Joined: Feb 2004 Gender: Male Posts: 962 Location: Orrville, OH
Re: TO: Earl, what is CP/M's attraction? « Reply #1 on May 29, 2007, 4:53pm »
I'm not Earl, but I can answer a little here because I also appreciate CP/M. For me CP/M provided DOS-like power when DOS really didn't exist. I recall using it on some early hardware for business applications, allowing me to do some things that I really couldn't do professionally on smaller platforms.
Joined: Feb 2006 Gender: Male Posts: 217 Location: Nebraska
Re: TO: Earl, what is CP/M's attraction? « Reply #2 on Jun 2, 2007, 10:47am »
thanks Jeff,
I searched the pod cast and found earls answer in Show 06 where he talks about the C128 doing CP/M. http://retrobits.libsyn.com/index.php?post_id=15457 Earl: is a "language nut". .....he said.. and use the CP/M for getting to different "BASICs" written in CP/M among other things.
I like to figure out what i could use it for and would it be worth the time.
Joined: Dec 2004 Gender: Male Posts: 1,565 Location: Kennedale, TX
Re: TO: Earl, what is CP/M's attraction? « Reply #3 on Jun 7, 2007, 4:11pm »
This conversation has come up a lot on various discussion groups I'm on. I think the general concensus is that it is useless for C128 users and that only about 1% or less of C128 users ever used the CP/M mode, hence it was probably a mistake by Commodore to include the mode on the machine in the first place. (Just think how much cheaper it would have been without the Z80 and other neccessary architechture.)
As far as what you can do with CP/M on machines that run only CP/M. Good question. The best I can tell about the only thing they were ever used for was word processing, spreadsheets, and occasionally telecom and some text games.
Joined: Nov 2005 Gender: Male Posts: 290 Location: Portland, OR
Re: TO: Earl, what is CP/M's attraction? « Reply #4 on Jun 7, 2007, 7:09pm »
Although the C128 has a very nice CP/M mode (made even better by subsequent speed-up releases), it probably was not very relevant. Interesting, just not relevant. I remember when a friend of mine bought one back in the day, and he talked a lot about it being able to run CP/M. Even being a Commodore fan, I thought, hmm, kind of retro Why would you want to do that?
The CP/M mode didn't run particularly fast because of some design decisions. And by the time the C128 came out, DOS was beginning to be interesting to people, and CP/M was well past it's prime.
It had (and has, if you're interested in the retro aspect) a bunch going for it - it could read many foreign disk formats using the 1571 (a quite versatile drive), giving it access to a fairly large, if aging, software base. CP/M 3.0 (CP/M+) is the cream of the crop of the Z80 CP/M offerings (though it also had more bloat than 2.2). And like I said on the podcast, it is a well designed and solid OS.
Still - even working for Epson in 1983, it was interesting to have a CP/M based desktop when the rest of the world was revving up to think about DOS. The PX-8 laptop somewhat made sense - because the software was often embedded (ROM chips), the OS didn't matter much. CP/M compatibility could only help, not hurt. A desktop - the QX-10 - that was a hard sell, and progressively more so as the years moved on. To illustrate this, Epson's follow-on, the QX-16, was a DOS and CP/M hybrid, like the DEC Rainbow. Later Epson models would abandon CP/M entirely, probably belatedly.
As for whether the C128 should have had CP/M? I don't honestly know how much it cost to put it in. If I meet up with Bil Herd sometime (which I really hope to do), I'd ask how much cost/effort the Z80 and CP/M mode required.
As long as I'm on that train of thought, I suppose I'd ask the question, why the C128 at all? This was 1985, the 16/32-bit systems were emerging. The world was changing - not much software was released for the C128 native mode. Why not just keep selling 64C models until the switch to the brave new world? I don't know what that alternate timeline looks like. I just know I'm glad they made the C128, because it's a smokin' 8-bit system...
Joined: Feb 2006 Gender: Male Posts: 135 Location: Nova Scotia
Re: TO: Earl, what is CP/M's attraction? « Reply #5 on Jun 7, 2007, 8:00pm »
Quote:
As far as what you can do with CP/M on machines that run only CP/M. Good question. The best I can tell about the only thing they were ever used for was word processing, spreadsheets, and occasionally telecom and some text games.
