Talk:Microprocessor

8 bit vs 32 bit

There seems to be a bit of conflict between fans of 8-bit chips and fans of 32-bit chips. Honestly, both sides seem biased to me, but I think it makes an encyclopedia article better to describe all the major viewpoints, even when some of them are biased.

Two areas that I think deserve a few more words in this article: energy and noise.

I've been able to add both points of view on energy:

• I added the Saether and Fredriksen reference. If I'm reading it correctly, it unambiguously says that 8-bit processors have several advantages over 32-bit processors, including less energy and lower noise.
• I added other references that say that it is "not true" that 8-bit processors use less energy than 32-bit processors.

Alas, those other references don't seem to mention noise or the other issues brought up by Saether and Fredriksen. So does pretty much everyone agree that 8-bit processors (when fabbed at the same technology level) produce less noise than 32-bit processors? Or is there a reliable source we can reference that has some other point of view, that we should mention in this article? --DavidCary (talk) 18:00, 8 July 2014 (UTC)

Those references were deleted in a later edit. --DavidCary (talk) 04:10, 16 December 2019 (UTC)

A proposal

I am going to link this here. I made an proposal for a new Project based on microprocessors. https://meta.wikimedia.org/wiki/WikiChips

Bobherry Talk Edits 01:53, 17 April 2018 (UTC)

About this

I have additional matter about this Nadimpalli bhargav (talk) 08:51, 24 April 2018 (UTC)

Microprocessor vs CPU

The article does a poor job of explaining the relationship between the two terms. As I understand it, a CPU is a broader term includes central processing units that are not all on one chip whereas a microprocessor is a CPU all on one chip. So we need to clarify that a Microprocessor is a type of CPU but not all CPU's are microprocessors. As such, we need to make this article more clearly a sub-article of CPU's. In modern times, almost all CPUs are microprocessors so the terms are, with few exception, used interchangeably these days. --Notcharliechaplin (talk) 15:26, 21 October 2018 (UTC)

Intel, AMD, Motorola et all refer to their bit-slice product systems as microprocessors. Many early Microprocessors 1800/1801 = 1802, 6500/6501 = 6502, etc. are one chip integration of multi chip products. TaylorLeem (talk) 22:33, 23 October 2022 (UTC)

In the 1950's and into the 1960's, the CPU was a box connected to memory in a separate box, and I/O devices somewhere else. Maybe the CPU and memory together is the processor. It is, then, not at all obvious what should count as "processor on a chip" or "CPU on a chip". The early chips required a separate chip to generate the clock, especially when it needed a higher voltage than other inputs. Does that make them not single chip? It seems that there is a lot of wiggle room in the name, and that was used by the companies. It is especially complicated by the early chips meant for embedded systems. That was true at least through the 8080, and yet the 8080 turned out to make a good general purpose machine. Gah4 (talk) 02:28, 24 June 2023 (UTC)

Gilbert Hyatt's Patent Battles on Microprocessors

Gilbert Hyatt actually had tons and tons of patent applications pertaining to the microprocessor, and his intellectual property strategy seems to have influenced processor engineering and law alike. According to several sources, a USPTO director wanted to have his blocked or deprioritized. CessnaMan1989 (talk) 16:39, 16 November 2021 (UTC)

Because his patent applications NEVER documented a working product which is theoretically a patent requirements. Mental masturbation should not be patentable. The result would be granting every perpetual motion machine a patent. TaylorLeem (talk) 22:36, 23 October 2022 (UTC)

RCA CDP1802

Integration of 4 CD4057 Accumulator/ALU and array of CD40108 triple port Ram and microsequencer. Orthogonal instruction set. TaylorLeem (talk) 22:25, 23 October 2022 (UTC)

India Education Program course assignment

This article was the subject of an educational assignment at Department of Electronics and Telecommunication, College of Engineering, Pune, India supported by Wikipedia Ambassadors through the India Education Program during the 2011 Q3 term. Further details are available on the course page.

