Talk:List of Intel microprocessors

From Wikipedia, the free encyclopedia

Contents 1 Hail the great 8085! :-) 2 80286 3 P4 -- a new generation? (G7?) 4 List clutter fixed 5 Listbox proposal 6 StrongARM/XScale here? 7 Why the aggressive cleanup? 8 Corrected 4004 clock speed again. 9 i860 (80860) 10 Chrono vs "class" ordering 11 Intel386 EX release date 12 Does the Intel 8051/52 belong here? 13 List split? 14 Inappropriateness 15 Pentium M in 64 bit section? 16 Which of Intel's line of processors was the turning point? 17 Why are Apple Macintosh computers called out for the Intel Core processors? 18 Virtual memory for Pentium I 19 80186 20 "Built-in multitasking" ? 21 IXP? 22 4004 NOT used in Pioneer

[edit] Hail the great 8085! :-)

The 8085 was the greatest, of course that's not NPOV :)!

Heh -- well, at least it was a better 8080 than the 8080 :-) --Wernher 21:44, 21 Aug 2004 (UTC)

Yes, it can be said that the 8085 is the final evolution of 8-bit microprocessors. Too bad it didn't make it into the early home computers (pre-IBM PC). The 8085 was eclipsed by the Z80 in the early home computer market. Many 8085s were manufactured but ended up as mainframe I/O controllers and in various early PC peripherals.

[edit] 80286

The 80286 is being given short shrift here. It was not just a faster 8086 or 80186! This was the CPU that made the 8 bit bus obsolete (there was no 80288) and introduced protected mode.

The i286 could run Windows 3.1 (in Standard mode) unlike the earlier 8086/8088/80186/80188. It shouldn't just be lumped in with those DOS real mode CPU's with no notations.

[edit] P4 -- a new generation? (G7?)

Wasn't the Pentium4 finally a leap into generation 7? ANyone care to put headings in, eg "486 generation" / "generation 4" / whatever the correct term is? -- Tarquin

Calling the Pentium 4 7th generation is a bit amusing considering it is inferior to the 'previous generation'. It was a redesigned core for the purposes of scaling high MHz, not speed. A Pentium III at 2.4GHz with 800MHz FSB would probably run rings around a P4EE. All creations for marketing, like how the Pentium Pro was sold under ~5 different names. Crusadeonilliteracy

[edit] List clutter fixed

The old list was pretty cluttered and hard to read (at least to me). I've formatted most of it as nested lists, which helps readability, but now we've got a pretty long page. Any ideas how to break this up? By year of production maybe? Or generations? -- Wapcaplet

[edit] Listbox proposal

How about something like this for the processors' individual articles? (486 example shown) Crusadeonilliteracy 14:39, 11 Dec 2003 (UTC)

Preceded by: Intel 80386

Intel microprocessors

Succeeded by: Intel Pentium

As you might have seen, there is now a listbox with roughly all the Intel µPs on it; hopefully that's just as informative (maybe even more?) as the proposal you mention above. --Wernher 18:58, 28 Jun 2004 (UTC)

[edit] StrongARM/XScale here?

Should the StrongARM/XScale chips be listed here since Intel makes them now? Ckape 08:22, 11 Jan 2004 (UTC)

I would say no, since those processors are based on designs from other companies (ARM both?), only manufactured and possibly developed further by Intel. Perhaps they should be put in a list or listbox for ARM? --Wernher 18:55, 28 Jun 2004 (UTC)

StrongARM was originally designed by Digital Equipment Corporation, then acquired by Intel. XScale was designed totally by Intel. Dyl 20:08, Aug 21, 2004 (UTC)

[edit] Why the aggressive cleanup?

The recent cleanup removed lots of useful information, and no reason was noted. My favorite quirky processor, the iAPX 432 went down the drain... Yaron 22:24, Jun 27, 2004 (UTC)

But that's simply outrageous. Doesn't people have any sense of tech history at all? I say we reenter the relevant 'cleaned up' information. Perhaps one should put it into a more legible form first (?), but reentered it must be. --Wernher 04:00, 28 Jun 2004 (UTC)

I agree. What's the purpose of an encyclopedia, if not to hold historical and perhaps some obscure information? If articles were to hold info only about the well-known, then why not just use Intel's website? Dyl 20:34, Aug 21, 2004 (UTC)

I added info on the iAPX 432, 376, 860, and 960, then noticed this discussion. There was no good reason for the information to be removed, and I would even dispute the value of having it in a "non-mainstream" section. If you're going to have such a section, arguably even the 4004, 4040, and 8008 belong there, as they never were widely used other than in embedded systems. --Brouhaha 00:21, 1 Sep 2004 (UTC)

[edit] Corrected 4004 clock speed again.

