AB084@DK0RRZK0.BITNET (09/07/88)
I'd like to report on the event mentioned in the subject line.
Please do not expect a profound analysis. I will give a collection
of personal impressions and observations.
1. Turbo C
It exists and they do already SELL it. There was a special offer for visitors
of the fair: The compiler at DM 150. Assembler and debugger exist but are
not yet released. We played around with the compiler for some time. Really
neat| Fast, easy to handle. Acceptable editor. Just real Turbo C.
We found a minor bug in the include library (a declaration missing in
<time.h>), so they have chances to further improve the product. A great
compiler im my opinion. We compiled dhrystone V1.1 (V1 would not give
any meaningful results for unknown reasons) and got 1706 dhr/sec.
I think I'll buy this package unless gcc proves to be comparable.
2. The Transputer (former ABAQ)
The box they showed there was obviously a prototype. No one would buy it
in that shape. I mean its look. The thing seemed to work, although as
it appeared, it isn't terribly stable. Dhrystones 3170 (V1.1) and 2270 (V1).
The shell is very csh-like, I found no obvious differences. The diff
doesn't do anything reasonable. We tried it on two files which happened
to be identical and it wrote kilobytes of differences to the screen.
Piping the same command through more made the system break down and weep.
It had to be booted a couple of times while we played with it.
Invoking cc with only a source file and no options as arguments wrote
assembler code to stdout. But one can do it in two steps, first call
cc -S source.s source.c
and then
asm -o text /lib/cstart.o source.s
Oh, and by the way, occasionally the source file had size zero after the
invocation of cc.
They had MicroEMACS 3.8 running. X-Windows do not work yet, that's why shell-
windows had a fixed size (a quarter of the screen, a little too small for
80 character display).
There seems to be no job control (according to the guy at the booth, he was
nice and helpful but I would hesitate to call him an expert).
cp'ing files did not work reliably, we found that the probability of
successful cp'ing was bigger after cd'ing to the target directory.
A FORTRAN compiler was said to exist, but it was not shown. Same for Pascal.
3. Beckemeyer products.
Gerd Sender presented Micro C-Shell, bdt's multitasking OS, and related
products. I Germany these are not so popular as in the States (as far as I can
tell from the net), and I saw them for the first time. I felt at home immedi-
ately, you know, I'm used to a UNIX environment. I had the impression, though,
that it isn't extremely fast, but that may have been the slow disk's fault.
They told me they could not sell us any campus licenses due to conditions
imposed by Beckemeyer Development Tools. (Is that true, David?)
4. Misc.
A Company named Pluennecke presented PROSPERO compilers. Knowing their FORTRAN
and Pascal, I would have liked to see their C. But the guy there was so un-
friendly and arrogant that I would not talk with him any longer.
We played with Laser C (833 dhr/sec V1.1), Lattice was also shown. I missed
MWC, don't know why it's so rare in Germany. I believe they are still selling
version 2.x here.
68030 box: Not shown, no comments.
5. Operating systems.
They showed an ST with TOS 1.4. Unfortunately I had no time to play with it.
Otherwise a total miss. I found people selling OS-9, but they would not show
it because they were demonstrating their network software.
I greatly missed Idris. I have never seen anybody in Germany to sell it.
I don't know why. This market is just as important for the ST as the States,
maybe even more. Maybe it'll be Minix then...
6. General impression
Nice coloured pictures, few hard facts. Games. A little disappointing.
When I talked about this to somebody at one of the booths, he said
to me: "What do you expect of a dying computer?"
