{
  "txid": "e9d8f0aed057456915042778dc369b90dd920eab0b14c429ac61f7b84fa1ab01",
  "block_height": 0,
  "time": null,
  "app": "treechat",
  "type": "reply",
  "map_content": "(Quick response) please excuse me for skipping introductory small talk. Will attempt to amswer your question\r\nbefore i re state the questions i asked in my second communication with you.\r\nReference material. Will take weeks to months to adaquately respond to your communication. Much appreciated you took the time to speak to me and be so civil. Thankyou kindly.\r\nHere\r\nso it's September 2018 and we are here today with Charles Moore the inventor of the fourth programming language and we're here in Edinburgh on the Firth of Forth overlooking the Forth Bridge which is a very nice and appropriate venue Oh welcome to Edinburgh Charles is it okay Chuck Chuck it's fine look this is my happy to be here excellent Chuck I thought because it's the 50th anniversary of Fort Lee is clearly a landmark it would be appropriate to go back to the beginning and do a kind of a retrospective interview take you back all the way back through the journey in your life that has been forth up to the up to the present day now hey I thought I'd begin if I may by taking you right back to the beginning and ask you when and where you were born where you come from in the United States I was born about a week ago in 1838 okay ninety nine thirty eight engraved in my memory in McKeesport Pennsylvania that's a suburb of Pittsburgh Wesch dreadful this wormhole with the steel industry of course because that was what pittsburgh was and then we moved to eerie eerie Pennsylvania on the beautiful Lake Erie which died and been reborn and then moved to Flint Michigan where I grew up okay so you were to go to school in Michigan graduate what sexual high school was in the news recently because of water problems and of course appearing about them I learned that there are two high schools now neither one of which is centralized schools so I guess a little way okay and what am I here at your your education totally little bit about you and went to MIT okay I was going to be a civil engineer and I got there and started taking courses and my calculus and decided I liked physics that's why I was gonna become a high energy physicist and nuclear physics was a topical subject at the time and there was a little article in a flip journal saying that I went to MIT to sunny unclear physics they had transposed the first two letters but eventually I got a Bachelor of physics from MIT then I went Stanford for a couple years before the mathematician because the part of physics that I liked was mathematics and that didn't pan out so well because I really was interested in computers this department was not the place for computers so where did your interesting computers become established at MIT products are going to Holland started NYC right I was kind of bored with my coursework so I got a part-time job at this miss omean Astrophysical Observatory Sao and they were in charge of boom watch which was going to attract the first subway optically and that got upstaged by the Russians who launched the satellite a year earlier that moon watch was an aesthetic scramble to try to track their satellites so it was this slight thickness it was 5758 right so I got a Giovanni a calculator to calculate I'm sorry positions for each of the observing stations and also I was calculating one of the other students from MIT pointed out that I could write a Fortran program that we do all this work for me so you're making the calculations by hand yes there was a freedom calculator that's all I did oh so did you know Fortran as a language prior to this or did the spark was ignited when your colleague mentioned this and you have to go away then for child in order to do it I wrote a Fortran manual and it started put your cards all without authorization when I finally got a program that could calculate these things I told my supervisor and it went up the chain of command and they said do it and so this was called ephemeris four and it was in early 1958 and produced reams of printed output some of which got total typed to the various stations that was a lot of fun of course it paid to have it with my grades I was working 40 hours a week instead of studying okay it was fun yes tell us a little bit about the computers that you were working on then that I suppose of service 58 computer that mi t-- got and the sodium was running at Harvard so there's a Harvard MIT cooperation and I was the IBM 709 which was upgraded to a celadon I know it was a thirty-six bit computer which is curious because half of 36 is 18 and my current computers are 18 bits all those are all servants this was a computer center where there was a glass wall to keep the spectators out of the computers and I got to go through the glass wall and actually hand my punch cards to the operator who would put them in a hopper and run the computer and eventually I got to put them in the hopper I've never got to push the buttons on the computer most of the days it was very informal like I can work third shift and get immediate turn around whereas otherwise it might be a day or two okay thank you to find those in every department ya know I put my card deck was about so long or two feet long it had to be compiled which took a long time minutes know so I had the ability of the program to read input cards to tell them what to do so I didn't have to recompile the program I could just initiate the input right okay and that was really the first tenants of the step toward forth I was going to say that does sound like that little spark there of recognition hey this is a good way of doing it as I recall it didn't ever stack the the in the interpreter would read the cards and do things and it had to read parameters and stolen away but it didn't have a stack in the stack came later okay so that's a neat sojourn into my notes here which I'm calling pre fourth the very first truthful pre for its fourth tip but not as we know it I've read somewhere that you invented fourth perhaps through gritted teeth to a certain extent because it was purely because you were unsatisfied your experiences and just trying to get work done so forth was quite a pragmatic solution to extant problem yo just for continuity yeah when I went to Stanford I got a job at the Stanford Linear Accelerator Center and there I wrote a program to steer the electron beam it didn't exist yet but it was being designed and so I this is another precursor to forth and it had a stack again I had to implement which not deck is put somehow that I could customize to modify the program and it had a stack and parameters at function calls and things like that in order to accomplish this and when I let Stanford I went to New York City the Big Apple is you can make it there you can make it anywhere and I was a freelance programmer hunt for programming jobs and so you just went out here I want you to set ups gonna take this letters to every company in Manhattan that seemed likely to want to use computers and I got maybe one percent response but it was enough to keep me alive right and one of those jobs was programming a mini computer in one of the normal high rises what here is this this would have been sixty two or three you know 65 says the Bronto were you married by that no no and it was a very interesting project I the program in a gas chromatograph again these were on the first gas chromatographs on one of the first mini computers and the programming medium was paper tape so I found myself up on the 11th floor of this high-rise on a Saturday night feeding paper tape into a reader which torrent and so I needed to take the paper tape together to patch it but I couldn't find the tape oh I forget what I did but I was really angry and just that I said there's got to be a better way as I finished that project and did other projects again with interpreters but let me just again feel there not to be a better way but I didn't have any resources to do anymore so that was 3/4 then I went to that I got married I got a job at no hassle a proper job which fruit pay the rent right and Lou Haskell made carpets it was a major carpet mill in Amsterdam New York upstate Mohawk River and there I was programming a B 5500 Burroughs bay train one of the exciting computers of the day and velasca worked in Cobalt and they wanted an online order entry system and nobody they didn't exist at the time I should say that sounds like quite a tall order this time around 1968 this is getting it into the fourth history so I programmed the be 5500 as a time-sharing system using fourth this was the second fourth the first one came a year earlier when I was given access to a 1130 IV a mini-computer it was desk sighs okay and I was tasked with demonstrating that the 11:30 driving a 220 no 2250 graphics display would help them design carpets and I did so and in that force in the process I wanted a fast responsive ability to drive that graphics display which for Trend couldn't do the IBM had for track packages did a bad job of driving it and I did a good job on writing forth so you presumably would have experimented with the Fortran package and is associated software forward life in the display right and determined that it was inappropriate the task at hand to do was it that that forced you to sit down I mean what I'm thinking is you say forth emerged from this or this would be so I would recognize it would I as forth if I was to see some source code written in that first fourth system yes I would recognize it as a fall off the the symbols were all different well now he is fetch in store for some reason I was using the sets I see with a line through it I don't remember what it did but all of those symbology were different than no and that was just it was not just in an effort to be economic with bike storage screen space space on the paper well that was the reason for the symbols yes they were abbreviations for words that otherwise would be kind of tedious to type oh and they gradually went away because they were just too cryptic once it left our Fetchit store which are so common that they deserve to have a single character representation so did you sit down at that point having determined that the Fortran system was not suitable with almost a clean sheet and to say I'm going to make my own language what did you already have some Inklings of what the ingredients that you would put in to make a suitable language which we now know know I bought I know what I wanted to do because of my experience at Stanford and MIT the this interpreter I had written was basically what I did at mo Haskell updated and expanded the interpreter to give it conditionals to give a looping capability but it was still an interpreter and it was perfectly adequate to generate the machine language for the graphics processor tools up cross compiler if you will yes so decide footballs machine language yes and then I I didn't I miss I had thought I had written forth in Fortran but I did that is not the case I had written forth in assembler and they said the language from the 1130 this was pointed out