Dammat. "were not limited" -> "were more limited" On Wed, Dec 21, 2016 at 9:15 PM, drlegendre . <drlegendre at gmail.com> wrote:

"The Z80 had more players and more names than all the rest" And yet it was essentially a bit-player in the days of the 'home computer' revolution - at least in the US. CBM, Apple, Atari - the three big names in home computers, all went with the 6502 family. And perhaps even more importantly, so did Nintendo, in the NES. The main use of Z80 in US home computing was in the absurdly small Timex / Sinclair ZX80 series - with their awful cramped membrane keyboards and seriously limited sound & video. The Z80 also showed up in the Osborne, Kaypro and TRS-80 models.. mostly due to the fact that CP/M was written to it. Commodore also put one in the C128, but by then, it was almost a dead letter. On Thu, Dec 22, 2016 at 9:49 AM, allison <ajp166 at verizon.net> wrote:

We all hang out with people who are smart enough to see things the same way that we do. Accordingly, our choices in computers, cars, cellphone providers always look to us like the MAJORITY. They are the BEST, and certainly the MOST POPULAR [among everybody that WE hang out with], but not necessarily the best selling. If the world were just, and the BEST outsold the worst, then we would all be using Amiga :-) On Thu, 22 Dec 2016, drlegendre . wrote: And yet, somehow, z80 was outselling 6502! http://jeremyreimer.com/m-item.lsp?i=137 Radio Shack, TRS-80, WAS one of the "three big names". It had a not-insignificant share of the market, and until 1982 was the best selling. Don't ignore the impact of having incompetents peddling in thousands of store, in every city and town! Atari took a while longer to get market share. http://www.trs-80.org/was-the-trs-80-once-the-top-selling-computer/ At the same time. Depending on how you define "first" ("first" to show V "first" to ship V "first" to be available for shelf purchase) will define whether Apple, Commodore, or Radio Shack was "the first". It is trivially esay to select a definition of "first" to make it your choice of those. Apple was the first of those announced and shown. I bought a TRS-80 ($400 (or $600 if you wanted their composite monitor and cassette player)) because it was the first one [by multiple months] that I could walk in the door of a local store and buy one. The more appealing Apple, which had been announced earlier that TRS-80, was hard to come by for several more months. That time differential of months seems inconsequential 40 years later, but it mattered to me right then. And, for most rational measures, Apple, TRS-80 and Commodore initial releases were a tie. (Was the photo finish by a nose, a whisker, or a hoof?) When the 5150 came out in August 1981, it was months before I could actually get one. AFTER the 5150 came out, people relized that TRS-80 was doomed, and in 1982, Apple 2 finally started to outsell TRS-80. It was LESS obvious that Apple 2 was doomed. But, within Apple, they knew there were troubled times ahead, and came out with the disastrous Apple 3, and disastrous [from point of view of SALES] Lisa. 'Course IBM poisoned the market for everything else, and nothing else sold like IBM. On August 12, 1981, I said "In 10 years, 3/4 of the market will be IBM PC and imitations of it." It is amazingly impressive that Apple (Mac) survived IBM! (If you think that Mac outsold PC, then you are looking at YOUR circle, and need to look at actual sales numbers) But, by the time that the Mac came out, TRS-80 was finally becoming that "bit player" that some assume that it was, or should have been. Which was years later, and WAS a bit player and absurdly small. It was NEVER the main use of Z80 in USA home computing. TRS-80 outsold them more than 100 to 1. Was that really a membrane keyboard, or was it just a PICTURE of a keyboard as a recommendation, like the "part of this complete breakfast". CP/M was written to 8080. Z80 was simply the "hottest" 8080 compatible processor available. Osborne and Kaypro were literally years later, and they did, indeed simply build clever, innovative CP/M machines. I've never been sure how much market share CP/M had, since that was a different circle than I was hanging out in. I'm sure that WITHIN that circle, it would seem like it was MOST of the market. Commodore's Z80 in the 128 was due to unnecessary fear that they might lose market share to CP/M, when IBM should have been their big worry. I don't know all of the details of the ST/Amiga technology swap, but BOTH were too late, if the primary goal was competing with IBM. What percentage of Apple 2's had Z80 cards added to them? (once estimated at an unbelievable 20%, and reputed to be why IBM thought that CP/M was a Microsoft product!) -- Grumpy Ol' Fred cisin at xenosoft.com

On 12/22/2016 11:37 PM, Fred Cisin wrote: As someone that was on the front end of the TRS80 it went from Zero to 60 in '77 as the first deliveries were a batch of 1000 then 5000 to the stores. they sold very fast far faster than expected. Indeed the first 12 months had a very steep curve. Most number and data I see to day in the popular media is just plain wrong. Allison

