rwa@cs.AthabascaU.CA (Ross Alexander) (11/25/89)
(my previous stuff starts w/ '>>') gl8f@astsun.astro.Virginia.EDU (Greg Lindahl) writes: >> [i have to toss my software investment to upgrade to current >> levels of CPU horsepower] >Well, this is a problem everywhere. Talk to all the science types who >bought VAXes in 1980 only to have to convert to Unix because DEC wouldn't >sell them fast boxes. And then DEC starts making fast VMS boxes again. Agreed. Mind you, we bought a VAX11/780 in 1981 and had the wit to run Unix on it from the start ;-). So it's been pretty painless. We're up to a DS3100 and Sun 4/370, so far, plus about 30 other machines. The VMS guys get all weird whenever I talk about running a few benches on our VAX 8820 :-) :-). Don't want me to install Ultrix on it, I guess. >>opportunity to ignore me, and to slant their market offerings to >>buyers at the low end, without losing the sale, and with no loss to >Er, the TT isn't really much of a "low end machine" in any configuration. >You can't sell millions of home computers if they cost over $2,000. So you >have a curious definition of "low end". Oh, sorry. My minicomputer/workstation background is showing... low end is anything under $5k, and is meant in relation to machines capable of running un*x comfortably ( > 4 Mbytes ram, > 80 Mbytes disk; I've seen it done with less [much less] but that's not my idea of a pleasant working environment). I think of all machines sold in units of a million as low end. Nothing curious about that - it's a pretty common opinion around here. >Atari's marketing policy was good. I've only been playing devil's >advocate and pointing out obvious flaws, hoping that the flame war >will end. Gee, I hope you don't think I was _flaming_! (1/2 ;-). 99.5% of what I've said is from personal experience, painfully acquired. The truth is never a flame. >And, yes, I have a big investment: about 1/3 of my total material >possessions is my ST. However, most of the software I use (terminal >emulation, LaTeX, Maple, Fortran) is portable elsewhere. I've spent >some time understanding GEM programming, but that is useful when I >program on other windowing systems. I got what I paid for in the first >place. Now, let me see, you claim to have portable Maple and Fortran compilers that run on other machines and on Atari STs? Nice, but hard to believe. And terminal emulation code that runs on both ST and other platforms? Well, it wouldn't be _that_ hard... the LaTex claim _is_ credible. As far as understanding wondow systems from GEM, ok - but I was talking about all the time I spent finding the bugs and devising workarounds. That's time lost if I port. I've got lots of source code too; I'm a programmer by trade; I'm bitching about losing my application packages that I have as binary licensed stuff - all the desktop publishing, for instance (yes, I have TeX and LaTex too, just don't use them much). I can port my sources easily, it's having to get new compilers and libraries that annoys me. Ditto my database stuff - bringin the data forward will be a pain, and I'll need to buy a copy of dBase(d) or somesuch for $$$. >>I want to buy a real high-end cpu/video section. I don't see it yet; >>the tt/p comes close but lacks expansion. >Wait until you see one. Then you'll know enough to make this statement. OK, flame on... This comes under the heading of teaching your grandmother to suck eggs. Your condescention really bugs me. I've been in this business since 1973, I program for a living, and a 2nd year student who never met me informs me that I don't know enough to form a personal opinion and loftily advises me to wait for more information. &$@#~!!! Have _you_ seen one? I'm looking at a 1024 x 860 display right now (Dec VaxStation 2000 + VR260A); I have a 1132 x 980 x 8 plane display sitting beside my left elbow (Sun 4/60-FGX-8). These machines are _mine_, I work with them all day every day and I don't share them with anyone. Ghods almighty, I've read the specs for the tt/p. I've read more than a few specs in my time, and _built_ a few computers too I might add, and how many displays do I need to look at before I decide that 640 x 480 x 4 planes isn't quite what I want !?!? Any competent [23]86 clone can do 800 x 600 x 4 planes with off-the-shelf hardware, and I don't think of something you can get at the local Clones-R-Us as high end. Mind you, the mono is fine - 1280 x 960 is pretty good. Flames mostly off... I'm used to high performance systems and good support. I bought my 1040 because it had a decent cpu in it (in 85, a 68k was pretty good given the price), it had a flat addressing space, and I knew it would be easy to shove in more memory. That, at least, has worked out. Besides, I loathe int*l ;-). And the mono display wasn't too bad atall, _in 1985_. But the software support has been utterly dismal, and we're still at 1985 performance levels. Meanwhile the rest of the world can run their 1985 code at 198[89] performance levels, if they want to run their 1985 code; or they can run new stuff if they've got it. They _don't_ have to pitch all their software/data/training investment to upgrade the hardware platform. The tt/p lacks expansion, period. I've seen the spec, and I _define_ that spec as lacking sufficient expansion capability _for me_. A SCSI port plus one partly-implemented VME slot, indeed! Hoo boy... I think the tt/p will be a real nice machine _for some people_. I'm just not one of those people. And since I want a TT for my house, that makes me a home user. So I feel my earlier statement that the tt/p is not right for some home users stands, and I still feel left out in the cold. This all goes to show that if you have personal opinion someone else is going to interpret it as a criticism of _their_ opinion, and as a command to change theirs toute de suite. Ross
gl8f@bessel.acc.Virginia.EDU (Greg Lindahl) (11/27/89)
In article <1273@atha.AthabascaU.CA> rwa@cs.AthabascaU.CA (Ross Alexander) writes: >Now, let me see, you claim to have portable Maple and Fortran >compilers that run on other machines and on Atari STs? No, that's not what I said at all. I can buy Maple and FORTRAN for other machines; so when I want to move the source code I have in FORTRAN (which is worth much more than the compiler) doesn't have to be re-written. This is a pretty standard definition of portability. As for the rest of your post: sheer flames in email please. I think that maybe joe average comp.sys.atari.st user isn't so interested in how many years you've been using what computers. Just because the TT/P isn't right for you doesn't mean you should tell the whole world about it in comp.sys.atari.st... We've already heard from Mr. Convert repeatedly on that subject. Your mileage may vary. Don't buy if you don't like it. ------ Greg Lindahl Astrophysicists for Choice