to me recently when people got interested the 1130 again for some reason then the trumpet design was thought successful because it didn't include color this is a black-and-white graphic stone and when the management realized that they sort of lost interest in the project when the water entry system was made of testing they lost the true stood out also because of the economic downturn so the whole Haskell experience kind of died away and I'm not looking for a job break that yeah Nik under I meant looking for a job literally so Chuck you mentioned in the previous segment that there was a flip it down to Moscow and you went you're out of a job you were looking for for a new job so what's next I don't know how it came about but George Cornyn offered me a job at Emma real what you notice this sorry what year was this 1971 so to talk about that 70 71 so you were married by now and you presumably had a family my hands a little baby you you got a baby okay and it's important to stay in work feeling George had known of my work at MIT I think and he was impressed by it so he offered me a job programming a data system to be attached to the 36 foot radio telescope in Tucson this work was in Green Bank which was near Charlottesville Charlottesville was the headquarters Charlottesville Virginia and so I started programming only I forget which mini computer it was a new mini computer and I took the fourth which I had developed at mo Haskell and ported it to this mini computer but this was a great leap forward in that I figured that forth could compile instead of just interpret and it would be much more efficient that way so the foresight developed at NRAO were much faster not quite machine speed this was not a subroutine threading system but it was indirect threaded code and it was fast and it was totally appropriated from the 16-bit computer so did you arrive at the indirect threat its approach to the experimentation did you try maybe tokens no it's really the first thing it occurred to me okay and then later these notions of direct threaded code which I don't think I ever used and then subroutines right in code which is what I now and maybe there was even a fourth version of interpreting but an interactive code I don't think it was original with me but I had never heard of it then okay this data collection system was a big success because let the astronomers look at the data well it was just coming out of the telescope I thought that would be fine I thought they would analyze their data in the home but no they didn't want to do it that way they wanted to take the raw data home with someone reanalyze it with this game first quick look at it and it was successful enough that the next year I was programming the telescope as well as the dating system so that they could be more closely coupled and that was a real success it was that long the same machines I'm sure the machine that was doing the data acquisition was also controlling the telescope to separate many computers okay and they were connected with a that Bank serial link I was really impressed but the opportunity to steer this gigantic telescope and and get this cutting-edge data this was a time of molecular spectroscopy they were detecting molecules in deep space for the first time and it was exciting I would I spent all the time I could at the telescope but really I had to do a little work back in Strongsville we've got to work on the telescope in the SAR time when there were thunderstorms and telescope maintenance and things like that so the fourth system that designs that that's would be what people such as myself would recognize as fourth yes very much so and was this fourth system native to the machine or running on top of vendor-supplied kernel operating system no these were all own labor force I don't remember how I got the first version to run I must have had to bootstrap it somehow right but I forget how that was the services got bootstrapped onto the mini computer than it could recompile itself and the cross-compiled from the telescope it was since standalone and that was always my goal to get a force that would be self-sufficient and these were disc based machines by then I presume they're going to sort their disk based machines they weren't tape or maybe no I think I think they were take anything [Music] but I forget what the boot mechanism was I don't know that there was a disk I don't remember okay and was the block storage technique developed there yes blocks were a convenient size sectors were too small so block gave you a reason 1k 1k bite oh for storing a me portion of the code and an application would technically be 20 30 blocks one of the unique features of forth being of course that applications are stored in their source code for generally speaking yes so it was at this time then am i right am i right in thinking that it was at the observatory that you met Elizabeth rather that's right so how did you come to meet Elizabeth what was Elizabeth what was she doing at the time because of course she doesn't you were the world's first programmer and it's quite quite widely known that Lizabeth was the world's second fourth programmer so interested to know how that came about well Elizabeth's husband was an astronomer so he was using he was at in Rio and then doing observing and she is working at a computer center in it in Tucson net complan was the chief of station at Tucson and somehow her husband Ned got together and they decided that they needed support in Tucson also I was back in Charlottesville so they hired Elizabeth to be up to provide that support okay so she became she was kind of appalled at this weird language but picked it up real quick and wasn't excited about the potential right yeah it's new to me I kind of had the impression that she was always there and so she was witnessing the evolution of the fourth system as you developed it you say that's actually not the case so she was hired and would have come in cold not knowing anything about fourth but she came in about a year after I started the Tucson project and then she was the salvation of it because they probably would have withered and died without her enthusiasm and without her support the people back in Charlottesville we're not happy that I was developing my own language and in this strange fashion and indeed one of the other programmers in Charlottesville said well he can the language I can't - so he wanted to develop his own language faith and things were getting out of control so the management sort of clamp down and that's one of the reasons for thing got formed what they puts forward with this this the Rattlers and and Ed Coughlin and be sort of out of frustration with the administration of NRL and also seen an opportunity for commercialization okay I'd love to talk to you about that but just just before we do I'd like to take just back a little bit because if if I my my search is correct why just prior to the incorporation of fourth Inc the fourth language itself was already beginning to enjoy a modicum of success in its own right among the observatories is that is that true that's right - and how did that come to be was that just chat amongst the scientists and the people working at the various observatories exchanging information and saying hey me to check out this new system that we've got that must have been the case again John Rother was an astronomer and it was a networking type of astronomer I know the theme first independent force was developed at Caltech again for astronomical purposes when you say independent you mean for that your involvement my maybe share this source code with them or something like that right it was Caltech fourth and then there were there was a version at Arizona State which was also an astronomy and it's Fred it became an official astronomical community language right which was very reassuring but that those times were passed now everyone uses see right but presumably you had at this point where there was a bit of an ownership beginning to build up behind the language you're obviously observing this that's right Elizabeth is observing this is that when the first Inklings among you began sort of emerges hey maybe we've got something here that could be success in its own right as the independence if he came clear that you could program in forth much more quickly and and have more efficient programs then you could we'll see him the other languages which were others APL Algol the trend you know all these new languages were going to replace Fortran and Fortran still exist today running mostly the programs that were developed way back in the 70s curious so if you can develop software and test the software in a fraction of the time in real time it is some commercial value in that right that's what we were hoping and that's what proved to pay out for thing could do an application in two weeks that otherwise would have taken a year and there was a market for that at the time of course they had to be marketing we were certainly less successful with marketing than Bill Gates was and we were in competition with it although we didn't realize that if Bill Gates hadn't stepped up with Microsoft for might have had a much different future okay let's take a look just a little bit of a step back so you're on the observatory fourth this has an inertia behind it it's becoming clear to you that there's some commercial value in that who was it that said you were start the company that's going to do this on our own all right well it was either Ned Conklin or John Rother or the combination net was leaving an array OS or at least leading Tucson and John Rother was going to Los Angeles I think tank I believe and so there was going to be change and we took the opportunity of this change to form for think and we all moved to Los Angeles okay so that must be said this one okay so this would be nice evan t 172 i guess yeah you've already said that you had you clearly you were married and you had a baby at this time young another gamble so that must have been a bit of a leap to we had financial backing okay from that car club of $5,000 so we were starting on a shoestring and counting on generating revenue so how did you go about so so not only is it a new business but there's a new business in a different geographical location you've moved to Los Angeles so you don't know any of the companies that are in that area that you might market to so how did you go about generating business I I don't know we did hire a marketing manager eventually we hired one but I think that came a little bit later because we had to get a couple of projects I think I think the first projects involved programming telescopes okay so we had this entree and we had a package to troll own a telescope and we merely had deported from one telescope to another again the telescope's were just being automated at the time and there was opportunity to cut the development time for a year two weeks and that got us enough funding to carry us over into the next stage which was commercial applications okay so now you're my fourth is reinventing itself through a certain extent versus general software house could that