Honoring Chuck Peddle; Father Of The 6502 And The Chips That Went With It

Chuck peddle father of 6502

Chuck Peddle, the patriarch of the 6502 microprocessor, died recently. Most people don’t know the effect that he and his team of engineers had on their lives. We often take the world of microprocessor for granted as a commonplace component in computation device, yet there was a time when there were just processors, and they were the size of whole printed circuit boards.

Chuck had the wild idea while working at Motorola that they could shrink the expensive processor board down to an integrated circuit, a chip, and that it would cost much less, tens of dollars instead of ten thousand plus. To hear Chuck talk about it, he got a cease-and-desist letter from the part of Motorola that made their living selling $14,000 processor boards and to knock off all of the noise about a $25 alternative.

In Chuck’s mind this was permission to take his idea, and the engineering team, elsewhere. Chuck and his team started MOS Technologies in the 1970’s in Norristown PA, and re-purposed their work on the Motorola 6800 to become the MOS 6502. Lawsuits followed.

If you asked Chuck about his contribution he would dismiss the claim that he was the father of the KIM-1 single board microcomputer of the early era, “Look at the keyboard, you can tell that was done by someone on the calculator side of things” I believe I heard him say.

When asked about the 6502, his reply was that he was more proud of the peripherals, the Input/Output (I/O) chips that supported a microprocessor. These support chips made it possible for a user to interact with the microprocessor and his reply was something along the lines of the fact that you couldn’t have a terminal or cash register system unless there was a way to read the user’s keys and display something back.

Bruno Gandpa, Al Alcorn, Chuck Peddle, Leonard Tramiel, Bil Herd at the release of Growing the 8 Bit Generation
Bruno Gandpa, Al Alcorn, Chuck Peddle, Leonard Tramiel, Bil Herd at the release of Growing the 8 Bit Generation. Photo courtesy of Robert Bernardo.

The story of the venerable 6502 is one of those that we will never know just how much influence it had on people’s computing experience, even today. You may have not played an Atari game yourself, but the chances are good that the architect/designer/programmer of your computer did. The designer of the game Minecraft started on a Commodore C128, a 6502 based system as was its famous predecessor the C64, and if you have ever seen a TV show about the 1980’s they inevitably show an Apple computer, based on the 6502.

Chuck and crew were gone by the time I got to Commodore, though I was officially hired by MOS first. We sat in their same chairs in their same offices and could feel their presence, I used to joke about a still warm cigar in the ashtray or spotting Peddle’s unicorn down the hall, a reference to the wizard-like aura we attributed to the early guys. I also used to call the early team the “Motorola 5”, in tribute to their “on the run” status from Motorola, though there may have actually been 6 of them.

In the video below, Jeri Ellsworth and Bil Herd ask Chuck Peddle if he is really an Evil Genius bent on world destruction.

I have seen the original 6502 schematic, it was as close to a religious experience as I have felt, it seemed to me that the lights dimmed and thought I could smell incense and hear chanting as the schematic was pulled from the lowest drawer… apparently early schematics were printed on parchment such was the condition of the old hand drawn schematic. My friend Benny, father of many a disk drive that you may never have heard of, brushed at something written lightly on one of the pad symbols and then laughed when he read what had been penciled in. What was written was in a sense a tribute to Chuck on a pin that he had lobbied for specifically to be included in the 6502, the Set Overflow (SO) pin, the pad had been renamed CPS for the Chuck Peddle Special pin.

To Chuck Peddle, one of the fathers of the modern processor AND the chips that made it usable: 1937-2019

77 thoughts on “Honoring Chuck Peddle; Father Of The 6502 And The Chips That Went With It

  1. Technically it was the 6501 that launched things. It ran 6502 opcodes, but was plug in compatible with the 6800. So it could directly replace the 6800 hardware wish. The 6501 was n the original release, but soon disappeared when Motorola objected. So when Byte ran an article in November 1975, it was more about then the 6501 (or so I remember), but as rarely mentioned afterwards. I don’t remember ever seeing a 6501 computer schematic.

    One thibg I wondered about in recent years. The 6501 was initially promoted as a very cheap CPU. I saw the Byte article, and $25 was within my realm , and that was the case for others. That made it great for hobbyists, who only needed one. The other CPUs were more expensive in single quantities.

