• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Michael Dell's "Lunatic Fringe" quote

Flapjack

2[H]4U
2FA
Joined
Apr 29, 2000
Messages
3,210
I remember about 20 years ago, good ol' Michael Dell said something like "People who swap their video cards in and out are on the lunatic fringe of computing".

It used to be pretty easy to find references to that quote, but now I can't find squat. Does anyone remember this (and hopefully have a source that still works)?
 
I can't say I remember that specific quote. However, I'm not totally surprised by it either.
 
1746823788102.png

https://www.zdnet.com/article/michael-dell-amd-environment-quottoo-fragile-quot/
1746823884754.png


first result on bing. it was 25 years ago and was about early amd chips....
 
To be fair to Dell, 2000 was a very different time in the PC space and Michael Dell was answering a question about offering AMD processors in his company's PCs. If you were using an AMD processor back then, you were typically building it yourself or finding a white label offering them.
 
That statement is familiar. To be fair, AMD's platform in those days was at best hit or miss. While the Athlon CPUs finally did away with the whole issue of AMD CPU's sometimes not running some software, problems with the platform were pretty significant. At the time non-Intel chipsets were all on a sliding scale of mediocre to dog shit. There was also a rash of VRM's in 1999 on many OEMs that were burning up and even catching fire. We replaced a lot of Slot A motherboards in OEM machines in those days. AMD also lagged Intel when it came to motherboard features most of the time. AMD's 7 series chipsets a year or two later even utilized a VIA south bridge to bring its features up to rival Intel's on paper but USB and SATA performance on those chipsets were still abysmal compared to Intel's chipsets. None of that even compares to the driver issues people often faced with AMD processor based systems prior to the availability of NVIDIA's nForce chipsets.

In truth though, the fact Dell never offered AMD CPU's in those days came down to business decisions more than reliability or perceived quality problems. It was widely reported that Intel was employing underhanded tactics to prevent companies like Dell (and others) from offering AMD processor based systems. The other prevailing theory at the time was that AMD simply couldn't supply enough chips for OEMs like Dell and HP. This was bullshit to some degree as that statement only held water if you were talking about replacing Intel in their lineups. HP offered AMD based systems but only on a limited number of models. Dell could have done the same at that time had it wished to do so.
 
That statement is familiar. To be fair, AMD's platform in those days was at best hit or miss. While the Athlon CPUs finally did away with the whole issue of AMD CPU's sometimes not running some software, problems with the platform were pretty significant. At the time non-Intel chipsets were all on a sliding scale of mediocre to dog shit. There was also a rash of VRM's in 1999 on many OEMs that were burning up and even catching fire. We replaced a lot of Slot A motherboards in OEM machines in those days. AMD also lagged Intel when it came to motherboard features most of the time. AMD's 7 series chipsets a year or two later even utilized a VIA south bridge to bring its features up to rival Intel's on paper but USB and SATA performance on those chipsets were still abysmal compared to Intel's chipsets. None of that even compares to the driver issues people often faced with AMD processor based systems prior to the availability of NVIDIA's nForce chipsets.

In truth though, the fact Dell never offered AMD CPU's in those days came down to business decisions more than reliability or perceived quality problems. It was widely reported that Intel was employing underhanded tactics to prevent companies like Dell (and others) from offering AMD processor based systems. The other prevailing theory at the time was that AMD simply couldn't supply enough chips for OEMs like Dell and HP. This was bullshit to some degree as that statement only held water if you were talking about replacing Intel in their lineups. HP offered AMD based systems but only on a limited number of models. Dell could have done the same at that time had it wished to do so.
yup, been there seen it. ive ran primarily amd since buying a compaq w/ a k6-2.
 
Guess it was because I was looking with DuckDuckGo (and then Google). Thanks!

I guess taken in context, it's not quite as abrasive as I remembered... but still pretty bad. I would definitely have considered myself "lunatic fringe" back then, but had plenty of friends who were happy to upgrade/install new hardware in their PCs that were most definitely not Hardcore or lunatics.