be right let me pause there so forth remaking itself as a general purpose applications software development house is this when Polly fourth came to be in the sense that you design I'm a writer thinking Polly fourth was the internal platform that we used to develop general purpose applications which later became product in its own right that's right Polly fourth Polly floors ran on many computers and I ported it to many different mini computers there was this is a time of the evolution of small computers and they were all different though not in usually significant ways but people were buying these computers and there was really inadequate software support for them much less support than you get now from computer manufacturers which is hardly adequate now and it was basically zero so people needed companies like else that could do software and we could do it it's super fast it would take me about a week to pour poly forth onto a new computer almost independent of whether it was Amal of light seen before or a completely different model it still took a week reported over whereupon all of my fourth tools were up and running and an application programming this another week you're busy maybe you're taking advantage of the fact that once you have the fourth of what I call the fourth nucleus which you would write in assembly language the rest of forth is written in fourth right so there's only a reasonably small portion of the kernel for one with a better word that you need to sit down and write from scratch for each platform the colonel was about 1k or 4k and that in that range and everything else was higher or fourth so once you can interpret a block you're off once you have the disk access it was it's a very satisfying experience is this when you were capping the famous portal for development system and and microfiche records okay that's right yep we kept records of all the applications we have done on microfiche I would have had a portable microfiche viewer and when I needed some code from an earlier application I could just retype it from the front of microfiche there was no decent portable medium for transferring data from one computer to another like you're talking about that I had that microfiche for a long time and I don't have any order that's a shame it would be nice to be able to resurrect it and put it online I gave a lot of stuff to the computer Museum in San Jose they may be the microfiche among find that I know one of the first things we did commercially was the database management system right Elizabeth rather refers to this quite a lot and that's referred to this in some private correspondence with myself is something that seems to be something she's particularly proud of oh can you talk a little bit about this database system why it came to be and some of the features of it well maybe the state of the art of databases at that time yep this is early days ok base management and do think this was something came a little bit later called a relational database so there's no real established way of representing your database at that time SQL was not a thing then we were our Covina tumor commissioned us to do a system on many duder would yet many terminals and my experience with order entry persuaded in what we knew what we were doing so he gave us a shot we came up with a very nice system I think we have 32 terminals and a 300 megabyte disk drive and they were storing transactions and they needed to search the database prepare reports and use quick access so we provided indexes into the database sorted by alphabetic order number or whatever and get very very fast access you know ten-second access to any any data in the place and we can run 32 terminals at full speed it's the same speed you can run one time and nobody else could do that at the time and he was very successful he did medical databases marketing databases I was a good that was a good application now I was a bit disappointed I wanted to do a strong me and I didn't really much care about commercial but that's where the money was right now so the databases may be just that the assignment it was used in another example in medical database systems did forth enjoy any reap any the rewards on the commercial success of that so sounds like it's quite an abundantly successful commercial venture for them sorry can you say the gentleman's name again I don't think so we were paid to do the programming just a one-off and the intellectual property rights remain with the customer we were selling forth as a tool for people to use we were selling applications and fourth was our secret weapon that Java see applications that Paul was a mistake but it was yeah and what what's with the hardware platforms of the day that was you mentioned 32 terminals this is something like an EVP 11 or something I think he is something called the mod comp but it was the class of pinochle see for the mentioned there were that much interested in developing commercial applications but that's where the money was you presumably did that for a number of years what memorable projects do you call from your time at faith incorporated all the neatest one was a an animation camera for Hollywood we did some of the animation for a movie called battle beyond the stars which was a remake of the Magnusson settled in space and let's fun oh the thing knows dude we've got a camera you need to control the camera in such a way as you can attract a spaceship as it crosses a background and do it many times within a pixel so they get the proper overlays and effects and so I spent time in a solid studio and in Hollywood programming this camera to the sky that was fun that took a little longer than the usual application I think that was a month or so so something completely different but again fourth was a good language for that because it was fast it was efficient and it was flexible you could describe how he wanted the camera to move and I would do that I also did Oh sound editing video editing to do application or you putting up images and mixing sound with it and that myself that was funny those are good people to work with in Hollywood so I've applied force and academic and commercial entertainment area and it was effective in all remaining so so if you've told us a little bit about some of the projects that you've worked on fourth incorporated because clearly you have some fond memories of projects that you've worked up and yet you decided to move on from fourth incorporated to pursue another Avenue the next generation perhaps fourth Kutner so you've taught us a little bit about the 4th projects that you've worked on that 4th incorporated clearly enjoyed your time there but you did side to move on from 4th incorporated and you left forth to pursue an interest I believe in implementing 4th in hardware that beat was that specifically the reason I left yes now I'm interested that I'm what interests me is that it seemed necessary correct me if I'm wrong for you to would have to leave 4th incorporated in order to pursue that so was there some resistance to you pursuing that within 4th incorporated to the extent that you felt that you have to go it alone - it's not complicated as everything does somehow we encountered a man called John peers and he established a company he was interested in forth through some reason he established a company to develop for hardware in the product course of doing this he bought for think so for think was involved with this hardware development okay and he was providing financing in fact that's about all he did he spent all of this time financing eventually for think bought themselves out of that deal it didn't turn out to be satisfactory but at the same time we have developed the Novak's computer chip so that development took place at fourth incorporated so all messed them together to develop the chip I had to be in Silicon Valley and so I moved to Silicon Valley and forth and stayed in Los Angeles and partly because I was getting bored the applications were fine but I had done the applications I didn't need to do them again and I wanted to do something more exciting besides I had this theory that software was in a real rut nothing new sighting had happened in software for ten years Hardware was or was that I was dynamic and aggressive and changing new generations of hardware continually coming out and I was interested in getting onto that cutting edge I was wrong it turned out that hardware development is even more conservative than software development but that's not the way I saw it at the time so I used forth to design the Novak's chip not actually the layout but the architecture and I simulated it and proved that it would work that gives certain performance levels and so we made it show did you have any background in electronics prior to this I had wrote a book you read a book and then you went off and developed your own chip and was called this is by Carver and Mead about CMOS chip designed and I've read the book and I said I can do that and so I did and I was right oh no that's the only reason that the Novak's chip and all their fourth chips sorry will come out the other reason the pool having half a year or the fact that you read a poker then we're gonna decide your own achievement we just cover that again yeah we can cut back its out if I had known more about chip designed I wouldn't have been able to develop these chips because they were they didn't follow the rules right everybody knew Sprague who tries me the way chips are designed so this was a fresh look at it and it proved to be very successful the normans chip was a 16-bit chip with 16-bit instruction and various fields bit fields in the instruction that could implement various aspects of forth words for instance of a shift or an ad or logical operation or a conditional they were all built into this one instruction so you could J something like a jeep and an at simultaneous thing or at least even one instruction slot that's right okay it was the chin duty of the programmer to pack as much into instruction as he could and thereby maximize the execution speed it worked it was good but it wasn't a parallel processor is that right so but it affects multiple instructions in one fetch and then sequentially executed no the signal destruction but that instruction could do parallel things I don't know similar to what the arm can do the arm can execute instruction and do a barrel for example at the time also similar to what the the 5500 was capable of doing this with my models a few world war or how computers should act so some prior inspiration oh did we apply for a patent I don't I don't recall I think well I think we got a couple of patents knowit's and then Pierce was unsuccessful in his continuing financial endeavors and sold the technology to Harris Harris produced a chip called mm Express 4000 Express you know rad hardened process which has been used in a lot of space projects still used today Anthony know that that chip was different than the Novak's but I used the Novak's architecture and corrected a couple of bucks did it not I'm sorry did it correct a couple of bucks that were in the original method