Worst CPUs ever, now with poll!

mikeymikec · Sep 26, 2024

igor_kavinski said:
On a related note, my idiot IT admin (who thinks formatting is the best solution) is trying to forcefully retire my trusty old i5-2400 work PC running Windows 7 that's been issue free for at least 14 years now. I will probably lose this battle and then I'll be out of a job soon because I don't see any point in accepting things for which there is no proper justification given. I would rather not work in a stupid organization like that.

Uh, retiring a Win7 install due to unpatched vulnerabilities is a pretty reasonable justification IMO. That's just aside from the fact that you don't have an up-to-date browser on there, many bits of new software will refuse to install on Win7, and for IT having a single PC that's different from all the others is just another complication to consider.... I'm saying all this even though I *like* Win7!

If I was the IT boss at your company I would have to hear a proper justification for keeping such an OS in play.

In my line of work (freelance computer fixer), I regularly let slide out-of-date versions of Windows, Microsoft Office, etc. I put the argument for the 'sensible ideal' security scenario but it's the customer's job to make the informed decision because they pay the bills, just like if I was the IT boss at your company, while I would expect my judgement to be more or less the final word on the topic, the higher-ups may have budget/other concerns and so therefore overrule me, but my word is sure as hell going to be on the record for if the excrement hits the fan; I wouldn't want someone with a soft spot for an unsupported bit of kit to be the reason I got fired for malpractice.

DZero · Sep 26, 2024

Guys... now I am thinking... the Transmeta processors would enter on this pool?

gdansk · Sep 26, 2024

No, they had a niche

Topweasel · Sep 28, 2024

igor_kavinski said:
So whose maintaining the updated list of worst CPUs?

It should be:

AMD Athlon The Smoky One

Socket A Thunderbirds where freakin awesome. I am never a big fan of these stacked comparisons. 6 months to a year before this Intel CPU's would do the same. Intel came up with their speed step for slowing down as tempatures got higher, first, and all of a sudden it was look how bad AMD is because their CPU's will burn up without a cooler. Instead of being a cool feature Intel CPU's it became proof that AMD chips ran hot and were bad.

igor_kavinski · Oct 13, 2024

The Q6600 makes everything before it look like the worst thing ever. I mean, yikes!

OneEng2 · Oct 17, 2024

Itanic .... definitely #1. VLIW and all kinds of "are you kidding me" design decisions.
Netburst. "Hey, clock speed sells, who cares about IPC"?
Bulldozer "Don't worry about the decode bandwidth. We will keep those execution units busy". Seriously, an AMD rep told me this.

I think it interesting some of the processors that are on the list. I consider some of those the BEST designs for their time.

DZero · Apr 23, 2025

OneEng2 said:
Itanic .... definitely #1. VLIW and all kinds of "are you kidding me" design decisions.
Netburst. "Hey, clock speed sells, who cares about IPC"?
Bulldozer "Don't worry about the decode bandwidth. We will keep those execution units busy". Seriously, an AMD rep told me this.

I think it interesting some of the processors that are on the list. I consider some of those the BEST designs for their time.

Sad story seems that Raptor Lake and AMD Thunderbird might compete in who can destroy the computer of the affected faster.

Panino Manino · Apr 23, 2025

OneEng2 said:
I think it interesting some of the processors that are on the list. I consider some of those the BEST designs for their time.

Someone explained those two PowerPC on the list?

Hulk · Apr 23, 2025

KompuKare said:
Ok, Intel is a monster now but I think if it wasn't for the huge revenue they got from IBM's pretty poor decision to chose 8088 (to think the 68000 was around back then and had 32-bit data and address registers) and they would likely be much much smaller today.

1. 8086. For all those who have struggling with 64KB segmented memory, DOS memory managers, 640KB limits (and millions did) there can be no other #1
2. 80286. For not fixing the above properly and introducing more confusion address modes.
3. Pentium 4. For being slower than the Pentium 3 and using more power plus who could forget RAMBUS.
4. Bulldozer. For almost the same reasons as #3...
5. Itanium. VLWI sounds good but $billions later: maybe it wasn't such a good idea.

8086? What was the alternative that was so much better?

A CPU must be judged by its contemporaries, not against what came later.

Have you ever struggled with other CPU's of that era? I have. The 8086 was a marvel.

LightningZ71 · Apr 23, 2025

The 8086/8 was certainly quite capable, but costlier than the 6502, not as performant as the Z80 on optimized code, and nowhere near as flexible or well thought out as the 68000. I've used systems with all four of them and coded on 3 of them. I rather enjoyed the 68000 Amiga I used the most.

DaaQ · Thursday at 12:46 AM

igor_kavinski said:
View attachment 109376

The Q6600 makes everything before it look like the worst thing ever. I mean, yikes!

OneEng2 said:
Itanic .... definitely #1. VLIW and all kinds of "are you kidding me" design decisions.
Netburst. "Hey, clock speed sells, who cares about IPC"?
Bulldozer "Don't worry about the decode bandwidth. We will keep those execution units busy". Seriously, an AMD rep told me this.

