PDA

View Full Version : Has anyone else seen or visited this site ?? (4004)



Marty
September 29th, 2016, 07:55 AM
Hi All;

Has anyone else seen or visited this site ??

I had kept up with it some years ago, and then it went Dark, but it seems to be active again..

http://insanity4004.blogspot.com/2012/06/worlds-first-microprocessor.html

THANK YOU Marty

KC9UDX
September 29th, 2016, 02:14 PM
I think the 4004 gets far too much credit.

It wasn't first, and it wasn't responsible for the advent of microcomputing.

RRPollack
September 29th, 2016, 04:21 PM
Yes, I've seen it. That probably has something to do with being its author. :-D

As with all hobby projects my interest in it waxes and wanes in phase with how much free time I have on my hands. Free time has been a bit scarce recently.

Marty
September 29th, 2016, 04:28 PM
Hi All;

KC, I am going to some what dis-agree with You..

Now, I could be wrong, but I think the 4004 was the first Micro Computer chip..

Before that was Dec and Data General, where what was done in a single IC took many IC's to do the same kind of thing, In like TTL..

And yes it was not "" advent of microcomputing"" but, from what they learned from doing it, helped alot in the Development of the 8008, and from that the 8080.. Which was used in the first wide spread Microcomputer machines..


THANK YOU Marty

KC9UDX
September 29th, 2016, 06:26 PM
The TMS1000, which as far as I know has greatly more capability than the 4004 (I know the TMS1000 pretty well, but don't know too much about the 4004), predates the 4004.

I think Fairchild had something earlier, and it seems to me that Uncle Sam beat them all to it.

The 4004 is only very influential in retrospect, I think. It's only significant due to the current popularity of Intel. Had Intel not existed, I have no doubt that things wouldn't have been much different, progressively speaking. The early hobbyist machines may not have been 8008 based, they may have been bit-slice at first, and then Texas Instruments, General Instrument, Fairchild, Motorola, RCA, or someone else's microprocessor would have been used. I think the 6502 happens either way: it was intended for microcontroller applications, which would have progressed without microcomputers. A microcontroller or the like would end up in microcomputers, and clearly the 6502 advanced the state of the art in microcomputers quicker than the competition, anyway.

Don't get me wrong, I think it's important to preserve the history of the 4004. But no one is preserving all the others, and 4004 is getting undue credit. The TMS1000 may be the Multics of microprocessors.

Tor
September 29th, 2016, 09:53 PM
The TMS1000, which as far as I know has greatly more capability than the 4004 (I know the TMS1000 pretty well, but don't know too much about the 4004), predates the 4004. I don't think that's correct. The 4004 was released in 1971, and the TMS1000 in 1974. But just as Intel used what they learned from their early processors to create the later ones (e.g. the 8080), TI created the TMS1000 from what they had learned by their own earlier chips (e.g. the '1795, see Ken Shirriff's blog). (That includes the TMS0100, which *was* from 1971 and was a calculator on a chip, and maybe comparable to the 4004 - but I don't know anything about the 0100 really).
[..]


The 4004 is only very influential in retrospect, I think. It's only significant due to the current popularity of Intel. That's a bit backwards, isn't it? Intel became what it is today *because* of its early experiments with the 4004 and the 8008, which paved the way for, eventually, the 8080. And from then on Intel became what it is today.

KC9UDX
September 29th, 2016, 11:05 PM
Intel became what it is today because IBM chose the 8088 for the 5150.

Tor
September 29th, 2016, 11:42 PM
Intel became what it is today because IBM chose the 8088 for the 5150. Yes. And the 8088, like it or not (I don't particularly like it), came out of the 8080 and the earlier efforts of Intel. So if Intel hadn't started with the 4004 and 8008 it's unlikely that an 8088 would have shown up when it did, and IBM wouldn't be able to choose it. So naturally the 4004 must be historically recognized. Intel became what it is because of those early processors. What would have happened if IBM had chosen differently we'll never know, but that changes nothing. The 4004 was significant in enabling Intel to be there when IBM looked for a processor.

KC9UDX
September 30th, 2016, 12:49 AM
I think we'll have to agree to disagree.

Marty
September 30th, 2016, 03:09 AM
Hi All;

"" That's a bit backwards, isn't it? Intel became what it is today *because* of its early experiments with the 4004 and the 8008, which paved the way for, eventually, the 8080. And from then on Intel became what it is today. ""

I totally agree, IBM choose Intel and the 8086/8088 because, of the sudden onslaught of 8080 based machines that swept the market taking everyone by suprise, including Intel..

