With Socket 370, you definitely have to watch out for FC-PGA and FC-PGA2 compatibility. I definitely wish I'd done more research before picking up that Abit BP6 years back, not knowing that board needs extensive modification to support FC-PGA Coppermine CPUs, especially in dual CPU mode as intended.

Meanwhile, more adventures with the Power Mac 9600:
-Going non-interleaved on RAM really helps with stability, that's for sure. FireWire transfers just work, the Finder hasn't crashed without input even once...
-Unreal Tournament is a real wake-up call, though. It's a slideshow on the 9600 with its Rage 128, moreso than I remember on my iMac G3 350, which wasn't all that great at running it, either. On top of that, 192 MB of RAM is not enough to actually get into a match without being slapped with an "out of memory" error. I guess this one's best reserved for late G4s that you can slap a fast AGP Radeon into - or you could just take the sensible route and use pretty much any PC from the last decade.
-I shudder to think of how much file system corruption I would've had by now in all the time I've been tinkering with this thing, had I not had Norton Utilities on hand. Seems like the worst of it happens if the system just seizes up during a FireWire transfer.

At the very least, I'm gonna cram more RAM in there, maybe a Radeon of some kind. I suppose it wouldn't hurt to borrow the 9250 PCI that I slapped into my MDD temporarily, and see how badly the 9600's dated architecture with the 50 MHz bus and EDO DRAM that I can't even run interleaved bottlenecks the whole thing...

dionb wrote:

TheAbandonwareGuy wrote:

Don't copper mines use a different voltage setting? I would think some sort of board revision that doesn't support it or a bad capacitor or something causing it to be too far out of specification to be the most likely cause.

Yes, they run on 1.6-1.75V as opposed to 1.9V-2.1V. There is an earlier revision of this board (Rev:1) that does not support them, but the Rev:2 definitely should, and all VRM-related components are identical down to IC or transistor part number. It also doesn't really look like a marginal/failing voltage issue; the error code is very reproducible and stable and behaviour is identical between a P3-600 and a P3-1000 that draws almost 50% more current.

I'm open to the idea it's something failing though, but question is where to start looking...

Stupid question but have you tried reflashing the BIOs back to the known good one anyways?

Maybe the board was originally used by an OEM with a BIOs modified in the backend that can't be seen. If it's not that or something failing I'm stumped.

I just removed two barrel batteries from 486 motherboards. The first went as usuall, little bit of new solder to make the old stuff flow and in 2 tries the battery already fell off the board.
But the second one, where the battery already had been changed once I had to put the solder iron up to 450 degrees before the solder after a long time finally started to flow. One side was neatly removed and resoldered by the previous owner, this side the battery immediatly "sprung" loose from it's hole. But I don't know what he had done on the other side because I needed to heat the solder blob more then 3 minutes straight before it did melt again a bit. After more then an hour I finally had removed most solder and then finally everything came out. It was not only part of the leg of the old battery that was still in it but also a piece of a needle or something that came out of the stretched up hole.
I guess he must have had issues removing the previous battery and use this needle thing to push it out and then solder the new battery in the same hole pushed against it. 😒

Never ever had such issues with a simple battery removal, really had to stay patient the whole time and what kind of pieces of art other people can leave behind.. I am also curious what kind of solder he used, even factory solder never required the full 450 degree Celsius of the 100W iron.. 😐

TheAbandonwareGuy wrote:

Stupid question but have you tried reflashing the BIOs back to the known good one anyways?

Maybe the board was originally used by an OEM with a BIOs modified in the backend that can't be seen. If it's not that or something failing I'm stumped.

I did more than that, I physically swapped the BIOS EEPROMs. No effect, regardless of which BIOS was in the board, one ran CuMines, the other one didn't.

But I already mentioned all this in the topic: Two identical boards, yet different CPUs working

Did a whole load of research to try and track down a R547 or R542, socket 7 board... Still no luck...

Totally not getting crazy 😵 😊

obligatory:

https://www.youtube.com/watch?v=ghrdSTC66MA

Had an interesting Dell GX110 pentium 3 machine come in at work for data recovery. Been running 24/7 since it was supplied new...there was enough dust inside to knit a jumper with!

With mine, decided to stop my SKT370 build as it wasn't going where I wanted. So decided on restoring and old Optiplex 330 i had laying about for WinXP stuff. Started by replacing the main CPU/case fan with a new 90m Arctic F9 and adapting it to suit the dell proprietary PWM connector. Dropped in a spare C2D 6750 I had and moved across my GT710 for basic GPU duty.

Total cost £3.40 so far for the fan...

GX110 cool one of my favorite OEM systems. I still need a 3 slot ISA riser for my Low Profile desktop GX110.

appiah4 wrote:

GX110 cool one of my favorite OEM systems. I still need a 3 slot ISA riser for my Low Profile desktop GX110.