    But in retrospect, I wonder about the price differential in larger quantities? It’s true that when the Altair 8800 hit, the whole computer didn’t cost much more than a single 8080. But MITS didn’t pay $300 for an 8080, they had to buy the case and other parts. So the 6502 pricing mattered to Steve Wozniak, who was initially buying his own computer. But surely in large quantities the other CPUs were either the same price or not much more than the 6592. The single quantity price meant little at that point.

    1. The 6501 was never meant to be a product. According to an interview with Chuck, they knew they would never be able to sell it and they never made more than a few. It was just a finger in the eye to Motorola.

    2. The 6502 was still mostly pin compatible with the 6800. You can see it on the schematic for the Apple 1. The most important difference was that the 6502 didn’t need an external clock generator, only a crystal.

      Obviously the 6502 used a different machine language from the 6800, so that (and the 6501 which was designed to be thrown under the bus from the beginning) is how MOS got away from Motorola.

    3. A price of $380 (IIRC) was actually a bargain compared to the prices of TTL bipolar transistor CPU chips that continued to be the standard for mainframe computing up until the end of the 20th century. No doubt there was a small discount for wholesale customers, but until the 650x and Z-80 came along to create price competition for Motorola and Intel, respectively, there was no reason for them to offer steep discounts. The pricing structure that currently offers CPU chips at one or two orders of magnitude less cost in quantities of 1000 simply didn’t exist back then. The big chipmakers were more than happy to take big profit margins (and fix prices in monopolistic ways) until little companies like MOS Tech came to compete with them on price. The 6502 was the original reduced instruction set computer CPU, cutting a few functions here and there to enable massive savings in manufacturing. The rest is history.

          1. This argument has been around for some time.

            CISC and RISC refer to architectures and not simply the number of instructions.

            A RISC architecture would be better described as reduced addressing modes and an orthogonal instruction set rather than simply reduced instructions. Of course if you have less addressing modes then you will have less instructions. Similarly orthogonal instruction sets reduce the number of instructions.

            I have even heard the argument that page zero addressing is like having lots of registers like a RISC. This is the opposite of the truth. They’re NOT registers, in fact it’s a whole extra addressing mode which is quite the opposite of RISC.

            If you put aside the number of registers, the number of addressing modes and the number of instructions and just look at the architecture diagram and compare that to both CISC and RISC architecture diagrams then you will see that the 6502 is clearly a CISC architecture.

          2. Yeah, probably. 6502 is a “single-address” architecture, which is pretty RISC. Or what the youngsters inexplicable call, a “load-store” architecture. And there certainly isn’t anything like single cycle execution. On the other hand, there is no reason not to think of Zero Page as 128 sixteen bit registers used for the various indexed/indirect addressing modes and stack pointers. An assembler or compiler for a 6502 is incredibly simple to write compared to the dual-address architectures like 8080 and Z80.

      1. I don’t know, but I can’t see why there wasn’t quantity discount back then.

        The microprocessor wasn’t introduced to be a “computer” but as a controller. Intel’s first CPUs were fallout from calculator and terminal projects. Why have a whole board of TTL logic when you could program a CPU to do the work? I think I still have that National book about the microprocessor as a logic replacement.

        Even Swiftwater Bill was briefly into using a microprocessor to track traffic.

        The hobbyists changed that. They/we wanted a comiuter at home and someone noticed that microprocessors were out, and could be fine for general purpose computers. Hence the Mark 8 and the Scelbi and the Altair 8800.

        If microprocessors weren’t sold with quantity discount, what was the value of using them to replace logic? For short runs maybe, but not to replace cheap or common items.

        My previous point was that even today people talk about how cheap tge 6502vwas at introduction, without thinking it through. How could the Altait 8800 be $439 on introduction at the end of 1974 if the 8080 was somewhere about $300? I certainly remember a comment that one could buy the whioe kit and throwaway all but the CPU and still be okay. Likely an exaggeration, but commentary on the single unit price which had to be reduced at larger quantities because how was MITS going to sell the CPU and the metal cabinet, the circuit boards and Ttl logic, the heavy transformer and rest of the power suppply, and all those S100 connectors, and still make a profit?

