A moment of silence for PixelView 6800 GT, died in the main menu of GTA: Vice City. I didn't even have a chance of doing any nefarious deeds before it went, I had only just installed the game and was setting up controls in options menu. That's like dying at the front entrance of a strip joint, what a way to go.

All hail Gigabyte 6800 GT, may its reign last longer than its predecessor. Like maybe a entire full year, that would be great.

Picked up an ASUS P2L97-DS dual slot 1 mobo and have been having a time with it. I was recently messing around with a baby AT slot 1 board and discovered the joys of the 440LX chipset. Now with this dually, I think I might actually hate the LX chipset - this thing has been fighting me every step of the way. I think many of my troubles have from just not RTFM, but still.

After a BIOS update, I got it to POST with 2 x 500MHz Katmai CPUs. The board actually supports 100MHz FSB, but I was worried doing that might be overclocking the other busses so I put it back down to 66MHz to try and get it working the rest of the way. I also reduced the RAM down to supported config by the manual. Still, I have not been able to get Win2K or XP to install - crashes on "Starting up Windows" just before you config the HD. I've tried the various kernal options - MPS Multiprocessor PC, ACPI Multiprocessor PC, etc and nothing. I was able to get Win2K working with a single Celeron 333, but that's about it.

So, I'm guessing it just hates Katmais and I'm out of luck for now (don't have any other CPUs to try out).

ubiq wrote on 2023-05-17, 23:40:

Picked up an ASUS P2L97-DS dual slot 1 mobo and have been having a time with it. I was recently messing around with a baby AT sl […]
Show full quote

Picked up an ASUS P2L97-DS dual slot 1 mobo and have been having a time with it. I was recently messing around with a baby AT slot 1 board and discovered the joys of the 440LX chipset. Now with this dually, I think I might actually hate the LX chipset - this thing has been fighting me every step of the way. I think many of my troubles have from just not RTFM, but still.

After a BIOS update, I got it to POST with 2 x 500MHz Katmai CPUs. The board actually supports 100MHz FSB, but I was worried doing that might be overclocking the other busses so I put it back down to 66MHz to try and get it working the rest of the way. I also reduced the RAM down to supported config by the manual. Still, I have not been able to get Win2K or XP to install - crashes on "Starting up Windows" just before you config the HD. I've tried the various kernal options - MPS Multiprocessor PC, ACPI Multiprocessor PC, etc and nothing. I was able to get Win2K working with a single Celeron 333, but that's about it.

So, I'm guessing it just hates Katmais and I'm out of luck for now (don't have any other CPUs to try out).

IMG_4538.jpeg
IMG_4531.jpeg

Maybe like Athlons, it needs more juice on the +5V rail? What are the Amps and W ratings of the PSU?
1 cpu is rateed 38W x2 might suck too much? Just an idea.

my activity today is selling a pair of gray 2GB DHX C5DHX corsairs 6400 (try turning evrything off in bios , on the Asus , set ram CL manually to standard settings , give 0.1-0.5V more on ram if need-be) , set jumpers manually if you have , and plese mount 1 HDD on IDE cable at a time (Preferably correct ide cable) with jumper settings put on Master , or slave if you have DVD-ROM in there. if nothing works cable select hdd , may be a hdd issue if DOS is stable.

ubiq wrote on 2023-05-17, 23:40:

Picked up an ASUS P2L97-DS dual slot 1 mobo

So, I'm guessing it just hates Katmais and I'm out of luck for now (don't have any other CPUs to try out).

the 440LX chipset doesn't support katmai P3s. they may function in that the voltage and multiplier is supported, but results would be unpredictable. 100MHz FSB P2 chips will work (at 66 or sometimes higher).

best option would be either celeron 300As at 75-83MHz FSB or P2 366 chips at 66MHz FSB. the onboard SCSI controller will likely stop working if the FSB is increased.

Also how does LX handle AGP speeds?
I mean BX does 1/1 at 66MHz FSB and 2/3 at 100MHz FSB.
Would Intel include 2/3 in LX if it would be limited to 66MHz FSB?