I think it interesting some of the processors that are on the list. I consider some of those the BEST designs for their time.

All I know is the Prescott processors blew chunks. IIRC was direct successor to the Northbridge CPUs.
I had the Kentsfield QX6600 and I doubt it would hold anything close to the FX8350. Not sure if Sandy or Ivy would still hold as well. Since I just retired my 8350 around Feb or March. It would run D4 and BG3 at playable rates. With a GTX780. Elden Ring would not load. The 780 was the 3GB version. Of course tunes settings. but avg in D4 was above 60 mostly up to 90fps since the driver would only give me option of IIRC 100 or 90 hz. ON a 144hz monitor. In Diablo I never hit system RAM.
I still have the Kentsfeild cpu, and an actual mobo to put it in even though it is a Biostar lga775, Striker Extreme failed. I suspect the Nvidia4 chipset. I degress.

My 2.6 Northwood would OC to 3.4 or very close to the EE at the time. I had SLI 8800 GTX in the QX6600 at the time and had loads of trouble. LOTR online crashed all the time. this was 2007.

After the QX6600 I got a Lynnfeild 750 and OC that to the top tier, ran it for years till D3 came out, and I threw together my FX8120 on an ASRock fx990 killer mobo with stock HSF. It ran for 2 or 3 years till mobo died from flat out being dirty and smoking in the room. Next machine was a laptop with 3200g for browsing.

Then when D4 launched I finally put together my was supposed to be for D3 rig. Which has now been updated to 9800x3d and 7900xtx. BUT, the FX 8350 was able to keep the 7900 XTX fed for D4 at around 100-110 frps avg. Give or take some dips. I did have alot of crashes though, not sure if game or pc stability.

So my vote for worst is the Pentium D that immediately followed Northwood I believe was NWc . I know there is 5 years between the FX and SB/IB but the FX is still usable if you can work around TPM 2.0

Thunder 57 · Thursday at 2:21 AM

DZero said:
Sad story seems that Raptor Lake and AMD Thunderbird might compete in who can destroy the computer of the affected faster.

The only way you would destroy your computer with Thunderbird would be to knock the HSF while it was running. With RPL it will kill itself over time running stock.

Thibsie · Thursday at 5:06 AM

Thunder 57 said:
The only way you would destroy your computer with Thunderbird would be to knock the HSF while it was running. With RPL it will kill itself over time running stock.

Absolutely. Can't be compared IMO.

Hulk · Thursday at 7:42 AM

LightningZ71 said:
The 8086/8 was certainly quite capable, but costlier than the 6502, not as performant as the Z80 on optimized code, and nowhere near as flexible or well thought out as the 68000. I've used systems with all four of them and coded on 3 of them. I rather enjoyed the 68000 Amiga I used the most.

Yet only the 8086 DNA survived.

Nothingness · Thursday at 7:47 AM

Hulk said:
Yet only the 8086 DNA survived.

Alas. BTW 8086 DNA can be traced back to 8085 (hell, Intel even had a macro assembler to translate 8085 to 8086) and I guess we can trace back the abominations of the 86 ISA to even older CPUs. Despite its issues, 68k was much better, but too late to the game; it's the only reason that this disgusting ISA survived: because Intel was first on the line for IBM.

Hulk · Thursday at 8:07 AM

Nothingness said:
Alas. BTW 8086 DNA can be traced back to 8085 (hell, Intel even had a macro assembler to translate 8085 to 8086) and I guess we can trace back the abominations of the 86 ISA to even older CPUs. Despite its issues, 68k was much better, but too late to the game; it's the only reason that this disgusting ISA survived: because Intel was first on the line for IBM.

So IBM did not even consider other options? Someone at the top said, "Just go with the first one at the door!"

Interesting. I did not know that. I thought there was some decision making that had gone on.

Hulk · Thursday at 8:14 AM

LightningZ71 said:
The 8086/8 was certainly quite capable, but costlier than the 6502, not as performant as the Z80 on optimized code, and nowhere near as flexible or well thought out as the 68000. I've used systems with all four of them and coded on 3 of them. I rather enjoyed the 68000 Amiga I used the most.

I had a 6502 (Atari 800). While it was capable for the time, in particular I did not enjoy 160x192 graphics with 2 bit color. While color registers could be changed during the vertical blank interrupt you were then limited to the "new colors" in horizontal screen sections. Or you could engage the 1 bit 320x192 graphics mode and fake two color using odd/even scan lines. Any real programming had to be done using the Assembly cartridge. Also if you started programming before connecting the floppy disc you would lose everything you did. I remember seeing IBM computers with 640x480 resolution with 16 colors and being absolutely amazed. Do you know how much video memory this thing must have I told my friends!

igor_kavinski · Thursday at 8:15 AM

Thunder 57 said:
With RPL it will kill itself over time running stock.