So they designed the 8085 and the Z-80 came out about this time, and with the need for more Memory and the competition from not only the Z-80, but the 6800, 6502 and others, they designed the 8086/8088 , which IBM saw as something that they had missed (MicroComputer Market).. And choose the up and coming 8088, as an edge over the rest of what was out there..
Altair, Imsai, Sol , etc..

THANK YOU Marty

RRPollack
September 30th, 2016, 10:16 AM
The TMS1000, which as far as I know has greatly more capability than the 4004 (I know the TMS1000 pretty well, but don't know too much about the 4004), predates the 4004.


According to all the references I found during my brief search, the TMS1000 became commercially available in November 1974. The 4004 became operational in January 1971, first production availability (to Busicom) in March 1971, and general availability in November 1971. Intel then released the 8008 in 1972, and the 4040 and 8080 in 1974. So Intel had four microprocessors in the field at the time the TMS1000 hit the street.

The innovative feature of the TMS1000 is that it was a complete system on a chip, not requiring external memory or I/O support chips. On the other hand, it was not expandable, as Intel's microprocessors were. Intel's 1976 entry into this market, the 8048, could also be interfaced to external EPROMs making prototyping and small production runs much easier. I developed products using the 8048 and 8049 for a small R&D firm in 1980-81 and am (was?) quite familiar with them.

I'm no great fan of Intel's. I think their design of the 4004 was greatly hampered by management's insistence on the use of 16-pin packages. The design of the 8086 was short-sighted, and it wasn't until the 80386 that a paged, flat address space was available. Nor were they the only developers of the microprocessor, and they would have evolved in some fashion even without Intel. But Intel did lead the way.

Corey986
September 30th, 2016, 10:16 AM
I totally agree, IBM choose Intel and the 8086/8088 because, of the sudden onslaught of 8080 based machines that swept the market taking everyone by suprise, including Intel..


Actually I used to work for the executive who negotiated the deal with IBM for Intel about using the 8080 for the IBM PC. If you recall at the time the Z80 was actually a more popular chip and would have made more sense for IBM to use and manage risk. According to him, the deal closer for IBM to chose Intel and the 8080 was a guarantee that they would maintain backwards compatibility for 20 years so that customers who had custom software written for the 1st IBM PC would be able to buy new IBM hardware without buying new software. If you think about the state of IBM's business at the time, that makes a lot of sense and would be a smart decision. I should add that my old boss said he had no idea if that was possible, but he wanted the deal so the product guys would just have to make that happen.

Cheers,
Corey

Corey986
October 1st, 2016, 11:11 AM
My bad. Freudian slip. I wrote 8080 not 8088. Was telling someone the whole story as it was told to me and I realized I posted the wrong CPU model.

Cheers,
Corey

Chuck(G)
October 1st, 2016, 11:18 AM
Sure, we all know how extended backward compatibility was vital to a company's success. That lack of extended compatibility (6502->68K->Power->Intel x86) is what killed Apple, right? :)

IBM should suffer such a death.

On the topic, I find old MCUs very interesting. I do have a TMS1000 from an old microwave oven; the circuity on the PCB tells me that the PMOS logic levels were a real challenge to accommodate; I also have an old sprinkler system control board with a COP444 (http://www.seanriddle.com/cop400.html) (National Semi) MCU which illustrates that it took quite a while for 4-bit MCUs to die out. The Apple Lisa also used the NS COP400 chip.

Corey986
October 1st, 2016, 03:44 PM
What I found interesting was Intel's promise was important. Zilog who should have been the top choice wouldn't commit to match intel's 20 year promise. My old boss used it as an example of how sales teams can negotiate a deal to steal a potential customer from the obvious market leader.

Chuck(G)
October 1st, 2016, 04:40 PM
You have to consider the time, also. Exxon had just purchased Zilog in 1980 and didn't have a clue about what to do with it (along with Qyx Qwip and Vydec).

I was told by employees at the time that leadership wasn't exactly strong. The Z8000 didn't make a big splash, in spite of the recruitment of AMD and Siemens.

It's a bit ironic that the Z80 (-ish) CPU is still around; while the 8080/8085 is history.

Dwight Elvey
October 1st, 2016, 08:23 PM
It Intel X86 family just recently stopped being able to execute 16 bit code.
It has been more than 20 years so Intel upheld their part of the deal.
Dwight

MarsMan2020
October 2nd, 2016, 08:55 AM
As far as I am aware, even modern x86 CPUs still startup in 16-bit real mode. FreeDOS will still boot on a new machine as long as there is a BIOS compatibility mode built into the EFI.