On 24/12/2016 8:32 AM, "Fred Cisin" <cisin at xenosoft.com> wrote: NO source is completely reliable. Most number and data I see to day in the popular media is just plain wrong. particularly on anything like this. Surely, there must have been some [relatively] objective compilations of the sales data? Interesting discussion. Of course sales was only part-reflective of price point and technical capability/suitabily for a particular market segment. There was also marketing and distribution. Get all four right and sales boom. Commodore seems to have excelled at all four by the time the C64 came along. Personally I never saw the C64 as one that was suitable for serious work however. When it came out I was using my own disk based TRS80 M1 clone (System 80) for writing my graduate thesis. I was also president for a local club so kept membership records. I'd also computerised our home finances and used to write my own statistic programs for the research work I was doing. The System 80 was great for these non-gaming productivity things. I had the opportunity to try the C64 when it came out. The high set keyboard. ..the mushy keys....the slow disk drive....the 40 column screen....the limited BASIC..the lack of a decent OS..the weird (to me) character set. I did not feel envious at all and would not have swapped my machine for a C64 given what I was doing. If I was a kid or gamer however the C64 would have won hands down! In the early 1980s period it was only the Apple II line that were true general purpose machines IMO. They could serve small business, the home professional (the kind of stuff I was doing ) and the kids/gamers. The flexibility was due to those slots and the open architecture. You paid through the nose for Apples though, at least here in NZ. One could argue that the BBC and Atari 800 line had similar general purpose capability but their pricing and market positioning kept them out of the small business market. Terry (Tez)

On 23 December 2016 at 05:45, drlegendre . <drlegendre at gmail.com> wrote: urs was no exception, and I thank you for it. There are only 8080 instructions in CP/M, not a single Z80-specific instruction. CP/M came about around 1974, the Z80 in 1976. All the CP/M computers I used had a Z80, but that's just shows what was popular at the time.

not informative responses are inevitable CP/M was 8080, and was intended for 8080/8085/Z80. But, as Allison pointed out, there was nothing stopping somebody from writing software that would run on CP/M that might require a Z80. Or write the BIOS for the computer as Z80. Therefore, a Z80 machine would be optimal, and if you wrote software to be sold into the CP/M world, it had to run on 8080, AND, if it made any BIOS calls, survive a BIOS that could be 8080, 8085, or Z80. Boring, certainly. Not sure if it was sophisticated enough to be stodgy. Ah! Therein lies the rub. The TRS-80 wasn't any good for some things. As a primarily entertainment machine, it was rubbish. No color, no sound, no joysticks, grossly inadequate graphics. (some of which could be worked around) If you were looking for an games/entertainment machine, it would be outside of the consideration set. Of the set of people looking for a games/entertainment machine, "NOBODY has one." Best game on it was "Adventure". Best graphics was Lim's "Android Nim" On the other side of the market, the TRS-80 did not have 80x24 text, and had a memory map that was incompatible with CP/M. Parasitic Engineering (Howard Fullmer!) and Omikron, both made [somewhat expensive] sandwich boards for the TRS-80 to change that memory map, and to add 8" SSSD drive support. Of the original "Big three" (Radio Shack, Apple, Commodore), who came first, Apple was the only one with entertainment capabilities, but they priced it out of your market. Later, the Commodore Vic-20 and C64 were aimed at your segment of the market, and priced appropriately. Radio Shack's later "Color Computer" (6809!) was far more appropriate, but it also suffered some crippling design decision errors. Apple II, TRS-80, and Commodore PET didn't have 80x24 screen text! THAT was why my cousin rejected them and went with a Heathkit. If you want to be able to use it as a terminal, . . . Later, add-on cards for Apple2 came out for 80 column. TRS-80 didn't get 80x24 text until the Model 4 (which is what the Model 3 should have been) (Model 2/12/16 was never intended to be a "Home" computer, and was marketed for small business) -- Grumpy Ol' Fred cisin at xenosoft.com

