What was more popular MSX machines or CP/M in Japan?
- VCF SoCal - February 15 - 16 2025, Hotel Fera, Aneheim CA
- VCF East - April 4 - 6 2025, Infoage Museum, Wall NJ
- VCF Southeast - July 2025, Atlanta, GA
- VCF West - Aug 2025, Computer History Museum, Mountain View, CA
- VCF Midwest - Sept 2025, Schaumburg, IL
-
Please review our updated Terms and Rules here
How long did CP/M have a hold before MS-DOS became a thing
- Thread starter alank2
- Start date
Agent Orange
Veteran Member
There were lots and lots of money to be made in the clone market. Two distinct markets were emerging; i.e. small business and the home/hobbyist crew. It didn't long for the entrepreneurs to start moving the arcade offerings over to the Apple and DOS platforms. Unless there was a bolt from the sky, CP/M didn't stand a chance without color graphics. One could buy PC-DOS for about $60 and move it around as desired. There were lots of good accounting software running on DOS, like Peachtree and word processors like WordStar. WordStar was pretty much the boss in the early going, which had its roots in CP/M, but also had the ability to be ported across many different platforms and was fairly dominate on DOS. I still have my original boxed set. Soon, WordPerfect for DOS moved in and remained top dog until the Windows era became dominate with Word. None of this is really news, but sort of a reminder of the presence and power of IBM's ability to shape and move an industry.
krebizfan
Veteran Member
DRI did promote GSX for a number of years not that it ever got much traction. It solved cross platform graphical programing with multiple colors.
Amusingly, the 1993 issue of a German user group newsletter I have been reading had an article on porting a DOS graphical program to CP/M. Just a little too late to help push CP/M back into a dominant position.
Amusingly, the 1993 issue of a German user group newsletter I have been reading had an article on porting a DOS graphical program to CP/M. Just a little too late to help push CP/M back into a dominant position.
Eudimorphodon
Veteran Member
The QX-16s sold in the United States, at least, had an additional board installed that gave them IBM CGA compatibility, and while they did use 80 track drives instead of 40 track they could mostly handle reading PC disks. Here’s a link to the tech manual, which has descriptions of the PC compatibility hacks.
It still wasn’t a particularly good PC compatible, but they did *try*.
In 1983, Microsoft had already a bigger revenue than Digital Research. However, Microsoft continued to sell CP/M cards for Apple II, my guess is Microsoft/IBM sold more MS DOS / PC-DOS units than CP/M sold CP/M.
CP/M-86 inherited from CP/M-80 and had many flaws (almost no error-handling, a very limited file system, no graphical support, data wriitting mechanisms less efficient...). When MS-DOS/PC-DOS 1.0 arrived, it already outperforms CP/M in many areas. With MS-DOS 2.0, CP/M-86 was outperformed and not quite a bit.
Per my understanding of the events, CP/M was released in 1974 and Gary Kildall had brillant ideas (BIOS for instance to standardize the operating system calls). CP/M was already dominant in 1977 but instead of keeping the momentum, Digital Research was evolving as a family company, without clear guidances, marketing strategy, and business orientation.
The most tragic mistake was to don't listen to the rapidly evolving market. When SCP required CP/M an operating system capable of handling 8086 processors, Digital Research (mainly due to Dorothy Kildall governance) simply ignored/snubbed them, leading them to hire a very smart and visionary engineer, Tim Paterson (SCP had at that time a BASIC OS from Microsoft but it was very limited). Tim Paterson managed to create an operating system (QDOS) over 4 months, that was similar to CP/M-80 (reverse engineering, same interface calls to allow easy portability from CP/M). But Tim Paterson also saw the shortcomings of CP/M and made many improvements (error handling, memory management, data writing, filessystem...).