7. Date
Sep. 2-4
That's it, friends
Yours
Michael
Michael Eibl, Inst. f. Theor. Phys., Universitaet zu Koeln
Cologne, West Germany
ab084@dk0rrzk0.BitNet
.t19@nikhefh.hep.nl (Geert J v Oldenborgh) (09/07/88)
> ... Prospero Fortran and Pascal ... I have heard nothing but horror stories from users of these (not the latest version, though). Many bugs and slow. Use Absoft for Fortran, don't use Pascal. > ... nice Unix feeling ... Try the GPshell in the Craft package. No multitasking, otherwise 99% cshell compatible and fast and small: you can still run LaTeX on a 1meg machine with foldr200. It also has nice VMS-like arrow commandline editing. Just some more info from G. J. van Oldenborgh
david@bdt.UUCP (David Beckemeyer) (09/13/88)
In article <8809061721.AA27884@ucbvax.Berkeley.EDU> <Micheal Eibl> writes: >6. General impression >Nice coloured pictures, few hard facts. Games. A little disappointing. >When I talked about this to somebody at one of the booths, he said >to me: "What do you expect of a dying computer?" >7. Date >Sep. 2-4 >That's it, friends >Yours >Michael > Michael Eibl, Inst. f. Theor. Phys., Universitaet zu Koeln > Cologne, West Germany >ab084@dk0rrzk0.BitNet I've been hearing this "dying computer" attitude from dealers, distributors, and even some magazine publishers (always the last to go) for several months, but I've been reluctant to say anything because I feared it would be construed as nothing more than an attempted stab at Atari. It isn't. It truly appears that Atari has played out their last hand. According to reports from dealers and distributors sales have fallen dramatically. Personally BDT sales of Atari products have dropped to 1/5 the size of 18 months ago (this is partially attributed to a major de-emphasis of the Atari product line by the company) and today Atari product sales represent less than 20% of BDT revenues. BDT revenues from other areas are now financing Atari customer support. We will continue to support our Atari customers. But how many other companies have alternate sources of revenue upon which to fund Atari support? I bet not very many. So what's going to happen? What about the end-user? I think the consumer who paid hard-earned money in hopes of obtaining a useful computer system is the big loser here, especially those that have invested lots of hardware, money, and time on the ST. I guess in all this rambling one thing I'm trying to say is that all of us invested in Atari. Now what are they going to do for us? A long slow death is the worst for everyone, including the stock-holders. It seems to me that this should be a lot more troubesome to all of us than whether or not they have fixed the Malloc bug. -- David Beckemeyer | Beckemeyer Development Tools | "Reckon the Ball's plumb open now, 478 Santa Clara Ave, Oakland, CA 94610 | and it's `swing partner'!" UUCP: {rutgers,sun}!hoptoad!bdt!david |
seitz@cory.Berkeley.EDU (Matthew Eric Seitz) (09/14/88)
In article <379@bdt.UUCP> david@bdt.UUCP (David Beckemeyer) writes: > >I've been hearing this "dying computer" attitude from dealers, distributors, >and even some magazine publishers (always the last to go) for several months, >but I've been reluctant to say anything because I feared it would be construed >as nothing more than an attempted stab at Atari. It isn't. It truly >appears that Atari has played out their last hand. According to reports >from dealers and distributors sales have fallen dramatically. > Originally, I flamed David for his negative articles. Now I believe he has a point. The computer store I work at during breaks has dropped ST hardware. I'm hearing a lot of the same negative feelings that Dave is. However, in the last month I've started hearing more good news. I'm seeing more new products and hearing more optimistic reports on Atari products and Atari's future. Some dealers are doing poorly, others are doing very well. Overall, I think Atari was dying. Now they've stablised, but they still have a long road to recovery. I don't think the ST will ever replace MS-DOS or the Mac, but it still may have a viable future. >-- >David Beckemeyer | >Beckemeyer Development Tools | "Reckon the Ball's plumb open now, >478 Santa Clara Ave, Oakland, CA 94610 | and it's `swing partner'!" >UUCP: {rutgers,sun}!hoptoad!bdt!david | Matt Seitz seitz@cory.berkeley.edu
trb@stag.UUCP ( Todd Burkey ) (09/14/88)
WARNING! My responses have only good things to say about the ST, so you Atari Haters (either the machine or the company) out there may wish to hit the 'n' key now... In article <379@bdt.UUCP> david@bdt.UUCP (David Beckemeyer) writes: >I've been hearing this "dying computer" attitude from dealers, distributors, >and even some magazine publishers (always the last to go) for several months, >but I've been reluctant to say anything because I feared it would be construed >as nothing more than an attempted stab at Atari. It isn't. It truly >appears that Atari has played out their last hand. And it could simply be that 1) May-September is a horrible selling time for computers and software anyway, 2) a lot of people are playing a wait-and-see game on things like the 68030 machine (or a drop in price in RAM which may drop the price of the Mega...), and 3) there is so much GOOD PD/Shareware software for the ST that it may really appear to be a dying computer from the point of view of a