The above message was substituted from {{IEP assignment}} by PrimeBOT (talk) on 19:54, 1 February 2023 (UTC)

Intel 4004/TMS1802NC Dispute

The beginning of the Intel 4004 (1971) section contains some strong statements such as it is "definitively false" that the Intel 4004 is the first microprocessor in favor of the TMS1802NC. This doesn't seem appropriate for this section and does not immediately cite evidence in favor of the falsity. I added a Disputed inline link here and a Citation Needed tag to these sentences. Aldaron lorem (talk) 16:38, 2 April 2023 (UTC)

All of the evidence is present in the TMS1802 and TMX 1795 sections.

intel only released the 4004 in '71 november. TI already filled the patent for the microprocessor in '71 august and released TMSC1802 a single chip microprocessor in '71 september. Also, the TMX 1795 has reached working prototype state at 1971 February 24. (https://www.righto.com/2015/05/the-texas-instruments-tmx-1795-first.html) 94.21.161.212 (talk) 00:57, 24 June 2023 (UTC)

This seems to have come up in the Intel 4004 article. Also, the 4004 is better known in a commercial context. Were the TI chips commercially available? Gah4 (talk) 02:32, 24 June 2023 (UTC)

No, they weren't, but nobody said, that the intel 4004 wasn't the first commercially available microprocessor. It was. Nobody denies it.

But it wasn't the first single chip microprocessor (that was the TMS1802 or the TMX 1795, if we accept working prototypes too and why we would not) and it was definetely not the first microprocessor overall (that was the CADC or the AL1). 178.164.222.20 (talk) 09:47, 24 June 2023 (UTC)

an update to the righto article is now saying the 4004 beat the TMX as the first microprocessor:

> "TMX 1795 almost became the first microprocessor. (Originally I thought the TMX 1795 was the first microprocessor, but it appears that the 4004 slightly beat it.)"

So I deleted the "definitively false" statement. I would also suggest "The Intel 4004 is often (falsely) regarded" be changed to "The Intel 4004 is often claimed". It is problematic for wikipedia to definitive say "true" or "false" cause as the many articles go into, the debate depends on precise definition of what is a single chip microprocessor. For instance, the TMS1802 is more specifically considered a "microcontroller" cause its entire program memory is self contained and it isn't programmable on the fly. And as the righto article argues, while the AL1 could be wired into receiving decoded microinstructions, it doesn't have instruction decode on chip and the "ROM controls memory read/write, selects an ALU operation, and provides the address of the next microcode instruction (there's no program counter)" even though the courtroom demonstration was compelling. Em3rgent0rdr (talk) 21:45, 30 December 2024 (UTC)

@Gah4 and Em3rgent0rdr:

The AL1 started design in 1968 and was working in March or April of 1969. It was disclosed publicly in April 1970 in Computer Design magazine. Contrary to the claims in this article, the AL1 did not require any external circuits to operate, and the ROM chips mentioned are to implement the code of the System IV, not the internal instruction set of the processor itself. This was demonstrated in rather dramatic form in 1995 when he built a complete computer using a AL1 fabricated in 1969 (stamp still on it) with pluggable (using NES carts!) ROM, RAM and an I/O interface which he used to drive a VT100 to run the exact same program TI and Intel used to validate their entries.

If anyone can offer a cogent argument why the AL1 should not simply be "the first", given it overturned both TI's and Hyatt's claims to such, I'm all ears. But for now, I'm going with Tredennick's statement which flatly gives priority to the AL1. The 4004 can claim to be the first to be sold as a stand-alone product, but that appears to be its only claim to "first" of any sort.

Maury Markowitz (talk) 18:35, 19 June 2025 (UTC)

Ken Shirriff argues the AL1 courtroom demo is a trick:

why don't I consider the AL1 to be the first microprocessor? It used an AL1 chip as the processor, along with ROM, RAM, and I/O and some address latches, so it seems like a single-chip CPU. But I've investigated this demonstration system closely, and while it was a brilliant hack, there's also some trickery. The ROM and its associated latch are actually set up as a microcode controller, providing 24 control lines to the rest of the system. The ROM controls memory read/write, selects an ALU operation, and provides the address of the next microcode instruction (there's no program counter). After close examination, it's clear that the AL1 chip is acting as an Arithmetic/Logic chip (thus the AL1 name), and not as a CPU.

