V4P895P3/SMT V1.0 is one sucker. Got it somewhat workable after trying for a long time, process detailed in a wall of edits. Even a semi-broken 286 board was easier to get working compared to that thing.
Plus it simply doesn't POST when EDO memory is inserted, as it requires FPM. That alone took forever to notice since i had never had to deal with a board that didn't like EDO. Gotta learn something new every day!

Back around the year 2000, I was working at a computer shop and the worst boards were anything made by SuperMicro. They literally had a 90% DOA rate.

Before I worked there, a few years before that I was upgrading from a 486DX2-66 to an AMD 5x86-133 and the board I bought from them, despite specifically saying in the specs that it supported that CPU would always BSOD when installing Windows. They ended up swapping the board out for a more expensive model for free after they tested my original board and another board of the same model had the same exact issue.

I had owned 3 different PCCHIPS SIS 730 Motherboards

The Rev 1.6 was the best despite only working with old Duron Chips

The REV 5 was OK but less stable and good with the Duron 1200 I had

The Rev 7 I wanted to use a 200mhz FSB 2ghz Athlon from a laptop and the board finally had official 266mhz FSB and supposedly supported higher FSBs but the damn thing was unstable even when using an older CPU

It was the first integrated 3D board I owned that was “useable “

And for an extreme budget I got tons of use out of the things but the final rev 7 was a real letdown

I always had issues with Slot A motherboards, I traveled with desktop systems for my display and if it got cold my slot A systems would never post, had to remove the cmos battery and reseat the CPU, ram and pci/agp devices and they would eventually post.

1 dimm slot was bad, agp eventually failed and it was hit or miss what pci slots would work,
Was like that from day 1 to scrap

IDK really, 70% of the Asus/Abit I've ever received as "working" turned out to be dead, and 70% of the PCChips I was given as dead turned out to be working, flakiest POS I've had new out of the box is a GA-7VRX rev 1.1.

Can't think of many major duds, thankfully. I do have one in particular though.

I bought a Gigabyte GA-P35DS3L (rev. 1.0) that was a bit flaky. Coming off of socket 478, going to 775 and C2D was a major step, but this board gave me all sorts of headaches for a while there and it was not easy to troubleshoot it due to a variety of weird/faulty parts. The issues were with stability, random file corruption and even inability to detect any HDDs at boot.
However, I did have a noname PSU in that tower that I replaced very quickly and the problems persisted. Then, Seagate's 7200.11 fiasco basically bricked my HDD which I then had replaced ASAP (thankfully I had all of the important stuff backed up). But even after I had all those faulty parts replaced, the motherboard still would pull off some weird behavior, I could never trust it. Thankfully, I managed to find an ASUS P5K for dirt cheap at the time and finally, after about a year, I had a system that was 100% to my liking (I also swapped out an aging 7800GTX for a Radeon 4850 during the summer of 2008, good times). That P5K ended up being my favorite motherboard of all time, so I guess every story can have a happy ending.

I put that Gigabyte board on another system I built for my sister and it seemed to go alright, at the time I figured if she complained of odd behavior I'd just go ahead and get another board and burn that Gigabyte to the ground or something. Weirdly enough, the board seemed to fly just fine there. Then again, that system never saw much use. Some years down the line, I decided to decommission that computer and tried messing around with that Gigabyte board again, mostly for testing C2D CPUs with various GPUs I had laying around. Wouldn't you know it, the weirdness started again. After a bit of troubleshooting, I gave up. I've had a few P35 boards throughout the years and they all performed with zero hurdles, so I decided to cut my losses short, put the board up for sale dirt cheap (I seem to remember 20E back in 2013-2014) and put a big disclaimer front and center just so I was level with everyone. Within the day, about 15 people had messaged me about grabbing the board and I ended up meeting with someone outside a church nearby, he was ecstatic for it.

All in all, many fun memories and stories, so I guess it's all fine in the end. I wonder if it was my board in particular that was faulty or it was something they found out and fixed in rev.2.0.

Any motherboard that I've had to repair varta battery leakage on. I've replaced DIP sockets, ISA slots, installed new vias and eyelets, put in new traces, stripped and applied new UV solder mask, etc. It is fun the first few times, but gets old quick. Now I'll only do it for rare boards if I can pick one up cheap that the seller thinks is beyond repair.

