buchholz@topaz.RUTGERS.EDU (Elliott Buchholz) (01/21/87)
I am currently in the process of buying a personal computer. I have
narrowed myself down to either a Mac or an Amiga, due to personal
preference, software availability, and flexibility. But I can't seem
to decide what would be my best decision. All my Mac friends scream,
"Buy Mac!!!" My Amiga friends retort, "Don't be silly! Amiga is
state-of-the-art."
So I ask, what do you people think is the better of the two, and
why? As far as price goes, the deals I have are:
Mac-plus: Est. $1550 for Mac (1 meg memory), Imagewriter II printer,
and second drive
Amiga: Est. $1650 for Amiga, Expansion board (512 k), color
monitor, and second drive (printer not in deal-- est.
another $450 or so)
I would appreciate some as-unbiased-as-possible opinions on what my
best deal is.
Please E-mail your responses to me, and I will tally up the responses
and post what the general majority opinion is.
Elliott Buchholz
Looking for a glitchy phrase
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
^ Bat-Locations... ^
^ ^
^ ARPA: buchholz@topaz.rutgers.edu ^
^ UUCP: ...!{harvard|seismo|pyramid}!rutgers!topaz!buchholz ^
^ PONY EXPRESS: RPO 4014 / CN 5063 / New Brunswick, NJ / 08903 ^
^ BAT-PHONE: (201)-247-6544 ^
^ ^
^ QUOTE: "Save the Texas Prairie Chicken" -- Michael Nesmith ^
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
Let the pigeons loose!!!brothers@topaz.RUTGERS.EDU (Laurence R. Brothers) (01/21/87)
To: buchholz@topaz.RUTGERS.EDU
"Don't be silly!"
Get an Atari ST! Like 1/2 the price of the amiga and essentially the
same features. OK, the amiga has a little better graphics, but then
the ST has a better OS..... Also, you can get higher memory models
of the ST cheaper still (1040 K ST is < $1000, including monitor, disc
drive)
Also, if you like Mac software, you can get a cartridge relatively
cheaply that plugs into the st case and presto -- color Mac.
Apparently, the ST actually runs many Mac program faster than the
Mac...
--
Laurence R. Brothers
brothers@topaz.rutgers.edu
{harvard,seismo,ut-sally,sri-iu,ihnp4!packard}!topaz!brothers
"I can't control my fingers -- I can't control my brain -- Oh nooooo!"ralph@mit-atrp.UUCP (Amiga-Man) (01/21/87)
In comparing the MAC vs. the Amiga, I just thought I'd mention a couple items which the Amiga offers which you should take into consideration: - A true multitasking OS. I find this feature very important. - Great graphics (fast!) AND NTSC compatibility. - You can mail order a reconditioned one for alot less than the price you mentioned. This is a temporary advantage. - The sidecar option (plugs into amiga and contains an IBM clone) gives you simple and effective total IBM compatibility (blek :-)). - Lots of Public domain stuff in SOURCE form. Great for budding programmers. Good luck on your choice !
daveh@cbmvax.cbm.UUCP (Dave Haynie) (01/21/87)
> > "Don't be silly!" "Don't be ignorant!" > Get an Atari ST! Like 1/2 the price of the amiga More like 3/4 or more of the price, depending on configurations. > and essentially the same features. Get a clue, pal... > OK, the amiga has a little better graphics ya, like 640x400x4 vs 640x400x1, etc. MUCH better, not a little better. > but then the ST has a better OS..... Technically, the Atari doesn't even HAVE an OS. A DOS, maybe, and a set of BIOS calls, certainly, but no OS. The Atari's BIOS is an attempted clone of MS-DOS, and causes lots of problems. Like hard disks that get unbareably slow with partitions greater than 5 Megs, and go ape on you when you get something over 40 subdirectories on them. The Amiga IS the only computer in this price range that has a real OS. If you don't know what a real OS is, I can suggest several books on the subject. Or ask a UNIX wiz why MS-DOS doesn't qualify as a real OS. > Also, you can get higher memory models of the ST cheaper still (1040 K ST > is < $1000, including monitor, disc drive) Of course, you can buy an Amiga and not need a "higher memory model". I've got one here with a megabyte, there's one in the next office with 512K, I've another around here with 4 megabytes on it. All the same model. If next year you decide that a megabyte isn't enough, you can add several more. And since the Amiga's OS multitasks, you can do real things with that memory, not just attempt to get by with kludges like "desk accessories" or "memory-resident utilities". The machine I'm typing on now has a CLI window open in the background doing a diskcopy, and as well I've got a screen blanking program, clock, print server, and a few other tasks running. Just try that on one of the other machines. > Laurence R. Brothers -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Dave Haynie {caip,ihnp4,allegra,seismo}!cbmvax!daveh "You can keep my things, they've come to take me home" -Peter Gabriel ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cmcmanis@sun.uucp (Chuck McManis) (01/21/87)
In article <8520@topaz.RUTGERS.EDU>, brothers@topaz.RUTGERS.EDU (Laurence R. Brothers) writes: > "Don't be silly!" > > Get an Atari ST! Like 1/2 the price of the amiga and essentially the > same features. OK, the amiga has a little better graphics, but then > the ST has a better OS..... Also, you can get higher memory models ^^^^^^ > of the ST cheaper still (1040 K ST is < $1000, including monitor, disc > drive) Great, and next your going to tell us it was all a big mistake and we really should go back to MS-DOS 1.1 for the "best" OS. Seriously the Atari OS was based on DRI's work toward an MS-DOS clone and it shows. One of the nicer things about the Amiga was that it's OS is based on TriPOS which at least was targeted for large systems and does not have artificial limits built in (like 40 "folders" or 36 Meg of disk space). It is also infinitely mor customizable than Atari's TOS. And you can't buy a 2 meg ST or even an 8 meg ST today but you can buy an Amiga with that much memory (and a 68881, and a 68020 if you want). -- --Chuck McManis uucp: {anywhere}!sun!cmcmanis BIX: cmcmanis ARPAnet: cmcmanis@sun.com These opinions are my own and no one elses, but you knew that didn't you.
woody@tybalt.caltech.edu (William Edward Woody) (01/22/87)
Uh, isn't this COMP.SYS.MAC, and not COMP.SYS.MAC&AMIGA&ATARI&Whatever else? I think the Amiga is cute, and the Atari is cute, and they both are great machines for certain tasks, just as the Mac is cute and great for whatever tasks it's good for. But can we leave the follow ups on if the Amiga is better than the Atari somewhere else? Thanks. - William Woody Mac! > ][n && /|\ woody@tybalt.caltech.edu woody@juliet.caltech.edu
djd@mit-eddie.MIT.EDU (Dave D'Souza) (01/22/87)
Come on, let's stop this ridiculous debate now. This subject has been beaten to death already. Please note that the original submitter of the question asked that resposes be e-mailed directly to him and NOT posted to the net. Also, a summary was promised and this will be more interesting than the individual letters. Please don't bother others with this cruft. I am quite sure many have had to replace worn out K keys after previous such debates. Here is is original author's address... In article <8515@topaz.RUTGERS.EDU> buchholz@topaz.RUTGERS.EDU (Elliott Buchholz) writes: > {stuff deleted} >I would appreciate some as-unbiased-as-possible opinions on what my >best deal is. > >Please E-mail your responses to me, and I will tally up the responses ============================ >and post what the general majority opinion is. >^ ARPA: buchholz@topaz.rutgers.edu ^ >^ UUCP: ...!{harvard|seismo|pyramid}!rutgers!topaz!buchholz ^
dillon@CORY.BERKELEY.EDU (Matt Dillon) (01/22/87)
Ha ha ha he he He ha Heh He. Have you taken a look at the Atari-St Newsgroup lately? Look closely, C-A, this is *exactly* why you should not put your OS into ROM. -Matt
woody@tybalt.caltech.edu (William Edward Woody) (01/22/87)
Apologies to those on comp.sys.amiga, comp.sys.m68k, and comp.sys.misc, for the previous flame I posted. I wanted it to only go to comp.sys.mac... I feel so stupid... - William Woody Mac! > ][n && /|\ woody@tybalt.caltech.edu woody@juliet.caltech.edu
ravi@mcnc.UUCP (Ravi Subrahmanyan) (01/22/87)
In article <1270@cbmvax.cbmvax.cbm.UUCP> daveh@cbmvax.cbm.UUCP (Dave Haynie) writes: >> >> "Don't be silly!" > >"Don't be ignorant!" I won't argue the price/os issues, that's subjective and it's your money if you get suckered into a bad deal.. however, some of these statements smack of ignorance on Dave's part (reminds me of the thing about fellows in grass houses not storing thrones..) >Like hard >disks that get unbareably slow with partitions greater than 5 Megs, >and go ape on you when you get something over 40 subdirectories on >them. Not true. I have 10 & 16 Meg partitions on my 30 Meg disk, and it books along quite well, thank you. This is really a moot point, however. I put together a 30 Meg hard disk for my Atari for $600. (You can buy a 20 Meg one for $650). *All* my friends with Amiga's are limping along with two floppies because they can't find real hard disks for comparable (affordable?) prices (even something that plugs into the printer port (for heavens sake!) is $850 or so around here) I get to see them eyeing my drive every now and then, and after that I don't even mind the 40 folder limit (for which Atari's worked out a patch-fix, by the way). > >Of course, you can buy an Amiga and not need a "higher memory model". The original poster didn't know about boards that can be bought for the ST that can be populated to go up to 4 Meg. > ... Amiga's multitasking OS ... don't need desk accs ... Dave's one ironclad point. The Amiga's multitasking *is* great. However, I look upon this as a difference in the machine's characters, not an issue for mudslinging. Does everyone want multitasking? I'm not sure.. (after all, it might slow my hard disk down :-) I imagine everyone is sick of this debate, so why doesn't someone direct followups to the place it all started (now that I've got my two bits in)? -ravi
fisher@elxsi.UUCP (Chuck Fisher) (01/22/87)
The following is excerpted from an article in the January, 1987 issue of Dr. Dobbs Journal of Software Tools. "Macintosh Buttons and Amiga Gadgets" was written by Jan L. Hamington who is an assistant professor in the Computer Science Department at Bentley College. She is the author of "Macintosh Assembly Language: An Introduction. "This article explores the details of the standard Macintosh and Amiga user interfaces, examines the system routines programmers use to create those interfaces, and discusses those system routines through a pair of sample 68000 assembly-language programs that support a portion of the standard user interfaces. "It might be unfair to base an evaluation of the system routines of the Macintosh and Amiga simply on the subset of the routines designed to manipulate the user interface. On the other hand, the programming strategies used to implement the standard user interfaces are similar to those required for other system operations. In general, the Macintosh's routines isolate programmers from low-level tasks such as list manipulation and initialization of data structures (File Manager parameter blocks are notable exceptions). Although this reduces the burden on the programmer, it can decrease the programmer's flexibility. The Macintosh routines are also more complete in terms of their support for the documented user interface. The effect is again to reduce the burden placed on the programmer. "On the Amiga, the Intuition library provides routines for the standard user interface. Although support for screen, windows, menus, and fonts is available, there is a great gap in terms of text editing. In other words, the Amiga does not provide system routines to fully implement its own standard interface recommendations. As someone who writes more programs that rely on text manipulation than on graphics, I believe that this is a serious deficiency. It is true that the Amiga performs some functions "automatically" for which a Macintosh program must include code (for example, moving and sizing windows). Nonetheless, the Macintosh does include system routines to handle those functions." What Jan seems to be saying is that the Amiga has an advantage to programming graphics and animation oriented applications over the Macintosh. But for text oriented applications the Macintosh provides better support in terms of system library routines. This sort of reaffirms the idea of the Amiga being a great game machine while the Macintosh is targeted more towards "business" applications. Looking at the software available for each machine also confirms this orientation. The bottom line is that in choosing between a Macintosh, Amiga, or Atari one should look to see whether the applications available meets one's expectations. Flame on: Although Amiga touts its multi-tasking capability, how many programs are written to take advantge of it? I would agree that "background printing" is desireable, but a Mac can do that as well as an Amiga. Most of the work that I do on the Mac are interactive in nature, and don't lend themselves that well to true multitasking. Switcher and Servant provide most of the capabilities needed to integrate foreground tasks in my opinion. Much as I hate to say it, but the success of a machine is not based solely on hardware, or the elegance of the user interface. Instead, what guarantees success is the name and backing of the manufacturer plus the willingness of third-party developers to produce software. When enough machines have been sold and enough software written, the machine then becomes a "success" and continues to generate momentum for some time. The Mac has achieved this level in the past 12 months, but I am afraid the Atari and Amiga have not.