Shame it will most likely have to go back to the client to be thown away, would love to restore this one...

Pentium iii Coppermine 1Ghz/133fsb installed aswell 😢

CarlHopkinsUK wrote:

Shame it will most likely have to go back to the client to be thown away, would love to restore this one... […]
Show full quote

appiah4 wrote:

GX110 cool one of my favorite OEM systems. I still need a 3 slot ISA riser for my Low Profile desktop GX110.

Shame it will most likely have to go back to the client to be thown away, would love to restore this one...

IMG_20180516_083753.jpg

IMG_20180516_083814.jpg

Pentium iii Coppermine 1Ghz/133fsb installed aswell 😢

I have the exact system, Low Profile. Mine came with a Coppermine 733 and 256MB ram but I upgraded it to a Coppermine 1000B/133Mhz and 512 MB. I have the Riser card populated with a Radeon 7000, a Terratec 128i and a USB 2.0 card, with a TP-Link WN821N Wifi adapter installed internally 😀 It is my Win98/Win2K/Warp4 PC.

gimme

also upgraded the compaq from 700MHz Duron turd to 1.1GHz Thunderbird

Nothing special, just replaced the cmos battery on my ThinkPad X20. Poor thing was flat as a pancake having been in the machine since it was built (28-August-2000, according to the time stamp in the main battery compartment). Nothing a new battery and some heat shrink couldn't put right. Its been bugging me for a while having to put in the time/date every time I powered it up but its sorted now.

Digitizing a Bee Gees' 1st Album cassette whilst eating scones my mom made, as well as watching Looney Tunes; and going to compose a CD for iTunes and for my mom's car radio and my car radio.

Disassembled one of my industrial systems that I purchased for the case itself and maybe to use some of the other hardware.

Installed the PIAGP backplane with the ISA bridge board (AGP, 4x PCI, and 6 ISA slots).

Hopefully going to be sweet when I get it all put together.

Next will be using one of the cases for an 80386DX-40 and ISA backplane setup.

Then I will still have one case (without a top cover) to use for something else... maybe testing or possibly for a build with an ASUS P5A-B if the mounting holes for the motherboard line up with the case.

I started the cleanout I'm having to remove redundant pieces of hardware + generations out of the mix. Anything newer than socket 478 is gonna be given away, including a HP branded core2duo motherboard and an MSI AM2 board full of bad caps (this might end up going to recycling since it does have issues, and still common as anything here), along with a couple dead 478 boards (bad PSU blew out the power rails before I received the systems, just wanted them for the GPU and other parts). I'm keeping all of the CPUs from the boards I'm tossing, since you never know with things 🤣

Experimented with the 486s again...

got a another LuckyStar LS486E board, this time FPM only, and drove it up to 66mhz FSB. It is a tad slower than the EDO one and also less stable. Former is rock solid(with heatsink on the Northbridge chip) while latter is flaky. A 50ns FPM stick did not change stability. The funny thing is that the chipset on the FPM only board is newer by around 20 weeks than on the other LS486E board - I wonder what criteria it is that enables a board to work withEDO ram besides the chipset?

I changed the BIOS but this did not enable "EDO support".

I have begun to take all my loose PCI and AGP gfx cards for a spin. So far no dead ones have been found.
Tested so far: Viper V550 TNT (aka TNT1), TNT2-Ultra, Geforce2 GTS ViVo, Matrox G400, Matrox G400-MAX and Voodoo3-3500. I do not have that many loose cards in the box, so I am halfway through testing. Must be enough for tonight, as it is already 1 am here.

Fiddled with my Windows 98SE machine (Athlon 64) by diagnosing my Quantum Fireball ST 3.5 Series HDD (not SMART compatible sadly) using a live USB of Linux and the benchmark crashed after formatting the drive, but, worked after that and it looked promising.

Then, Windows 98SE on the main drive was getting errors left and right (IDE controller drivers stopped working, installed Windows over the current files, then it worked, but my sound card had no MIDI and reinstalling the sound card's drivers gave a program error, and a blue screen after rebooting Windows). I diagnosed the issue, so I pulled the 512MB RAM out and kept the 256MB in place, re-installed Windows 98SE after running the degree command to remove the Windows directory, and no program errors or crashes (blue screens), and everything on my sound card was working once again.

This page is helpful for using 512MB RAM or more in 9x http://www.thpc.info/ram/vcache98.html

Ran some CGA calibration utilities on my XT build to help calibrate the LCD monitor I'm presently using.

I'm not entirely happy with the image output from the VCR, as it makes the image a bit dark with some slight ghosting and loss of some fine detail. Going to have to try a few other things to try to get a reasonable composite output to this monitor.