^^^ My understanding is that LX predates 100MHz+ FSBs, so didn't include any dividers for the other busses; anything other than 66MHz will always be overclocking on an LX.

libby wrote on 2023-05-18, 21:31:

the 440LX chipset doesn't support katmai P3s. they may function in that the voltage and multiplier is supported, but results would be unpredictable. 100MHz FSB P2 chips will work (at 66 or sometimes higher).

best option would be either celeron 300As at 75-83MHz FSB or P2 366 chips at 66MHz FSB. the onboard SCSI controller will likely stop working if the FSB is increased.

Thanks, that was pretty much what I figured - a hard lesson in that just getting something to POST only means so much.

I don't have any SCSI devices - and from what I know I may as well stick to IDE if I'm using a CF/SD adapter; not messing around with device ids and whatever.

Anyway, I managed to find a P2B-DS that should be arriving in a week or two so this has all be good experience that will be useful when it arrives. 🙂

Thermalwrong wrote on 2023-05-14, 00:48:

Looks good, how is it in terms of heat? I've got a very similar Dell pizzabox case with the Netplex 4/25 which right now has a s […]
Show full quote

LewisRaz wrote on 2023-05-13, 11:53:

Benched my DX4 Overdrive against the stock DX2 in my dell 466i.

https://youtu.be/yL86qGZdJW0

Looks good, how is it in terms of heat? I've got a very similar Dell pizzabox case with the Netplex 4/25 which right now has a somewhat slower sx25. Does your one have a fan in the front possibly?

The other day I was going through some of my old boards that were put away as 'for parts', since I've found more than one board I thought was dead actually works sometime later because of a dumb thing I missed.
A Toshiba Satellite 110CT board that I thought had been broken when I tried to clean its corrosion, turned out to work quite well but I was using the wrong power supply which it would refuse to power from. (it needs 18.2v internal AC>DC PSU, it was mixed up with a 15.2v PSU from another model). Put it back together and found the internal keyboard / mouse were all messed up, but external kb/mouse worked. Known good laptop keyboards wouldn't work on it and it appeared that the battery leak had got to the keyboard area first. Decided to pull off the keyboard connector and found it was so badly corroded and stuck on there that in the end I just cut it up and broke chunks of it off. Some traces had gone bad:
Satellite-110ct-keyboard-tracing.jpg
For reference, from left to right everything except the last 6 pins is the keyboard connector which integrates the upper and lower layers into one connector. Then the last 6 are the Toshiba Accupoint mouse which is a 4-way variable resistor between 5v and gnd to make up the left/right and up/down directions. The keyboard connector is mostly the same on the Toshiba Satellite T2100 to Satellite 400 / 410 / 420 / 430 series laptops.
I was able to repair most of the traces underneath and put solder mask on everything to keep it in place and not melt off:
sat110-kb.JPG
Took a connector from a board I thought was dead (hot air 250c at full speed) and soldered it onto this 110CT. That got the keyboard working then the mouse was not detected unless an external one was plugged in. That turned out to be the Alps chip which is the internal pointer controller had gone bad. Borrowed that from the same board and now the mouse works. It got downgraded to a Satellite 110CS with a DSTN screen in the process because that's what was left and now it's a pretty nice laptop that was a pile of junk parts a little while ago.

In the process, I learnt that using hot air soldering on a Toshiba Satellite motherboard when it's not secured to anything is a *terrible idea*. These motherboards have large QFPs on both sides and if the board flexes while the PCB is heated, legs from those QFPs will come unstuck. I thought I'd re-killed this Satellite 110CT but through careful resoldering of the QFP legs and inspecting it was working again.

That got me thinking, I had bought a Toshiba Satellite 430 motherboard some months ago which I thought I'd killed because I put hot air on it after failing to get the VGA working right, thinking it was a badly fitted VGA card connector, I tried resoldering the QFPs and eventually it stopped posting at all after I'd worked on it a bit. I thought that maybe some internal vias were broken and these boards are fragile but no. These boards just have lots of pins and lots of things to get wrong.
Looking at it a few months later and in bright daylight I made tons of mistakes, bridged legs on some chips and the corner of one QFP had pins just completely off the board.
Delicately resoldering those and cleaning up my mistakes the board now works again and with a scrap Satellite 400CDT case, along with a mostly broken video card that can't run an LCD display, it's a working system again - til I get some better parts for it:
sat430-pcb-1.jpg
The speaker is from an HP laptop and the buttons are hot glued into place. I'm upset about the arrow keys, notice they're a different colour? Turns out melting hot glue with a hair drier also melts keyboard keys 🙁
sat430-pcb-2.jpg
The 400CDT's case is smaller than the 430CDT so uh, I made it fit.
sat430-pcb-3.jpg
The last Satellite 4xx power supply I tried fixing had a load of corrosion in it and went bang when I last powered it, so stuff it, now the laptop runs from DC instead. This could be useful in future when more of the internal AC>DC power supplies on these 400 series laptops fail to capacitor leaks.