        And yes, when the 6502 came a!ong, “RISC” didn’t exist. It was just another CPU, that in retrospect was simple since that’s what was being done. It was no breakthrough design. But ever since RISC was introduced as a concept, people have been repeating tge notion that the 6502 is “RISC” because someone tossed out the idea long ago, and others keep repeating it.

  2. 6502 tore them uppity overpriced manufacturers down several notches. Sporting peripherals support chips that easily pushed the prices down and making computing hardware available to ‘the masses’. Motor O la bitched and moaned but surfed the wave in an almost embarassing catch up BIT STILL OVERPRICED. KIMs rocked the process control and the true first 100$ computer. VIC-20 record breaker PC often missing from the conversation. crApple IBlue wish they had the numbers.
    So RESPECT.

  3. I’d just like to throw a salute to *all* of you C= guys — Bil, you and your coworkers and those who preceded you, as well as those few that came after. You literally changed the world, and I, for one, admire you for that.

    You are well deserving of the title of giant in this industry and world, and I can but hope to be tall enough someday that you might, upon leaning down and peering intently, see me, in a sort of “Horton Hears A Who” sort of manner.

    Godspeed to all of you – and Merry Christmas! (…or whatever you celebrate.)

    1. I remember those. When I was in college (UMASS/Amherst), one of the guys in my class sent away to Intel and got one. He gathered up all the Continental Specialties breadboards he could find (I think he had 3 or 4) and built up a complete 8080 system, with keyboard I/O and later added video.

      The lab manager was not pleased, as the breadboards were a very limited resource and needed for labs, etc. But he got a lot of attention, and a job offer from Intel when he graduated. This would have been 1974 or 75.

    1. The grudge with Motorola was over the pinout. The 6501 was not a 6800, they had to put some effort into the architecture. And history shows that the 6502 was at least as successful as the 6800. The single quantity price helped the 6502 get started, but the architecture seems to have kept it going. Nobody talks about the Fairchild F8.

      At best Chuck Peddle learned from his time at Motorola, but even then he seems to have offered improvements to Motorola and they weren’t interested, so he took his marbles and went elsewhere.

      The analog is Steve Wozniak, making his own home computer and offering it to HP. When they weren’t 8nterested, he started Apple with Steve Jobs

      1. I’m asking, not telling. What piqued my interest was “Chuck and his team started MOS Technologies in the 1970’s in Norristown PA, and re-purposed their work on the Motorola 6800 to become the MOS 6502.”

        “Re-purposed” may have been a poor word choice, but I didn’t write it. I now lean to Michael Black’s way of seeing it, if I lean at all.

  4. If the 6502 never existed:
    Apple would probably be a very different company.
    Acorn Computers Ltd, Apple and VLSI Technology would probably never have created Advanced RISC Machines Ltd, and the world would be different.

    Did Chuck Peddle help change the world, probably more than most people will ever realise.

    1. I was starting to get worried hackaday was not going to report on this event, but I’m very happy with the words Bil Herd provided. You guys will always be legends in my book!

      Thanks Chuck, and also thanks Bil for your contributions to computer history.

  5. Great article !
    Time now to go get my Apple ][ from the closet and have a LodeRunner play in memory of Chuck Peddle.
    Always very impressed by people like him who are able to change the world.

  6. Chuck Peddle…. Thank you Sir! Rescued me from the 1802 and provided years and years of extreme hacking enjoyment. OSI/Aim/Sym,/Kim/Apple, plus made a good bit of royalties from a mem board for hobbyists I taped up the artwork for in the motel room while stuck in Sunnyvale at a Mass Spec class ’cause the Univ would not even pay to rent a bicycle and I was bored! KFC every day for weeks ’cause only restaurant in walking distance. DATAK was a valued friend!

    Those were good days! Hope you’all get a wave like that to ride in your travels! It’s been a wonderful hobby.

    1. What’s wrong with the CDP1802? Both it and the MK6502 have their strengths and their weaknesses. In fact I got my start on a Commodore Pet machine, and then switched to the Apple 2Plus,, and then to the Apple 2E design. Even though I am surrounded by the PC, I still think of applying one or the other to problems.