sansom@trwrb.UUCP (Richard Sansom) (01/22/87)
In article <1270@cbmvax.cbmvax.cbm.UUCP> daveh@cbmvax.cbm.UUCP (Dave Haynie) writes: >> "Don't be silly!" >"Don't be ignorant!" >> Get an Atari ST! Like 1/2 the price of the amiga >More like 3/4 or more of the price, depending on configurations. ^^^ O.k., I know of several places that offer a 520 ST (monochrome, 512K, 1 drive) for ~$500. That sounds closer to 1/3 the price of the Amiga's $1500 (color monitor, 512K, 1 drive). And I've never heard of a single monochrome-ST owner trying to find a replacement for their original monitor. >> and essentially the same features. >Get a clue, pal... >> OK, the amiga has a little better graphics >ya, like 640x400x4 vs 640x400x1, etc. MUCH better, not a little better. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Purely subjective. The Amiga can display more colors then the ST - if you need more color, get an Amiga. However, I've yet to see a _single_ program on the Amiga which could not be done on the ST (I stress single because of the next bit...). >> but then the ST has a better OS..... >Technically, the Atari doesn't even HAVE an OS. A DOS, maybe, and a >set of BIOS calls, certainly, but no OS.... >...The Amiga IS the only computer in this price range that has a >real OS.... Agreed. The ST's OS leaves alot to be desired. Atari should spend more time writing a real OS for the ST than coming out with new computer lines. In my opinion, the Amiga's only real advantage over the ST is in it's ability to run muliple tasks. Even though there are ways to multitask on the ST (MTC-shell, OS/9, FORTH, etc), it would've been nice if Atari had taken care of this for us. Sorry to contribute to this silly discussion, but you knew someone was going to rebut that last article. How did this latest bout of Amiga v.s. ST get started anyway? -Rich -- __________ ______ ____ _____ ___ /_________//___ ||__|/____|/__/ Richard E. Sansom ___ ____/ / ____________ TRW Electronics & Defense Sector / / / /\ < | /| / One Space Park Drive, R3/1028 / / / / \ \ | / | / Redondo Beach, CA 90278 /__/ /__/ \__\|__/ |__/ ...{decvax,ucbvax,ihnp4}!trwrb!sansom
jafischer@watrose.UUCP (01/22/87)
In article <636@mit-amt.MEDIA.MIT.EDU> ralph@ATRP.MEDIA.MIT.EDU (Amiga-Man) writes: > ... [talks about advantages of Amiga] >- A true multitasking OS. I find this feature very important. I'm really curious: just how great is multitasking? I.e., how much do you, the typical Amiga owners, use this feature? Even on Unix I find that the only thing I consistently use '&' for is printing, and on my ST I have a wide range of print spoolers to choose from. And when I'm in the edit-compile-edit cycle, the speed of Megamax compile-time, coupled with a RAMdisk, makes multi- tasking a 'yeah it would be nice, but not utterly world-stopping' luxury. (There is a multitasking C-Shell, but it's basically non-GEM, and still pretty buggy). But then again... I _don't_ envy your high (until recently) prices, nor your apparently buggy OS (see message 1292 in comp.sys.amiga!), but... I _DO_ envy your multitasking! I'm just curious how useful it actually is. I.e., is it that necessary in a personal computer, and is it worth the slow-down in execution? Even though I would really _like_ it, would I really use it that much? Uh oh, this could be taken as a typical my-PC-is-better-than-yours posting. It isn't. The main intent of this posting is the question of multitasking. - Jonathan Fischer
jdm@gssc.UUCP (01/22/87)
*** tongue in cheek flame on *** ok, enough of this crap about which is better: the amiga, mac, or atari. they're all toys!! a buddy of mine has an amiga, so i went over to have a look-see and find out what all this psychobabble was all about. i found this puny litte box that was hardly as big as my vt100! he mumbled something about a 68 or other "micro-processor" being able to outstrip a vax "on some operations." like what? certainly not disk accesses!! these newfangled "boxes" don't even have *real* disk drives. i mean, what good is a little three and a half inch jobbie compared to my 12-platter rm-05's? i had brought some "real" software over to test this thing - you know, fire up a couple makes an troff jobs and see what happens - but it seems that he didn't have no nine-track for his little pee-cee. hrmph. so he showed me some graphics. who needs 'em? i don't program by drawin' perty pitchers. hell, i dunno about you, but my compiler uses TEXT, fer crissake, and when the compiler don't do what i want, i talk to the assembler, and if *that* don't fix it, i use od(1) and ipatch(1) to party on the bits as they lie there sitting pretty on the disk. graphics, ha!! real programmers don't *need* graphics. better put a tourniquete on it before y'all BLT to death! i always knew you boys had some strange, genetic DDA problems. anyhoo, if y'all own one o' them pee-cees, don't fret too much. you can still impress your ignorant friends. unless they been down to the video arcade and seed some *real* neet pitchers. that's okay though. real programmers didn't make them, either. you can still dial-up to a real computer and do some real work, i suppose, till yer damn kids wanna use the phone. or, you can do your checkbook on it, if you got the extra time to write all that stuff down twice. well, now wait a sec - you can use it as a typewriter. yeah, that's it. that's the ticket. a $2000 typewriter!! you're all set!! i *knew* there was a reason you had to have that thing. add a few bucks more for a cuisenart word processor and you've got it!! me? no thanks, i'll stick to troff. real programmers don't need wizzy wigs. unless their hair is falling out. *** flame off =) *** -- jdm NOTE: the above opinions may or may not be my own. think what you will.
cmcmanis@sun.uucp (Chuck McManis) (01/22/87)
In a previous article, dillon@CORY.BERKELEY.EDU (Matt Dillon) writes: > Ha ha ha he he He ha Heh He. > Have you taken a look at the Atari-St Newsgroup lately? Look closely, > C-A, this is *exactly* why you should not put your OS into ROM. > -Matt But Matt they don't have the nifty SetFuncion() call nor the RomTagsInRam stuff that makes little fixes easy. With a "Rom conversion" to 1.2 and the Additional 256K ram you get out of it (your WCS becomes fast ram) You can put simple changes for the ROM into RAM. With the proper Tags the ROM will pick them up as replacements for its internal routines. Really an elegant compromise. (You can even go back an load your 1.1 Kickstart to run brain damaged EA programs, but I don't think it will be write protected then. It is the Atari (publicly stated) we have better things to do than fix bugs in the Rom attitude that is killing them. -- --Chuck McManis uucp: {anywhere}!sun!cmcmanis BIX: cmcmanis ARPAnet: cmcmanis@sun.com These opinions are my own and no one elses, but you knew that didn't you.
