Worst CPUs ever, now with poll!

Page 18 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

What's the worst CPU ever? Please explain your choice.

  • Intel iAPX 432

  • Intel Itanium (Merced)

  • Intel 80286

  • IBM PowerPC 970

  • IBM/Motorola PowerPC 60x

  • AMD K5

  • AMD family 15h

  • AMD family 10h


Results are only viewable after voting.

bbhaag

Diamond Member
Jul 2, 2011
4,036
36
126
^^^Agreed. In all likelihood it came down to price just like it always does. IBM wasn't expecting the PC to be the booming success that it was so it makes sense that they would choose the lowly Intel 8086.
 

superstition

Platinum Member
Feb 2, 2008
2,219
0
101
^^^Agreed. In all likelihood it came down to price just like it always does. IBM wasn't expecting the PC to be the booming success that it was so it makes sense that they would choose the lowly Intel 8086.
Choosing an outdated gimpy CPU and low-grade OS also made it less likely to compete with their lucrative mainframe business.
 

bbhaag

Diamond Member
Jul 2, 2011
4,036
36
126
You are absolutely correct. Corporate customers were and to a point still are their bread and butter. I do find it ironic that the "low-grade os" as you put it came back to bite them in the ass later down the road. I often wonder what it would be like if OS/2 would have taken off and been a success in the consumer market.
 
Apr 27, 2000
10,758
518
126
I wonder what it would have been like had IBM simply chosen CP/M instead.
 

bbhaag

Diamond Member
Jul 2, 2011
4,036
36
126
That would have been interesting. I recently purchased a used copy of Peter Nortons DOS 5 Guide and he goes into a few details about the early history of the IBM PC and DOS. According to him the DOS variant of the PC was quickly preferred by users and vendors over the CP/M and p-System versions and since this was a PC targeted for home and small business' IBM listened.
 
Oct 9, 1999
11,232
38
126
Cyrix PR-300.

They didnt even use the proper PCI bus divider so the clocks on the PCI bus were way out of spec and alot of
PCI cards would not even work on a cyrix system.
 

JoeRambo

Senior member
Jun 13, 2013
638
42
136
Cyrix PR-300.

They didnt even use the proper PCI bus divider so the clocks on the PCI bus were way out of spec and alot of
PCI cards would not even work on a cyrix system.
Not to mention that PR300 was just imaginary as those were hilariously slow CPUs. Tons of compatibility problems too, some programs just crashed on them. My classmate had one, and regretted it each day. Nothing "6x86", nor "300" about it, solid piece of you know what.
 

sirmo

Golden Member
Oct 10, 2011
1,008
0
136
I had a Cyrix 233 MMX.. that thing was great.. ran Linux like a champ too.
 

superstition

Platinum Member
Feb 2, 2008
2,219
0
101
That would have been interesting. I recently purchased a used copy of Peter Nortons DOS 5 Guide and he goes into a few details about the early history of the IBM PC and DOS. According to him the DOS variant of the PC was quickly preferred by users and vendors over the CP/M and p-System versions and since this was a PC targeted for home and small business' IBM listened.
There were better options when IBM picked DOS than CP/M.

I do find it ironic that the "low-grade os" as you put it came back to bite them in the ass later down the road. I often wonder what it would be like if OS/2 would have taken off and been a success in the consumer market.
Windows isn't what I was referring to, as it came out later and was a success because of the IBM deal. However, early versions of Windows were certainly primitive.
 

bbhaag

Diamond Member
Jul 2, 2011
4,036
36
126
There were better options when IBM picked DOS than CP/M.
Those were Peter Nortons observations not mine. I honestly don't know enough about the subject to form an honest opinion on it. I just thought it was interesting when I read it and he felt strongly enough about it to mention it in one of his books. An older book I know and yes opinions can change over time as more information becomes available.

Windows isn't what I was referring to, as it came out later and was a success because of the IBM deal. However, early versions of Windows were certainly primitive.
I wasn't necessarily referring to Windows. More to the fact that IBM put their faith in Microsoft and once DOS 5.0 became available the partnership broke down. It was poor judgement on my part to mention OS/2 because it was released many years later.
 

dark zero

Platinum Member
Jun 2, 2015
2,497
4
61
Reviving this thread..

I want to add some candidates:
- The whole Carrizo uArch.... It was a massive regression of Kaveri and it got worse on laptops.
- On ARM I can confirm that Snapdragon 810 and Mediatek Helio P10 were the worst of the bunch. Those things overheats a lot.

Also I can add the whole Mediatek Helio P series... What a massive letdown.
 