software developer trying to make a living off of the computer. >I think the consumer who paid hard-earned money in hopes of obtaining >a useful computer system is the big loser here, especially those that >have invested lots of hardware, money, and time on the ST. Fooey. I've wasted more money on IBM PC's and Mac's than I ever could on my ST (or Amiga for that matter). A lot of us, early on, put $3000 plus into an IBM PC that within years was worth less than $800. I also bought a Mac that became worthless at an even more alarming rate as Apple nickle and dimed me with their upgrades (128K->512K->??? actually cost more than the original Mac ($1900). The IBM PC lost its appeal with its 64K limitations and the increasingly buggy software that started coming out from companies like Borland, Microsoft, etc (it almost seemed to be becoming a fad as to who could rush out their product first and still get away with supplying the user with the most bugs). And the Mac was just too slow and messy to develop code on (and still be able to remember how to write code on Unix systems). From the home user point of view, I still think the ST is best, since it is cheap, has great games :-), has better quality PD software in most areas than the IBM PC or the MAC, can integrate better with 'other' disk formats (and OS's) that the users may be using at work, and is very easy for the novice user to 'learn' to program on (via fast but simple languages like GFA Basic.) Note I haven't said anything 'nasty' about the Amiga...I like the Amiga, but it was always just a little behind (about 6 months to a year) the ST in the areas I was interested in (languages, ports of PD packages, certain games, etc). Plus when I got rid of the Amiga, I still had to deal with having to have three OS's within reach depending upon which game or application I wanted to run...which I hear isn't a problem any more. >It seems to me that this should be a lot more troubesome to all of us >than whether or not they have fixed the Malloc bug. I agree, specially since I have never really seen how the 'Malloc' bug has really prevented me from doing anything on the ST as a developer. I have never lost any sleep over it or even seen the problem when using MWC. I actually have wasted more time in the other mundane bugs (like having to put character eater loops into my code to handle users that like to hold a scroll key down...else bconin bombed). -Todd Burkey "a member of STdNET - The ST developers' Network" trb@stag.UUCP -> to join, mail to ftg!dwm@stag.UUCP <-
dsmall@well.UUCP (David Small) (09/26/88)
(base note is Dave Beckemeyer's note about the Atari ST being a dying machine). Dave, Neil Harris would probably agree with you. He just quit Atari. You can bet that if there were a future there, he'd have stayed. Neil is headed for GEnie network management, I understand. I wish him luck. I personally feel (and this is just my personal opinion) that this is a good event to mark as "the end of Atari US". Atari has lost far more than just one employee-slave. Dave Small Gadgets By Small, Inc.
AEE05@DK0RRZK0.BITNET (10/26/88)
In article <0907881307> mcvax|np1|nikhefh|t19@uunet.uu.net <Geert J v Oldenborgh> writes: >> ... Prospero Fortran and Pascal ... > >I have heard nothing but horror stories from users of these >(not the latest version, though). Many bugs and slow. Use >Absoft for Fortran, don't use Pascal. > ^stuff deleted! I'm using ProPascal and ProFortran for approx. 2 years now. Due to my positive impression I feel obliged to contradict Geert in his opinion concerning these compilers. Many bugs ... Latest versions (i.e. 2.13) are free of fatal bugs, maybe they're still a bit rough around the edges. 2.11 had some bugs, yes. ... and slow Compiler front end is slow, linker is even slower. But the code is quite fast| I'm sceptical about compilation time enthusiasts. They all liked T*rbo Pascal 2.x and 3.x on the small I*Ms. They are totally happy about new concepts in Rev. 4.0 now, i.e. the modular compilation feature ("units"), for example. Well, Prospero compilers did it all the time ... Prospero are the only ones who offer not only 68881-libraries but even 68020/68881-libraries for all their compilers (i.e. Fortran, Pascal, C). You just have to relink your programs, no source modification at all. They offer a real language family. You may use Assembler routines in Fortran functions called by Pascal procedures re- ferred to in a main program written in C and vice versa. Saves lotta time for a team of programmers, really| Their interest in feedback from users is big, their hotline is as good as the rest of their service, as updates are free of charge, for example. Summarizing, ProFortran and ProPascal are no tools to write quick'n dirty hacks, ok. But at least for any other, bigger job they're the best I know - up to now... Disclaimer: These are the opinions of a satisfied customer - nothing more. ____________________________________________________________________________ Udo Keller Dept. of Theoretical Biology Inst. of Developmental Physiology BitNet: aee05@dk0rrzk0.bitnet University of Cologne ------------------------------------ SloNet: Gyrhofstr. 17 "While Eeyore frets... 5000 Cologne 41 and Piglet hesitates FRG and Owl pontificates ... Pooh just _is_." __________________________________________________________________________ ?