tim@ism780c.UUCP (Tim Smith) (01/22/87)
In an article, Dave Haynie writes: > > The Amiga IS the only computer in this price range that has a > real OS. If you don't know what a real OS is, I can suggest > several books on the subject. Or ask a UNIX wiz why MS-DOS > doesn't qualify as a real OS. Please explain how the Amiga has a real OS, but the Mac does not. While you're at it, please explain why Amiga-DOS is a real OS, but PX/IX is not. -- Religion: just say "no" Tim Smith USENET: sdcrdcf!ism780c!tim Compuserve: 72257,3706 Delphi or GEnie: mnementh
ihm@minnie.UUCP (01/23/87)
> >I am currently in the process of buying a personal computer. I have >narrowed myself down to either a Mac or an Amiga, due to personal >preference, software availability, and flexibility. But I can't seem >to decide what would be my best decision. All my Mac friends scream, >"Buy Mac!!!" My Amiga friends retort, "Don't be silly! Amiga is >state-of-the-art." > So I ask, what do you people think is the better of the two, and >why? As far as price goes, the deals I have are: > >Mac-plus: Est. $1550 for Mac (1 meg memory), Imagewriter II printer, > and second drive >Amiga: Est. $1650 for Amiga, Expansion board (512 k), color > monitor, and second drive (printer not in deal-- est. > another $450 or so) > >I would appreciate some as-unbiased-as-possible opinions on what my >best deal is. > >Please E-mail your responses to me, and I will tally up the responses >and post what the general majority opinion is. > > Elliott Buchholz > Looking for a glitchy phrase > Well, you know the old phrase "you can wait forever", but I am going to suggest exactly that. Personally, I think the Amiga has certain hardware and OS advantages, but the Mac is better supported and likely to stay around longer. I would however, suggest that you wait a few weeks and see what Apple is about to announce. I don't know much about the projected availability dates, but there are supposed to be several really super new products being announced VERY shortly (like perhaps this weekend). If they announce what I think they will, I will probably buy the high-end 68020(30?) system for myself. It'll be a nice replacement for my current intel/MS-DOG machine, and rumor has it that there will be a vmunix (probably berkeley) available for it (although not immediately). --I -- uucp: ihnp4!nrcvax!ihm
grr@cbmvax.cbm.UUCP (George Robbins) (01/23/87)
In article <241@elxsi.UUCP> fisher@elxsi.UUCP (Chuck Fisher) writes: > >Flame on: Although Amiga touts its multi-tasking capability, >how many programs are written to take advantge of it? I would >agree that "background printing" is desireable, but a Mac can >do that as well as an Amiga. Most of the work that I do on the >Mac are interactive in nature, and don't lend themselves that >well to true multitasking. Switcher and Servant provide most >of the capabilities needed to integrate foreground tasks in my >opinion. You were doing good up till this, but now it looks like you are maybe missing an important point. When the system software is designed with multi-tasking in mind, programs don't have to be written to take advantage of multi-tasking. It's just an option that exits when the *user* needs/wants it. The is one place where the MAC blew it, using an event loop philosophy that sounds an awful lot like some 60's style* design to implement state-of-the-art human interfaces. * primitive online systems required batch programs to test for special events in their main processing loops, and branch to a special inquiry routine/program to handle requests from the 'online' terminal(s). We're talking 20-25 years ago, IBM 1401's and stuff. Can you say 'desk accessory'? -- George Robbins - now working for, uucp: {ihnp4|seismo|rutgers}!cbmvax!grr but no way officially representing arpa: cbmvax!grr@seismo.css.GOV Commodore, Engineering Department fone: 215-431-9255 (only by moonlite)
grr@cbmvax.cbm.UUCP (George Robbins) (01/23/87)
In article <1549@trwrb.UUCP> sansom@trwrb.UUCP (Richard Sansom) writes: > >Sorry to contribute to this silly discussion, but you knew someone was >going to rebut that last article. How did this latest bout of Amiga v.s. >ST get started anyway? >-Rich Some poor person asked for help in deciding which system to buy, sigh. Enuff... -- George Robbins - now working for, uucp: {ihnp4|seismo|rutgers}!cbmvax!grr but no way officially representing arpa: cbmvax!grr@seismo.css.GOV Commodore, Engineering Department fone: 215-431-9255 (only by moonlite)
tim@ism780c.UUCP (Tim Smith) (01/23/87)
In article <5240@ism780c.UUCP> I wrote: >OS, but PX/IX is not. That is, of course, PC/IX, not PX/IX. Oops. -- Religion: just say "no" Tim Smith USENET: sdcrdcf!ism780c!tim Compuserve: 72257,3706 Delphi or GEnie: mnementh
jtr485@umich.UUCP (01/23/87)
In article <1270@cbmvax.cbmvax.cbm.UUCP>, daveh@cbmvax.UUCP writes: > > > but then the ST has a better OS..... > > Laurence R. Brothers > Technically, the Atari doesn't even HAVE an OS. A DOS, maybe, and a > set of BIOS calls, certainly, but no OS. The Atari's BIOS is an > attempted clone of MS-DOS, and causes lots of problems. Like hard > disks that get unbareably slow with partitions greater than 5 Megs, > and go ape on you when you get something over 40 subdirectories on > them. The Amiga IS the only computer in this price range that has a > real OS. If you don't know what a real OS is, I can suggest several > books on the subject. Or ask a UNIX wiz why MS-DOS doesn't qualify > as a real OS. > Dave Haynie {caip,ihnp4,allegra,seismo}!cbmvax!daveh So, tell me. What makes a DOS not an OS? AmiganaworkDOS (am i gana work DOS) doesn't quite cut it either. But, for the cost of the Amiga you can get an ST, OS/9 and OS/9 C. OS/9 is a serious multitasking OS. Built small but built properly. FYI: 1/2 the price is closer to the mark when you compare the 520 Mono ST to a 512K Amiga. Lest you get me wrong, I do think as marketed the ST does not compare to the Amiga. But, neither is what is could very easily be and should be. What is hurting both machines is the need to pull good margin to hold up corporate ships with large holes in their hulls. --j.a.tainter
mwm@eris.BERKELEY.EDU (Mike Meyer) (01/23/87)
In article <5240@ism780c.UUCP> tim@ism780c.UUCP (Tim Smith) writes: >In an article, Dave Haynie writes: >> >> The Amiga IS the only computer in this price range that has a >> real OS. If you don't know what a real OS is, I can suggest >> several books on the subject. Or ask a UNIX wiz why MS-DOS >> doesn't qualify as a real OS. > >Please explain how the Amiga has a real OS, but the Mac does >not. While you're at it, please explain why Amiga-DOS is a real >OS, but PX/IX is not. I didn't make the first quote (and I disagree with it), but let me answer the questions anyway: First, a definition. A "program loader" is a program that puts other programs in memory, and passes control of the machine to them. If you enhance these with a set of routines for doing hardware-dependent things in a consistent manner, and make the entry points the same on all hardware, you have something like CP/M. Those who deal with large computers don't consider these to be a "real OS" (I deal with large computers, and have dealt with such creatures; I agree with the snobs :-). My understanding is that this *IS* what the Mac provides. Once an applications program has control of the machine, it's got control of the machine, and has to wait for the program to give it back. There's no way for it to run code - either it's own, or another applications program - unless the application permits it. Am I wrong? Now, the questionable statment also said "in this price range." Gotta be carefull how you define price range. Since peripherals can be added arbitrarily, let's look at the smallest machine that will run the OS in question. ST's apparently start at about $500, and an Amiga can be put together for $900 or so (one drive, 256K and a TV for a monitor). Let's arbitrarily label the price range as $500-$1000 (I assume that minimal Mac's can be found in that price range, also). If you can build a system that runs Pick-Axe for < $1000, then there's a second system with a real OS, even by strict definitions (well, I do know people who don't consider variations on Sys III & V as real OS's, but we'll ignore them for now). On the other hand, I know that the RS Coler Computer runs a real OS (OS/9, either Level I or Level II; the latter meaning you have an MMU and protected memory, even!). I think a low-end CoCo, even enough for level II, is still under $500, so it's not in the price range I chose, either :-). But add a requirement for a good monitor, and it'll be in that price range. This is why I disagree with the above statement; it's caused by under-exposure to the micro market. <mike
dillon@CORY.BERKELEY.EDU (Matt Dillon) (01/23/87)
To Everybody in General To Mr Chuck Fisher Specifically The excerp that CF posted originally (which started this war) is essentially correct. One of your added comments (below), however, is completely incorrect and has already been brained to death in previous wars between these two newsgroups: >Flame on: Although Amiga touts its multi-tasking capability, >how many programs are written to take advantge of it? I would >agree that "background printing" is desireable, but a Mac can >do that as well as an Amiga. Most of the work that I do on the >Mac are interactive in nature, and don't lend themselves that >well to true multitasking. Switcher and Servant provide most >of the capabilities needed to integrate foreground tasks in my >opinion. You obviously don't understand the concept of multitasking. To-wit, the idea is that you can run several completely unassociated programs at the same time without any of them knowing or caring that the others are also running. There doesn't have to be any 'software' that 'takes advantage of multitasking', the multitasking allows the user to fully use the machine and however much memory it has in any way he chooses. For instance (and I take a standard example), I frequently use my terminal program while compiling large sources. There seems to be the misconception that multitasking presumes a slowdown in throughput. Certainly running two compiles simultaniously from RAM disk would cause each compile to run half as fast, but consider more common occurances: Taking the former example one notes that the terminal program takes almost no CPU (it's continuously waiting for keystrokes or chars from the modem), and thus both tasks run at essentially full speed. The same goes for most other programs one might be using... the USER is still a single tasking entity with limited interrupt capability. The USER knows exactly what is running when, and why, because he has full control over all the programs he runs. But you should remember that what this user views as a sequential singularity usually requires multitasking on the computer in question: "Ho Hum reading my mail with my terminal program. Let's see, I want to do a compile. OK, start the compile, go back to my terminal program..." Switcher and Servant attest to the fact that most MAC people understand what multitasking is and want it on their machines. I can but agree with them. -Matt
dwb@well.UUCP (David W. Berry) (01/23/87)
C'mon folks. We just went through all this. And the original poster specificall requested comments by >MAIL<. Let's can it huh? At least don't distribute to multiple groups... David -- David W. Berry dwb@well.uucp dwb@Delphi dwb@GEnie 293-0752@408.MaBell
flaps@utcsri.UUCP (Alan J Rosenthal) (01/23/87)
In article <241@elxsi.UUCP> fisher@elxsi.UUCP (Chuck Fisher) quotes from an article in the January, 1987 issue of Dr. Dobbs Journal of Software Tools, "Macintosh Buttons and Amiga Gadgets" written by Jan L. Hamington: ... >The Macintosh routines are also more complete in >terms of their support for the documented user interface. ... >It is true that the Amiga performs some >functions "automatically" for which a Macintosh program must >include code (for example, moving and sizing windows). I find it necessary to point out that, like many other things on the mac, windows are not completely implemented. Just like multitasking, which is not really implemented on the mac in the sense that a task always has to voluntarily give up its control of the cpu (by calling SystemTask, I think it is)*, with windows the mac doesn't store what's obscured when a window goes on top of another window. Therefore the user program must store this at all times, because it is wiped out before the user program gets notified that the window has been obscured. Then the user program must redraw the contents of the window that was obscured when the window is brought to front again. Therefore I think it is a gross understatement to say that the Amiga merely performs these functions automatically rather than the user program having to do it itself. The Amiga supports windows completely and the Mac does not. On the Amiga, for example, you can draw into a backgrounded window, one of the fundamental things you might want to do with a window. ajr * It's true that on the Amiga you can block task switching. I don't know how common this capability is in general, but this is not the same. A looping program on the mac blocks task switching, for example, and does not on the Amiga. Anyway, my favourite large-computer-that-must-be-kept-in- an-air-conditioned-room was the dec10 running tops10, which also had this capability to ignore task switching, and it ran hundreds of users at a time & no one ever said it wasn't multitasking.