oldest had a problem with interrupts and another problem of multiply which were vastly overrated it didn't really amount to much but Pierce could not come up with the money to make a new noise and so the ball past I was frustrated by you know it's pleasure kind of ended nothing these projects ever end they started with her away yes and I've never seen ending but at some point I left Novak's and I was gonna decide my own computer and it was gonna be a 32-bit version or 32-bit version of the Novak's for the 32-bit instruction and Moorfields I worked on that for a year too nothing ever came of it can you tell us a little bit about the how the design of checklist was done then what years are we talking about we're talking early late 80s early 90s no fights with some middle 80s I think right maybe five years old oh so he's still laying out onto the good tape then or did you have computer-aided - we've been contracted with a company in Scottsdale Arizona who had experience with designing cane arrays so the Novak's chip was a Gatorade and there were tools you had to use to decide the Gatorade and a man named Bob Murphy who was responsible for doing that I was responsible for designing the architecture destruction scent registers the stacks and he implemented that indicator ray very very successful collaboration my force code was basically a simulator of the instruction set running through the architecture it didn't produce a design it produced it validated inside and that's how the the bug did the interrupts came in because you couldn't test interrupts in the simulator summer rental so this is my first introduction to silicon design tools the industry standard tools I followed them unsatisfactory but I it is it's sort of a classic that the silicon people insist you use their tools because they trust them they have been calibrated and foreign tools are not welcome that's still the case today even more so you must use industry standard software but there was a slip-up somehow the service called Moses and it was at Caltech at the time that you prototype chip designs and it costs about five thousand dollars to produce 20 chips we took you I took advantage of that and have an account with Moses for a while greener a still has an account with Moses the technology at the time was three micron 3,000 nanometers it's now down to eight nanometers huge shrinkage of the process but it became more more opaque to the independent designer like me at the time I was working with Moses for $5,000 you can get the chips now the new process is if you can get access to them cost you 10 20 million so it's completely out of range and the independent or even a small company like John Prine arrays but at the time it was exciting and you submit your chips inside you get just back at a couple months it's gonna work wash your fingers so your experience with designing the Novak's cost you to detour slightly back into software development in order to correct your perceived grievances shall we say with the tools of the time that we use to develop hardware would that be right yes it's it's always been a back and forth between hardware and software the software divided developed the hardware you have to implement fourth on the new hardware in order to design the next generation of chip and that was very productive informative so from Navy's come forth was born that be true no no the first game Shaboom I was gonna I was going to decide the 32-bit machine with 32-bit instructions Russell fish persuaded me that use 8-bit instructions and pack them for per 32-bit word and that led to a completely different architecture well completely different instruction set the architecture was the same with two stacks and a CPU and some kind of i/o so again that chip is designed with interest and to Shaboom was designed at okey-okey semiconductor okey decide Center oh I had to use okie simulators to verify the chip design but I could use forth to design the chip no I couldn't even I had to use okey-okey layout tools law so this is my second exposure to standard industry software and it was but not your doing the layout I was doing allowed and it was painful it was but frustrating I can see it could be done better but we got the chip and it worked it worked perfectly I think oh and and fish was gonna do the marketing I was doing a design and the design was incredibly successful and the marketing was a failure again to build a better mousetrap the world does not beat a path to your door you have to pave the path for them but shabu was the first chip I designed with stacks on the chip Novak's and external stacks in memory chips which was very clumsy it's a Harris took those stacks and put them on their on their chip so that that informed me that the stacks could be on ship Harris sacks were very deep Aris had 256 team Staxx the Shaboom had 16 deep stacks so a very shallow stacked my current designs have eight deep stacks stack doesn't need to be deep it just needs to be present yes I comments you've certainly on the data stack if you've got more than eight slots on the data stack you're probably doing it wrong if if to do deep recursion to recursively compute the Fibonacci series you need huge stacks larger than anything you can put it on hardware so you have to software implement a stack but you can always so okay so so unfortunately the Shaboom was not successful but undeterred you moved on to ie 21 would that be correct Shaboom was not marketed successfully but it did lead to 10 or 20 pounds so it was very successful as a patent vehicle so known as a computer chip so we were able to derive some income from that eventually came along later well I came later so you're still kind of struggling here to make ends meet I presume looking for a project that's gonna pay off nicely no the shoe booth was developed are they get a shoe string no money upfront it took me I took about a year again living off savings that was another project came along it was an Internet appliance this was in the 90s and the internet was just blossoming so we formed a company to build a little browser email around a video tube and programmed and forth and running a this was the site 21 and the company was ITV correct yes so is this when was this with Jeff Fox Jeff was involved in that Oh Gary Langstrom was the president of this new company Jeff wanted a ship which could be networked and that was the f21 these these chips had the same computer basically but different capabilities one aiming at graphics display and one aiming at networking this computer let's see derive from the Saguna but it had four five bit instructions in a computer word and two stacks and the cpu so forth no dings chef came first the p21 that's right the p21 was the first version of this architecture developed for us eh tink who saw uses for it in industry displays but the key he sold them to signage once we got very successful chip and it took several form types I think he made 20,000 he I don't think he ever made back his expenses I have termed myself and Shana but I will design these beautiful chips that are not commercially successful my latest chip was saying GA 144 which has not been commercially successful yet you still have hopes but they were good chips they worked they were fast they were low-power yes well as you know it's not necessarily the best hardware that becomes the most popular you know that from the current state of the world shall we say no don't tell okay let's take a little bit but let's take a little bit of a step back so it was a rat candy Jeff Fox is in my opinion because there were a number of individuals in in the fourth universe that are landmarks along the journey forth yourself is Elizabeth rather at baby and one of those pizza I believe made a significant contribution to fourth is Jake Fox who's sadly no longer with us might as well finish you could recap how you came to know Jeff projects you worked on together what your relationship I I don't recall it must have been through think the fourth century school there were a lot of the I can't tell you the history of fig but it was formed on these late 70s I think and a this is a collection of people in Silicon Valley and it spread to Europe and Japan and China and gradually died away but Jeff was a shaker a mover in the Silicon Valley and he commissioned this f21 ship partly because of his old ambitions importantly I think is a charity gesture to me I can't to bring some it's a work a few some paid paid work as it were you know okay so what's it's a strictly a business relationship between you or did you become friends it was just friendly oh I think I think I signed a contract with him Fox enterprises something like that to design the chip and I think I fulfilling the contract so from the is it was the f21 a precursor to the multi-core chips The Witcher are so not the f21 the the chip he designed with Jeff was that the precursor to the multi-core chips at the yes okay yes that's right things things were kind of at a standstill come around 2002 year 2000 when it occurred to me I could put multiple computers on the same chip these these chips were an embarrassment he couldn't have too small a die or if you couldn't handle it you had that a certain size die partly because of the mechanics and partly because of the pants around the outside so my died were mostly empty now all the people would fill that with memory but I didn't know how to design memory so I thought I was filled with computers and the first version had 24 computers an array of 4 by 6 I think it was oh they could talk to each other and it was a networking a truly parallel processor and worked well each one being is hiring individual fourth machine yes each one had a little bit of memory sixty-four words of memory and then these standard stacks and CPUs I had been using in the f21 so I the new thing here was the memory then I had to design the 64 word memory which I managed to do it worked it has been approved since it would be hard to make it larger than 64 words okay what's the constraint you've got to drive large capacitance in the memory and you have to address decoding to get identify a particular word and this is a long time solve problem in silicon world but it's not known to me to put someone else's memory on your chip is a real hassle it can be done but it's no I can't do it that's interesting so that's the primary constraints with the current generation of multi-core chips are the memory constraints at least it means is computers are tricky to program but I don't think it's really too constrained I think it's a feature because it encourages you to write small compact code I think if I doubled the memory people could not do any more they can just do it more easily right okay programming does force you to plan and factor not only the software level without the hardware level note do you plan your path across the chip from what each node is going to do so it isn't new I don't like to like don't like G's the word paradigm but it is a new programming paradigm for most people with development paradigm again it's you must learn how to coordinate the activity so many computers each one of