[net.micro.cpm] Is CP/M dead

rbloom@apg-1.ARPA (Robert Bloom AMSTE-TEI 3775) (10/26/85)

                               Is CP/M-80 dead?
                                      or
                  Fun in buying TurboPascal for my NorthStar


     When Adam Osborne was asked if he saw any future in CP/M-80 at a recent
seminar I attended, his answer was short and to the point: "None.  Next
Question?"

     Of course everyone laughed at that and I even kinda aggreed with the
statement.  While I thought that CP/M-80 would never actually die, it probably
would become a secondary option to IBM in small machines.  Then I tried to get
a discount copy of TurboPascal for my NorthStar Horizon running N* CP/M.

     Those that have organic memories going back a couple months remember a
want notice I posted on "Wanna buy TurboPascal for CP/M-80."  To all that
replied, thank you; I was surprised at the response.  Of the thirty or so
replies:

     - about half recommended going directly to Borland and buying it at the
retail of $69,
     - the other half contained various suggestions, some good, some not so
good.  There were two recommendations for "Spite Software" in, I think,
Washington State and two for "The Programmers's Shop" in Maine.
     - I even got several offers for pirate copies.

     All-in-all it took me about four weeks to actually receive a copy.  I
think I tried just about *all* of the 800 numbers in most of the popular
computer magazines.  The results of this informal survey in the basis for the
title of this note:  4/5ths of those called only offered MS-DOS software - the
other 1/5 would order CP/M stuff for a two to four week delay.  Only one had
it in stock and only one copy at that.  The magic words were *IBM*compatible*.
Yes, I knew that most micros sold are Irish Barf Models, but several sales-
people even asked if CP/M-80 was *IBM* *compatible*!  sheesh.
 
     I ended up buying the in-stock copy from the Programmers Shop and got it
10-14 days later.  (The guys there knew what CP/M was - and even asked the
right questions about disk format without prompting.)  A good place to deal
with.

     Through a bit of stupidity on my part, I actually brought and paid for
TWO copies, the other from a local dealer.  When one gets desperate and mad
one makes bad decisions.  (Anyone need a [factory-sealed] of Turbo copy for
CP/M?  And no, Frank, this is not an ad!)  The sum result was that my effort
to get a fast cheaper copy by going through mail-order backfired by being even
slower and more expensive.  My recommendation: buy directly from Borland.

     All-in-all, Osborne's comment kept coming back to me - despite CP/M-plus,
Echelon, ZCPR3, SB180, Richard Conn (sorry Rich), the dedication of thousands
of hackers, and the archives of Simtel20, CP/M *is* dying.  Even Digital
Research is no longer supporting it!  (That might not be a big loss though.)

     Is there anything we can (or should) be doing about that?  I just bought
an new z80 s-100 (N*) box - is it obsolete even before I pay it off?  The cost
to convert it to the magic IBM-compatible cost more than a separate new Irish
Barf Model PC.  Are we all going to go the way of the TI99a people?  And if we
go, aren't the 6502 Apple hackers right behind?  Or in front?

     As the cheerleaders say: 'Lets hear some *NOISE*!'

-bob bloom  (the thoughts above are mine and may not reflect reality as seen
by anyone else but me.  I have no relationship with anyone except my spouse.)

ABN.ISCAMS@USC-ISID.ARPA (10/26/85)

Bob,

Enjoyed your trials narrative better than you probably enjoyed the trials!
Wish I had an answer to the fading popularity of CP/M .. I'm sitting here
with my good and faithful Toad (Morrow Decision I, S100) and the stolid and
mysterious Beast (CompuPro S100 multi-user with Gifford MP/M 816), yet all
my most recent developmental work is in the flashy Gulfstream..
NOT because it's a better programming environment (ugh), but only because
(1) that's where the markets are
(2) bloody people seem to want flash out there (e.g., popup windows,
    fancy screens, everything at their fingertips at the same time but
    "don't expect ME to read a manual", etc.

Sigh ...

Regards,
David Kirschbaum
Toad Hall
ABN.ISCAMS@USC-ISID

WANCHO@SIMTEL20.ARPA (Frank J. Wancho) (10/26/85)

Bob,

First, let's clarify one point: if you judge the vitality of an
operating system, or anything else for that matter, on the basis of
vendor support, then there are many popular, but dead products out
there.

In the case of CP/M and DRI, it was CP/M-Plus that was put out to
pasture under the heading of Level "D" support "for mature products
that are not actively supported".  CP/M 2.2 (CP/M-80) has been
assigned Level "C" support, which is limited to questions submitted on
CompuServe and phone calls.

Now, that is not as good as Level "B" ("active support", meaning
they'll eventually answer letters, too) nor Level "A".  But, it's not
dead, or in the "mature" catagory as far as DRI is concerned.

Of course, the counter argument is that CP/M-80 was never supported by
DRI in the first place.  Ever try to call in the early days and get an
answer?