wagner@utcs.UUCP (01/23/87)
In article <11902@sun.uucp> cmcmanis@sun.uucp (Chuck McManis) writes: >But Matt they don't have the nifty SetFuncion() call nor the RomTagsInRam >stuff that makes little fixes easy. With a "Rom conversion" to 1.2 and >the Additional 256K ram you get out of it (your WCS becomes fast ram) >You can put simple changes for the ROM into RAM. With the proper Tags >the ROM will pick them up as replacements for its internal routines. >Really an elegant compromise. (You can even go back an load your 1.1 >Kickstart to run brain damaged EA programs, but I don't think it will >be write protected then. A while ago I asked if anyone had any documentation other than the teasing reference in the Enhancer manual for RomTagsInRam. I don't think I ever saw any answers. Someone out there must know something about this feature. People from time to time have spoken about writing their own device drivers. I spoke of perhaps front-ending the trackdisk.device. We would need this information to do the job properly. Michael
cmcmanis@sun.uucp (Chuck McManis) (01/23/87)
In article <241@elxsi.UUCP>, fisher@elxsi.UUCP (Chuck Fisher) writes: > > [ .. Actually he quotes from Dr. Dobbs but I've deleted that part ..] > Flame on: Although Amiga touts its multi-tasking capability, > how many programs are written to take advantge of it? I would > agree that "background printing" is desireable, but a Mac can > do that as well as an Amiga. Most of the work that I do on the > Mac are interactive in nature, and don't lend themselves that > well to true multitasking. Switcher and Servant provide most > of the capabilities needed to integrate foreground tasks in my > opinion. Unfortunately the above is a misconception held by many people. One of the nice things about multitasking on the Amiga is that it is built into the system. That means YOU DON'T HAVE TO DO ANYTHING SPECIAL TO TAKE ADVANTAGE OF IT. Every program you run is added to the list of things to run as far as it is concerned it has the whole machine to itself. When you were in college did you have to do anything to make your programs take advantage of multitasking on the mainframe? Of course not, you ran your programs and other people ran their programs, and sometimes you would send things out to the printer while continuing to work on your code and the program printed *while you worked*. It is the same on the Amiga, if you want to start something from the workbench then just double click on it's icon, whatever you are currently running won't stop, it just keeps on chugging and the new application starts up. The system makes sure both programs don't grab the disk at the same time or write to the same file at the same time etc, you code just runs blythely along. And if your code wants to talk to other tasks on the machine it opens a message port and gives it a name, then any other task can look for that message port and start conversing. The only limit on all of this is memory, since there is no virtual memory capability if you run one program that takes all of memory then you cannot run another if it won't fit. > Much as I hate to say it, but the success of a machine is not > based solely on hardware, or the elegance of the user interface. > Instead, what guarantees success is the name and backing of the > manufacturer plus the willingness of third-party developers to > produce software. When enough machines have been sold and > enough software written, the machine then becomes a "success" > and continues to generate momentum for some time. The Mac has > achieved this level in the past 12 months, but I am afraid the > Atari and Amiga have not. This is an entirely true and entirely misleading statement. Yes in the last year the Mac has finally broken through that undefinable area that makes it a "real" machine. This was three years after it's introduction! One year after the Amiga and Atari were introduced *both* Atari and CBM sold more of their respective machines in the first year than Apple sold Mac's. They also have more software a year after their intro than the Mac did, and more peripherals. If they follow that trend, then in three years their combined installed base will be *eight* times the current Mac installed Base. But everyone knows projections are wildly unreliable things, it could be worse, it could be better. Apologies if the SNR of this message is below 50%... -- --Chuck McManis uucp: {anywhere}!sun!cmcmanis BIX: cmcmanis ARPAnet: cmcmanis@sun.com These opinions are my own and no one elses, but you knew that didn't you.
ross@ulowell.UUCP (01/23/87)
In article <8520@topaz.RUTGERS.EDU> brothers@topaz.RUTGERS.EDU (Laurence R. Brothers) writes: >"Don't be silly!" > >Get an Atari ST! Like 1/2 the price of the amiga and essentially the >same features. OK, the amiga has a little better graphics, but then >the ST has a better OS..... Also, you can get higher memory models *flame on* sigh, here we go again. Ok, so you say the Atari has a better OS. Case in point. I'm kermiting down a whole directory of public domain stuff from a BSD system. When, I suddently realize, oh crud, not only am I going to run out of RAM, I am using the dynamic RAM drive, but I don't have enough space on the floppy when I take the files out of ram and move them to the floppy. Solution, bring up another couple of MULTIPROCESSING windows. One runs compress on each file on the RAM disk as it comes in, and another copies it out to disk and deletes it, after it has been compressed, hopefully :-). It worked, I got 1.5M of compressed stuff out onto that disk, and then took it home to deal with it. The point of this is, that a statement that multiprocessing is what makes the Amiga. The statment that the ST has a better OS is a sign of true ignorance. Take a course in OS's before you speak, and you are guilty of misinformation. Personally I dislike Micro's in general, but the Amiga and its OS are the first of more to come that are usable. (Micros I mean, no no not a rumor) I just can stand doing one thing at a time slowly. All right, this is the first time I gave into tempation and replied to one of these, but I can't stand ignorance. *flame off* >of the ST cheaper still (1040 K ST is < $1000, including monitor, disc >drive) You get what you pay for, generally. > >Also, if you like Mac software, you can get a cartridge relatively >cheaply that plugs into the st case and presto -- color Mac. >Apparently, the ST actually runs many Mac program faster than the >Mac... Yea, and the amiga can be turned into an IBM. I have seen an ST emulator on the Amiga also, but never say it work. A c64 emulator also exists, and there are constant rumors of a Mac emulator. But, notice that no one is emulating the Amiga. Two possibilities: 1.) The Amiga is failing, naw, to much loyalty, and commodore stock continues to rise. 2.) The Amiga is truly a cut above, or state of the art, and can't be emulated by the more primitive hardware and OSes. Ross Miller uucp: ross@ulowell.uucp csnet: ross%ulowell.csnet@csnet-relay.arpa disclaimer: These are not the opinions of my employers. They don't care. This was written by the same monkeys who came up with shakespere. (sp?) :-)
tim@ism780c.UUCP (Tim Smith) (01/24/87)
This comparison ( which seems to favor the Mac ) is probably about
as relevant as the Byte one ( which favored the Amiga ). In other
words, it is not relevant. A sane programmer only writes this
stuff once. From then on, he/she re-uses it rather than re-write
it. Thus, it doesn't matter if one of the machines provides slightly
more support for windows or slightly more support for text.
--
Tim Smith USENET: sdcrdcf!ism780c!tim Compuserve: 72257,3706
Delphi or GEnie: mnementhtim@ism780c.UUCP (Tim Smith) (01/24/87)
Compiling and reading mail at the same time are not good examples
for why one wants multitasking. On the Mac, I use LightspeedC.
The compiler and linker are so fast, I don't have time to read
my mail! :-)
A better example would be doing some slow operation on a large
database while reading mail.
--
Tim Smith USENET: sdcrdcf!ism780c!tim Compuserve: 72257,3706
Delphi or GEnie: mnementhtim@ism780c.UUCP (Tim Smith) (01/24/87)
In article <2312@jade.BERKELEY.EDU> mwm@eris.BERKELEY.EDU (Mike Meyer) writes: >> >>Please explain how the Amiga has a real OS, but the Mac does >>not. While you're at it, please explain why Amiga-DOS is a real >>OS, but PC/IX is not. > >My understanding is that this *IS* what the Mac provides. Once an >applications program has control of the machine, it's got control of >the machine, and has to wait for the program to give it back. There's >no way for it to run code - either it's own, or another applications >program - unless the application permits it. Am I wrong? > If I am running Switcher or Servant on a Mac, then I may have multiple programs loaded. Context is switched by explicit action on my part, while, say, on my UNIX system, it can also be switched by explicit action on the part of the clock. I don't see an essential difference as far as the operating-systemness of either system goes. Also, I thought that the Amiga only switched tasks when the currently executing task does something that would block ( IO, or something like the UNIX sleep or pause system calls )? Is this so? If someone can come up with a good definition of "operating system", then it would be possible to decide which computers have them. One OS textbook I saw defined an operating system as the software that takes control when a user program makes an error, but that definition seems kind of worthless. -- Tim Smith USENET: sdcrdcf!ism780c!tim Compuserve: 72257,3706 Delphi or GEnie: mnementh
korn@eris.BERKELEY.EDU (Peter "Arrgh" Korn) (01/24/87)
In article <8520@topaz.RUTGERS.EDU> brothers@topaz.RUTGERS.EDU (Laurence R. Brothers) writes: >"Don't be silly!" >... >Also, if you like Mac software, you can get a cartridge relatively >cheaply that plugs into the st case and presto -- color Mac. >Apparently, the ST actually runs many Mac program faster than the >Mac... I hate to continue this discussion, but the above assertion is an outright LIE! Some friends of mine who own Atari ST machines came over to my house one evening with their STs and their Mac-Cartridges, and we systematically tried my entire software collection (over 100 programs, public domain and otherwise) . LESS THAT 5% OF IT RAN! These friends, who had been chiding me continuously for the past few weeks that their ataris "would be a mac for $100's less, and faster too" quickly changed their tune after actually testing the product. Which brings up a very important point. If you haven't used a program then don't spend the UseNet's money telling me what it will do, and that it's a great program and will make other things worthless, etc. etc. You are welcome to announce that "I've just heard of something interesting, and the marketing folks _claim_ that it will do this--looks like it's worth checking out". That's fine, if not welcomed. But don't waste my time and other people's money making claims that you know nothing about. Unfortunately, I've noticed this phenomenon occuring a lot more in people who are Atari fanatics. I trust that this is just a statistical anomoly... So, if you want a Mac, buy a Mac. If you want something else, then don't get a mac, get that something else. But don't be fooled into thinking that you can get a mac w/out getting a mac... Peter "try it before you preach about it" Korn ----- Peter "Arrgh" Korn Hacker? Me? A hacker? No, actually korn@ucbvax.Berkeley.EDU I'm a mac-er. All's we do is {decvax,dual,hplabs,sdcsvax,ulysses}!ucbvax!korn make library calls.