CP/M was an operating system, pure and simple. It could control devices like disk drives. That was a big deal for microcomputers in the late 70's early 80's. Micros like the C64 didn't have an OS, they simply booted into BASIC. So for micro manufacturers (and there were a bucket load of them in the early, early days) looking to compete in the small to medium sized business market, CP/M provided them with an operating system to match their machines with disk drives (and later hard drives) rather than inventing a DOS from scratch. Of course some manufacturers did go that route (TRS-DOS, Apple DOS, etc.) Unlike the DOSes that followed it, CP/M didn't have much in the way of anciliary programs. Most of those were left up to third party programmers.
During the early years CP/M was quite successful. And thousands (yes thousands) of programs were written and sold for it, more than word processors, spread sheets, etc. Much more. But I think because most CP/M machines were marketed to businesses these types of programs are what are remembered. Who hasn't heard of WordStar? That program started on CP/M.
I don't think CP/M sparks the same nostalgia though as other OSes. Maybe because it was so simple it was almost transparent. I have several DOSes and CP/M for my Model 4. But the later versions of TRS-DOS (actually made by the makers of LSDOS and not Tandy) are much more enjoyable for me to use than CP/M. I think this is because, well I'm a computer geek (aren't we all?) and I am interested in the OS as well as the applications and hence like a featurefull OS. Whereas CP/M does the job as OS is supposed to do and nothing more.
Should it be relegated to annals of history as unworthy of a retro look? Definitely not. But has it? Probably.
Holy crap! Did I just write an essay? Sorry 'bout that!
Joined: Nov 2005 Gender: Male Posts: 829 Location: Hampshire, England
Re: TO: Earl, what is CP/M's attraction? « Reply #6 on Jun 8, 2007, 4:37am »
Cpm 2.2 and Cpm + were bundled with the Amstrad 6128. I always wondered why at the time. The occasional game would make use of CP/M and i used it mainly for file management with my disks.
i think it was also used to help with bank switching between the two blocks of 64k ram you had.
Joined: Jan 2006 Gender: Male Posts: 55 Location: Melbourne, Australia
Re: TO: Earl, what is CP/M's attraction? « Reply #7 on Jun 8, 2007, 11:54am »
Quote:
As for whether the C128 should have had CP/M? I don't honestly know how much it cost to put it in. If I meet up with Bil Herd sometime (which I really hope to do), I'd ask how much cost/effort the Z80 and CP/M mode required.
I can't remember if it was in the book On The Edge or one of the "Commodore128 Animals" videos that Dave Haynie and Bil Herd did, but somewhere he explained that the Z80 was hacked in to get around some timing issues which caused problems with the C64's Magic Voice cart which did something nasty on boot up. It was just a quick hack that got turned into a sales feature.
Joined: Jul 2006 Posts: 87 Location: Mississauga, ON Canada
Re: TO: Earl, what is CP/M's attraction? « Reply #8 on Jun 8, 2007, 1:03pm »
Quote:
CP/M was an operating system, pure and simple. It could control devices like disk drives. That was a big deal for microcomputers in the late 70's early 80's. Micros like the C64 didn't have an OS, they simply booted into BASIC.
Granted, the 64 booted into BASIC, but is that not a form of an operating system? Can you access disks from the command line? Yes. Is there a kernal that presents a standard API for accessing things like disks and other devices (modems, printers, etc.)? Yes.
I might be just being picky, but that sounds like an OS to me. BASIC prompt, bash shell prompt, command.com prompt - all different interfaces to basically the same idea.
Joined: Dec 2004 Gender: Male Posts: 1,565 Location: Kennedale, TX
Re: TO: Earl, what is CP/M's attraction? « Reply #9 on Jun 8, 2007, 2:19pm »
Quote:
Granted, the 64 booted into BASIC, but is that not a form of an operating system? Can you access disks from the command line? Yes. Is there a kernal that presents a standard API for accessing things like disks and other devices (modems, printers, etc.)? Yes
I agree. The C64's "operating system" was just as legitimate to me as DOS or CP/M.
As far as "why the 128" somebody asked.. This question has merit in that the 128D was costing almost the same as an Amiga. Nobody will question that the Amiga was the superior machine. However, had the 128 been updated in a more logical manner such as I've mentioned before.. For example, adding a hi-res mode to the existing VIC-II chip and adding RGB output to it, plus the extra RAM... that would have made a system that cost more than the C64, but not terribly more. So an Amiga would have still been the next logical step up from the C128... but they put in all these extra features into the C128 like the VDC chip, plus its own RAM, plus the Z80, etc.. which were probably unneccessary.