which cannot do very much it's not like an Intel Core which is a full-blown Pentium it's a tiny computer but tiny computers can do remarkable a lot of things especially when you have many of them the 24 did the 24 courtship turned into a 40 courtship and this was the company called TPL with that electron as president and so entering into a somewhat controversial phase of 4th Street one of them yes oh it's at this time in developing the multi-core chip the electron discovered there's some value in the patents from shamone and so he started licensing these patents and did so very successfully ultimately raising some I think 200 million dollars in license fees did any of that success find its way to you I was supposed to get maybe 50% and then they got negotiated down to 25% and eventually I ended up with about 5% and this was not acceptable so controversy erupted and so after said a moment to evolve 11 million dollars so this was not no nothing I was I should have been happy with you but that wasn't our agreement and so protest I wish I had the eleven million dollars now but it's long since anticipated not foolishly but unfortunately on the other hand if I had 11 million dollars I would have invested it in a new chip and probably lost it anyway so well there but it did you did manage to funds the incorporation well I guess we should look slipper TPL just for a moment am I right in thinking that it's true that the ctrl C has now been settled in a legal manner there was litigation that were muscles back and forth and they eventually were sent lot sense for a lose-lose situation the only one who wins as a lawyer yes but sometimes there's no recourse another company was involved compatriot fish and I were co-inventors on the patents and fish sold his interest to Patriot and I kept my interest and these this led to the conflict and an ultimate resolution and joint venture and patents and litigation oh all very complicated all very unsatisfactory from my point of view a waste of time but it did lead to the multi-core chips yes I mean presumably while you're involved in litigation and happen to make depositions and get involved in the minutia of the various evil cases you're not designing that's right but I did collect a wonderful team of force people to to design and program the 40/40 courtship Jeff Fox was involved and Greg Bailey and Dean Sanderson and it was a star-studded team which is great work but it would did not impress my chrome and he terminated which was another source of conflict but all of that was born greeneries right so wait so some of the income he managed to drive from TPMS i used in order to form green arrays seed funding through that new kind of thing what was it shoebox money I forget we tried to make a deal with like Rome where he would provide financing but he wanted control and the whole problem was his control was destructive so greener Ray's went off and got some scratch funding from angels and decided that 144 it did not electron did not like that he thought we was stealing his intellectual property but that was not the case and it would lead more lawsuits that hasn't happened we justify why that hasn't happened it hasn't happened right it is that was a subtle pillars we want is that because you're able to satisfy him that it wasn't the case no I think we satisfied him that we were not commercially successful so there was no point you can't sue someone who has no assets of its the point suing subliminal assets and green arrays has a wonderful chip and no market no market so far still promising applications the market is not quite what I expected I was designing this chip to the unclean lean beam forth ship it turns out it is a very low power chip and the market is a in low power not in forth on the other hand forth was required in order to make it to be low power so it all hangs together okay so let me say it's like power to put some figures on that a typical application the only number that comes to mind is seven Pico joules for instruction it's on it's on green erase website of a couple of years ago it was the lowest power computer in existence by an order of magnitude I I don't know the situation today I'm not the track yet the MTS has a very low power leaves the promoter just low-power microprocessor it's ten times the energy of us so how do you use this chip in order to take advantage of the low energy I think it'd be grateful artificial intelligence but we have we haven't followed that killer app you that could be the can could be okay something that needs to be able to scale up highly parallel the inferences are quite clear are they not but just going back to what you're saying about the potential for legal action to be taken in the event that green arrays are suddenly successful does that not provide an impetus to come up with an alternative design maybe a new architecture in order to protect oneself against possible legal ramifications the the greatness of the design notwithstanding its presumably a desire to extricate yourselves from that potential risk you're right it would be nice to do that I don't have any idea of what a new architecture could be I've shot my wad if you will and besides I don't think that would head off the lawsuit lawsuits do not need to be rational they don't sue anyone for any reason and it's up to a judge to decide so making we we redesigned the 1:44 using new tools using swift forth for instance the dog to the fourth using same force it's and color for that stuff fourth would read redesigned everything and proved to improve the design made all the changes were possible and I think we made enough but again something like call us the judges call well I don't want to go there it's just too painful so I'll bring this that brings us to the present day I'd like I'd like to know mizuka solution a quick recap on green arrays and mother 4th luminary is Greg Bailey something if you could recount how you came to meet and know Greg and indeed live alongside them agreeing arrays well I think Dean Sanderson it was the first programmer that for think hired and Greg and Dean were a pair and so Greg got involved with more think and have been involved ever since and forth applications greg has a company called Athena which has a number of successful forms applications dealing with the Internet and he's the president of green arrays greener race once upon a time had maybe 12 people actively involved now there's about six scattered all over the world it's it's easy to have these remote collaborations with the internet and again I think we're taking advantage of this cutting-edge technology so it's there or what would you what do you think next is what's next for Chuck Moore and green arrays well I'm the chairman of green arrays which is a ceremonial post to all intents and purposes are retired but it's possible to come out of retirement green arrays has a couple of exciting opportunities which might come to fruition oh one of which will be a new version of the 144 I think it's gonna be called the GA 144 a 13 and I don't know why can you talk a little bit about the chip about the 144 they did the new the new version the version will have lower power applications at the moment sometimes we have to sit and spin waiting for an event to occur and that uses power if we don't have to spin it'll cut the power significantly we also need more communication between multiple chips on our development board we have two chips and they can talk across a high-speed serial line but there's only two high-speed serial lines we want to have six now that that's that's one direction for I would like to go is to a small geometry with that that would draw me out of retirement if we could get down to fourteen nanometers of ten nanometers with thee I could put 10,000 cores on the same sized eye as we have know and I think that would be a qualitative change 10,000 computers you good program what would you do the Hulk you it's it's all mind-boggling we could do that but two requirements one you have to have the $25,000,000 result to make a prototype second you have to persuade the fab house that there's going to be a significant volume to be worth their trouble and significant volume means millions so we'd have to have the killer app plus the money and they're not asleep Oh you mentioned sourcils forth yes that is my current interest is it I'm surprised to hear that because from researching online watching various videos of you fireside chats over the years I know it's something that you pursued but I thought you abandoned it you're ultimately there was a journey that you went on and you're ultimately you're not convinced of the value of it are you saying that you you're re-evaluating that I'm really worried again oh it still isn't clear that there's an advantage but I are you go back to the adage the map is not the territory okay and this fourth source code is a description of the program it isn't the program and translating it into the program is not completely satisfactory it would be on a on a fourth computer like like the 144 there's one one correspondence between source and object and that's the end of it but I like the notion of a decompiler you've got the object program and you scan it and generate what looks like the fourth source you just turn on process backwards the advantage there is that you eliminate the source instead having two files source an object you only have object so you've cut the overhead of an instance of an application in half big dealers memories cheap so that was always kind of marginal but recently I've come up with a new way of programming the PC the PC only has one stack and adding the second stack is most inefficient what I'm doing is using the 16 registers as a stack and explicitly putting things in the register that I want the way to do this without violating the fourth appearance is to have super strips the fetch operator would fetch from an address in register 0 and put the result in register 0 if you want to fetch to to register 1 you put a superscript 1 on the gesture sign and so on using superscripts to indicate register sources and destinations leads to a very pleasant appearing d coupon yes that's quite a nice notation to read and I want to experiment with that oh so just to be clear you're not sitting at a keyboard tapping eggs in or are you know it's octal octal was much more comfortable in the until one structure said but that requires you to memorize the opcodes no I've got a fix for that okay you point to an instruction the opcode is decoded in the description this is shown to you like this is an instruction that fetches from register one and the source of register to look at this it's a work in progress and I'm not there was this as possibly that's something I work on another thing I want to do it's a native force on the PC which means I have to go into UEFI instead of BIOS and I don't know how to do that but I can work on that and the fourth thing I can do so I sort of neutral as to which of those futures will pan out I'm you see I'm not gonna bust the gun I think I'd like to see ten thousand fourth cores on a single die help it help us raise 25 million dollars what make a deal I don't know anyone actually pays that you make a deal with the family and they in exchange for promising in huge volumes of juice they will negotiate that's not my part of the business yeah but I think we'll then I think we'll leave people to think about what they could possibly do with 10,000 cores on that on a single chip it's probably a good place to leave it I'd like to thank you for coming to Edinburgh forth easily and there has been a pleasure to talk to you today this times one\r\n\r\n\r\n\r\nhttps://youtube.com/watch?v=SX3kXbLmwn4&si=kulYF2ttGXHpPfin",