andrewjp@uoregon.UUCP (01/24/87)
I apologize in advance for not letting this subject die gracefully, but I get a bit tired of hearing things like: In article <1974@alvin.mcnc.UUCP> ravi@alvin.UUCP (Ravi Subrahmanyan) writes: > Dave's one ironclad point. The Amiga's multitasking *is* >great. However, I look upon this as a difference in the machine's >characters, not an issue for mudslinging. It may surprise many people to learn this, but the ST operating system handles a sort of multitasking very well. It has to be written into the software, but I personally prefer it that way (I am saddened every time I see how slow this VAX is even when I'm the only user logged on, thanks to the multitasking ovrhead). One of my roommates is a developer for the ST and has nearly completed a whole line of desk acceessories that multitask beautifully. As I understand it, the same thing is possible with TOS applications, though a bit trickier. So while I admit that, for most consumers' purposes, there are not multitasking applications available for the ST, I stolidly assert that the machine does support it very nicely, so a programmer/developer who wants multitasking capability shouldn't look elsewhere so quickly. > I imagine everyone is sick of this debate, so why doesn't >someone direct followups to the place it all started (now that I've >got my two bits in)? > Ditto. Cygnus. Eric Swanson c/o andrewjp@drizzle.uucp P.O. Box 30098 Eugene, OR 97403 (503) 484-2790 or (503) 484-4184
jimomura@lsuc.UUCP (01/24/87)
Pete Korn's point about Magic Sac only running 5% of Mac software may
be right. I don't know. All I know is that I've seen it run Excell, Macpaint
and one of the top Mac word processing packages (and a handful of other
things). I knew well that it didn't run everything, but frankly I'm not
that worried. It seems to run the majority of the most important programs.
As far as I know, D. Small is still working to make improvements to improve
the emulation. So far, what it really shows me is that the Mac programmers
don't seem to write very good programs. If Apple brings out a new Mac with
improved resolutions, the same problem programs will likely not work on
it either.
I haven't bought Magic Sac myself. I have no use for it. Nor do I
bother with the CP/M emulator (which I *would* have use for--having much
CP/M software). There's quite enough good software for the Atari ST already
and more will be available as time goes on.
Cheers! -- Jim O.uh2@psuvm.bitnet.UUCP (01/25/87)
Someone asks, "I'm curious. Just how great is multi-tasking?" Now I am no wizard, but isn't the real beauty of multi-taksing a lot more than just being able to print in the background. For example, device drivers, print spoolers, communication programs, fast database systems, and a lotta other keen stuff are much easier to develop and debug if they can be implemented as a bunch of simultaneously runnig tasks on a multi-tasking machine. In other words, if 6 chimpanzees, programming at random, work on a mu;titasking system they will produce all the worlds great programs a lot faster than if they were working in MSDOS. 8-) Hey--I could be wrong.
jtr485@umich.UUCP (01/25/87)
In article <969@ulowell.cs.ulowell.edu>, ross@ulowell.UUCP writes: > Yea, and the amiga can be turned into an IBM. I have seen an ST emulator > on the Amiga also, but never say it work. A c64 emulator also exists, > and there are constant rumors of a Mac emulator. But, notice that no > one is emulating the Amiga. Two possibilities: > 1.) The Amiga is failing, naw, to much loyalty, and commodore stock > continues to rise. > 2.) The Amiga is truly a cut above, or state of the art, and can't be > emulated by the more primitive hardware and OSes. 3.) The Amiga is not stable enough for an emulation to be a viable product! > Ross Miller --j.a.tainter
rjg@nis.UUCP (01/26/87)
Why do these silly arguments always seem to end up in "but THIS machine is CHEAPER!" ? I determined all my needs. I looked at many machines. I compared those machines to my needs. I examined areas that the inspected machines lacked. I determined ways to work around it, if necessary, or whether these were wishes instead of needs. I examined areas that the machines came out as overkills. I tried to determine the advantageous as well as disadvantageous aspects of each machine. After deciding on a particular machine, I price shopped for the best price (which includes service, customer support from the seller, etc.) The fact that I bought an Amiga is beside the point. So is the fact that other machines are cheaper (and many are more expensive, by the way). The point is that it fits my needs on a long term basis. One needs to examine all the possibilities. Didn't this all start with someone asking for opinions? Instead of blasting one machine or another, and pointing out deficiencies in the other guys machine (many of the deficiencies listed are more in the area of what you would call "personal opinion deficiencies": They don't suit your purposes, but they might just fill someone elses.) Let's try and point out the _positive_ aspects of the Amiga, the Mac, the ST, and anything else, and let the people who want to buy a machine base their opinion on the positive qualities, not the negative qualities. You'll notice that there are a lot more satisfied people who a machine on positive aspects rather than buy someone elses machine because they only have negative aspects of the first. But then again, this has been repeated over, and over, and over again, and I am sure that it will be again, and again .... -- Robert J. Granvin UUCP: ihnp4!meccts!nis!rjg Programmer/Analyst - Technical Services ATT: (612) 894-9494 National Information Systems, Inc. "Let's see who's up the creek without an overthruster NOW, Space Cadet!"
stever@videovax.UUCP (01/26/87)
In article <5283@ism780c.UUCP>, Tim Smith (tim@ism780c.UUCP) asks for a definition of the term "operating system," along with other questions: John J. Donovan's definition of an operating system is: An operating system is concerned with the allocation of resources and services, such as memory, processors, devices, and information. The operating system correspondingly includes programs to manage these resources, such as a _traffic controller_, a _scheduler_, _memory management module_, _I/O programs_, and a _file system_. _Systems Programming_, McGraw-Hill, New York (1972), p. 15 (emphasis in the original) > . . . > If I am running Switcher or Servant on a Mac, then I may have > multiple programs loaded. Context is switched by explicit action > on my part, while, say, on my UNIX system, it can also be > switched by explicit action on the part of the clock. I don't > see an essential difference as far as the operating-systemness of > either system goes. The purpose of an operating system is to manage the machine's resources, making most efficient use of them without human intervention. In this case, you are playing the role of "traffic controller." Because you do not have full knowledge of the internal state of the machine at any given time (and because humans act on a timescale far longer than that of the computer), you cannot manage the system resources on a millisecond-by- millisecond basis. According to Donovan, this is a function of the operating system, so in this case the Mac is deficient. Steve Rice ---------------------------------------------------------------------------- {decvax | hplabs | ihnp4 | uw-beaver}!tektronix!videovax!stever
lsr@apple.UUCP (01/26/87)
In article <3986@utcsri.UUCP> flaps@utcsri.UUCP (Alan J Rosenthal) writes: >voluntarily give up its control of the cpu (by calling SystemTask, I think it >is)*, with windows the mac doesn't store what's obscured when a window goes >on top of another window. Therefore the user program must store this at all >times, because it is wiped out before the user program gets notified that the >window has been obscured. Then the user program must redraw the contents of >the window that was obscured when the window is brought to front again. > It is possible to have the system remember the contents of a window and refresh it automatically. Most applications don't do this because the data needed to refresh the window is readily available, and it is easy to just redraw it as necessary. If your application had multiple disk-based documents open at once, you might prefer to have the system do the refresh, since that might avoid disk acesses needed to bring the data back into memory. >Therefore I think it is a gross understatement to say that the Amiga merely >performs these functions automatically rather than the user program having >to do it itself. The Amiga supports windows completely and the Mac does not. >On the Amiga, for example, you can draw into a backgrounded window, one of >the fundamental things you might want to do with a window. > You can draw into a background Macintosh window as well. The system handles all the clipping for you. As far as I can tell, programming the 2 machines is equally difficult (or easy). The Macintosh has built in text editing primitives and a wider variety of graphics primitives, but the Amiga handles resizing, moving, and updating windows automatically. -- Larry Rosenstein Object Specialist Apple Computer AppleLink: Rosenstein1 UUCP: {sun, voder, nsc, mtxinu, dual}!apple!lsr CSNET: lsr@Apple.CSNET
hadeishi@husc4.UUCP (01/27/87)