On 12/23/2016 02:30 PM, Fred Cisin wrote: As Z80 machnes of the day the Kaypro was meant for work and screen games like Adventure played well on it as it was responsive. But there was always a divide between games/graphics/sound and those with good text terminal. An example of oddballs was S100 Cromemco with a dazzler and terminal board . Or maybe Startrek. TRS80 was one of the "packaged systems" aka allinone that tried to be broad and mostly was limited at all. All of them were trying to find a market and it was a market that was changing at the same time. And it was late to the game. Cool machine but it misses again. Yes the H89, DEC Vt180 (costly though), and many others had decent terminal, Good processing and disks. Games not so much. The TRS80 later mods were late enough to loose presence. Apple With 80x24 helped it firm up its footing as a useful business machine. I still go back at if the desired software was available then the machine was suitable for the Commies and Apples that included games. The other is was Editors, high level languages, spreadsheet, and database, that group favored machine with z80, 6502 and good 80x24 screens. That last group barely included TRS80, Apple with Softcard and all was in, then there were the multitude of other z80 systems that nearly all ran CP/M even if they came with something else. If I had to call a year 1980 was a breakpoint for systems. I will call it the year of rising expectations. It bridges from early 1979 though 1980 in a broad way. If you ran a widely used OS your had a good start. If you ran the popular and desired software you had a good start. The number of CPUs in the running were few and usually Z80 or 6502. The 16biters were being talked about as the next thing but not really there yet. Then it was more about what you did and how well. An example of that is 80x24, if you had sparklies and flashes it was less desirable for example. it is was a TV monitor vs built in and good quality. If you did a weaker display scheme but it could be improved by adding a card that was a plus (S100 and other buses had strength here). Same for available memory and adding more in 1979 64K was expensive and uncommon but a year later it was nearly required. Same for mass storage audio cassettes were on the way out and floppies and hard disks by 1980 reaching the not so obscenely expensive to have and very much needed for the software expected. The key there is EXPECTED. People were buying machines and had for the most part left the mad hacker trying to make it work behind and had clear wants and needs. Thats the first generation Home/personal computer user now looking for the upgrade machine. There were also the new customers wanting to do specific things and didn't care about specs, boards, and bells and whistles. Its no small comment that in that time frame the big mags like BYTE, Kilobaud, and Interface Age were setting those expectations. Oddly games had variations of the above but expectations were also higher and things like better graphics, color and complex sound were important. Oddly they had the same limitations and dependencies that were being address with CPU speed and a need for more bits. Looking back it was not so much the good old days as they days of chaos, innovations and new and watching what would stick to the wall. Allison

On 12/22/2016 11:04 PM, Fred Cisin wrote: Both had a very long life in systems and embedded products. It wasn't cost, I truly believe it was the generations that had trash 80 , apple][, and others containing them and were familiar. For a brief moment in the history of computers that was indeed true. Then other better developed systems displaced it. Marketing over actual design plus plain availability. Not true in its entirety. IF you look at the 1980 and 1981 issues on Byte, interface age, Kilobaud, and Dr Dobbs it was clear 16 bits was the next step. There were systems delivering that but it was all very new despite the 8088 being already old! The 5150 was not so much the great 16bitter as it was IBM. Having those three letter made it official. What interesting was I had at work a multibus system with 8086 running at 8mhz the summer of 1981 to do work and IBM was running at 4.77mhz. At that speed it was not that much faster than Z80. The life of TRS80 was from introduction in '77 and was doing the downhill slide by 1980. In those days that was a long life for a point product. The T/ZX80 was mostly viewed as a toy. It was late here for that level of a machine. Still it was small and cheap and a more than a few were embedded into systems... ;) CP/M was 8080 code save for one thing by 1980 finding 8080 compatible aps was getting difficult. Also most of the CP/M implementations were hybrid code, CP/M core was 8080 and most everything else including the bios was Z80. Also it was the introduction period for ZCPR, and P2DOSThat spawned a generation of improved CP/M and then CP/M3. It was the 8080/8085//z80 market. The only other that had volume but was very hardware specific was TRS-dos or the kin LDOS. Granted Cormemco had their unix like os (chromix) and ther ewer others like NS*DOS. But if you wanted the miainline aps it was CP/M. Even Microsoft was a major supplier of aps and language tools for that base. Late and strange. Likely that 20% was close to right. IT was a good base to have A z80 in as a full up apple had good display (high res version) and disk and tools support. It was also around the time when the ap was more important than what CPU you bowed to. I have the advantage of being there not as a kid but as a professional trying to be bleeding edge and grapple with all the different machines. My history includes designing with 8008 commercially in late 72 into 73 while still in college. I ordered my Altair 8800 in December of 1974 (1 week after I received the copy of PE). I would retire that before the end of '77 for a NS* horizon crate to replace the very flaky Altair (who would think gold over copper without nickel would be so bad!). I would move to extending that NS* to multiple cpus and softsector floppies. At the same time I played with COSMAC, SC/MP/, 8035, TI9900 (technico Super starter system), 6800 (d1), Kim-1.... I watched the computer world evolved from my exposure to PDP-8 in 69 though 32bits (VAX) by 1982 and got to participate and play with most all of it. The biggest thing was not what CPU you had but what could you do with it. It was the appearance of the applications, programming tools and languages (BASIC, assemblers, higher level languages), screen editor (electric pencil), spread sheet (visicalc), the data bases (DBASE), Those were the things that made a Apple or TRS80 of CP/M box useful and desirable. No coincidence that mass storage specifically floppies then later hard disks were a factor. If your system could not run those aps and connect those devices it was replaced. Its why apple ][ and Z80 based CP/M had the longer life (presence) as the initial 5150 was a few years getting to that point both in capability, aps, and cost. To my experience The PC did not displace existing Apples][ and CP/M until its cost and advantage was well established as people already had sunk cost and functional aps. For those that were new to it all the PC was the TRS80 warts, incompatibility, and all and thats why I felt the PC was a long step backward until the 286 and 386 appeared. The PC world to me by then was moribund as it was not better and I had 16 and 32bit machines that were both faster and better. I was a late adopter of PC as it was compatibility with the ecosystem that forced it as PDP11 and VAX was still better and functional until IP based networking was important in the late 90s. Allison