The worst one I ever had in my main system was a PC Chip M741LMRT. I misdiagnosed a dead motherboard as dead Celeron CPU, bought a new CPU, then found out that wasn't the problem so needed an So370 or Slot 1 motherboard for less than EUR 50. Hopeless thing, with bad drivers, constant resource conflicts between the onboard/integrated devices and sometimes just randomly refusing to boot. After a few months I'd saved up enough to get an Asus CUBX. That was a MASSIVE relief.

Honorable mention: my housemate's Asus A7N8X-E Deluxe. Beautiful board. Fast board. Stable board. But the most picky thing when it comes to RAM I have ever encountered. In the end the only modules we could find that worked in his board were my Infineon DIMMs. Every single DDR module we had available worked on my board, a Gigabyte GA-7N400-L I bought as scrap and turned out just to have had a BIOS misflash- with exactly the same nForce2 Ultra 400 northbridge. So it was clearly wasn't the chipset but Asus' engineering that messed it up. To add insult to injury, my board benched about 1% higher too, so it wasn't even a matter of pushing the performance envelope.

I've been burned by more Asus motherboards and other parts then I care to remember. I avoid them like the plague.

chrismeyer6 wrote on 2021-05-30, 01:04:

I've been burned by more Asus motherboards and other parts then I care to remember. I avoid them like the plague.

That is interesting. I have 3 Asus Socket 7 motherboards and 1 Asus Socket A motherboard and all have been trouble free for me.

Yeah for some reason Asus products just don't like me. Hell I had a CD come apart in a 32x Asus CD drive.

jasa1063 wrote on 2021-05-30, 01:36:

chrismeyer6 wrote on 2021-05-30, 01:04:

I've been burned by more Asus motherboards and other parts then I care to remember. I avoid them like the plague.

That is interesting. I have 3 Asus Socket 7 motherboards and 1 Asus Socket A motherboard and all have been trouble free for me.

Except during the "Capacitor Plague era" (https://en.wikipedia.org/wiki/Capacitor_plague) have never had much issue with most any board maker (except some Pcchips and Compaq's) as nearly all brands fell into the problems with early cap death in early 2000's depending on the board. Asus was not the only one that had a problem and some of my best boards are pre-2000 Asus and some later ones did have some engineering issues as dionb mentioned above but so did others as far as ram compatibilty (ECS sure did). During that "cap plague" period switched to some Asrock, a spin off of Asus (new at the time in early 2000's) and all those boards from mid 2000's still have and work w/o a re-cap except one.

Lucky did not have to work with many SuperMicros but have heard horror stories of them long ago as cyclone3d mentioned....
So overall if I had to comment on the worst motherboards to work on since mid 1990's I would put some PcChips and some Compaq at my top of the list as for worst, but maybe I just forgot about the other bad boards 🤣.

chrismeyer6 wrote on 2021-05-30, 02:06:

Yeah for some reason Asus products just don't like me. Hell I had a CD come apart in a 32x Asus CD drive.

Asus does not make CDROMS, they just rebadge them, most were Lite-On but then again it could have been a defective CD too

I got an amptron 8600 thingy with TXpro chipset on it.

Got it working once, kinda, after modding the Dallas battery.

Decided to add it to the donor board pile, as I had plenty of other better socket 7 boards and couldn't be bothered fixing it.

My worst right now is what is found in Olivetti Envision P75. This thing never caught on and I can see why, at least as far a PCI slots are concerned. I couldn't get ANY PCI LAN card to work in it. They either stop POST (Realteks), freeze the machine during load (some intels), get recognized but fail to work (3com) or are completely misregonized. ISA cards do work but there's very few free IRQs, only IRQ3, 10 and 11. 5 is used by LPT port of all things, 7 and 9 by onboard sound

In addition it doesn't run any boot ROMs so stuff like XTIDE and RAID cards are out of question. The BIOS doesn't support LBA and doesn't seem to automatically translate large drives, I had to manually enter CHS values with translation to get a 4GB drive to work.