See my comment on this topic in the discussion section of the 4004.

--Colin Douglas Howell 00:34, 9 Oct 2004 (UTC)

[edit] i860 (80860)

The i860 (80860) is currently listed as a 32-bit processor, but it was in fact Intel's first 64-bit microprocessor. I'd have fixed this myself but am not sure how best to do it. The processors are currently listed in order of introduction, and I didn't want to change that section heading to say "32-bit and 64-bit microprocessors". --Brouhaha 19:33, 17 Oct 2004 (UTC)

Thanks for that info. I think the 860 was more of a 32/64-bit processor, given its 32-bit integer ALU(s); I guess Intel would want to market it as a 64-bit one due to the FPU and buses. Anyway, the 64-bit feature should be noted by the 860's location in the listing, as you suggest.

To address the article's chrono vs "class" ordering problem, I'm currently working on a combined scheme: the µPs will be ordered by "class", i.e. wordlength as well as "mainstreaminess" (the latter an issue with the 432, i960 and i860), but within each "class" there'll be release date notes (between the processor items) for those other-class processors released in the same time window. --Wernher 23:14, 17 Oct 2004 (UTC)

See the following thread for the continuation of the chrono vs "class" ordering discussion. --Wernher 02:30, 20 Oct 2004 (UTC)

[edit] Chrono vs "class" ordering

So there; I've made the change to "class" ordering with chronological entries. Phew, some job, if I may say so. :-) Please go check whether I've introduced any typos/brainos (might just happen, you know). --Wernher 02:30, 20 Oct 2004 (UTC)

[edit] Intel386 EX release date

Title says it all... Anybody have definite info regarding that date? The earliest possible date I've got so far, after googling around for a bit, is August 1994 (from an Intel386 Embedded Processor Update) --Wernher 15:58, 20 Oct 2004 (UTC)

I have now also found refs to September 1994 in misc datasheets (i.e., date of initial datasheet publication -- rev 0), so I think the Aug '94 release date may be correct. I'll settle for that one until told otherwise. --Wernher 15:58, 20 Oct 2004 (UTC)

[edit] Does the Intel 8051/52 belong here?

I noticed the page about the Intel_8051 isn't connected to this collection of Intel microprocessors. Is that an intentional distinction between microprocessors and microcontrollers? Or an oversight?

An oversight, I think. Mirror Vax 02:28, 5 Jun 2005 (UTC)

Er, nope. That is, I, for one, in my work on this list, have consciously made a distinction between µPs and µCs---in line with the Intel µP list box. We might perhaps add a separate List of Intel microcontrollers, or maybe it would be better to put the µCs in a separate section in the present µP list. For the Motorola chips, we have a common list but two distinct templates, 1 and 2. --Wernher 01:07, 6 Jun 2005 (UTC)

I'd like to see the microcontrollers, too - 8031, 8748, and all the other little chips. For a while every Macintosh sold had an Intel chip in it...in the keyboard. Guess I'll have to take a rainy weekend and hit my old Intel manuals. --Wtshymanski 17:55, 12 July 2005 (UTC)

[edit] List split?

This page is now 31kb, and ~45 page-downs long on an 800x600 screen, if no-one objects, I think I shall split it into a few slightly overlapping lists(eg: list of intel x86 processors, list of non-x86 intel processors, chronological list of intel processors, etc.).

Does anyone have any ideas as to what good lists would be? --Boffy b

Completely off the top of my head (so please don't take it to conclusively), I'd say that one suitable split would be a 'techno-chronological' one, resulting in the following articles, for example:

• "List of Intel 8-bit and 16-bit processors"
• "List of Intel 32-bit non-Pentium processors"
• "List of Intel 32-bit Pentium processors" (split into the following two lists if necessary)
  • "List of Intel Pentium I, Pro, II, and III processors"
  • "List of Intel 32-bit Pentium 4 processors"
• "List of Intel 64-bit processors" (or "List of Intel 64-bit Pentium and Itanium processors")

The split of the 32-bit Pentium lists may be necessary due to size, and technically desirable anyway since the P4's microarchitecture is quite different from the PII/III's. I'm not sure if I like the (overly long?) article names, but it's just a thought... At least, these titles are reasonably intuitive (or?).