Sorry to carry on this Mac vs. Amiga comparison, but it can sometimes be instructive. In <5283@ism780c.UUCP> tim@ism780c.UUCP (Tim Smith) asks: > [ comments about Mac Servant and Switcher ] >Also, I thought that the Amiga only switched tasks when the currently >executing task does something that would block ( IO, or something like >the UNIX sleep or pause system calls )? Is this so? No, the Amiga switches tasks every few milliseconds or so. A task gets the processor only if it is not in the Wait() state, however, so if a task is just sitting there waiting for input, it does not take up any processor time. When the operating system gets a signal (such as in input event for the task, or whatever), it will wake up the task and tell it that an event has occurred. This is very different from the typical Mac program which sits there and loops until an event occurs. For example, on the Amiga, even when there is only one user program running, there are many tasks running. There is the task that takes care of the disk drive, the task taking care of input events, the Intuition granddaddy task, and so forth. (In fact, on the Amiga the OS is just a task, a task that runs at a VERY high priority and sleeps most of the time, waiting for input events. This is different from the UNIX OS which has a magic supervisor that takes care of system stuff.) Though there are many tasks in the system, in general they are all in the Wait() state until the user program or the user asks them to do something. For example, when the program asks for a read or write on the disk, that request is processed in a multitasking mode (i.e., the program can go do other things while waiting for the disk.) The Amiga has very fast context switching, since the whole OS is designed for multitasking performance. On the Mac, since the OS is not designed for multitasking, a context switch has to save a whole lot of information; thus Switcher only does a context switch when you ask it to do so, not automatically all the time as the Amiga Exec does. Thus using Switcher it is impossible to have two programs actually doing something at the same time, whereas on the Amiga this is trivial. On the other hand, Switcher does provide 75% of the functionality of multitasking, which is quick switching between multiple programs all loaded at once. For example, when I am using my Amiga the thing I do most often is switch back to my CLI window to do some disk operations while I'm running my editor or vt100 emulator or whatever. This does not really require multitasking, except when I am downloading I can still do it on an Amiga but not on a Mac. Also since the OS is always there as a task, even if the program is off doing some complicated thing, I can still front/back my windows, switch to a different screen, etc., much more rapidly on an Amiga than on a Mac. Servant was designed to support true context switching through what is called a "Servant task". Unfortunately there are almost no programs which run as "Servant tasks," so this is rather moot. I am told that Hertzfeld considered trying to implement true multitasking for regular Mac programs, but gave up when he realized that to do real-time context switching would slow down programs by an unacceptably huge factor (since you have to save almost all of the machine's state, rather than just the registers, as you would in a true multitasking environment.) It may be that more and more programs will be compiled to run as "Servant tasks" in which case the Mac will slowly become a multitasking machine. The Amiga has been multitasking from day one. The Amiga Exec is designed in a very beautiful way. The library system, etc. have been designed for both elegance and speed, as well as extensibility and backward-compatibility. The only real dog in the Amiga OS is the DOS, which was written by non-Amiga people, and thus is a total dog. It runs OK, but sort of breaks all of the OS rules, but it IS a multitasking DOS, so it does file locking and that sort of stuff. It works, but I wish it were as elegant and powerful as the rest of the Amiga OS. (Thanks Carl Sassenrath, RJ Mical, Jim Mackraz, etc., etc., etc.) One thing I have noticed is that the Amiga windowing system is lightning fast, in comparison with something like Suntools on the Sun. I mean Suntools practically rolled over and died while I was running a simple fractal generation program. I mean this is on a SUN/3, which has "real" hard disk drives, costs $30K, and two 68020s running at 14 Mhz to bench in at about 1 1/2 times faster than a VAX 11/785. Just trying to resize a window under Suntools became a monumental task, and the whole system hung when I tried to rewind the tape drive (it came back when the tape drive finished rewinding, and then proceeded to try to take care of all the mouse clicks and stuff I had been doing while it was dead, all at once. Ugh.) I mean, which is the REAL OS here, Amiga Exec/Intuition/DOS or the Suntools/UNIX? -Mitsu
jxc@rayssd.UUCP (01/27/87)
Jim Omura writes: > Pete Korn's point about Magic Sac only running 5% of Mac software may > be right. ... It seems to run the majority of the most important programs. Jim, are you implying that of all the wonderful Mac software that we hear Mac users boasting about constantly, that only 5% of it is made up of important programs? Hmmm... :-) /^^^/ / _/__________________________________________________ / !/ Jeffrey Jay Clesius !\ / / Raytheon Submarine Signal Division !_\ /^^^/ / /! 1847 West Main Road, Mail Stop 188 !/ / ^^^^ ! Portsmouth, RI 02871-1087 (401) 847-8000 (X4015) / / ! {allegra, decvax!brunix, linus} rayssd!jxc /! / !____________________________________________________/_! / / / ^^^^
robert@sri-spam.UUCP (01/27/87)
In article <1105@husc6.UUCP> hadeishi@husc4.UUCP (mitsuharu hadeishi) writes: > > One thing I have noticed is that the Amiga windowing >system is lightning fast, in comparison with something like >Suntools on the Sun. I mean Suntools practically rolled over >and died while I was running a simple fractal generation program. >I mean this is on a SUN/3, which has "real" hard disk drives, >costs $30K, and two 68020s running at 14 Mhz to bench in at about >1 1/2 times faster than a VAX 11/785. Just trying to resize a >window under Suntools became a monumental task, and the whole >system hung when I tried to rewind the tape drive (it came back >when the tape drive finished rewinding, and then proceeded to >try to take care of all the mouse clicks and stuff I had been >doing while it was dead, all at once. Ugh.) I know this isn't comp.sys.suns so I'll limit my comments on this; The tape drive problem you describe is dependant on the tape controller card in use (I think you have the SCSI), but is not really the fault of the computer per se. Also, "two 68020's"? Isn't there only just one, on the CPU card? Reply by mail if you want to. I do agree that there is a problem with window command queueing under Suntools. Does the Amiga have any problems in this area? Do you get instant response to the mouse? In other words, does the mouse have the highest priority? As for fractal generation on a Sun 3, did you use SunView? If you did then that was probably a big % of your problem. Robert Allen, robert@spam.istc.sri.com
keithd@cadovax.UUCP (01/28/87)
In article <8392@watrose.UUCP> jafischer@watrose.UUCP (Jonathan Fischer) writes: > >In article <636@mit-amt.MEDIA.MIT.EDU> ralph@ATRP.MEDIA.MIT.EDU (Amiga-Man) writes: >> ... [talks about advantages of Amiga] >>- A true multitasking OS. I find this feature very important. > > I'm really curious: just how great is multitasking? I.e., how much do >you, the typical Amiga owners, use this feature? Even on Unix I find that the >only thing I consistently use '&' for is printing, and on my ST I have a wide >range of print spoolers to choose from. I use it frequently when I'm telecommunicating. Uploading and Downloading of software is a relatively boring and time consuming event. While my terminal emulator is doing this, I'm off in another window working on something else. As long as I'm not doing down and dirty debugging of new software that is prone to crash the machine, It works pretty good, and I do this quite often. Other things that are handy, are formatting disks in background while tele- communicating, loading up ram disks, etc. I'm spoiled now, and wouldn't want to do without it. >cycle, the speed of Megamax compile-time, coupled with a RAMdisk, makes multi- >tasking a 'yeah it would be nice, but not utterly world-stopping' luxury. Well, it's not world-stopping, but I sure miss it when I'm on a machine that dosen't have it. >and is it worth the slow-down in execution? Even though I would really _like_ >it, would I really use it that much? As Matt mentioned, there is not that much slowdown, many programs spend an awful lot of time doing nothing, or waiting for something to happen, which dosen't take up any CPU at all. Keith Doyle # {ucbvax,ihnp4,decvax}!trwrb!cadovax!keithd # cadovax!keithd@ucla-locus.arpa
woody@tybalt.caltech.edu.UUCP (01/28/87)
One advantage of the Mac over the Amiga: It certainly is a lot cuter sitting on my little dorm desk than a big, clunky Amiga would be. But I suppose that only applys for college students who move somewhere else every nine months or so, and who likes to take his computer when he travels (and tries to travel light, too!) Of course friends of mine complain when they try to use my Mac: they hate squinting at the little screen... - William Woody Mac! > ][n && /|\ woody@tybalt.caltech.edu woody@juliet.caltech.edu
grunau_b@husc4.UUCP (01/28/87)
In article <596@rayssd.RAY.COM> jxc@rayssd.RAY.COM (Jeffrey J. Clesius) writes: >Jim Omura writes: >> Pete Korn's point about Magic Sac only running 5% of Mac software may >> be right. ... It seems to run the majority of the most important programs. > >Jim, are you implying that of all the wonderful Mac software that we hear >Mac users boasting about constantly, that only 5% of it is made up of >important programs? Hmmm... :-) . . . > ! {allegra, decvax!brunix, linus} rayssd!jxc My estimate would probably be more like 1%. grunau@husc4.UUCP OR { seismo | rutgers | decvax!ihnp4 } !husc6!husc4!grunau
blgardne@esunix.UUCP (01/28/87)
in article <8392@watrose.UUCP>, jafischer@watrose.UUCP says: . . . > I _DO_ envy your multitasking! I'm just curious how useful it > actually is. I.e., is it that necessary in a personal computer, > and is it worth the slow-down in execution? Even though I would really _like_ > it, would I really use it that much? > > Uh oh, this could be taken as a typical my-PC-is-better-than-yours > posting. It isn't. The main intent of this posting is the question of > multitasking. > > - Jonathan Fischer Multitasking is like a PC: if you don't have one, the question is "what good is it?" after you've used it the question is "how can you live without it?". Multitasking is one of the reasons I bought my Amiga, it sounded like a great idea, and none of the other systems offered it. Of course the question from my friends was "what will you ever use multitasking for?". How about this: I'm online with a local BBS, downloading a bunch of neat new public domain software, usually in ARC format. After I've got one file downloaded, I can deARC it, and start fiddling around with it from a CLI window (or Workbench if you wish) while downloading the next file. Or I can type up a message to be posted, or create an ARC file to upload, while still downloading. Ever have a HUGE file to download or upload? Play a game of Othello, Missile Command, or Asteroids while the modem is busy. It's true that some form of the above is available on other systems with careful selection of your software. But the point of the Amiga's multitasking is that _any_ combination of programs is possible. I guess I'm addicted to multitasking now. Last week I was downloading some stuff from Usenet, and had a couple of files in my Amiga, when I thought "hey, I ought to upload this stuff to the local Amiga BBS!" I had it all ARCed up and ready to go when suddenly I realized "Oh, I've only got one modem, and it won't multitask. Darn!" You can get REAL used to multitasking! Help stamp out PC wars: stamp out crossposting! -- ================================================= "The Admiral is well aware of the regulations..." ================================================= Blaine Gardner @ Evans & Sutherland {ihnp4, decvax}!decwrl!esunix!blgardne 560 Arapeen Drive Salt Lake City, Utah 84108 (801) 582-5847
sebruun@ihlpa.UUCP (01/29/87)
