I would assume majority of ISA cards were more expensive and had better manufacturing than majority of PCI cards.
Also the on-board parts are larger and simpler, which tend to be sturdier compared to small and complex parts.

Barring capacitor failure and rust, I'd say ISA cards are tougher than PCI cards.

I wonder what people do with their hardware. Since 1997, not a single of my ISA/PCI/whatever cards have failed. Well, one did, an S3 ViRGE. However, when I tested it years later, it was working fine. So I don't count it.

I only notice video cards becoming unreliable when you get into high powered 3D gaming cards. Those cards live a stressful life and are built to be as cheap, fast, and quiet as they can get away with. That set of priorities tends to conflict with longevity.
BGA packaging and lead free solder are also more touchy than older packaging/solder types.
I'll crudely generalize that the more power a chip uses, the less reliable it and it's associated circuitry end up being in the real world.
Low end cards of a given generation are usually quite reliable. The high end models of 3D gaming cards are built to last 1-2 years and then be thrown out. The gamers don't care after that, but they do care about initial cost, size, and noise. The manufacturers want to meet those concerns, and they have no interest in helping old high end models float around in resale.

I haven't encountered problems with cards from the 2D era, but I don't have as much experience with them. I've never had a card from the 2D era that was anything fancy, just basic consumer cards.

ISA predates the change over to lead free solder and BGA in the mid 2000's that shamino mentioned above I suppose this could still affect PCI cards from the same period.
I would also say its down to numbers, I'd bet for 1 isa variant of a card 5 or more PCI variant were produced, which naturally means more to fail in the long run.

Lead-free solder is a most likely cause of failure. My second guess would be cheaper caps that dry and die without any noticable symptom.

For me ISA and PCI video cards have been pretty reliable. My issues come from AGP era gaming cards most likely caused by heat and manufacturing issues. Newer PCIE GPUs are even worse.

The ISA era lasted a long time. Early ISA video cards were a mile long and had a million chips on them. Late generation Trident and Cirrus Logic cards were just the main chip, BIOS and RAM (pretty much the same in ISA and PCI).

I was going through a junk box of cards a few weeks ago (actually somebody elses junk box I snagged a long time ago) and found a dirty rusty IBM VGA card (8 bit ISA). This card even had the paper over the BIOS chip window missing. Cleaned it a little and stuck it in an old 386 and it worked just fine.

shamino wrote:

The high end models of 3D gaming cards are built to last 1-2 years and then be thrown out.

I don’t know what you’re on about, none of my mid to high end hardware mysteriously died after only one to two years of regular use. Strangely the same hardware usually died pretty quickly in the hands of others, such as my little brother. But never while I was using it. I’ve even had to downgrade sometimes to older cards, if only to test other configurations. Still working, nothing just “died” in my use cases.

Heat is the big killer. Keeping a dust-free, well-ventilated case will increase longevity of all components.

Durability could be greatly improved with thermal throttling. It's hard to kill an Intel CPU through overheating (or excessive thermal cycling) because Intel takes this seriously. By the time Intel chips were dissipating enough heat for this to be a big concern, they were on top of it. But it's easy to kill a graphics card this way.
Supposedly there is a thermal limit in nVidia's drivers (don't know about ATI). But the threshold is set so high that it's of no value.
Do modern cards detect fan failures yet? Older ones certainly don't.
I had a Ti4200 die because a heatsink push pin popped loose. That's an older card, but build quality on 3D gaming cards was like that for a long time. Newer cards at least have screws, and not just two of them, so I guess that's progress.
My 9800 Pro started artifacting at boot. It probably ran too hot. I don't think it even had a temperature sensor. Next time I want to use it I'll install an aftermarket heatsink and see what happens, but it's clearly damaged and will probably be flaky from now on.
A relative's 9800GT died after about 2 years of heavy gaming. Typical story for those cards.
At the other extreme, my Geforce2 MX cards were/are just about unbreakable. I don't think they're any more failure prone than older 2D era cards, because they don't use enough power to put much stress on themselves. They don't even need a heatsink.

I still use GT200 cards on modern systems, but I modified the fan profile to keep them at or below 75C. As shipped, they want to run at 90C+. Cleaning them is also a PITA and most people (who don't value old hardware like we do here) wouldn't bother with it. I also don't play games nearly as often as I did when I was younger. One year with a heavy gamer is probably equal to 10-15 years for me.

I bought a used GTX 480 from a guy who was a real hardcore gamer and I never had issues with it during the 2 years I used it. It always ran at around 90 degrees Celsius (190-195F) while gaming and never dropped below 70 degrees even when idling. The guy I sold it to never complained about it failing.

VLB MASTER RACE

I bought a used GTX 480 from a guy who was a real hardcore gamer and I never had issues with it during the 2 years I used it. It always ran at around 90 degrees Celsius (190-195F) while gaming and never dropped below 70 degrees even when idling.

Constant high temperature is not a bad thing actually. It's much more problematic when temperature fluctuates from low to high wildly like 35-90 degrees.

High temp will cause electron migration in the GPU and will short out sooner or later plus you are heating the board (RAM) and caps which shorten their life as well. Thermal cycling will affect the BGA solder more then the chip itself.

Unknown_K wrote:

High temp will cause electron migration in the GPU and will short out sooner or later plus you are heating the board (RAM) and caps which shorten their life as well. Thermal cycling will affect the BGA solder more then the chip itself.

Operating within normal temperature will not shorten the life of a cap, a shitty cap will bust even if it does not get any electricity, sooner or later; a cap rated to work at 1055C will work at 1055C as long as it would at 85C.

appiah4 wrote:

Unknown_K wrote:

High temp will cause electron migration in the GPU and will short out sooner or later plus you are heating the board (RAM) and caps which shorten their life as well. Thermal cycling will affect the BGA solder more then the chip itself.

Operating within normal temperature will not shorten the life of a cap, a shitty cap will bust even if it does not get any electricity, sooner or later; a cap rated to work at 1055C will work at 1055C as long as it would at 85C.

No, reduced operating temperature does increase component life. However, operating well below the max ratings will likely result in a different aging mechanism being the dominant path the failure.