Topweasel

Diamond Member
Oct 19, 2000
4,622
288
136
Worse is an understatement. A Willamette P4-1400 was about as fast as a P3-900. The first P4's were dreadful. Once Intel beefed up the cache, better compliers were written, and the clock speeds went past 2 GHz, it was a competitive part, but still was terribly inefficient. I voted for the Itanic in the poll, but I would have voted for Netburst as the worst CPU architecture. Even as bad as AMD K15h was when it first came out, it looks like masterful execution compared to early Netburst. Intel even sold Willamette systems with SDRAM memory for a while, because of the Rambus fiasco, which took away the only redeeming quality of the first P4's, their enormous memory bandwidth.
Oh god yes to all of that. Bulldozer has it's issues but it really comes down to not being very fast it's a bad design and it predates the shift they expected by probably about 10 years. But the release of the P4 was an epic failure. Costly systems with costly Ram, limited to two slots and the platform as a whole was a mess Intel figured out late that it wouldn't take 3 sticks so OEM's were selling systems with a sticker over the ram slots. The fact that it's overall performance was poor and Intel had to dump it pretty quick (quicker than they could even come up with an alternative) is a lot like Bulldozer. But Bulldozer didn't have nearly the rocky road at release and on.

But the winner has to be Merced/Itantium as a whole. Intel wasted 10's of Billions on that platform. It was going to replace X86. It was going to break all their cross licensing deals. It convinced several other chip companies that they weren't needed any more. It absolutely devastated the high end server market and had absolutely nothing to show for it. Hell Intel is probably likely to sell more 7980xe's than all AI64 sales put together.
 
Apr 4, 2010
41
0
71
How about the pentium 4 willamette. The other Pentium 4s weren't so bad, but the willamette was beyond a disaster, Northwood was decent, Prescott and Cedar Mill were ok.
 

whm1974

Diamond Member
Jul 24, 2016
6,259
239
96
Oh god yes to all of that. Bulldozer has it's issues but it really comes down to not being very fast it's a bad design and it predates the shift they expected by probably about 10 years. But the release of the P4 was an epic failure. Costly systems with costly Ram, limited to two slots and the platform as a whole was a mess Intel figured out late that it wouldn't take 3 sticks so OEM's were selling systems with a sticker over the ram slots. The fact that it's overall performance was poor and Intel had to dump it pretty quick (quicker than they could even come up with an alternative) is a lot like Bulldozer. But Bulldozer didn't have nearly the rocky road at release and on.

But the winner has to be Merced/Itantium as a whole. Intel wasted 10's of Billions on that platform. It was going to replace X86. It was going to break all their cross licensing deals. It convinced several other chip companies that they weren't needed any more. It absolutely devastated the high end server market and had absolutely nothing to show for it. Hell Intel is probably likely to sell more 7980xe's than all AI64 sales put together.
It could be argued that Intel made way more money back then they spent on developing Itantium by all of the x86-64 Xeons they sold to companies that were using SPARCs, Alphas, MIPS, etc previously. Intel just didn't count on AMD extending the x86 ISA to 64-bits, and Microsoft supporting it.
 

scannall

Golden Member
Jan 1, 2012
1,311
136
136
I wonder what it would have been like had IBM simply chosen CP/M instead.
If Gary Kildall had shown up, they would have. They had to settle for a quick and dirty hack from a garage company call Microsoft.
 

PeterScott

Platinum Member
Jul 7, 2017
2,605
226
96
Nice way to cover up your own inability to Google.
Why should I have to google obscure code names in a poll?

It is clearly a biased attempt to tilt the poll results.

Also why no Pentium 4? What a strange poll, about "worse processors", that leaves out what many consider the worse processor like Pentium 4 and obscures Bulldozer.
 
Last edited:

Gikaseixas

Platinum Member
Jul 1, 2004
2,806
4
106
I had a P4 1.8Ghz Northwood and a Athlon 1900+ 1.6Ghz and the AMD chip run faster in every application and games i used. Willamette was actually much worse than Northwood so...

P4 Willamette wins this contest
 
Apr 27, 2000
10,758
518
126
Why should I have to google obscure code names in a poll?

It is clearly a biased attempt to tilt the poll results.
Um. I don't think jhu was that kind of poster? Besides, if you just put "Bulldozer" then you open up the possibility that people aren't voting for Piledriver, Steamroller, or XV. 15h is the whole family.
 
Apr 27, 2000
10,758
518
126
If Gary Kildall had shown up, they would have. They had to settle for a quick and dirty hack from a garage company call Microsoft.
Pfft, all they had to do was just buy out Digital Research, which would be the modern Google way of doing it. Instead they wanted to be bitchy because Gary's wife wouldn't sign an NDA or some such horse manure. Anyway that whole negotiation seems unprofessional in retrospect.
 


ASK THE COMMUNITY

TRENDING THREADS