      1. The 1802 came a year later. It was a static CMOS design, which meant that (other things being equal) it was slower. It’s been a long time since I seriously considered it, but I suspect that the reasons I didn’t use it paralleled the decisions of other engineers.

      2. We spec’ed out that to do the same set of floating point operations (laplace transform for digital filter and other manipulations) in the 1802 would take a couple of seconds vs. being able to get them all (6-8 operations) in about 250ms on the 6502.

        1. Thanks again for your perspective Bil. It’s important that their stories get told by the friends that remember them! Their accomplishments live on.
          We all stand on the shoulders of giants.

    2. Thanks for the great nostalgia! CDP1802 was my first, but later 6502, F8 and others. The good old days of Byte. If only we knew then how the hardware and compilers would evolve! I heard once that the original space shuttle main computers ran on 1802’s. My first microelectronics professor made the whole class by a CDP1802 based kit by Netronics. After assembling the thing my wife couldn’t stop laughing that all I could make it do was turn on a ‘stupid little light’! 7A 3f 7b 30 31 was the hex machine code if I recall correctly – can’t believe I remember it 45 years later!

    3. I learned a lot from my AIM-65, including extending the assembler for floating point in Forth. I love the sculpted keyboard in the Apple IIe or Platinum. And I have an OSI Challenger II and dual 8″ drives. It won’t boot from the drives now, I think from a clock drift problem. It ran a big spectrophotomoter with photon counter and shutters and all that at one time and did the analysis and plotted on a storage display and very early pen plotter. Might have been 2MHz? Anyway, huge amount of work done fast enough by 1 or 2MHz 8 bit systems.

      Now it takes 8 or 16 MHz to blink a light? Or 32 bits at 40 to 200 MHz for MicroPython, which I like a lot, but still….

  7. Sad news, just hearing about this now. Time to break out your KIM-1 and slosh some of those electrons around inside it.

    Back in the day I wrangled my share of both 6502 and Z80 machine language, and will admit that, at the time, I didn’t appreciate the ’02 as much as it deserved. Something about the size of the register set, the quirky tiny stack, the (apparent) lack of pointer registers, led me to favor the zilog part. But in retrospect, I achieved far more working with the ’02. Funny I didn’t see it at the time.

    Recently it only finally hit home that, well .. the zilog part was just not all I used to think. A couple of decades of C coding later, I realized the Z80 register set really doesn’t stick the “address register” concept. To the extent that you *can* implement C on a Z80, it’s cumbersome to work with. The Intel lineages didn’t get things right (re pointers) until the 8086/8088 came along.

    Now granted, the 6502 has no pointer registers at all, but that’s what zero page was for. I don’t think I’d care to try to implement C on an ’02, bcz I think that’d be missing the point. When the world consisted of sub-64K systems, you didn’t need C. What you did need, especially given the low clock rates of the time, was efficient memory access. The 6502 had this in spades. Just consider that most 02 instructions ran in 3 or 4 clock cycles, compared to z80 taking a minimum of like 11. I didn’t understand the dominance of the 6502 back in 1980, but it’s plain as day now that that’s why — the efficient memory access. Thanks CP (wherever you are) for a cpu that could get out of it’s own way!

    Also would like to add, love the comment about the I/O chips. Woz (respect, man) put something like 110 MSI parts into the ][ but look at the C64. Roughly same level of peripheral, vastly reduced chip count. (Yes, IWM in the 2e got there, too) Knowing that, ok, once you’ve integrated the processor, next integrate the I/O, was spot on. Everyone loves the C64s SID and VIC, but those 6526s did a boatload of the work, and they’re just an evolution from the 6522.

    Anyway, sad that Chuck has taken the big NMI. Hopefully he doesn’t mind if we keep on embedding 6502s in our FPGA projects ..

    ps thanks Bill, yer no slouch yourself =)

    1. I started on a TRS-80 so Z80 CPU and before I learned basic I found out how to program in Z80. I remember got a book and could not get it till I got some single step program that showed every step on the screen could press the down arrow to go to the next Z80 command.

      Then go the C=64 and wanted to learn the CPU in that. From the Z80 with a lot of double registers and the 6002 with just about one the A some X and Y but I was thinking how do they program on it. Then seen how it has like 255 zero page registers and so that’s to me like 255 registers.