It also saved the 100% compatibility problem (in this case, with Commodore's CP/M cartridge by designing it onto the board, and with Commodore Magic Voice, which fouled banking by altering the memory configuration lines in realtime: the 8502 would crash, but the Z80's activity would not be detected and the C= key could be checked to force a C64 memory map). -- ------------------------------------ personal: http://www.cameronkaiser.com/ -- Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser at floodgap.com -- FORTUNE: You learn from your mistakes. Today will be very educational. -----

On 24 December 2016 at 05:02, geneb <geneb at deltasoft.com> wrote: Seconded. I don't think CBM was scared of CP/M at all. I think it maybe thought it was a handy extra. Some of the story is here: https://hackaday.com/2013/12/09/guest-post-the-real-story-of-hacking-togeth… I am somewhat irritated by Bil Herd's claim that it was the last 8-bit computer. The MSX TurboR was arguably the greatest Z80 home computer: https://en.wikipedia.org/wiki/MSX#MSX_turboR It came out in 1990: https://www.msx.org/wiki/MSX_Turbo_R But its R800 CPU is arguably 16-bit although MSX-DOS doesn't use that. https://en.wikipedia.org/wiki/R800_(CPU) It is in part based on the Z800: https://en.wikipedia.org/wiki/Zilog_Z800 The SAM Coup? was pure Z80 machine, a lovely British design, a much-enhanced ZX Spectrum 48, and it came out in 1989 and went on sale in 1990: https://en.wikipedia.org/wiki/SAM_Coup%C3%A9 The Acorn BBC Master wasn't all-new but neither was the C128. The Master was an elegant upgrade to the BBC Micro, and was released in 1986: https://en.wikipedia.org/wiki/BBC_Master I'm sure there are many more. So no, Herd is definitely wrong. The C128 was _not_ the last new 8-bit computer. It wasn't even the last new Commodore 8-bit computer -- the C65 was arguably that (and a more logical successor to the C64, IMHO). -- Liam Proven ? Profile: https://about.me/liamproven Email: lproven at cix.co.uk ? Google Mail/Talk/Plus: lproven at gmail.com Twitter/Facebook/Flickr: lproven ? Skype/LinkedIn/AIM/Yahoo: liamproven UK: +44 7939-087884 ? ?R/WhatsApp/Telegram/Signal: +420 702 829 053

On Fri, Dec 23, 2016 at 3:59 AM, Peter Corlett <abuse at cabal.org.uk> wrote: The same trick works perfectly well with a 6502, and in fact was invented by Don Lancaster using a 6502 years before the ZX80 was designed. That doesn't really explain the choice of the Z80. Also, by the time the ZX80 was introduced, both the Z80 and 6502 were basically dirt cheap. Any premium price the Z80 had once commanded had long since evaporated.

On Sat, Dec 24, 2016 at 8:10 AM, Peter Corlett <abuse at cabal.org.uk> wrote: The details aren't identical, but on the 6502 it's abusing the fetch of both an opcode and an immediate operand. If there was nothing in it, then it's a mystery why the Z80 was selected. As much as I like the 6502, the Z80 is better in some ways, particularly if you want to do a lot of 16-bit add and subtract.