The rest is history. After losing the IBM deal, the only way that could have changed the trend would have been to dramatically change the concept of CP/M (especially for graphical support, and also on a decent filesystem). Gary Kildall failed to realize that consumers did not care about multitasking and efficiency. What consumers wanted were to use they preferred applications and to get fancy games just to relax.
When GSX came in 1984/1985, it was already too late. Had GSX been developped in 1981 with GEM, maybe (I say maybe) they would have been the natural choices for many PC Compatible manufacturers (IBM had an exclusive agreement with Microsoft till 1990 so IBM won't have put forward CP/M).
Last edited:
rlauzon
Experienced Member
Sadly, all too common with businesses run by tech, rather than business, people.
krebizfan
Veteran Member
IBM wasn't exclusive with MS. CP/M-86 was sold by IBM. FlexOS (Concurrent DOS) was a staple behind many of IBM's POS systems. Had Concurrent DOS 286 multitasked DOS applications in protected mode, IBM would have probably dropped MS right then.
Last edited:
Microsoft was already a software giant in 1984. Even if IBM decided to drop MS for another operating system, other PC vendors would have still chosen MS-DOS (even if IBM was still dominant).
as I said, if CP/M-86 was delivered sooner to fulfill SCP request (early 1980) and better designed for 16-bit computers with hard drives (well-designed filesystem with timestamp and directory support and graphical support with GSX or just allowing applications to directly target hardware especially on PC & Compatibles), CP/M-86 would have still won even with the IBM-MS agreement.
Bad decisions (engineering, business, marketing...) from DRI led to this scenario. I remember Gary Kildall doing a demo of multitasking CP/M in early 1984, with command lines and calculations in the background using different terminals. No graphical demo, no business user application demo. An attendee from HP simply talked about how operating system should not be prevalent and just let applications run smoothly with the hardware provided. In early 1984, MS-DOS fully supported directories, large hard disks, timestamping and abstracted layers to allow hardware manufacturers to build dedicated drivers. CP/M-86 was nowhere near that (only basic timestamp support).
Also in 1984 there were already a few dozens of CGA games available, CP/M had only text adventure like games. Of course CGA games where nowhere near the quality of C64 games (or Amstrad CPC games in Europe) but at least, it was still a compelling feature to let kids play when daddy was not here.
DRI just lacked of a Tim Paterson to simply modernize completely CP/M-80 and to stand up against Gary Kildall who simply wanted an OS for developers and computer scientists.
cj7hawk
Veteran Member
I don't think DR made any mistakes. Not even with IBM. Nor did Microsoft gain market dominance for another decade. Back in the 80s, a lot of people thought Apple would show the way forward, but the existing base of PC users who often had technical applications were still predominant until the end of the 80s and continued with DOS.
IIRC, Gary Kildall was very, very wealthy despite evidential alcholism, and even had his own lear jet - A major achievement in the 80s / early 90s. He never lived long enough to see his empire's demise, and despite Windows taking off, the use of DOS and similar systems was still massive and the future of PCs undecided by 1995.
That DOS survives until today is remarkable, and mainly it's only DOS data structures that remain.
PCs didn't become mainstream until well into this century, and didn't become a common part of society until around 10 years ago. The death of the "Computer Shop" that used to assemble pieces signalled the end of that phase ( yes, they still exist, but very few people buy a PC from a computer shop now ).
It's still too soon for our era, but one day, names like Kildall, Gates, Allen, Wozniak, Sinclair, Tramiel, Bushnell etc, will be revered as those who ushered in a new generation much as people like Newton. I wonder which name will be remembered the most?
IIRC, Gary Kildall was very, very wealthy despite evidential alcholism, and even had his own lear jet - A major achievement in the 80s / early 90s. He never lived long enough to see his empire's demise, and despite Windows taking off, the use of DOS and similar systems was still massive and the future of PCs undecided by 1995.
That DOS survives until today is remarkable, and mainly it's only DOS data structures that remain.