It also uses VLSI Wildcat chipset and there's no documentation available for it, and BIOS is garbage with almost nothing in it. I.e I cannot turn off COM and LPT or change any resources of onboard sound.

The board also has a spot for COAST but when I looked at the pinout it didn't match anything I have around, so there was no point giving the board cache. It has none onboard.

On the positive sides, with lack of LAN, it is a self contained machine that you can connect to TV via SCART (no RGB sadly) and you can have a sort of DOS multimedia PC something something. it is 430mm wide too and fits into a HiFi stack, if your other devices are black and curvy too 🤣.

One time I found this mATX PC-Chips with a Socket A and universal AGP for really cheap, it was my first time using a motherboard with SiS chipset. I don't know the exact model, but it is very common, but if you happen to find one, just run as far away from it as possible!
The design is so stupid, it has these two "big" electrolytic capacitors on the very left of the AGP slot, making it impossible to plug in some AGP cards.
Although the speed wasn't great, the major problem was compatibility with some AGP cards, especially with 3dfx and Matrox cards, where after installing the Video drivers, after a reboot you'd get just a white screen. This problem was documented on the internet, and apparently there was a BIOS Update that would fix this issue, but even after updating the Bios the problem would persist, at least with the Voodoo cards.

Pretty much every nForce chipset board I encountered, one ECS (nForce 630A) and two Asus (first one was a 570SLI-based board, the other was a nForce 430-based) they'd do weird funky stuff, as in, they'd fire up just fine, but would just not even start POSTing.

On the NForce 430 board I noticied that if I applied pressure on the board around the chipset heatsink or the RAM slots, that was enough to coax it into firing up.

I also had a NEC PowerMate VL370 tower that gave me the same grief (I really liked this machine despite its odd quirks), one day the board just died like that, it wouldn't even fire up.

Edit: the nForce 430 board is a Asus M2N-MX, despite its crummy, puffed up capacitors (Yum! Err rather... Yuck!) it somehow still works... I can't even fathom why. Might take a stab at recapping it at some point.

The 570 board is an Asus M2N-SLI Deluxe, compared to the MX it doesn't seem as temperamental (though it sometimes complains about a USB overload... Thing? Weird).

out of the box in general? the worst by far were the DFI motherboards, mainly because shitty BIOS and virtually no updates for them, it was an excercise of patience to get a fully working and stable computer with this motherboards.

from a retro stand point and to use today? The Soyo 5EHM: very picky with AGP cards, usb adapters and jumper settings, the usb's 1.1 onboard are awful. It was OK for basics system back in its day but many suffered from bad caps (mainly around the CPU). It took me many months to find the right combination of parts and now its working as my main retro machine.

slow piece of crap?: all the opti viper based motherboards and all the pcchips 530 and 630 based motherboards. Slow, unstable, very picky with CPU's

most unstable board?: ECS K7s5A by far, i had to work in 3 (thank god that they were uncommon here!), all of them had problems with memory, no matter what i tried, they were super unstable. unHonorable mention: MSI 745 Ultra

guarantee of bad caps: Soyo 7VBA 133, i never found one in good shape, they all died from the same capacitor plague

Living wrote on 2021-05-30, 17:10:

most unstable board?: ECS K7s5A by far, i had to work in 3 (thank god that they were uncommon here!), all of them had problems with memory, no matter what i tried, they were super unstable. unHonorable mention: MSI 745 Ultra

I know the ECS K7S5A got a lot of hate, and they definitely had some quality issues back in the day. Regarding this board, I seem to be the odd one out: My K7S5A never made any trouble (if you don't try to enable STOP/GRANT mode that isn't officially supported by that board anyways), neither with an Athlon Thunderbird 1000 and 256MiB SDR RAM, nor later with an XP 2200+ and 1GiB DDR RAM. So the board is that bad that you can't even rely on it making trouble.

mkarcher wrote on 2021-05-30, 17:25:

So the board is that bad that you can't even rely on it making trouble.

That's gotta be the quote of the day 😀

jasa1063 wrote on 2021-05-29, 15:25:

PC Chips M599LMR with SIS 530 chipset

Disregarding the missing AGP slot it's a really decent board...you just have to mod the latest BIOS with AMIBCP, there are some settings to unlock or/and improve.

kind regards
soggi