It showed that combined with a microcode controller, the AL1 could be used as a barely-functioning processor. In addition, you could probably use a similar approach to build a processor out of an earlier ALU chip such as the 74181 or Fairchild 3800, and nobody is arguing that those are microprocessors.

Shirriff describes how the ROM control logic is "running the show", not the AL1:

The AL1 chip is an 8-bit arithmetic/logic chip with some registers. It does no instruction fetching, no instruction decoding, and doesn't have an instruction set. It doesn't implement any control functions. It doesn't perform memory or I/O operations. (It's similar to a 74181 ALU chip with registers.) So how can it operate as a CPU?

The trick in the demo is the ROM has a "ROM memory address registers", which seems like an innocent latch. But this latch is under the control of the ROM, not the "CPU". The ROM and latch form a state machine that is controlling the system, performing RAM and I/O operations and directing the AL1 chip to perform ALU operations. The ROM is not sending instructions to the "CPU"; the ROM is running the show. The ROM doesn't hold AL1 instructions; it holds crazy microcode-like sequences that get the system to stumble through operations. There's no program counter as such, just the ROM jumping from address to address.

In other words, if you think of a microprocessor as ALU + control, the AL1 chip has only the ALU half, not the control half. The ROM-based state machine provides the control half.

Em3rgent0rdr (talk) 16:05, 20 June 2025 (UTC)

But by that token, the F8 isn't a micro either. I have yet to see anyone make that argument, however. Maury Markowitz (talk) 00:36, 21 June 2025 (UTC)

Naming

I think the article should also explicitly mention that microprocessors are also simply called processors, like saying in the lead: "A microprocessor (also called a processor) is a computer processor..." i think i added it to the article but for some reason it's not in the article. i have mostly seen the term microprocessor being used in the 1980s but not now Pancho507 (talk) 17:35, 13 February 2024 (UTC)

Well, it was needed in the 1970's when big processors were still popular. There are still big processors, though usually made from many microprocessors. Otherwise, it seems to me that processor is just short for microprocessor, in many cases. It can also be the generic word, when one doesn't really need to know or care about the size. It is now not so unusual to generate a processor inside an FPGA, in which case it is called a soft-processor. (And not micro.) Gah4 (talk) 19:13, 13 February 2024 (UTC)

We already have Processor (computing) which appears to consider itself a superset of Microprocessor, Central processing unit, Graphics processing unit among others. Do you think this organization should be improved somehow? ~Kvng (talk) 13:56, 16 February 2024 (UTC)

Fairchild 3804 claim

The 1970 Fairchild Semiconductor Integrated Circuit Data Catalog (printed 1969) mentions an upcoming IC, the Fairchild 3804, on page 7-65 (577 in the PDF) as a "4 bit CPU element," and that "the 3804 is an MOS/LSI device that functions as a 4 bit slice of a Central Processing Unit." Lee Boysel is mentioned designing this chip at Fairchild (as well as their 3800 ALU) in a 2018 CHM blog post, which cites an unpublished 1998 paper by Boysel for the 3804 claim, which I was unable to find online. However, it's unclear if the 3804 ever actually released; it only has a few mentions on the internet as far as I can tell, most of them mentioning it as an ALU. The 1970 Fairchild Catalog was unindexed by Google, and I only found its mention of the 3804 through a citation on this patent. I still think this claim should be investigated.

(Also, a scan of the April 1970 issue of Computer Design, which contains an article Lee Boysel often cites for his AL1 CPU claim, would most likely help with a proper description of the AL1, since Boysel claimed the article "...[described] the AL1 chip and its usage in considerable detail". From what I can tell neither this article nor the issue of Computer Design it came from are online anywhere.) Vintageperson (talk) 03:57, 20 January 2025 (UTC)