PCs didn't become mainstream until well into this century, and didn't become a common part of society until around 10 years ago. The death of the "Computer Shop" that used to assemble pieces signalled the end of that phase ( yes, they still exist, but very few people buy a PC from a computer shop now ).
It's still too soon for our era, but one day, names like Kildall, Gates, Allen, Wozniak, Sinclair, Tramiel, Bushnell etc, will be revered as those who ushered in a new generation much as people like Newton. I wonder which name will be remembered the most?
pbirkel@gmail.com
Veteran Member
The Businessmen Bills, Jobs and Gates. Because they were customer-facing and therefore have the most name recognition by the most (people) memories. No one on your list revolutionized our understanding of the universe like Newton. The same can be said for anyone in the discipline of "computer science", although arguments will continue over the parentage of "digital computers".
If we're looking for Newton-comparable understanding then we need to look at the genesis of "computation" and "information theory" rather than commercial developments. So Alan Turing, Claude Shannon, and a few others. Newton may have mass-market name recognition but I expect that very few could state his three laws of motion.
cj7hawk
Veteran Member
Turing is to computing what Pythagorus is to Newtonian physics.
Federico Faggin comes to mind as the inventor of the microprocessor decades ahead of it's time.
Hmm, this could make good card game. Print cards with the names of computing pioneers and group them according to whatever criteria matches all the cards. Smallest grouping wins. Skill at arguing is a necessity. Like a geeky version of poker, that no one else could or would understand.
Eudimorphodon
Veteran Member
Uhm... what?
I'm sure it depends on where you're living and all that, but even little old grandmas were stuck learning how to use PCs by the middle of the 1990's in the United States (to use this newfangled "email" thing). Ten years ago was actually past the point where PCs were getting displaced *out* of home/recreational use by appliances like smartphones.
Not really? Sure, people were still using plenty of DOS programs in 1995, but they were inevitably running them on PCs with Windows of some form preinstalled, and it was game over when Windows 95 showed up. If you worked in a law office or something in 1996 you might still be plugging away at WordPerfect 5.1 in a DOS full-screen session because you have literally thousands of documents in that format (and that was still the standard if you logged into Pacer with your modem and downloaded something from the court), but you had Windows 95 lurking in the background, and it was only a matter of time before you found yourself also having to deal with Microsoft Word docs sneaking onto your docket.
I mean, believe me, I'm saying this grudgingly. There was a lot of hype and optimism in the early-mid-90's about these amazing "new" technologies like RISC CPUs that were going to spawn a completely new generation of PCs (with better operating systems to match) and how Microsoft and Intel had no hope of stemming this incoming tital wave of competition, but... no, it didn't happen. And ironically we can probably drop a hunk of the blame for this on MS-DOS; at this point they had held an absolute stranglehold on the PC market for nearly a decade and even though, yes, the requirements for a new PC had changed from a machine with a couple floppy disks and a CGA card to hundred-plus-mhz CPUs and megapixel color displays between 1985 and 1995 what hadn't changed was the customer expectation that when they walked home from the store with this new computer they'd be able to grab the installation disks for whatever old software they had that they wanted to keep, install it, and keep on running it for as long as they wanted to. Every company that actually put an "all new" machine out in the 1990's failed hard and fast; not that there were many that even really went on sale, most of these ideas ended up stuck in development hell (See: Taligent) and only eventually emerged as bits and pieces of tech welded onto other existing operating systems.
(Apple was the only company that managed to survive switching CPU platforms, and they were only able to pull it off because they A: had such tight control of their platform, B: had (luckily) designed their base OS to be so hardware agnostic, and C: the CPU they switched to was so much more powerful than the CPUs Motorola had essentially abandoned development of around 1990 that customers didn't perceive their new computer as being slower than their old one when running old software, which was the case with any RISC computer that tried emulating DOS software. The hype around RISC CPUs was 100% dependant on the assumption that Intel wouldn't be able to apply enough in the way of go-fast hardware techniques to their CPUs to keep up (or even stay within an order of magnitude) with the performance of new designs; that assumption was already questionable when the Pentium came out, and was pretty much kaput by the end of the 1990's no matter what Apple bravely tried to pretend into the early 2000's with their rigged Photoshop benchmarks.)