> >Jim, are you implying that of all the wonderful Mac software that we hear > >Mac users boasting about constantly, that only 5% of it is made up of > >important programs? Hmmm... :-) > . . . > My estimate would probably be more like 1%. > Damn, I'm really getting very, very tired of all this "My computer is [bigger/meaner/better/faster/more lovable] than yours" crap. Could all of you little boys and girls please go outside and play, and let the rest of us get some serious work done here. I have a Macintosh, and I wouldn't have anything else. *You* have an Amiga/Atari/whatever and wouldn't have anything else. Fine. Variety is the spice of life and all that stuff... I remember sometime back when the Motorola vs. Intel wars were raging in these sectors and someone suggested the formation of a new newsgroup, net.stupid.penislength or something. Such a group, should it exist (regretably, it does not) would be the perfect place for this current nonsense. ENOUGH, PLEASE!! (*Whew*, just had to get that off my chest...) Steve Bruun AT&T Bell Labs, Naperville, IL ihnp4!ihlpa!sebruun
jtr485@umich.UUCP (01/29/87)
In article <596@rayssd.RAY.COM>, jxc@rayssd.UUCP writes: > > Pete Korn's point about Magic Sac only running 5% of Mac software may > > be right. ... It seems to run the majority of the most important programs. > Jim, are you implying that of all the wonderful Mac software that we hear > Mac users boasting about constantly, that only 5% of it is made up of > important programs? Hmmm... :-) I don't know about him, but I am saying that! --j.a.tainter > / !/ Jeffrey Jay Clesius !\
daveh@cbmvax.UUCP (01/29/87)
> Keywords: Mac, Amiga, ST > Xref: cbmvax comp.sys.amiga:1618 comp.sys.mac:937 comp.sys.atari.st:1063 > > > In article <636@mit-amt.MEDIA.MIT.EDU> ralph@ATRP.MEDIA.MIT.EDU (Amiga-Man) writes: >> ... [talks about advantages of Amiga] >>- A true multitasking OS. I find this feature very important. > > I'm really curious: just how great is multitasking? I.e., how much do > you, the typical Amiga owners, use this feature? Even on Unix I find that the > only thing I consistently use '&' for is printing, and on my ST I have a wide > range of print spoolers to choose from. ... > > I _DO_ envy your multitasking! I'm just curious how useful it > actually is. I.e., is it that necessary in a personal computer, > and is it worth the slow-down in execution? Even though I would really _like_ > it, would I really use it that much? > > - Jonathan Fischer Well, first of all, even if you never experience a need for multitasking, you get two things from it. First of all, the Operating System uses it. System devices, like disk driver, have a server task that operates them. The DOS will generally send messages to these tasks, which then "take it from there", often asynchronously. The nice thing about this is that a calling program isn't required to wait for the disk operation to finish before going on. Disk I/O is mostly waiting for the heads to move, so as a background task you use very little CPU time on disk access (on the Amiga, the floppy reads and writes are done by a custom chip during an interlaced DMA slot, so you have even more CPU time available. But on any machine this can be a big win). The multitasking system lets you very easily take advantage of CPU time that would otherwise be wasted in some kind of a wait loop. Another example of system multitasking is Intuition, the graphic interface manager on the Amiga. Intuition exists as a high priority background task. It monitors input events, like keyboard or mouse actions. Programs send messages to Intuition, asking for any events Intuition has noticed; Intuition tells each program about only those events its specifically concerned about. As a separate task, Intuition will never miss mouse movements or key strokes, while in a single tasking system you can very easily miss such things, or slow your program down trying not to miss any. Also, the Intuition process as a high level I/O server makes alot of sense when you consider that several different tasks may all be asking about these events. Since Intuition is there, no task need know about any other running with it. Now there's the multitasking I use myself, out of choice versus system design. Truthfully, I use multitasking on the Amiga much more than on UNIX, mainly because the windowing/screening capabilities of the Amiga let me run simultaneous interactive things much better than I can on normal UNIX (may X-Windows on a UNIX system will change this). I do the same amount of multitasking on the Apollo workstations I use here, for basically the same reasons; the windowing. To start off, I usually run a few background tasks, the print server, PopCLI (a new CLI window at the touch of a key, just like on the Apollo), and Emacs if I'm on a machine with enough memory. I can start up Kermit and download from our VAX, click that screen to the back (it keeps running, of course), and edit or compile something while the download is taking place. The ability to multitask also completely eliminates the need for specially written "desk accessory" or "RAM resident" programs, since any number of programs can run together or stick around in memory; none of them have to obey any special constraints. And the result is much greater efficiency; instead of my application program having to poll the system whenever possible to find any request for a desk accessory, my program just does what I intend to do with it. And other tasks like WorkBench or CLI can load an accessory program if I like. And finally, individual programs can benefit from multitasking. First of all, there's never any reason to kludge "DOS" type commands into a program, since you can always fire up a CLI window to do the job the right way. A program can implement things with tasks much more efficiently and effectively than with standard polling methods. One controlling task can operate with an arbitrary number of servent tasks using "software buss" messing protocols. I think most folks feel satisfied with single tasking because they've always done things that way. The more you use a system that multitasks well, like the Amiga (or an Apollo or Sun), the more you feel frustrated with the old single tasking systems. At least, that's been my experience. -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Dave Haynie {caip,ihnp4,allegra,seismo}!cbmvax!daveh "You can keep my things, they've come to take me home" -Peter Gabriel ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
daveh@cbmvax.UUCP (01/30/87)
> Xref: cbmvax comp.sys.amiga:1637 comp.sys.mac:951 comp.sys.m68k:157 comp.sys.misc:245 > ST's apparently start at about $500, and an Amiga can be > put together for $900 or so (one drive, 256K and a TV for a monitor). > Let's arbitrarily label the price range as $500-$1000 (I assume that > minimal Mac's can be found in that price range, also). If you can > build a system that runs Pick-Axe for < $1000, then there's a second > system with a real OS, even by strict definitions (well, I do know > people who don't consider variations on Sys III & V as real OS's, but > we'll ignore them for now). > <mike I guess I'd have to agree on this one, OS-9 68K on an ST is certainly what I consider a real OS (as opposed to program-loader/BIOS), and actually from what I've seen of OS-9 it seem to fix quite a few things that UNIX doesn't like about running in a small environment like a micro. Even OS-9 Level II on a CoCo I'd have to conceed as a real OS. Is there one for a Mac? The only problem with all of these is that they are add-ons, not supplied by the vendors of the machines in question. Which means that, for the most part, there will be little or no commercial support for the real OS; everything will be written to run with the program loader. Now if you could run TOS in a OS-9 window (port X-Windows or something) on an ST, I might take a second look at the machine. A Sys III or Sys V implementation would be real nice, providing you've got hardware powerful enough to run it at any decent speed (like, an MMU at least). -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Dave Haynie {caip,ihnp4,allegra,seismo}!cbmvax!daveh "You can keep my things, they've come to take me home" -Peter Gabriel ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
dvmark@cca.UUCP (01/30/87)
Now, now, now. Lets not get context switching confused with multi-tasking. On the Mac with swither, only one program is running at a time. You can't start a program compile, and switch to another program and do something else in the mean time. When you switch to another program, the other programs are not running, just the one in front of you. To illustrate this point even better, I have a friend that runs a BBS on his Amiga, and while it is running, he can run a text editor to develop programs, compile and link-edit them *while* the BBS is running, allowing some dialed- in user to read the BBS mail, up/down-load programs, etc. This cannot be done on a Mac or Atari-ST. A question was raised about whether a compute-bound program can monopolize the Amiga and take up all of the available CPU. Here's the story: there are 256 priorities for tasks in the Amiga OS. If a task is ready to run and there are no other tasks at it's priority, it is just handed control and that is it. All higher priority tasks must be in a wait state (or else they would be running, instead of the job at the current priority). When there is an interrupt that makes a higher priority job ready to run, the current job that is running is stopped, and the higher priority job is given control at that point (basic tricle-down dispatching). Here is the good part: If a task is ready to run, and there are more than 1 task ready to run at its priority, the EXEC routines set a hardware timer to interrupt in 64 milliseconds (that's about 1/15 of a second in case you don't have your calculator handy) and then give the task control. If that task does not give up control of the CPU in 64 ms, *BOOM* out he goes, and the next task at that priority is given control the same way. Since all user tasks in the Amiga run at the same priority, this effectively time-slices use of the CPU, and things continue to run. (There is a new command in 1.2 AmigaDOS that allows you to change priority, so you could, in 1.2, block other programs with a compute-bound program, but its your own fault if you ask for it to be done.) The time-slicing that is described in the prior paragraph is very sophisticat- ed, and is *not* done if only one task is running at a given priority, so no overhead is incurred if only one thing at a time is runnable. The only machine that I know of that uses this round-robin time-slicing method could be effectively used as a boat anchor for the USS Nimitz (aircraft carrier). There is certainly no micro that does this. When the Mac and Atari-ST can do the above, drop me a line, and we'll talk about multi-tasking operating systems.