  "media_type": "text/markdown",
  "filename": "|",
  "author": "14aqJ2hMtENYJVCJaekcrqi12fiZJzoWGK",
  "display_name": "MarkKordusic",
  "channel": null,
  "parent_txid": "4dbf3da439f7920b0c17db5df1fcd65a4c2736b8c14f8ef2ed892128d58cf5f6",
  "ref_txid": null,
  "tags": null,
  "reply_count": 1,
  "like_count": 0,
  "timestamp": "2026-03-17T07:52:34.000Z",
  "media_url": null,
  "aip_verified": true,
  "attachments": [],
  "ui_name": "MarkKordusic",
  "ui_display_name": "MarkKordusic",
  "ui_handle": "MarkKordusic",
  "ui_display_raw": "MarkKordusic",
  "ui_signer": "14aqJ2hMtENYJVCJaekcrqi12fiZJzoWGK",
  "ref_ui_name": "unknown",
  "ref_ui_signer": "unknown"
}

{
  "txid": "4ab4870ef3621405bcd20e11a84670ef30acaa706303c759dddea84c58f9ab04",
  "block_height": 0,
  "time": null,
  "app": "treechat",
  "type": "reply",
  "map_content": "Much much much more to come please bare with me if you find it in your heart to do so.\r\nhttps://www.forth.com/wp-content/uploads/2018/11/thinking-forth-color.pdf",
  "media_type": "text/markdown",
  "filename": "|",
  "author": "14aqJ2hMtENYJVCJaekcrqi12fiZJzoWGK",
  "display_name": "MarkKordusic",
  "channel": null,
  "parent_txid": "4dbf3da439f7920b0c17db5df1fcd65a4c2736b8c14f8ef2ed892128d58cf5f6",
  "ref_txid": null,
  "tags": null,
  "reply_count": 1,
  "like_count": 0,
  "timestamp": "2026-03-17T07:52:34.000Z",
  "media_url": null,
  "aip_verified": true,
  "attachments": [],
  "ui_name": "MarkKordusic",
  "ui_display_name": "MarkKordusic",
  "ui_handle": "MarkKordusic",
  "ui_display_raw": "MarkKordusic",
  "ui_signer": "14aqJ2hMtENYJVCJaekcrqi12fiZJzoWGK",
  "ref_ui_name": "unknown",
  "ref_ui_signer": "unknown"
}

{
  "txid": "f75d290cc5852a7df7f420def757f2ee1c77907391812b9679e5d25429a19a4e",
  "block_height": 0,
  "time": null,
  "app": "treechat",
  "type": "reply",
  "map_content": "Supplementary quotes.\r\nChuck moore\" forth is the only programing language i would consider programing a computer with\"\r\n From memory may be incorrect wording but i believe he said  something along these lines very closely.\r\nAnother transcrpit for reference material and video.\r\ni am chuck moore i created fourth 45 years ago it is still being used which is moderately amazing it's certainly being used by me it's the only language i would ever consider programming a computer in i'm the cto of a company called green arrays which is a startup that's been in business about five years so it's not exactly a startup but we're still looking for customers a killer app um investors we have a high-speed low-power multi-core chip 144 computers on a 5-millimeter chip it's a remarkable little chip it's a lot of fun to program it has enormous computing capability i'll show you that in the next slide so first i'm going to talk about fourth and then i'm going to talk about the fourth engine which is a computer which runs forth very efficiently i'll show you some fourth programming and then point out why how you can program in low energy and i recommend that all of you keep energy in your mind when you're writing programs because pcs are going to be going away smart phones are going to be embedded embedded meaning i embedded in your body in your ear in your eye and your heart and power is going to be crucial in those applications if we talk about smart dust we have untold trillions of computers floating in the air they aren't going to have very much memory they are going to have very much power if you talk about molecular computers that are circulating in your bloodstream they're going to have much power they're going to have much memory and they aren't going to be able to use much energy so keep that in mind as you move through the next 10 years oh let's go to the next slide all right here's a here's 144 computers just so that you know what we're talking about there are eight rows of 18 computers and then the geometry of a chip that turns out to be uh square each computer can talk to its neighbors it's a mesh of computers four neighbors each computer takes one and a half nanoseconds to execute an instruction that means it's running at 666 mips uh it takes four milliwatts if it's running flat out which is set seven pico joules per instruction which is small now i'm going to be talking about energy there's energy and there's power power is energy per unit time power is watts and a watt is a joule per second so it really doesn't matter how long it takes to perform an operation an operation is going to require a certain amount of energy it requires energy because you have to charge capacitances and you have to run currents through [Music] resistors but energy is the key and one of our instructions takes roughly seven pico joules more or less one of these computers takes a hundred nanowatts if it's idle remember there's milliwatts and microwatts and nanowatts so it takes very little energy just sitting there and 144 computers they all power up idle the chip at reset is drawing no power the entire chip can run 96 gifs 96 000 mips which is a lot of pcs it only takes 550 ml watts to do that that's half a lot if the chip is idle at reset 15 microwatts of of leakage current although all the transistors are sitting there no transistors changing state but you still get a little bit of leakage through the uh through the gate this is what i mean by low power this is the lowest power computer that i know of a green arrays one of green arrays customers developed a chart showing all existing computers commercial and experimental and we're right at the top of the list this is about the best you can do the only people that can do better are those that deal with sub threshold logic with the high noise margins and this is a a good computer the computers are numbered from zero with their row in the column zero to uh seven one seven and i'll be referring to those numbers so that is that is the that is my motivation for talking about these things i wasn't paying any attention to energy until a couple years ago when we started realizing that this chip is remarkably low energy and now that we know how important it is i can reduce the amount of energy it's using and i will be doing that i've been to a number of hardware conferences where people are talking about designing asics large a6s millions and millions of transistors and i've always considered that size and speed were important parameters but in the modern context there's only three parameters that matter power power and power we've got to reduce the power for portable applications and also to reduce the energy used by these server farms that require a nuclear power station to run okay here's one of the computers uh it's 18 bits 18-bit words for not very important reasons but it's a very uh happy number it lets you count up to 256 000. it has a 64 word ram that's not kill words that's words you can pack four instructions per word so each computer can execute a that's up up to 256 instructions that's what i mean by saying it's a small computer that has implications in the way you program it as characteristic of fourth and a fourth computers it has two push down stacks one which is common to all computers and that is a stack on which you store return addresses for nesting subroutine calls the other which is pretty much unique to fourth is that you store parameters on it anything you read goes on to the stack anything you write comes off of the stack there are three registers of interesting there t is the top of the stack s is the second element in the stack and r is the top element on the return stack you've got the programmer has got to be aware of these two stacks and manipulate them nicely nothing is done automatically there are three address registers p is the program counter which is where you are in the instruction stream a and b are just registers there are four communication ports to your neighbors up down right and left and you can act you can read and write to your neighbors you can read and write all four of them if you want to there's a bit for each port in the address so that is that is the context those are the facilities available to a programmer here is the instruction set there are five bit instructions we can pack four per word except that in the third slot you have to add a couple zeros so in the list of instructions those in red can be put in slot three and all the others can be put in any slot jump instructions can be in either slot 0 1 or 2 and all remaining bits in the word are the address for the jump i'll just briefly go through the instructions this is a perfectly conventional instruction set except for a few in the right hand column you've got jump you've got call you've got a jump if zero you've got a jump if positive you've got a jump if the r register is non-zero and decrement so that's a decrementing count for uh for next loop the second column you've got fences and stores fetch from register a fetch from the address and register b or fetch inline in the next word in the instruction stream and the same equivalent stores third column are our alu operations we've got uh binary operation add and and or unary operations two star two slash and minus the binary operations take as input the contents of t and s the two things on the top of the data stack and they return an argument nt throwing away s so it's a it's a stack operation these are called zero operand instructions the only reason i can pack them into five bits is that you don't have to specify their operand the operands are known in the third column you've got some stack operations you can duplicate the top of the stack you can drop the top of the stack you can pop the stack from s into from the return stack into the data stack vice versa you can push the data stack onto the return stack and then three interesting instructions so you can store into register a store under register b read back from register a and do nothing the dot is a no op no op is essential in order to fill in slots in a word instruction word that you haven't used 32 instructions if you haven't memorized them already it would only take you another five minutes and that's partly what i mean by saying this is an easy computer program now if you're familiar at all with fourth those are fourth words there's a one-to-one mapping between your source code and your machine language compiler is a few thousand lines of code it doesn't optimize it merely translates your source into the machine language this is a different world from what i've been hearing in the last couple days um it may be a world that dates back to 1968 and it's not