Again, man, I dunno. Maybe if your memories of the 1990's are from a less developed/third world market where various 8-bit computer cast-offs managed to hold on longer because of local economic conditions your perception of just how dominant MS-DOS/PC Compatibilty was by the turn of the 1990's might be a little clouded. But at the global level the writing was thoroughly engraved on the wall well before 1995, whether you wanted that future or not.
And that proves the whole point; the "PC Compatible" became so embedded in the very DNA of the computer industry bits of it still linger to this day. I mean, sure, you could actually more broadly argue that applies to CP/M, since some of the concepts that modern OSes use (IE, the separation of roles between BIOS and OS level APIs, etc) were embodied in CP/M and MS-DOS started as a broadly CP/M-compatible workalike, but if you play that game then you're perfectly free to just keep notching that bar down to the DEC PDP-8 OSes that CP/M freely borrowed from, and from there to various research OSes dating back to the early 1950's... etc. The thing about MS-DOS' DNA is the majority of computers sold up to this very day can natively run an MS-DOS binary without CPU emulation. There has yet to be a clean break with a computer architecture first sold at retail in 1981.
CP/M was certainly an important and influential operating system of its time, but that time was between 1974 and the very early 1980's. The modern paradigm of a completely binary-compatible-across-hardware-vendors "universal" computing platform didn't happen en-mass until the PC Compatible, and it's kind of interesting that it was a thing that was arrived at almost accidentally. If history had gone just a little differently MS-DOS could have ended up like CP/M, a core OS that ran on a bunch of mutually incompatible (in terms of disk formats, graphics hardware, etc) business computers that might have all found their own little niches but never actually gave the home customer what they really wanted. In *that* world, sure, maybe we all would have migrated to something all new in 1995. But in our world, nope; the organically spawned PC compatible, with all its ugly warts and goofy idiosyncracies, was forced to keep on evolving.
krebizfan
Veteran Member
DRI was reported to have been working on a number of OS concepts before the success of PC-DOS forced a realignment to making CP/M somewhat compatible with PC-DOS including the adoption of FAT. It would have been a challenge for DRI to also introduce another new file system in the mid-80s. Unreleased software is always expected to have turned into absolute masterpieces if it wasn't for nefarious forces.
Agent Orange
Veteran Member
I retired from the feds in 2007. My agency always had the latest and greatest in hardware and software. However, at the time that I check out, the software travel program that you scheduled all your official government travel on was a form of dBase III which was compiled and run off of a secure server at HQ. No graphics whatsoever. And guess what? - - - it wasn't a kludge either, and if you ever used a typewriter, you could log on and do your travel request with just your unique password. Also, there were a host of other LEO programs of the same ilk.
cj7hawk
Veteran Member
In 1995, you didn't need a computer to interact with government. You didn't need to use a computer to obtain reasonable prices for products or services. You didn't need to have a computer to be in the majority.
For most, computers were something they used in the office, then went home and watched the television. There were no smart screens (PC in a TV) or Netflix. People didn't use social media as a majority - in fact, outside of younger people, few did.
Sure, computers were accessible to little old ladies in 1995. but I wouldn't say that you could expect to find most people in nursing homes using computers every day - that is most certainly the case now.
Smartphones are still a PC - just the most recent form factor of the PC.
So yes, I do believe that in 1995, that while PCs were ubiquitous, they were far from mainstream. Though it may be that we are just discussing difference in what "Mainstream" means.
Not in 1995. Let's have a look at the historical facts from the era.
1995.