      Looked at others code and it was not long till I know how to program in 6502 and doing that in Machine language it’s about down to the hardware and I did some of that too.

      Thanks for post this about Chuck peddle. I did see that video years ago with the 3 of them.

      -Raymond Day

  8. if you were in the UK in the 80s then you probably learned 6502 on an Acorn BBC model B. Ironically it was the limitations of the 6502 that what led Acorn along with Steve Furber and Sophie Wilson to develop the ARM processor.

  9. Trivia – 1802 was radiation hardened so could be used in space. It also had one illegal instruction that would attempt to do both an input and an output at the same time and potentially render it inoperative though I never saw this happen despite trying. Hand coding machine code those first years. When got the very first program up and running via 8 toggle switches and an enter pb then showed the wife…. the broom nearly got me in the head. You paid $100 to make a light blink!!

    6502 eased physical handling and was easier to implement. Take a good long look at the Disk I/F card for the Apple ][, a genius piece of incredible HACKER garage engineering. OSI had the good hardware for hackers on the electronics side,

    No microscope needed and you could hand-solder everything. Wife even got involved when I was svc mgr of an Apple dealership… had her replacing soldered in chips, repairing motherboards… and making shift-key mods.

    Thank you for the moment of nostalgia! Electronics hacking was a lot more fun back then… IMHO.

      1. My wife learned to hand-assemble 6502 code for the KIM-1 when she was in college.

        I was there, I saw it, though she will probably deny it to her dying day. She took an Assembly Language Programming course to satisfy her “foreign language” requirement, as she was/is no good at learning languages.

        So I have a soft spot in my heart for the 6502, though I never owned an Apple II or used the 6502 in any design. All my early assembly code was 6802. But I did draw schematics, not on parchment, but on what we called “vellum”, also “drafting vellum”. It’s a plastic material that tolerates multiple erasures without going through. We drew with pencils and “decals”; pieces of sticky vellum with common MSI logic functions preprinted on them. You’d stick them down, draw the lines to them and add pin numbers. Then the people in the “schematic digitizing” department would put these C and D size sheets up on big digitizers and click on all the lines and junctions, and enter all the signal names. You’d get them back in a few days. This was in the early 80s at Data General. It wasn’t long before we got graphics terminals and a home grown schematic capture system (the name of which I have conveniently forgotten).

  10. R.I.P Chuck, he gave a lot of us our first start into computing!
    And the famous SO pin. The 1541 drive ROM contains a lot of BVS instructions, if you’ve ever wondered why… ;)

  11. “…yet there was a time when there were just processors, and they were the size of whole printed circuit boards. Chuck had the wild idea while working at Motorola that they could shrink the expensive processor board down to an integrated circuit, a chip, and that it would cost much less, tens of dollars instead of ten thousand plus.”

    That makes it sound like he invented the microprocessor. Not meaning to reduce his impact, it was this:

    https://en.wikipedia.org/wiki/Chuck_Peddle

    “Peddle worked at Motorola on the development of the 6800 processor. Peddle recognized a market for an ultra-low-price microprocessor and began to champion such a design to complement the $300 Motorola 6800. His efforts were frustrated by Motorola management and he was told to drop the project. He then left for MOS Technology, where he headed the design of the 650x family of processors; these were made as a $25 answer to the Motorola 6800.”

  12. I recently worked for a low-cost toy company doing electronics. We used several sub-$.15 IC suppliers with data sheets only available in Mandarin to do simple MIDI audio. Some came in fancy SOIC-8 packages but most were CoB/DICE. Once day I was able to take a peek at some internal docs from the distributor that revealed the control core in most of these low cost MIDI ICs was in-fact a modified 6502; completely unlicensed I am sure – as with most things made of Chinesium.

    But the 6502 lives on in the most unexpected places in 2020 – like the latest talking action figures under Christmas trees this season. Quite the legacy indeed.

    1. Marco Polo is the original IP thief. Is your spaghetti a licensed copy or an illegal Chinese clone? Do Americans pay royalties to the Chinese for gunpowder? Who exactly is the IP thief here?

      1. No one uses gun powder any more. It is all smokeless powder these days. And I am pretty sure the the patents on noodles and gunpowder had long since expired if they had such a thing at that time before Marco Polo showed up unlike much of the IP that Chinese companies violate today. So if the 6502 is still under copyright and or patents then it would be the Chinese company that produced that chip if they had not paid for the right to produce it. You silly little thing you.

