its a 820 chipset board... My personal feelings tell me to burn both boards in a fire, but that's not really helpful. Hot flashin bios'es has NEVER worked for me in the past. Ironically using a 3com nic to flash a bios chip has. I can help you with that if you have one and the bios chip fits into the NIC. Other than that, if the bios isnt totally bricked, follow the directions here: http://www.biosman.com/biosrecovery.html I've saved one board before like this.

Thanks! Yes, Having an RDRAm retro-board is losing it's glitter from so many of these little niggly problems. I'll check out the link. I've got a 3Com card somewhere, let me check it out.

ASUS has always been my favorite boards, they have always been high quality. Check also that the board is not flexed or warped, and check the caps for leaks/bulges too. If your patient enough and have a magnifying glass you can look around for cracked solder joints too.

No joy, everything I've tried. I'm probably going to file the socket 370 project and switch this over to a socket 423/850 chipset board w/ the rest of my setup remaining the same. Thanks for all the help everyone, I gave it an effort anyway.

Just buy an Eprom programmer on Ebay.. Those things arent expensive these days.. Then try to write an new bios on the Cmos chip.

I have been toying with the idea of getting a 820E based board but as far as I can see the 815E is actually faster. 820 can run 1GB but that's only so exciting.

BTW I have hot flashed a pair of nforce2 boards before. Booted up the board with the intact BIOS and swapped the chips and flashed the borked chip. It worked.

In my experience, the 820 is a lot nicer if you plan to run XP. The increased memory bandwidth helps a little and the lack of a 512mb ceiling makes the thing more practical. With 512 or 1gb memory and a 1ghz CPU, the 820 can still hold it's own.

It seems like most of the negative remarks about the i820 either refer to the version with MTH and SDRAM (those were buggy), is a repeat of old rumor or is no longer relevant to today's software.

All I know is back in the day I had the P3C-LS running a 1.267GHz in a slotket and 256MB RAM and it was FAST. I used to run all the sims in FBPRO 99 for our on-line league b/c compared to others systems it blistered the paint off the case.

You can tweak similar speeds from pc-133 on an apollo 133T chipset. Unless you go with a DDR board you won't really get much better. I have 1.5GB ram on mine w 1GB/sec read/write and about 690MB/sec copy @ 3-2-2-6. It's an ASUS TUV4X board, works quite nice really.

My first P3 ever was an 820 system. Had it for a long time... and can't say it was any faster than my subsequent 815 or Apollo 694X systems. FSB is the key to a fast P3- at around 150mhz even a lowly Coppermine 1ghz matches Athlon 1400/ P4 2ghz.

In my experience w/ a Rambus system, you really don't get any performance boost out of the chipset until you get >1.6ghz. Then it really moves. We did a wholesale PC upgrade at my employer around the RDRAM time-frame. With the initial CPU's at 1.3+ ghz they were moderately faster than the slot 1's we replaced, however, about a year later when we upgraded to 1.8 and 2 ghz cpu's (b/c they got dirt cheap after the 478 socket came out) the performance increase was more significant than just a processor speed jump would indicate. In this 423 system I'm building it will be a 1.8ghz cpu and a dual channel setup.

I liked RDRAM, aside from the lousy job Rambus did at marketing and licensing. Let's face it, that's really what killed RDRAM.

P4 and P3 are a lot different regardless of them both having RAMBUS chipsets. P3 is limited by its FSB (64-bit 66-133 MHz). This matches up perfectly to single channel 64-bit SDRAM. It can't use any more memory bandwidth.

I thought maybe RDRAM might help with AGP Texturing, but from what I found looking at reviews, 815 is faster at it than 820. Oh well.

P4 on the other hand has 4x the FSB bandwidth and so RAMBUS and dual channel DDR are ideal. But if you are running a low end Willamette the CPU is too slow for it to matter.

Digging through my stash, I've found an old Intel VC820. Perhaps I could use that for my socket 370 project, but now I'm kind of excited about this socket 423 project.

7cjbill2 wrote:

Digging through my stash, I've found an old Intel VC820. Perhaps I could use that for my socket 370 project, but now I'm kind of excited about this socket 423 project.

socket 370 has been done from hell to back. It's about time someone did an alternate retro build.

I had the same problem with a P3C-LS and no booting with no beeps. After cleaning the mainboard, all runs fine again.