Just because Adam Osborne says there's no future in CP/M-80 doesn't
mean that it is dead.  There's a difference.  The market for CP/M is
simply mature and mostly saturated.  If you're going to write
software, don't bother writing for CP/M - write for the hungry mass
market that's still trying to catch up with the rest of us using
tight, functional, and efficient programs.  There's a future and a
fortune there, not in CP/M.

There are about maybe 2 million CP/M users out there, many of them
perfectly content with what they have and don't need anything else.
See Jerry Pournelle's column this month for one, possibly prevelant
viewpoint, at least among us diehards, and note that Jerry has a
choice.

--Frank

ACB.COR@OFFICE-1.ARPA (Alan Bomberger) (10/28/85)

I join those who believe in CP/M.  Not that it is the greatest etc, but because
it is functional, simple, and predictable.  I have done some rather 
sophisitcated things for CP/M including POP UP windows and ThinkTank like 
things.  But people are killing it.  The users are killing it.  How?

Simple.  As a commercial CP/M developer and vendor I find that it is nearly 
impossible to find any publication read by the "2 million" CP/M users.  While 
there are many publications that CP/M users read most of them are dominated by 
"IBM" users.  Advertising rates are based on total circulation and thus 
advertising for the 3% of the CP/M readers is prohibitive.  This applies to 
distributors as well who must examine their advertising costs.  So since the 
publishers killed off the likes of Microsystems there is no central place to 
advertise new CP/M software.

I am sorry if that sounds like a very commercial reason and I would surely 
appreciate feedback.  I find it frustrating.  CP/M users should rally behind 
some publication and boost its readership to 500,000 CP/M readers.  That ought 
to create the atmosphere for developers to reach the "2 million" users.

jgilbert@WSMR08.ARPA (John Gilbert CD) (10/28/85)

I think Alan's message raises a matter of great concern to the users of CPM...
there simply is no magazine that deals principally with CPM matters and hence
can create a forum for information exchange concerning it.

I was startled to learn recently that KayPro's magazine Profiles, with 100,000
circulation is that largest magazine appealing mostly to CPM users and it is
undergoing an identity crisis since KayPro is no selling MS-DOS based
computers.

John Gilbert

Burton.osbunorth@XEROX.ARPA (10/28/85)

Dave,

I think you're pretty much right, at least for us unrepentant and
unreconstructed hackers.  But for my wife, DOS is better, because the
bigger memory space of a 16 bit system makes for "friendlier" programs.
And my wife is part of that great mass market that will buy a machine
with 3 initials on it because they feel good about those three initials
(beats me why, with Compupro's around, but that's another story).  

Even CP/M 86, had it caught on, would have been better than CP/M 80 for
this reason.  

If Gary Kildall made only one mistake, it was in putting work into an
8-bit operating system, and not the 16 -bit version.  Thus, he wasn't
ready when IBM *wanted* to use CP/M 86.

Sigh ...

Phil Burton

RCONN@SIMTEL20.ARPA (Rick Conn) (10/29/85)

Hello, Bob,

	"Is CP/M Dead?" is a matter of perspective, as I see it.  It also
depends on what you mean by "Dead."  I offer the following:

Meaning of "Is CP/M Dead?"			Answer to "Is CP/M Dead?"
--------------------------			-------------------------
1. Is anyone making commercial software 	No, CP/M is not dead; there
 packages for CP/M now?				 are many new commercial
						 packages out there (1)

2. Are the majority of the users of all		I don't know, but I think
 microcomputers running CP/M?			 50/50 chance that IBM has
						 not reached 2 Million users;
						 but how many old CP/M users
						 switched to IBM?

3. Are new users of microcomputers buying	Yes, the new users are largely
 CP/M systems (counting derivatives)?		 buying IBM, so CP/M is dying
 (this refers to personal computer users)	 if not already dead

4. Are total sales of new microcomputers	I don't know, but I am aware
 leaning toward CP/M (counting derivatives)?	 of many, many embedded micros
 (this refers to PCs and embedded computers)	 running CP/M et al

5. Is the flow of public domain CP/M software	No, as is evidenced by the
 stopping?					 latest SIG/M releases I saw
						 today, CP/M is still thriving