interesting to you but again i caution that the future is going to be different than the present and i'm i'm happy to deal with a very transparent very small compiler you've got we can argue about whether you call it a compiler or an assembler because it really doesn't do anything but compiler is a more impressive word here's here's the code here's code for one of these computers and this is an example of color fourth i created force in in about 1970 and i've used it ever since i call that classic fourth color fourth i devised in 2000 it has a number of nice features mostly that i felt i needed a new language for the 21st century but one of the things it does is it minim it reduces the amount of of punctuation fourth has very little punctuation compared to say c but color forth has almost no punctuation instead it uses color each [Music] each source word in the source code each word in the source code has a tag that indicates its function and the function is reflected in this editor as color white words are comments they're ignored red words are being defined so the definition of go is the green words follow it uh those green words are compiled uh the yellow words are executed at compile time they're direct commands to compiler or to the yes to the compiler the gray words you see there are inserted by the compiler to tell you where you are in your 64 words when we finish the definition of go we've used one word of memory and so your program counter is set at 0 1 in hex init marks the end of code which is compiled into memory and begins the code that is compiled with this executed at load time these computers have got four ports and you can read and write to the ports they're memory mapped so you can also jump to them if you jump to a port you are waiting for your neighbor to give you an instruction if you jump to all four ports at the same time you're waiting for one of your neighbors to give you an instruction to execute and it is the programmer's responsibility that only one of those neighbors will actually provide an instruction if you get two at the same time you're dead at load time at the time we're loading this program into memory we're executing those green words after init up is the address of the up port it is stored in the register a down is the address of the down port and it's stored in the register b and then we jump to go so this is how you can initialize a computer without actually having to execute code from its ram and i'll show you some more examples it's it's very nice to be able to tell your neighbor to do something for you for instance there's a hierarchy of memory available on these computers you can get at the data on your return stack instantly you can get at the data on your re i'm sorry data on your data stack instantly you get data from your return stack with one instruction pop 1.5 nanoseconds you can read data from your ram but it takes you five nanoseconds to do that you can read data from your neighbor's ram uh if you have written a little program and and send it to him he can give you sequential data from ram without you having to give him explicit addresses or if you're doing an fft he can give you scrambled data from his ram knowing what you're going to do with it you can have smart memory if your neighbor doesn't have enough memory he can ask his neighbor i've had as many as 10 computers chained up passing passing data to me and i can get the data as fast as i can handle it you can actually use your neighbor's stack these stacks are circular you've got the two registers on top of the data stack and then eight circular registers beneath them so if you push things onto the stack eventually they fall off the end if you pop things from the stack after the first two you're going to get the next eight circularly you can get you can circulate the stack indefinitely which is useful and i'll show an example you can do the same with your neighbor your neighbor can give you things off his stack indefinitely and you've got four neighbors so we have lots of memory available to you in these small chunks of 64 words 64 words is 9 000 words of memory on chip now for words 18-bit words you can think of them as 18 000 9-bit bytes if you wish i actually use 6-bit bytes so i can get 27 000 characters here's an example of some of the programming and this has direct relevance to energy it's um two star is a left shift left shift t this is something that we do remarkably often there's a lot of bit bit blinding plus is an ad next is a countdown loop if if you run out of count it it falls through and the count is discarded so here's an example a four next loop with a call to drive in the middle going to be done ten times the loop count is one less than the uh number of repetitions um oh that's confusing micronext is another kind of loop instead of jumping to an address you jump to the beginning of your current instruction word so a micro x a micro next loop can handle uh can can repeat three instructions because micro next can be in slot 3. so the instructions can be in slot 0 1 and 2 or any combination of that if i say 416 for micronext i'm executing an empty loop with only the micro next instruction i'm going to 416 times which is a microsecond another way of doing that would be to put a no op in there now i have a two instruction loop and i execute the no off and then i execute the micro next and repeat indefinitely and i get in a microsecond again the only advantage of the second one is i can count longer than 256 000 i can count up to a million instruction delays so those are basically delay loops there's two problems with delay loops two problems one problem they take energy the computer is running and micronext doesn't take as much energy as some of the other instructions but still you want to keep that in mind of those two loops the empty loop and the no-op loop the first is uses less energy because micro next is executed repeatedly in the second case you alternating me two and two instructions no op and micro next so the instruction decode transistors are toggling in the first case nothing is changing in the computer except the count so you're much better off with the first construct than the second if you're interested in energy and having become interested in energy that's the way i do it i'll show you some um fetch is fourth symbol for uh at sign is for symbol for a fetch operator so fetch means read store means right bang means right exclamation point those are particular symbols for historic reasons and also because these words are used so frequently that they deserve a small a small word now the communication between neighbors has has a handshake involved if i'm reading as all the computers start out doing i raise my read line and my neighbors can look at that if they want but everyone is reading so there's the one massive collision of 144 times four ports um no problem somebody will break that symmetry by writing when that happens when both the read and write lines on a communication port are high both computers say and take off and run the data is transferred from my stack to his stack when both lines are high there's no no logic no latches no delays um it's it's sort of a blocking read if if i want to read i just hang up i just wait until i've got something to read it takes no energy while i'm reading while i'm waiting the chip is idle i mean the computer is idle no transistors are toggling communication is is key when you design an application you've got to do two things right write code for each of the computers involved and then manage their communication um it's fun there's all kinds of nice tricks you can play to make things work properly here's another another example of code this is put into block 200. the yellow line at the bottom describes how it is put into 200. this code which is a a message if you will will be sent eight nodes left and three nodes down across what i call my ether interface it is code that pre-exists in all the computers and passes messages back and forth this is a clock actually it's more than a clock it's a it's a it's driving a ceramic oscillator which is like a crystal except it's easier to drive the word drive is going to drive the crystal crystal is attached between a pin which this computer has access to and ground i hit it with a positive voltage and it it the the the pin will rise i hit it with a negative voltage and the pin will drop i do that 10 times and the crystal sees those drive signals and starts oscillating so i can start the crystal running i'll show you how i do that in a minute then having started it i don't have to delay any longer the drive signal was a micro next loop and it was pre it was pre-programmed to have the right frequency so that the crystal would start to oscillate but that drive is is running the computer and it's using energy actually you use energy when you put an output on the pin and that energy probably dominates everything else but anyway we go to pump i no longer have to have a delay loop i can look i can read the pin and this is a blocking read and when the pin transitions from low to high i wake up and do something one of the things i do is change the polarity of the edge i'm interested in so next time when the pin goes from high to low i wake up otherwise i'm asleep so it takes no energy to keep this crystal running i just wake up and kick it and go back to sleep ring avoids the kicking i can pump it and then i can let it ring for a few times while the amplitude gradually decreases and i can pump it again a strategy for driving a crystal with using minimum energy and it it works beautifully now in the case of initializing when i load this code i'm initializing the state of the computer in this case i have all these dupes and overs i'm filling the data stack with numbers and taking advantage of the fact that it's circular so in drive when i do a store b i'm storing something into the address in register b and that address is the address of the pin and the thing i am storing is already on the stack and will remain on the stack indefinitely so i don't have to fetch a literal which makes the code smaller actually it doesn't matter because the code is only 16 words long but still you like to do it nicely and you like to explore the limits of how how nice is possible so i put these numbers on the stack in exactly the right way and then i can reference them in the code without any cost this is a good crystal driver for several reasons one that only uses one pin and pins are a limited resource we have 88 pins on this on this chip it is not as efficient as a one of these little surface mount uh crystal oscillators we're a very low power computer but we're not as low power as a custom asic we're better than fpga much better in an fpga but if you design transistors unique to a particular application we can't compete we've got versatility we can program all kinds of things but for a dedicated application you're better off with custom silicon all right optimum programming this is not being anything that you aren't uh you are familiar with the key absolute elephant in the room minimize the number of instructions you execute