And at the end you say, "Thank you, Great Warrior, for your most excellent service in giving me 6+ GHz of undisputed performance at unbeatable price!" and then you have to move to AMD...

igor_kavinski · Thursday at 8:21 AM

Hulk said:
I remember seeing IBM computers with 640x480 resolution with 16 colors and being absolutely amazed. Do you know how much video memory this thing must have I told my friends!

It's the same reason why x86 continues to live. Despite its IPC superiority, Apple M4 cannot run a Geforce 4090 or 5090 or 9070 XT. And suppose Apple came out with a micro-ATX board for it tomorrow (very unlikely), how interested would AMD/Nvidia be in rewriting their GPU drivers for it? Apple would have to PAY both for the driver development and then hope the cost pays off in the end. They could totally go down this path but they are rolling in so much cash that they go "meh" at even the thought of it.

DZero · Thursday at 9:33 AM

igor_kavinski said:
And at the end you say, "Thank you, Great Warrior, for your most excellent service in giving me 6+ GHz of undisputed performance at unbeatable price!" and then you have to move to AMD...

Or even more... reuse the older Alder Lake you discarded to be "slow"

LightningZ71 · Thursday at 9:43 AM

Hulk said:
I had a 6502 (Atari 800). While it was capable for the time, in particular I did not enjoy 160x192 graphics with 2 bit color. While color registers could be changed during the vertical blank interrupt you were then limited to the "new colors" in horizontal screen sections. Or you could engage the 1 bit 320x192 graphics mode and fake two color using odd/even scan lines. Any real programming had to be done using the Assembly cartridge. Also if you started programming before connecting the floppy disc you would lose everything you did. I remember seeing IBM computers with 640x480 resolution with 16 colors and being absolutely amazed. Do you know how much video memory this thing must have I told my friends!

I had a Laser 128 EX with the 65C02. It was an interesting thing to program on, being ALMOST an apple IIe/c. In that implementation, it had a few more video modes accessible and more than 2 bit color. It was a solid machine though with lots of expansion capabilities (though, much of it was difficult to source). I had a friend with a C64 and we would compare notes a lot.

Nothingness · Thursday at 10:26 AM

Hulk said:
So IBM did not even consider other options? Someone at the top said, "Just go with the first one at the door!"

Interesting. I did not know that. I thought there was some decision making that had gone on.

Some history of the birth of IBM PC is told here: https://spectrum.ieee.org/the-insid...ts-biggest-blunder-the-tms9900-microprocessor

Of course that's not a proof, I'm not sure any IBM insider ever told anything about that.

igor_kavinski · Thursday at 10:50 AM

Nothingness said:
Of course that's not a proof, I'm not sure any IBM insider ever told anything about that.

https://www.pcmag.com/news/why-the-ibm-pc-used-an-intel-8088

In Hard Drive: Bill Gates and the Making of the Microsoft Empire, by James Wallace and Jim Erickson, Syndes said IBM considered several different chips, including the 68000, but the 68000 was still six to nine months later than IBM needed it to be.

I guess the only reason they didn't wait was because they never thought the IBM PC would be the success it became. If they had known, they would've "done it right". Probably even the guy in charge of the project thought of it as a long shot so the choice of CPU was kind of an afterthought. His thinking probably went, "Who cares about this thing having the best CPU? It may get canned or be a huge market failure. Why waste time waiting for a better CPU?". As someone who is forced to take on projects (unfortunately this seems to be the only way to have any chance of seeing more money in my life), I can attest that you want the project to be over ASAP. It's just too much stress.

Panino Manino · Thursday at 12:32 PM

Never see anyone mention that the 8086 was the best option. It just sort of happened and it worked, the success of the IBM PC with the 8086 was also an "unfortunate" success for Intel. It was good enough, but not their best.
Also, I have this impression that people brainwashed themselves into misremembering how good Intel was at making CPU. I don't think they where the best, that they designed the best CPU and ISA, but you know the history, isn't always the best that comes out on top.

Nothingness · Thursday at 12:35 PM

Panino Manino said:
Never see anyone mention that the 8086 was the best option. It just sort of happened and it worked, the success of the IBM PC with the 8086 was also an "unfortunate" success for Intel. It was good enough, but not their best.
Also, I have this impression that people brainwashed themselves into misremembering how good Intel was at making CPU. I don't think they where the best, that they designed the best CPU and ISA, but you know the history, isn't always the best that comes out on top.

I think IBM made the right decision given their schedule, too bad 68k was late.
Nitpicking: IBM chose the 8088, not the 8086 😉

Worst CPUs ever, now with poll!

Lifer

Senior member

Diamond Member

Diamond Member

Lifer

Senior member

Senior member

Golden Member

Diamond Member

Platinum Member

Golden Member

Diamond Member

Golden Member

Diamond Member

Diamond Member

Diamond Member

Diamond Member

Lifer

Lifer

Senior member

Platinum Member

Diamond Member

Lifer

Golden Member

Diamond Member