At the time, I believe the C128's main attraction for me was the fact that it has a readable (and fast - compared to software rendering) 80-column screen that I could use to call ANSI BBS systems.
Re: TO: Earl, what is CP/M's attraction? « Reply #10 on Jul 26, 2007, 8:57pm »
Quote:
As for whether the C128 should have had CP/M? I don't honestly know how much it cost to put it in. If I meet up with Bil Herd sometime (which I really hope to do), I'd ask how much cost/effort the Z80 and CP/M mode required.
- Earl
I think I can lay some of background about how things like the Z80/CPM got into the C128 and describe the cauldron that was engineering at CBM in the early 80’s.
First a comment about the timeline, we started the C128 in late summer and it was due in CES on January 5th or 6th. A good part of the design was like lobbing real high angle artillery shells that would land three months later, we had several custom IC’s including a complete from scratch IC (the MMU) that had to be completely specified and into chip layout and design basically within a month. This meant that the core design and specs had to all be done basically within that same month and that we would live with any mistake or oversights possibly for the life of the product.
That’s not to say I couldn’t grab the chip designer and add features for when the chip came out, such as a Z80 clock coming out of the VIC chip, we just wouldn’t know if it worked in time to change the chips much certainly not the architecture, in time to be meaningful.
So with regard to the Z80, this wasn’t a single problem/decision or a simple thought to increase features at a cost. One thing that was in my head was the power requirements of the system vs. the C64. The C128 was the first with a switcher supply and so had better thermal characteristics and better life from the fact that to pull more current from the analog supply of the 64 you just ran it warmer (and lost some load regulation). The half amp of requirement that the CPM cart represented probably was officially “too much” for the C64 supplies but in real life who would notice or correlate more power supply failures since those things failed at high rates anyways.
However I had a switcher and a fuse, I HAD to design for worse case load which included the CPM cart and the modem cart simultaneously (as an example).
Now when I started the C128 I stated that we should make it C64 compatible. That turned into I _could_ make it C64 compatible.
And then marketing announced it was 100% compatible. Now I never would have said that because I don’t think anyone knew what that meant. I remember when I heard that my product was now 100% compatible that I just grinned and said okay,
So I had dozens of problems at any one moment and I tended to try and resolve as many problems as I could per solution. Meanwhile a little research showed that the CPM cartridge wasn’t “compatible” with all rev’s of the C64 which they thought was attributable to the revision of Vic chip. (they were in part wrong if not totally wrong in this assumption). I found a memo just a month ago while getting ready for the VCF where production was complaining about 20,000 or 40,000 failed CPM cartridges sitting in production. I should scan that and post it some time.
So now one night during the mad rush of initial design, I had a cartridge that drew too much power, unless I put additional money into the power supply which means I would have increased the base cost by up to a couple of bucks (crossed a magic threshold for the time) that would literally not be used on hundreds of thousands of system (multiply waste times 1 million), I had a CPM cart that may not work and cloud the definition of compatibility (I didn’t think I could sell the argument that it was SO compatible it failed on the C128 also) and then there were design issues that may not be over-comable at all which is the real reason the cart didn’t work.
So I tried on the statement “I could build the Z80 into the board”. Nothing happened. The ground didn’t shake the skies didn’t split open so I tried the statement again this time louder and more serious. It seemed to fit enough that it was worth spending the night understanding the consequences. Nobody in management really knew what I was doing nor were they usually nimble enough to stop any of the real design work, by morning (all good things got done in overnighters) I had researched cost, availability and design (we were doing work with Zilog for Z8000’s so I wasn’t afraid of the new vendor) and the power consumption was a fraction of the cartridges total power when designed onto the board. Also timing got better, much better.
The PCB’s were cut with Z80’s on them before management had a chance to say anything.
Management did do a smart thing when told about it, they got Von Ertwine involved full time to port the base system. My original vision was you just ran the original CPM but there was more to be had by integrating to new hardware.