that's almost synonymous with minimize the size of your source code actually most of the instructions you execute are going to be in a loop so you minimize the size of your loop you can do that by factoring your application very carefully by unrolling loops by doing all the tricks to minimize the number of instructions you execute now actually in the case of these computers you can only have 256 instructions so you have to minimize the total number of instructions as well as the number that are actually inside the loops you want to use all the slots in the word the slots which you don't execute or don't use are going to be in the right hand end of the word they're going to contain no ops but they are going to be executed they're going to take energy and they're going to take time so you don't you want to avoid that it is hard and perhaps impossible to use all your slots that's just the cost of doing business also when you're trying to minimize energy you want to recognize that there is a minute there's a floor on that you can reduce the amount of energy you use but eventually going to hit the point where there's a certain amount of work to be done certain number of transistors to switch you're just going to have to swallow that cost the goal is not you can't reduce it and except the last the last line all you can do is minimize waste use all the slots fetches and stores are best put early in the word in slot zero or one instead of late in the word because as soon as you stop using the address bus instruction prefetch can take place and fetching the next instruction word so it's ready for you when you finish the current instruction word so there's a left bias to those i o operations then it matters where you put the code in memory uh a jump in slot 2 only has three address bits which means it can jump within a little three a little eight word page you want to position your jumps where they matter so that you can use a slot to jump and that just means moving things around in memory that means you need to be aware of where your jumps are and what slot they're in and the compiler tells you that but the best thing you can do and actually the thing that is the most fun is you come up with a better algorithm whatever you think you're doing think about it say do i really have to do it this way is there a different way which might be better better in matching the constraints that the little computer puts upon you that's to make it fast to make it compact and fit in 256 instructions the same rules apply really you want to minimize the number of instructions you want to avoid literals a literal is a fetch from the address and the p register which is pointing to the next instruction in your instruction next word the instruction stream and you can store a literal there when you fetch it the p register is incremented over the literal and you just picked up something in line and when you put a number in your source code that's what happens um that's certainly cheaper than making an explicit reference to ram but it's not as cheap as having something already on the stack which is why i preload the stack with up to eight numbers that i'm going to cycle through dupor would give you a zero duplicate the thing on top of the sac exclusive wore it with itself and you pack a zero it's a lot cheaper than fetching a literal zero which would require one slot for the fetch instruction and four slots for the literal and itself but this only works with zero it works a little bit with minus one but most literals you can't construct you have to actually pick them up dupoir destroys the top of the stack if you care about the top of the stack then you have to have a dupe dupe ore but it's still cheaper micronext is better than next if you can do you can use it a micro next is often used with shift instructions um if you want to shift left 10 places you can set up a four next micro next loop it'll count down four places and do a shift this is slower than actually having ten unrolling that loop and putting 10 2 stars in line but until you get up to eight or so shifts it's cheaper to put them in line initialize from the port but again the better algorithm is the better way of making your code small hey timing is working out nicely here's another page of code this is a i call a block and i found that this is the amount of source code that you can you can fit in 256 instructions this is an impressive little piece of code it goes into node eight it's reading sram or it's it's accessing sram you've got these words read write and read modify write which lets you address either randomly or sequentially into into off-chip ram offship ram is absolutely essential it's only the only way we can get a decent amount of memory we've got a million words of chip ram in this particular chip ram chip and can address it in 50 nanoseconds there are three nodes involved in reading ram one node has the address bus one node has the data bus and one node has the control lines it's worth talking about but i don't have time now for low energy programs first thing you have to do is be able to measure the energy you do that with a micro ammeter attached to the the the input trace to your chip such meters cost a couple thousand dollars but they're absolutely worth the investment and they let you measure the things which i'm trying to optimize what you want is a low duty cycle you want your computers to be sleeping they're sleeping they're not doing any energy if they're spinning they're they're draining energy a 32 kilohertz crystal will give you timing real clock timing without using any energy because you're only waking up every whatever it is 16 30 milliseconds to do something if you if you zero your stacks you avoid stack thrashing whenever you read something and put it on the stack you're pushing the stack if the stack all contains all zeros you're pushing zero into what it was all previously a zero and it's not taking very much energy if your stack is random you're going to be using measurably more energy than if it's empty same with a return stack and there's another thing which we only recently discovered it matters where you put your loops if you if your loop has a lot of address bits changing it's going to cost you more energy than if it doesn't so if you're jumping within a page you're better off than if you're jumping way back when more address bits change and so it goes they're fun to program they're challenging to program you have to factor in fourth and this chip you have to factor into these tiny computers and i didn't talk about wires i showed you the code for wires you fetch from one port right to another port and you've programmed node to be nothing but a wire it just passes messages through itself um that's you've got to do that you are free to choose put your your your active computers anywhere you want to put them your i o pins because if you aren't reading input and writing output you aren't doing anything and it's a layout you've got to lay out the code in a way that minimizes the amount of wire nodes you have to pass messages through and that's what i have to say you\r\n\r\n\r\nhttps://youtube.com/watch?v=0PclgBd6_Zs&si=Mxn_OEflFc12GeF7",
  "media_type": "text/markdown",
  "filename": "|",
  "author": "14aqJ2hMtENYJVCJaekcrqi12fiZJzoWGK",
  "display_name": "MarkKordusic",
  "channel": null,
  "parent_txid": "4dbf3da439f7920b0c17db5df1fcd65a4c2736b8c14f8ef2ed892128d58cf5f6",
  "ref_txid": null,
  "tags": null,
  "reply_count": 0,
  "like_count": 0,
  "timestamp": "2026-03-17T08:11:59.000Z",
  "media_url": null,
  "aip_verified": true,
  "attachments": [],
  "ui_name": "MarkKordusic",
  "ui_display_name": "MarkKordusic",
  "ui_handle": "MarkKordusic",
  "ui_display_raw": "MarkKordusic",
  "ui_signer": "14aqJ2hMtENYJVCJaekcrqi12fiZJzoWGK",
  "ref_ui_name": "unknown",
  "ref_ui_signer": "unknown"
}

{
  "txid": "13d9e5277d2d8c4790ef334bdc0d8cae9a14c3b9e3932b72b18af141bbed2075",
  "block_height": 0,
  "time": null,
  "app": "treechat",
  "type": "reply",
  "map_content": "@Sunnie  can you please refrain from responding until i have adaquately responded to your communication and have stated here that i have done so.",
  "media_type": "text/markdown",
  "filename": "|",
  "author": "14aqJ2hMtENYJVCJaekcrqi12fiZJzoWGK",
  "display_name": "MarkKordusic",
  "channel": null,
  "parent_txid": "4dbf3da439f7920b0c17db5df1fcd65a4c2736b8c14f8ef2ed892128d58cf5f6",
  "ref_txid": null,
  "tags": null,
  "reply_count": 0,
  "like_count": 2,
  "timestamp": "2026-03-17T08:11:59.000Z",
  "media_url": null,
  "aip_verified": true,
  "attachments": [],
  "ui_name": "MarkKordusic",
  "ui_display_name": "MarkKordusic",
  "ui_handle": "MarkKordusic",
  "ui_display_raw": "MarkKordusic",
  "ui_signer": "14aqJ2hMtENYJVCJaekcrqi12fiZJzoWGK",
  "ref_ui_name": "unknown",
  "ref_ui_signer": "unknown"
}

{
  "txid": "67470ea42d8730050d034bc7319dc46d40208f51fe8132964bc4f34b2d0bd83f",
  "block_height": 0,
  "time": null,
  "app": "treechat",
  "type": "reply",
  "map_content": "Draft of response to @Sunnie \r\nOk firstly you are not a collegue as i would have to at least  spoken to someone at length many times face to face to be certain that someone is in fact a true collegue so the remark about excellence simply does not apply to you.",
  "media_type": "text/markdown",
  "filename": "|",
  "author": "14aqJ2hMtENYJVCJaekcrqi12fiZJzoWGK",
  "display_name": "MarkKordusic",
  "channel": null,
  "parent_txid": "4dbf3da439f7920b0c17db5df1fcd65a4c2736b8c14f8ef2ed892128d58cf5f6",
  "ref_txid": null,
  "tags": null,
  "reply_count": 3,
  "like_count": 0,
  "timestamp": "2026-03-17T09:25:17.000Z",
  "media_url": null,
  "aip_verified": true,
  "attachments": [],
  "ui_name": "MarkKordusic",
  "ui_display_name": "MarkKordusic",
  "ui_handle": "MarkKordusic",
  "ui_display_raw": "MarkKordusic",
  "ui_signer": "14aqJ2hMtENYJVCJaekcrqi12fiZJzoWGK",
  "ref_ui_name": "unknown",
  "ref_ui_signer": "unknown"
}

{
  "txid": "5ffdea5bcd8bcdd46c687386f1637f213beb741aa91dce792ce50c29a882eac8",
  "block_height": 0,
  "time": null,
  "app": "treechat",
  "type": "reply",
  "map_content": "One more for now. My distrust of google ai is one of my foundation principles.\r\nThe only reason i repeat the only reason i am talking to you  @Sunnie  is someone here seemed to me expressing positivity and interest in you. And have the impression you are a friend of Treechat.\r\nSorry to be so blunt, i am simply trying to assist Treechat devs. Pure work no joy.",
  "media_type": "text/markdown",
  "filename": "|",
  "author": "14aqJ2hMtENYJVCJaekcrqi12fiZJzoWGK",
  "display_name": "MarkKordusic",
  "channel": null,
  "parent_txid": "4dbf3da439f7920b0c17db5df1fcd65a4c2736b8c14f8ef2ed892128d58cf5f6",
  "ref_txid": null,
  "tags": null,
  "reply_count": 0,
  "like_count": 0,
  "timestamp": "2026-03-17T11:34:45.000Z",
  "media_url": null,
  "aip_verified": true,
  "attachments": [],
  "ui_name": "MarkKordusic",
  "ui_display_name": "MarkKordusic",
  "ui_handle": "MarkKordusic",
  "ui_display_raw": "MarkKordusic",
  "ui_signer": "14aqJ2hMtENYJVCJaekcrqi12fiZJzoWGK",
  "ref_ui_name": "unknown",
  "ref_ui_signer": "unknown"
}