        1. Just FYI Black Powder / muzzle loading is still a big thing in the US, with BP hunting seasons etc. Synthetics, (not smokeless powders) have made headway.. Industrial espionage is older than patents having probably existed certainly since “States”, and industry have existed, and likely before..

  13. Literally been in a jungle for a few weeks and missed his passing, such a pity a very interesting fellow indeed. worked for ACT when they came to the UK which is my closest association. Thanks Bil for a very good article!

  14. Thank You Mr. Peddle for a career in engineering. In 1975 or 1976 I learned microprocessors by learning the 6502 in college. In the lab was the Kim-1. It was the start of a love affair with processors and computers. Thanks again.

  15. My first PC was a Sirius I (aka Victor 9000) designed by Chuck Peddle. It had a 1.2M variable speed 5″ floppy and a 10MB HDD at a time when most people were still dicking around with 160k floppies. Great piece of hardware compared to the IBM PC, but might beats right and it faded into obscurity. Not before I ran it to death programming parts for a lasercutting machine (in 1984!) though, and later upgraded the memory so that it could emulate an IBM machine to run Fido BBS software.

    As an old computer engineer I have high regard for his achievements. Rest in peace, Chuck.

    1. Finally someone remembers the Victor 9000. I worked at Victor Technologies for 7 years in the 80’s and had about 20 of those beauties. I still have one Vicki portable as a souvenir. Fun times.

  16. Charlie Peddle’s Lowly Parts Club Plan

    It was 40 years ago today
    That computers taught the kids to play
    With chips to make a new kind of game
    And the world has never been the same
    So let me introduce to you
    The micro called the 6502!

    (With apologies to Sgt. Pepper’s Lonely Hearts Club Band and the Beatles… and Chuck, who told me that Charlie’s OK; it fit the rhythm of the Beatles song better.)

    We made a 6502 badge as a 40th anniversary tribute to Chuck and his 6502 (http://sunrise-ev.com/6502.htm). We sent one to him with a real MOS Technology 6502, and got a nice letter of thanks. He was still hard at work on a multicore 6502 chip!

  17. We owe a lot to Mr. Peddle. I sent him an email once many years ago, and he was kind enough to respond to some random then-teenager’s questions. Thanks for everything, Chuck. Your 65XX family of parts will stand as little silicon monuments.

    Well written, Bil.

  18. Someone mentioned their AIM-65. That was produced by Rockwell International, the giant aerospace company of the era who also helped put things into space and invented the practical modem that 3M used in their fax machines. I know because I sold them. Just think why Rockwell licensed the 6502 technology when they had their own PPS-4, PPS-8 that went into Mattel games, calculators, microwave oven controllers at Litton – yes I assisted in those too. Rockwell engineers knew that the 6502 was the best and cheapest to manufacture at the time amongst all the other CPUs. I think Balley and other arcade manufacturers also converted over from the PPS families in due time. I sold numerous AIMs to be embedded because the electro-mechanical engineers of that era found themselves being replaced unless they could learn how these microcomputer chips could do better in new designs for the companies that employed them. Weightroniics, Silent Knight, Litton, Honeywell, 3M and many others in my sales area. The 6502 and the industry it inspired allowed many, many innovations and products now accepted as mainstream but only pipe dreams back then. Bill Mensch carries on the legacy at WDC, Western Design Corporation, which licenses, manufactures, and supports to this day, most of the hardware and software development embedded in products such as those mentioned that might be pirated. Long live the 6502, long live the story of these GIANTS and their courage to break away and change theirs and our world.

  19. Great post by Bill. Even I have personally been more Intel 80x side – the 65x side was always in My mind as well – I was just too young that time to understand the real value.

    As the RCA 1802 was also mentioned here I have to share one thing about that. In Finland – Lappeenranta was a genius called Tauno Rupponen who created educational SBC with CDP 1802. That is still only (simple) hw device I have seen with possibility to run the CPU by clicking button for clock pulses. It was possible to see bits flowing at data, address and control buses 😃 You could also input code using hex pad without any SW at board. What a great device that was…

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.