So during the course of the project the Z80 saved my tail when a problem cropped up with one of our own cartridges. WHat would happen was Magic Voice cartridge would dynamically pull a line that up until then had been static, and then just grab the reset vector (“wait… we are in C128 mode, there is an MMU and some other things that have to be initialized…. <crash>) The cool thing is that Magic Voice was done by the guys that did the TI Speak and Spell and it was a real honor to meet and work with them, I didn’t mind what they had done, just no one expected it. Everyone went home thinking that once again Herd had failed to live up to his promise of 100% compatibility (whenever anyone attributed the bold claim to me I would just grin and say okay). The next morning (did I mention all good work gets done during overnighters) the Magic Voice cartridge was purring merrily along. I had flipped the polarity of the processor reset line which means the Z80 woke up first and it reset to $0000 and would initialize the MMU before jumping into C64 mode and the reset vector simultaneously. Later we would find a cart that was hard to detect as needing C64 mode so we built the safety switch of holding down the Commodore key to force C64 mode. (we found that poetic at 3:00 am)
One of the real stories behind the story was that it turns out we had pretty much lifted the CPM cartridge design from a Microsoft Z80 card for the Apple II. I found the schematic in the guy who “designed” the cartridge’s stuff a year later. This would explain why it wasn’t really designed right for the C64 in the first place, it wasn’t designed in detail, it was adapted by someone not around to live with the consequence.
Now as to the likeability of CPM for an as of yet produced machine. I simply couldn't predict which of the features we were building in would be the useful or popular. I couldn’t solve for the equation of “what would they say a year from now”, but I knew that if a feature was in they might use it and if it wasn’t they couldn’t. I used to describe the C128 as a 5 pound bag with 9 pounds of sh*t in it, I couldn’t quite get 10 pounds in. I knew that for almost no per unit cost, (the power supply cost savings paid for the $1 Z80) and in one motion we DID have access to true business 80 column software for the time. Now I could tap a base of already written software for the 80 column app when there was nothing for native C128 mode… if nothing else it would allow use of 80 column mode for the first months of the product until other apps could be written, for almost no cost. In fact the CES booth’s demonstration of 80 column mode was pretty-much CPM-centric which is why the 8563 problems as related to Z80’s stuff was so important. Von came through making a last minute change to CPM using a disk editor and hand calculating bytes so that checksums still came out as he hadn’t lugged his dev system to CES.
So the goal for CPM was just to be in there, to be an option, to demonstrate more capability while offering a software base of existing applications meanwhile keeping us closer to our goal of compatibility. For zero dollar end unit cost it was an easy call that night.
There was one production complication involving the Z80, if you open a early/mid C128 you will see a thick wire jumper on the top of the board. There was a reflection on A10 or something like that, but only on the trace stub that involved the Z80, there was no reflection on the 6502 stub. The jumper squashed the reflection by offering a second path, you can actually remove the jumper and there is full connectivity between the endpoint… except for a signal rising in 8 ns that originates from the Z80.
There is more to the story but this was long enough, sorry for the length.
Joined: Jan 2006 Gender: Male Posts: 55 Location: Melbourne, Australia
Re: TO: Earl, what is CP/M's attraction? « Reply #11 on Jul 26, 2007, 11:17pm »
Quote:
Quote:
As for whether the C128 should have had CP/M? I don't honestly know how much it cost to put it in. If I meet up with Bil Herd sometime (which I really hope to do), I'd ask how much cost/effort the Z80 and CP/M mode required.
- Earl
I think I can lay some of background about how things like the Z80/CPM got into the C128 and describe the cauldron that was engineering at CBM in the early 80’s.
...
There is more to the story but this was long enough, sorry for the length.
Bil Herd
Thanks for the story Bil! It is always interesting to hear tales from the trenches and I just can't get enough - don't apologize for the length. I don't think anyone here will be annoyed that you took the time to post this explanation for us.
Welcome to the forum and I hope we get to hear more tales from you.
Joined: Nov 2005 Gender: Male Posts: 829 Location: Hampshire, England
Re: TO: Earl, what is CP/M's attraction? « Reply #12 on Jul 27, 2007, 6:06am »
Yeah thanks Bil ! Great stuff, really appreciate you taking the time to post. Its great to hear stories from the people who were part of the legends and myth!
Joined: Jan 2005 Gender: Male Posts: 171 Location: South Carolina USA
Re: TO: Earl, what is CP/M's attraction? « Reply #13 on Jul 27, 2007, 6:48am »
Yes thank you very much for the "insider story". I know I personaly love things like this and I am sure everyone on here does as well. I mean the Dave Haynie video of the last days of commodore has a special place on my shelf.