would stay put) was burnt. Oh no! Took the wire off. Hmmm, may not need it after all since some people said it wasn't needed and others argued over which pin to connect. Still didn't POST. Tried the CPU in a Shuttle AV18ET (supporting Tualatin, but wouldn't overclock the 1.2 that well either) and it POSTed. Whew, I didn't burn out the CPU. Put the old P3 back in the BX133, POSTed. Okay, broke the last stub off the CPU (if I needed to connect the two pins, then I'd have to get some conductive ink). Put it in, it POSTed. YEAH! Hmmm, it only showed as a Pentium III 172B or something like that. Checked BIOS, reports 1.30 volts as default, even with my wire trick to up the voltage. Okay, remove the wire, still reads 1.30V default. Okay, re-do the wire, but this time do VID1, VID2 AND VID25 for slightly higher default voltage. SUCCESS! Now got a default of 1.65 or so. Bump the voltage a bit above that, set to 133FSB, VOILA, boots into Windows. My findings so far... in the SoftMenu there is some latency setting for the CPU that can be 1 or 8, with 8 being better and 1 being slower but more stable. Set to 8, sometimes Windows doesn't finish booting or applications crash when I try to run stuff. Set to 1, seems stable after a few hours of benchmarking. I'm wondering how much performance I'm losing from that, and if connecting those two pins, connecting VttPWRGD (one of the pins I had to break off) to either PWRGD or VTT would allow me to run at 8 instead of 1?
In any case, I'm pretty pleased that my BX chipset motherboard is running a 1.6GHz CPU, even though at POST the BIOS reports it as being something like a Pentium III 576B Processor, hehe.
I've been wanting to try this for months. The system used to be one that I used daily, so I didn't want to mess with it. However, it has been replaced. The old system is a P3-700@933 on an Abit BX133-RAID plus the usual assortment of other stuff. The replacement system was a Celeron 1.0A@1.33 on an Abit VH6-T (plus the rest of the system). I gave away the 1.0A to a friend and got a 1.2, which would not hit 133FSB, only working stable at 130FSB and a somewhat high temperature (wanted to keep the system fairly quiet, so had a slow speed CPU fan). So, I backed down a step to a Celeron 1.1A@1.46 (does go higher, but I want the easy 133FSB with everything else running at spec). Now I had a replacement system, what should I do with the old system that I no longer used daily? Put the Celeron 1.2 in it, of course! Since it wasn't my "main" machine, if I happened to break it, oh well. So, I go to my old bookmarked pages and re-study all the information. Alright, voltage tweak, mask those three pins, connect those two pins... flash the BX133 BIOS to the newest version (I was still running the original release BIOS, hehe, but the system had been stable and overclocked for 25 months using Windows Me without any reinstalls, used daily for almost 20 of those 25 months), ream out three holes in the CPU socket so the masked pins can fit... Didn't work out too well because I had problems masking the pins and trying to get the CPU to sit properly in the socket. Finally I got fed up and just got some needle-nose pliers and broke those three offending pins off by rocking them back and forth. Okay, after straightening the pins around those three that I bent, I inserted the CPU (had to rock it a bit since everthing had been bent up and restraightened) and hit the power switch. Nothing. Took the CPU out and noticed that the wire connecting the two pins (one of which I broke off only half, so the wire would stay put) was burnt. Oh no! Took the wire off. Hmmm, may not need it after all since some people said it wasn't needed and others argued over which pin to connect. Still didn't POST. Tried the CPU in a Shuttle AV18ET (supporting Tualatin, but wouldn't overclock the 1.2 that well either) and it POSTed. Whew, I didn't burn out the CPU. Put the old P3 back in the BX133, POSTed. Okay, broke the last stub off the CPU (if I needed to connect the two pins, then I'd have to get some conductive ink). Put it in, it POSTed. YEAH! Hmmm, it only showed as a Pentium III 172B or something like that. Checked BIOS, reports 1.30 volts as default, even with my wire trick to up the voltage. Okay, remove the wire, still reads 1.30V default. Okay, re-do the wire, but this time do VID1, VID2 AND VID25 for slightly higher default voltage. SUCCESS! Now got a default of 1.65 or so. Bump the voltage a bit above that, set to 133FSB, VOILA, boots into Windows. My findings so far... in the SoftMenu there is some latency setting for the CPU that can be 1 or 8, with 8 being better and 1 being slower but more stable. Set to 8, sometimes Windows doesn't finish booting or applications crash when I try to run stuff. Set to 1, seems stable after a few hours of benchmarking. I'm wondering how much performance I'm losing from that, and if connecting those two pins, connecting VttPWRGD (one of the pins I had to break off) to either PWRGD or VTT would allow me to run at 8 instead of 1?
In any case, I'm pretty pleased that my BX chipset motherboard is running a 1.6GHz CPU, even though at POST the BIOS reports it as being something like a Pentium III 576B Processor, hehe.
Facebook
Twitter