I had nothing but AMD CPUs and never had significant issues with them, all the way back to 66mhz or so. I've had less problems building AMD systems than Intel systems, especially when it comes to Intel and all the driver issues on the chipset side. Never had any AMD chipset issues that I can remember... if you even had to install any at all (Windows picked up most just fine). Hell, the early 2000s were all about Intel's first copper interconnect CPUs crashing on heavy tasks (discovered here, actually) trying to catch AMD after they got to 1000mhz. Intel issued recalls for those CPUs, IIRC. Then there was the dirty Rambus/RDRAM scandal and systems crashing when you populated the 3rd RIMM slot. Definitely agree they were some filthy times for Intel. Kinda nice to see them so behind AMD now (and having yet more problems with their current CPUs).

Yeah, I don't know what Michael Dell was talking about.
 
Last edited:
Guess it was because I was looking with DuckDuckGo (and then Google). Thanks!

I guess taken in context, it's not quite as abrasive as I remembered... but still pretty bad. I would definitely have considered myself "lunatic fringe" back then, but had plenty of friends who were happy to upgrade/install new hardware in their PCs that were most definitely not Hardcore or lunatics.

I had nothing but AMD CPUs and never had significant issues with them, all the way back to 66mhz or so. I've had less problems building AMD systems than Intel systems, especially when it comes to Intel and all the driver issues on the chipset side. Never had any AMD chipset issues that I can remember... if you even had to install any at all (Windows picked up most just fine). Hell, the early 2000s were all about Intel's first copper interconnect CPUs crashing on heavy tasks (discovered here, actually) trying to catch AMD after they got to 1000mhz. Intel issued recalls for those CPUs, IIRC. Then there was the dirty Rambus/RDRAM scandal and systems crashing when you populated the 3rd RIMM slot. Definitely agree they were some filthy times for Intel. Kinda nice to see them so behind AMD now (and having yet more problems with their current CPUs).

Yeah, I don't know what Michael Dell was talking about.
The problem with AMD CPU's from the 386 onward to the K6 is that there was software out there that flat didn't run on non-Intel CPU's. This wasn't specifically an AMD thing as Cyrix had the same problems. There were also plenty of cases where non-Intel CPU's could run the software but did so at a significant performance penalty. Clock for clock all the synthetic benchmarks of the day (Norton Utilities, SpecPerf, etc.) would show Intel CPU's were considerably faster. I tested an AMD 486 DX2 80MHz and it only matched an Intel i486 DX2 66MHz CPU. The Cyrix 80MHz CPU I had at the time was even worse. Back in those days, everyone used the same motherboard socket so there weren't significant platform issues though some motherboards were designed with jumper settings for write through and write back cache modes.

When the K5 came out, it was too late to the party and much slower than NexGen Systems NX586 and Intel's Pentium lineup. Not to say Intel's lineup was totally free of problems. The main thing being the recall on the Pentium 60/66 and the FDIV bug. However, when their shit worked it worked better than anything the competition had. The Pentium and Pentium Pro were way ahead of anything Cyrix and AMD did until Cyrix's 6x86. AMD never really had a good answer to that. After buying NexGen Systems the NX686 became the AMD K6, which was pretty well received but still didn't run some software or ran it badly.

The K62 was better received, but this is where their platforms diverged. While Super 7 boards could run Intel Pentium MMX CPU's and the like, Intel's Pentium II ran on the slot 1 motherboards. For a time, AMD was stuck on the aging super 7 and the garbage 3rd party chipsets of the day with their AGP compatibility problems, questionable VRD implementations, shit BIOSes and awful fucking drivers. Frankly, while AMD started to shake off the knock of CPU chipmaker rep during the Athlon era, it was still plagued by motherboards with bad VRMs and awful chipsets. Sadly, it's NVIDIA that did the most to help here with its nForce 2 line. AMD's own chipsets were decent, but weren't without their share of problems.

AMD had its fanboys back then but don't kid yourself. AMD was a shit show of a CPU maker until the Athlon era and it wasn't all sunshine and rainbows on the platform side. Even in the Socket 939 days, AMD still had its share of platform problems.
 
I get what you're saying, but we're talking about the 2000s, not 1996 (AMD K5). If you're saying you had those problems with the CPUs that came out in the 2000s (K6, K6-2, K6-3, Athlon), you had some issues no one else had.

The Athlon was the first great time to be an AMD fan. I also had a Cyrix 686 (a 66mhz version that ran at 150mhz all day long). No compatibility problems there, either.

If a program didn't run on a K6-3, which was available in 1999, that is a software problem. Both used x86. If something wasn't optimized for SSE or something, that's understandable.
 
Back
Top