Also hopefully just helped epictronics with overclocking an IBM 486 motherboard

Wow. I-m impressed... I wish I had those skills.... Sadly I was unsuccessful in fixing stuff with hot-air station , Tryed one time to change a PowerManagementChip on a HP , but traces just colided and nothing worked , keep't shooting down the powersupply. I must Admit I-m not that good. however you? to save that old Japaneze Marvel... epic.

Meanwhile yestrerday I got some supply-es : (they where cheap 0.1 euro/piece)

While sorting out my ISA sound capable producing digital out I realized that my CT4500 have also SPDIF contacts on PCB. So after some DuPont Connector and custom DIY RCA adapter cable fun ended up with following contraption (attached pic). Well it ended up working in DOS partially. Should have done some research first. AWE SPDIF is only 16bit/44.1khz compatible. So tracker music is N/A. Forced myself trough my FM music library (CQM yay!) . 😀
Midi is more enjoyable by EMU8000 in crisp noise free digital format. I had AWE32 back in a days so it hit my nostalgia feel a little bit.

I'm just wondering how should I set cabling out in more permanent manner. Probably via separate daughterboard...

I'm making boxes for motherboards. It's really time-consuming work, but if I don't make them, no one else will.

Turbo -> wrote on 2023-05-20, 16:30:

I'm making boxes for motherboards. It's really time-consuming work, but if I don't make them, no one else will.

Great ! a place near me used to have good generic "assemble your self" boxes that fit most all AT/ATX boards like that for very cheap, but they quit carrying them.
I can get free ones from the post office but they have usps logo's all over them and are not as good imho

Today decided to test a few HD's and DVD-roms and found issues, was not the drives but my SATA test cable went bad some how......try a diff cable and the drives are fine 😀

Got friends coming around in a few days for drinks & 1 of my retro gaming nights that I hold each year. So today I've been doing some spot checks on the hardware I'm going to be using to ensure there's no problems. Mostly Microcomputers & a couple of PCs for this night.

Good job I did test in advance, as 2 of the TVs I intended to use my TI-99/4A on did not like it's Component output and I didn't fancy switching to RF. Found my oldest LCD TV happily accepts the output and the TI-99/4A looks pretty fantastic on it, super sharp and vibrant colours.

The only other issue I encountered so far was my XT's MFM controller card had been dislodged from the ISA slot. I noticed right away when the system booted super fast from the CF card instead of the slower HDD. I reseated the card and all was well. It does beg the question of why I still bother with the old slow & noisy MFM Hard Drive when the CF card does a much better job without making a sound - but ultimately nostalgia and the novelty factor wins out. 🤣

Yay, my suggestions for Epictronic's 486 motherboard worked 😀

Also lately I've been running low on 1MB 30 pin simms, I even found a couple of dead ones - but I've got these stupid 256kb simms and they have extra solder pads...

60ns and made in 1990? That's pretty fancy actually but 256kb sticks aren't really useful to me since I'm usually setting up 386/486 that want 32 bits and I've only got enough here for 16-bits. Even then, that's not much RAM for a 386sx either.
These ones with two sets of solder masking are wired up for either 2x 256Kx4 chips or 2x 1024Kx4 and the parity chip differs - I have two sets like this, there are these CuBig 60ns modules and a set of 4 Micron 256KB 100ns modules that are laid out pretty much the same.