1997:
Source:
You aren't wrong in 1997. But Gary Kildall died in 1995. DOS was still the biggest OS and Windows was uncertain. It was right on the edge of what you describe as happening, The two charts show a very sudden change - but in 1995, it was a promise of change yet to be fulfilled. Gary never saw a time when DOS wasn't dominant - After 1995, the future of OS direction became a lot more certain.
I am thinking that the data we accessed in 1982 on hard drives is still accessible under Windows 11, even if the disk interfaces have changed so radically. These interface standards didn't exist in CP/M, but I can't disagree on the path that technology development took. In many ways MSDOS and the PC hardware non-agnostic meant that DOS was a step backwards from CP/M - And I knew a lot of people who used GEM in the early 90s and believed it to be superior, but the lower cost technology usually prevails.
cj7hawk
Veteran Member
If it ain't broken...
DRI went the way of more advanced OS' in a world that just wanted to load up a program. I wish Windows 11 wasn't trying to be an operating system for more than one person. I liked Windows 95. 98 was pretty good too. XP was still my favorite.
But if we examine why DOS hung around for so long, it was the drivers... You could patch the living daylights out of it. And it all worked. Bolt on technology as an OS.
Patching CP/M? Well, while not impossible, it was a concept that evolved later on. It's not that you can't patch CP/M or include drivers - but they don't fit the TPA concept well. The TPA stretched from 0100 to the BIOS base. There's no room in that model for a driver system.
1944GPW
Veteran Member
It's odd and maybe even misleading that 1995 chart doesn't have Windows 2 or Windows 3.0 on it as separate bars, only NT. They were pretty big by that time. I can only say I was flat-out developing Win2 then Win3 applications in C in the early 90s for architectural space planning (Win2 interfacing with AutoCAD through AutoLISP) and real-time graphical chart plotting of options traded stocks for Win3 and OS/2 PM. I recall running Windows 1.0 on a PC/AT then and it was pretty much the fastest GUI I'd ever seen.
Eudimorphodon
Veteran Member
Uhm… you understand that the difference between 1995 and 1997 on those charts is one of them lumps Windows (specifically 3.x) into “DOS” and the other splits out 95 separately, right? I’m sorry, but you’re totally deluding yourself if you draw a line trying to distinguish computers running 3.x as something other than “Windows PCs” and that the future was somehow murky at this point.
Windows 3.1 had a market share of over 70% when measured by the number of PCs it came preinstalled on, pretty much unchanged for its entire market life. They sold three million copies the first month it was out *alone*, and this was on top of the over 10 million units they’d moved of Windows 3.0 since 1990. (And, also, Windows 3.x was *extremely* widely pirated. I think any relevant statute of limitations is long enough expired I can probably admit to have witnessed more than a bit of it going on.) And Windows 3.x was, in fact, still a viable desktop operating system up until the late 1990s (I used to get on the Internet with it, I used Windows for Workgroups to set up file sharing for older 486 computers in office environments, etc, well after 95 was the standard pre-install on new PCs.), so sure, in 1997 quite a few people were still plugging along waiting to get Windows 95 (or 98) with their next PC purchase. But that does not by any means demonstrate that the future was some kind of a mystery *well before* your 1995 date.
All your chart shows is that 95 definitively murdered what little momentum OS/2 had left in 1995 and peeled another 20% off of Apple’s flailing garbage fire of a product line.
I’m not even going to pretend that Microsoft didn’t achieve these numbers without cheating at least a little; as early as 1991 Microsoft started rewriting their DOS OEM licenses to encourage companies to bundle Windows on basically every PC they sold, putting products like OS/2 in the position of essentially competing with a free alternative. But it 100% worked.
Plasma
Veteran Member
- Joined
- Nov 7, 2005
- Messages
- 2,309
Computers were definitely mainstream by 1995, at least in the United States. Everyone I knew had one, even people living in trailers. Might have only been a 286, but they had a computer.
I'm also not sure why you are lumping in smart devices, social media, and Netflix with computers. Those all came much later.