Of course, the x86/non-x86 dividing line that you propose is quite logical too, but some intra-"Pentium X" list splits like I have suggested above would also be needed if list size are to be significantly reduced. --Wernher 5 July 2005 02:05 (UTC)

[edit] Inappropriateness

I have deleted the title of section one of this article, as it was wildly inappropriate. I am not sure of what it is meant to be, so I have temporarily assigned it a title noting the change. The original title was: KEVIN SUCKS BALLS A LOT AND HE LIKES IT A LOT HE ALSO EATS CRAP

I think you'll agree with my change.

NOTE: More cleanups now...

===Pentium ("ClassicDCRAP")===(ALSO CRAP)

has been changed to ===Pentium===

Section 7 has been removed. The full text was as follows:

===well this is as follows===

Oh my god!! THIS IS CRAP

Not needed, guys.

===KEVIN SUCKS CRAP KEVIN GO SUCK CRAP===

Removed from See Also section

80486DX's speeds were listed as crap speeds, now fixed. Also, gigaCRAPS changed to gigabytes. "Used in Desktop computing and crap servers" changed to "Used in Desktop computing and servers". "Level 1 crap cache on chip" changed to "Level 1 cache on chip".

Pentium, "Bus width 64 craps" changed to "Bus width 64 bits". "Number of transistors 3.1 million craps" changed to "Number of transistors 3.1 million".

Someone has a crap fetish, so says Bryan Jones

Thanks for trying to fix things here. However, there's still a lot that 64.30.49.146 has removed today that's not been added back, so I'm just going to revert things to the last good version. Jgp 01:35, 4 October 2005 (UTC)

[edit] Pentium M in 64 bit section?

Why to put Pentium M (32 bit only) in Section 9: The 64-bit processors: IA64 and EM64T?

It's one of the "chronological entries"; I guess the idea is to show what other stuff was going on at the time. Guy Harris 08:10, 21 January 2006 (UTC)

I also find this really confusing. I suggest moving it. Riki 13:49, 25 January 2006 (UTC)

I think you mean removing it; there's already an entry for Pentium M in the 32-bit section - the entry in the 64-bit section indicates what time it came out relative to the times various 64-bit processors came out, and that entry points to the main entry. Guy Harris 19:55, 25 January 2006 (UTC)

The organization is terrible. It should either be strictly chronological (which I personally thing makes the most sense), or strictly by functional groupings (4/8/16/32/64 bit, etc.), but not both mashed together. Maybe it should be split into two separate lists. --Brouhaha 07:09, 7 March 2006 (UTC)

The reason I made this ordering was that which Guy Harris mentions above: showing which other processors were made at the same time as those in the given section. I think this is a valuable piece of information, and I honestly thought any confusion about the ordering would be reduced/eliminated by the phrase "chronological entry" being included in all the pertinent entries. --Wernher 04:46, 8 March 2006 (UTC)

It might be less confusing than some past organization, I haven't checked. But it's still very confusing, even to someone like me who is very familiar with Intel product history. I still believe that it should be one way or the other, or two lists. If it was by functional grouping, it would be adequate for each entry to have an intro date or year, but having one list with multiple entries for many products is absurd. I suppose it's possible that there may be other Wikipedia pages like this, but I haven't seen them, and I've NEVER seen such a thing in a dead-tree encylopedia. --Brouhaha 22:04, 20 March 2006 (UTC)

Personally, I would have no problem with having two lists. My only concern is that some people might update one page and forget to update th other, but that's minor. Jgp 01:14, 21 March 2006 (UTC)

And it already happens with the confusing conbined list. --Brouhaha 02:33, 21 March 2006 (UTC)

[edit] Which of Intel's line of processors was the turning point?