6. Where is the majority of the public domain	No, CP/M is not dead; looking
 software?					 at the libraries, CP/M is
						 first, IBM is second; but how
						 do the "good/junk" ratios
						 compare?
--------------
(1) Some examples of new commercial packages for CP/M (or Z System):
	What				Author			Sold by
	----				------			-------
Z-MSG (Electronic Mail)			Tim Gary		Echelon
ZAS (Z80/HD64180 Assembler)		Patrick O'Connell	Mitek
DSD (Dynamic Screen Debugger)		John Otken		Soft Advances
ZDM (Z80/HD64180 Debugger)		Robert Doolittle	Echelon
ITOZ/ZTOI Assembly Code Translators	Robert Doolittle	Echelon
ZRDOS (Z System BDOS Replacement)	Dennis Wright		Echelon
REVAS (Disassembler)			Al Hawley		REVASCO
DISCAT (Disk Catalog System)		Richard Conn		Echelon
TERM III (Communications System)	Richard Conn		Echelon
MIX C Compiler				MIX Software		...
MIX Editor				MIX Software		...
KAMAS (Outline Processor)		KAMASOFT, Inc		...
Write-Hand Man				Poor Person Software	...
REL/MAC (REL-to-source translator)	MicroSmith Computer Tech  ...
HiSoft C				HiSoft			...
HiSoft Pascal				HiSoft			...
DateStamper				Plu*Perfect Systems	...
ConIX Operating System			Computer Helper Industries, Inc ...
RED Editor				Edward K. Ream		...
Z80ASM (Assembler)			SLR Systems		...
MEX Communications Program		Ron Fowler		NightOwl

[I'm sure I left someone out - data is from the Echelon Newsletters and
 the latest issue of Dr Dobbs; also, "..." above means the same as the
 author; apologies to anyone I excluded]

------------

	So, the above sums up my response to your question.  If you look
at the "new user" community, the answer is what you expected - Yes, CP/M is
dead since the majority of new PC users are not buying it (my meaning 3 above).
There are lots of applications, especially embedded applications, where a Z80
running CP/M is a marvelous, simple, and cheap solution (try embedding an IBM
PC in a communications controller, if you see what I mean).  In the other
four categories, the issue is not at all clear.  CP/M (and the Z System)
is certainly alive as I see it.

	As for myself, I am running ZCPR3/ZRDOS and do not intend to switch
at home.  This combination meets my needs nicely.  At work, I use TIPCs
(IBM workalikes), and I have no complaints, but I don't do anything there
that I can't do at home.  I write code in assembly language and C at home,
and exclusively in C at work (except on the VAX, where I write exclusively
in Ada, and the VAX 11/785 makes a nice PC when working with Ada).

	My next hardware move is to upgrade to an S-100, 64180 board
(one is out now).  I plan to follow that with more software, particularly
a multi-tasking ZCPR4 with a large TPA.  After that, I will probably move
to a good, very fast UNIX system (68020, 32032 class).  Essentially, I plan
to skip the IBM PC market entirely ... if I need a powerful processor, I'll
go with one backed by a good operating system.

	For further discussions, see Jerry Pournelle's column in the
November "Byte" and the "Of Interest" column by Alex Ragen in the November
"Dr. Dobb's Journal."  Quoted from Alex' column:

	"Electronic Business, a magazine devoted to the dollars and
cents issues of the computer business, recently published a long article
analyzing the microprocessor market.  It carried the surprising message
that 8-bit processors make up by far the lion's share of that market,
with 16-bit processors trailing far behind.  Moreover, most industry
observers expect this situation to continue for many years.  The reason
is that the now inexpensive 8-bit processors and their peripherals are
perfectly adequate for most applications.  Where 16-bit processors hold
sway is in the prestige conscious world of the personal computer, where
only the latest fashion is marketable."

		Rick Conn

Disclaimer: While affiliated with Echelon (and thereby Hitachi [makers
of the 64180] and many others on the above list), this text expresses
my personal opinions and is not intended to be an advertisement or the
opinions of anyone else.
-------

tom@utcsri.UUCP (Tom Nadas) (11/01/85)

For advertisers wanting to reach a good hunk of the CP/M marketplace,
try FOGHORN the dynamite monthly newsletter of the FIRST OSBORNE 
GROUP (now expanded to serve all CP/M owners):  P.O. Box 3474/
Daly City, CA/94015-0474.  Advertising rates vary, but I think 
you'll find their CPM (the other one, advertising COST PER THOUSAND 
(M)) to be much more attractive than thhose of the glossy magazines.

RJS in Toronto
c/o
-- 

					Tom Nadas

UUCP:   {decvax,linus,ihnp4,uw-beaver,allegra,utzoo}!utcsri!tom
CSNET:  tom@toronto

lowans.wbst@XEROX.ARPA (11/08/85)

Alan
	Look into "User's Guide" "The Magazine For CP/M & MS-DOS computer
Users". Issue #16 is a good 50% full of CP/M (8bit) info. It is published by
TUG Inc six times a year. Subscription is $21 for 6 issues; $40
for 12. To subscribe send to:

		User's Guide
		PO Box 5245
		Redwood City, CA 94063

		It's looks like a good magazine for 8 bit CP/M users.
								Paul

RCONN@SIMTEL20.ARPA (Rick Conn) (11/08/85)

I saw a copy of User's Guide last night, and the main topic was that
"CP/M Lives" -- so much so that they are changing format, and
User's Guide will go back to a CP/M-only mag and a new mag, PC User,
will come out for PC-only users.
	Rick
-------