On 2016-12-23 12:16 PM, Liam Proven wrote: The Acornsoft games were very high quality (hard to distinguish from their arcade inspirations). But I was mostly interested in programming, so I loved our BBC Micro Model B to bits. A far superior machine to the Apple and Commodores. --Toby

On 2016-12-23 2:00 PM, Liam Proven wrote: The Acornsoft games included clones of Defender (Planetoid), Scramble, Pac-man, and many others. They were all written in assembler by apparent masters of the craft and were highly sought after. It's probably true that there were fewer independent games than the American platforms, but Acorn invested hugely in their ROM, cassette and disk software library (including a number of language ROMs). --Toby

On 23/12/2016 17:00, "Liam Proven" <lproven at gmail.com> wrote: Acornsoft Planetoid *IS* Defender, looks and plays the same as the Williams original. -- Adrian/Witchy Binary Dinosaurs creator/curator Www.binarydinosaurs.co.uk - the UK's biggest private home computer collection?

On 12/23/2016 10:16 AM, Liam Proven wrote: Back then about 200$ US?? Depends on when. The average Kim-1 here was 300$ Motorola 6800D1 was 289$ in 1977. The RCA cosmac (Quest kit) was still over 100$ at introduction (the board and by the parts your self would cost me 45$ quantity 2). NS*Horizon S100 with Z80, MDS, 2 floppies and 32K ram and DOS was about $2000 in 1978 and a year later about $1700 as the price of ram dropped. BY 1979 a Horizon with 32K and 2 drives was about $1500. The biggest thing was ram. In the 1975 time frame 4K was about 300$, by 1978 that would be 16K at maybe $300 and by 1980 64K was under $300 and going down fast. The other was mass storage where floppies at intro were 350-450 for the bare drive and a controller was a few hundred more. The first break was $599 for a complete NS* (Thats Northstar Computers inc) had the MDS drive and controller package with DOS and Extended basic and that was mid 1977. Two years later two drives in a box with power supply was about 199$. There were others pushing prices down too. it was the innovation, production volume curve at work. When the timex/sinclair with membrane keys got her eit was around 99$ and immensely unpopular the later chicklet keyboard version was better accepted. BY then people wanted printer and mass storage and that machine was 2-4 years behind the expectations curve. I will not argue that, nothing was cheap but I also was buying raw chips and the prices for one or two were not always cheap. Boards especially two sided were not cheap either. The problem ooutside US for US products was import VAT. I was told back then a 2000$ NS system was easily 3000$ over there. Also getting systems here were interesting. Oddly it was similar in Canada for the same reason VAT/Import Duty. They spawned a few cool systems there too. However much of the US hardware was known and found there or they did S100 with local designs. Then again a few Canadian companies like Matrox made a good presence here in the USA. As did Epson who I consider one of the first japanse. I'm fortunate to have two views, that of the hobby small systems market, and that of being inside DEC during the 80s. The big vendor (DEC, DG, Prime, IBM) machine world was a bit different. If I can associate a factor in that its DEC internal had a world wide computer network such that I could talk to Palo Alto via Email just as easy as Valbonne or Galway and as fast even in the early 80s. That didn't happen till the 90s for those not inside a company network. The internet was a killer ap! Allison

On 23 December 2016 at 19:56, Liam Proven <lproven at gmail.com> wrote: Sorry -- wrong currency sign. $1390. -- Liam Proven ? Profile: https://about.me/liamproven Email: lproven at cix.co.uk ? Google Mail/Talk/Plus: lproven at gmail.com Twitter/Facebook/Flickr: lproven ? Skype/LinkedIn/AIM/Yahoo: liamproven UK: +44 7939-087884 ? ?R/WhatsApp/Telegram/Signal: +420 702 829 053

On 23 December 2016 at 19:34, Michael Holley <swtpc6800 at comcast.net> wrote: Very interesting -- thanks for that! I'm intrigued that although you prefix notate dollars ($300), you postfix-notate pounds (125?), when Brits prefix too. Any particular reason? I suspect that that was before the home computer gaming boom, and 2-3y later things would look rather different. As a student (1985-1988) I attended launches of the CBM Amiga and Apple ][GS -- I don't recall exactly when -- but they were vastly out of my price range. In '85 I was still using a ZX Spectrum with ZX Microdrives; by '87 or '88 I'd managed to get a floppy interface and a single 5?" drive (as the media were much cheaper than the new 3?" diskettes, which were 5?-10? the price). So even after 16-bit machines started to appear, even fairly serious hobbyists such as myself continued to use 8-bits & other old-fashioned tech, as they were massively cheaper. -- Liam Proven ? Profile: https://about.me/liamproven Email: lproven at cix.co.uk ? Google Mail/Talk/Plus: lproven at gmail.com Twitter/Facebook/Flickr: lproven ? Skype/LinkedIn/AIM/Yahoo: liamproven UK: +44 7939-087884 ? ?R/WhatsApp/Telegram/Signal: +420 702 829 053