Intel has developed so many processors today and is the most successful company in this field. Which processor do you think was the turning point in the life of Intel? --Susam Pal, Infosys Technolgies Ltd., India

Hello Susam, In my opinion, it can be said that the original Intel 4004 is the turning point for Intel because the 4004 was the original integrated microprocessor. It can also be said that the 8086/8088 microprocessors were the turning point because they were used in the IBM PC (and clones) - a computer design that has changed computing worldwide since the early 1980's. --Anonymous

A very diffucult Q to answer, but perhaps the 8080 is as much of a candidate as the 4004 and the 8086/88, since the 8080 was the one, that, for all practical purposes, started off the microcomputer industry as a whole. That processor inspired the Zilog Z80 and the Motorola 6800, the latter of which in turn inspired the MOS Technology 6502. The rest, as they say, is history... --Wernher 02:27, 1 February 2006 (UTC)

The first Intel CPU can't really be called a turning point for Intel because there was nothing before it to turn from! I suppose the 8086 and/or 8088 would qualify since it's the x86 series that has made Intel a household word or maybe the 80386 since that was the first 32 bit CPU and made Microsoft Windows the standard for PC's but truly Intel has been on a course straight up since it began and didn't need a turning point.

[edit] Why are Apple Macintosh computers called out for the Intel Core processors?

Most other system vendors (e.g., Dell) don't get links here, and certainly not one link per Intel processor variant they use. Why should Apple? If they have to be listed at all, it should be only one link. --Brouhaha 07:06, 7 March 2006 (UTC)

Looking back, an anon added the individual models and added the Apple links. I've deleted the Apple links and have added an end-note to the T1200. Intel do not actually list it as a product; nor is it listed in their datasheet. -- iici 15:15, 9 March 2006 (UTC)

[edit] Virtual memory for Pentium I

What does this mean? "Virtual Memory 64 terabytes". Please, give the sources of information or delete this. -- A5b 16:34, 11 March 2006 (UTC)

-DeathKnight: This means that the processor is capable of supporting that much virtual memory. However, I edited much of the 386/486 info regarding how much memory they could handle. Due to the fact that it was wildly incorrect. I will look all of the specs for the Pentium to confirm.

I think the lines that say Virtual Memory (Insert ammount here) Should read as

Support for up to (Insert Ammount Here) of Virtual Memory

Its a bit longer, but it less likely to cause confusion.

"Addressible memory" refers to the amount of physical memory the processor can handle; it's 2^(the size of the physical address bus in bits). It's:

• 4 gigabytes in most of the older 32-bit processors;
• 16 megabytes in the 80386SX, which had a 24-bit physical address bus rather than the 32-bit physical address bus of the 80386DX, 80486, and original pentium;
• 64 gigabytes in the later 32-bit processors with PAE, as the physical address expanded to 36 bits.

It might be bigger still in the EM64T processors.

"Virtual memory" means the size of the virtual address space; unfortunately, there are arguably two useful values in the 386 and later non-EM64T processors, namely the linear address space (which is, as far as I know, 4 gigabytes in all of those processors, even the 80386SX, as limiting linear virtual addresses to 24 bits would probably have caused compatiblity problems with a lot of operating systems) and the segmented address space (which, if they can support up to 8192 segments in the GDT and 8192 segments in the LDT, is 64 terabytes, i.e. (8192+8192)*2^32, with 2^32 being the maximum segment size). Guy Harris 05:36, 6 June 2006 (UTC)

-DeathKnight: Thank you for fixing my mistakes. I should have double checked as the sources I had originally gone with for some of the info was wrong.

[edit] 80186

It should be noted here that despite everyone saying it was only available as embedded systems, there were still lots of desktop systems (like the siemens pc-d I have in my cellar). it was available with a 80187 fpu and a revolutionary bus system that allowed up to 127 extension cards.. if your case was big enough... and then later I got some 14mb hard drive... ah, those were the times... —The preceding unsigned comment was added by 84.176.232.207 (talk • contribs) 19:11, September 10, 2006 (UTC)

[edit] "Built-in multitasking" ?

I removed this completely nonsensical statement. It was perhaps the intention that some reference be made to support for _virtual memory_, so it's fair to say "had built-in support for multitasking o/s's that require vm".

[edit] IXP?

Shouldn't the Intel IXP product line be listed?

[edit] 4004 NOT used in Pioneer

The claim that the 4004 was used in a Pioneer spacecraft is a myth, per Dr. Larry Lesher of NASA and Steve Short of Intel Press Relations: [1] --Brouhaha 08:06, 15 November 2006 (UTC)