My Quick 861DA hot air station got the chips off without issue at 280c with high flow rate - the chips are stored away and might even be useful for fixing video cards or older systems.
These CuBig were the second set I did where I was much happier with the process: Take off the chips from the 72-pin and 30-pin simms, put them all aside and tin up the pads with fresh solder & flux on the SIMM PCBs. Move the SMD capacitor that sits under each RAM chip to where the larger parity chip sits. One SMD cap broke while cleaning it and the 72pin donor RAM was obviously in a smoker's PC last.
Once the PCB's pads were tinned up with a bit of solder on each pad, the cheap flux was cleaned off and I put some chipquik smd no-clean tack flux on there - much nicer for hot air soldering / component placement and easily cleaned, while my cheaper fluxes leave rosin all over which tough to clean and maybe impossible once SMD chips are soldered into place.
Put the 1MB chips onto the PCB's pads which is a bit tricky since the pads are now raised. This is where solder paste would be nice but it needs a stencil or better application skills than I have. Hot aired them on at 280c again with a lower flow rate so the chips don't blow away and wait for each RAM chip to drop into place.
Then installed the SIMMs onto my test computer which was set up on a floppy to boot memtest and these CuBig modules now pass memtest on a 386dx-40 😀
This was the 2nd set, the first set I tried re-using the solder already on the pads and had to manually re-work about half the RAM chip legs to get the computer to detect / work with them. These Micron sticks also had a resistor connection A9 I think to Gnd? That was okay when they were 256KB simms but now they're 1MB the computer hates that. So I just took off the resistor:

There we go, now I have 6x more 30-pin SIMMs that are somewhat more useful than before.

I could save a bunch of Athlon MP systems, too much for me alone but I needed to rescue them. Some are damaged (DIMM slots), one doesn't POST and a mosfet smells like burned electronics and 2 POST but crash on Windows boot. But I'm slowly booting every one of them into Windows and running some 3Dmark loops to verify stability...

Fortunately they are the B2 chipset revision of the 760MPX platform.

I got an Asus p3v133 board a while ago that wouldn't post and it always hang on code "0B" on the post card.
I couldn't figure it out what is wrong with it until I decided to give it another try,
this time I find a 22 ohms resistor array on top of the agp slot with broken corner and it was open,
so I soldered a 20 ohms resistor in parallel with the broken one and that fixed the issue.

this resistor is on the stop signal between the north bridge and south bridge/pci slots.

Karbist wrote on 2023-05-21, 20:25:

I got an Asus p3v133 board a while ago that wouldn't post and it always hang on code "0B" on the post card. I couldn't figure it […]
Show full quote

I got an Asus p3v133 board a while ago that wouldn't post and it always hang on code "0B" on the post card.
I couldn't figure it out what is wrong with it until I decided to give it another try,
this time I find a 22 ohms resistor array on top of the agp slot with broken corner and it was open,
so I soldered a 20 ohms resistor in parallel with the broken one and that fixed the issue.

RN15.jpg

this resistor is on the stop signal between the north bridge and south bridge/pci slots.

boardview.jpg

Where did you get the schematics?
I'm über interested in going to that place. 😀

Minutemanqvs wrote on 2023-05-21, 20:13:

I could save a bunch of Athlon MP systems, too much for me alone but I needed to rescue them. Some are damaged (DIMM slots), one […]
Show full quote

I could save a bunch of Athlon MP systems, too much for me alone but I needed to rescue them. Some are damaged (DIMM slots), one doesn't POST and a mosfet smells like burned electronics and 2 POST but crash on Windows boot. But I'm slowly booting every one of them into Windows and running some 3Dmark loops to verify stability...

Fortunately they are the B2 chipset revision of the 760MPX platform.

This is simply great!!!

Nexxen wrote on 2023-05-21, 20:41:

Where did you get the schematics?
I'm über interested in going to that place. 😀

https://vinafix.com/resources/boardview-asus- … top-tsict.1223/
Download link at top right.

Karbist wrote on 2023-05-21, 21:12:

Nexxen wrote on 2023-05-21, 20:41:

Where did you get the schematics?
I'm über interested in going to that place. 😀

https://vinafix.com/resources/boardview-asus- … top-tsict.1223/
Download link at top right.

Thanks mate! This is actually a goldmine!

Karbist wrote on 2023-05-21, 21:12:

Nexxen wrote on 2023-05-21, 20:41:

Where did you get the schematics?
I'm über interested in going to that place. 😀

https://vinafix.com/resources/boardview-asus- … top-tsict.1223/
Download link at top right.

Nice!
Do you have other links to such old hardware?
Could be saving many from being scrapped.