foy@aero.UUCP (01/31/87)
In article <2815@ihlpa.UUCP> sebruun@ihlpa.UUCP (Bruun) writes: ... >Damn, I'm really getting very, very tired of all this "My computer is >[bigger/meaner/better/faster/more lovable] than yours" crap. Could all >of you little boys and girls please go outside and play, and let the >rest of us get some serious work done here. I have a Macintosh, and I >wouldn't have anything else. *You* have an Amiga/Atari/whatever and >wouldn't have anything else. Fine. Variety is the spice of life and >all that stuff... I remember sometime back when the Motorola vs. Intel I recently purchased an Amiga. One of the things that helped me to decide which computer I thought best for my purposes was some of the discussion of "My Computer is better than yours". Perhaps a news group comp.micro. comparisons would be helpful. I personally don't find the ongoing discussion too disturbing. Frequently I learn something about my computer that I would not have known without reading the comparative discussion, ie the uses of multitasking. Richard Foy
fry_b@husc4.UUCP (01/31/87)
In article <12679@cca.CCA.COM> dvmark@CCA.UUCP (Mark Jay Jones) writes: >Now, now, now. Lets not get context switching confused with multi-tasking. On >the Mac with swither, only one program is running at a time. You can't start >a program compile, and switch to another program and do something else in the >mean time. When you switch to another program, the other programs are not >running, just the one in front of you. I respect the Amiga's multitasking, and I certainly don't mean to imply that Switcher is multitasking, but I think there is a point that should be made. In addition to the much-discussed background down- loaders and printers, Thunderscan software (written by Andy Hertzfeld, Switcher's author) is designed to scan under Switcher while you are doing something else in another partition. This is of course a tribute to Andy's abilities, but it also demonstrates that multitasking is not IMPOSSIBLE on the Mac, merely inconvenient. And there was also the great MultiMac mystery, which seemed a good step in the right direction. David Fry fry@huma1.harvard.EDU Department of Mathematics fry@harvma1.bitnet Harvard University fry%huma1@harvsc4.bitnet Cambridge, MA 02138 ...!harvard!huma1!fry
jkg@gitpyr.UUCP (02/01/87)
(hey - does it feel a little warm in here to you?...) In article <6707@aero.ARPA> foy@aero.UUCP (Richard Foy) writes: >In article <2815@ihlpa.UUCP> sebruun@ihlpa.UUCP (Bruun) writes: >>Damn, I'm really getting very, very tired of all this "My computer is >>[bigger/meaner/better/faster/more lovable] than yours" crap. Could all >>of you little boys and girls please go outside and play, and let the >>rest of us get some serious work done here. (I think I smell smoke.....) > [...] I personally don't find the >ongoing discussion too disturbing. Frequently I learn something about my >computer that I would not have known without reading the comparative >discussion, ie the uses of multitasking. (what's that crackling noise?...) You WOULD be disturbed if you subscribed to comp.sys.m68k and didn't give a rat's ass about the Mac OR the Amiga. I apologize for being so churlish, but I have read enough of this insipid, useless prattle in a newsgroup where it doesn't belong. I get sick and tired of seeing "<N> articles in comp.sys.m68k - read? [ynq]", typing "y", and finding that <N-1> of said articles are of the form "my computer happens to have a 68000 and can beat up your computer". Can we PLEASE, PLEASE, PLEASE take this discussion elsewhere (I personally nominate talk.religion.really.stupid.meaningless.arguments.about.computers <- no emoticon - I'm serious, folks). This is not a personal attack on Mr. Foy - his article just prompted it. Believe it or not, there are people in the world who use the 68000 and do not own a 68000-based computer. How about putting this discussion to bed? Jim Greenlee -- The Shadow...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!gitpyr!jkg Jryy, abj lbh'ir tbar naq qbar vg! Whfg unq gb xrrc svqqyvat jvgu vg hagvy lbh oebxr vg, qvqa'g lbh?!
jimomura@lsuc.UUCP (02/01/87)
Ross Miller mentioned the ST Emulator for the Amiga. This program is a fake and meant as a joke only. It draws the Atari screen and then does nothing. You have to reboot the computer. Cheers! -- Jim O.
jimomura@lsuc.UUCP (02/01/87)
Last night I was particularly busy and I had an excellent chance to
confirm OS-9 Multitasking on the Atari 1040ST. I did a mass download with
Kermit from my Unix node with the following command line:
kermit rl /t1 >>>temp &
The '>>>temp' is a redirection of 'stdout' and 'stderr' to a file.
While that was going on (over a half hour's worth of download at 1200 baud)
I began a formatted printout with the 'pr' utility as follows:
pr -l=5 >/p &
That printout was something like 20 pages and ran for a fairly long
time in itself. It was done before the download.
While doing all that I edited a couple of files with the Scred full
screen text editor. I found no apparent degredation, even with cursor
movements and screen redrawing. Memory left over tended to be around
250K to 350K depending on what I was doing at any given time. This is
typical (I was using 64K for Hard Disk Cache and 256K RAM disk and had
booted from TOS indirectly leaving all the TOS/GEM memory intact *beside*
OS-9 in RAM).
I also did some re-arranging of files on the Hard Disk without problems
(ie. No interference with disk accesses from Kermit's download).
In fact, it's my impression that if I had a 2nd RS-232 port I could
easily have started another connection to another system for Telecom or
even more crossloading.
Cheers! -- Jim O.braner@batcomputer.UUCP (02/02/87)
[] In the Atari ST (or the Mac, as far as I know) there is NO polling of the keyboard (nor the mouse): They are INTERRUPT-DRIVEN. Since it takes very little time to process the incoming keystrokes relative to the time in-between keystrokes, that does not degrade performance. And it has absolutely nothing to do with multi-tasking. Beyond that, the ST (or GEM, rather) has a "limited multi-tasking kernel" which lets one 'application' and several 'desk-accesories' (in other words, several programs) run simultaneously, each being reactivated when a message is addressed to it (e.g. a keystroke or a mouse click), but not bothered when not necessary. As for background reading of the disk: Usually a program NEEDS the data it requested from the disk drive BEFORE it can do anything else, so no amount of multi-tasking will speed that program up while it's waiting. On the other hand, SOME specially-designed programs sometimes read the disk BEFORE the data is needed, i.e. invite an overlap of reading the disk and some other processing. On the ST that is possible too, since the disk I/O is done via the DMA chip, which (as far as I know) CAN take a command and then execute it independently. It does steal CPU memory cycles for that, though (or does it? experts, step up?). - Moshe Braner PS: I am just trying to clarify the notion of "true multi-tasking", and NOT to continue religious wars about which machine is better.
bj@well.UUCP (02/02/87)
Multi-tasking is a super important part of the Amiga, if you care to take it as an advantage. You can have a single program commanding any number of processes and tasks to create the user inferface and magic of the program. If you want to add in sound to a program it becomes a seperate application, seperate to code and debug and use, but can coexists with the main program. If you care to you can really do a lot of stuff with the different chips and capabilities of the Amiga without slowing the main body of your program, just have different processes and/or tasks that interact using messages or signals. I come from the big computers to the Amiga, because there is SOOOOO much potential for great programs. It reguires that the developer think about different interactions between processes, but in the long run it is an advantage that can really make the Amiga a bigger success than anything else out there --- assuming that people here about the Amiga in a positive light from those that write about computers. Thank you Jerry about the Commodore blurb in InfoWorld, I also think that the "unshown machine(s)" are wonderful, if they get them all working and to the people and get the trade presses to like Commodore again. Lots to do, many hills ahead. -Jim Becker Terrapin Software
bzs@bu-cs.UUCP (02/09/87)
Several years ago I remember a very popular INFO node (for Emacs/Info) was the TERMS topic which was a comparison of dozens of terminals, in those days the discussions comparing terminals were at least as heated as these micro discussions (who remembers what N-key roll-over means?) I suggest that these flame wars are not idiocy but reveal a real and serious need of the technical community; to choose between various products. The problem right now is that it's high volume with little direction or organization, although some of the comments are obviously important even they are being buried and lost in the noise. It does seem to somehow fit the criteria for a new group (high volume, technical content, appearing undirected in other groups causing complaints etc.) I think what is sorely needed is a moderated group with a highly motivated moderator who could glean the comments with the goal of producing summary comparisons by users of various products which could then somehow be retrieved at a later date by interested individuals, like a source archive. I'm not sure how it should be organized, that would be the moderator's job (anything would be better than this.) Any volunteers for a moderator? Or are we doomed to just see these comparison wars break out with little utility in random groups forever? (I *don't* volunteer, I don't really use micros much and don't have the time.) As a rallying cry I suggest 'comp.reviews' with the possibility of later forming sub-groups if needed (comp.reviews.68k or whatever.) -Barry Shein, Boston University
ihm@minnie.UUCP (02/11/87)
> >Several years ago I remember a very popular INFO node (for Emacs/Info) . . . >The problem right now is that it's high volume with little direction >or organization, although some of the comments are obviously important . . . >I think what is sorely needed is a moderated group with a highly >motivated moderator who could glean the comments with the goal of . . . >As a rallying cry I suggest 'comp.reviews' with the possibility of later >forming sub-groups if needed (comp.reviews.68k or whatever.) How about talk.religion.microcomputers > > -Barry Shein, Boston University -- uucp: ihnp4!nrcvax!ihm
werner@aecom.UUCP (02/17/87)
In article <4031@bu-cs.BU.EDU>, bzs@bu-cs.BU.EDU (Barry Shein) writes: > > I suggest that these flame wars are not idiocy but reveal a real and > serious need of the technical community; to choose between various > products. In this month's Dr. Dobbs Journal, an article on Editors begins with the following comment which is just as applicable to computer systems, etc ...: "Editors? You wanna talk editors? How's about something trivial like life after death, religion, or politics?" - Don Watkins, sysop on CompuServe's IBM NET. -- Craig Werner (MD/PhD '91) !philabs!aecom!werner (1935-14E Eastchester Rd., Bronx NY 10461, 212-931-2517) "Why is it that half the calories is twice the price?"
bill@tolerant.UUCP (Bill Arnett) (05/22/87)
Can someone tell me what the new INIT "Easy Access" does?
--
Bill Arnett {ucbvax,nsc}!tolerant!bill
Tolerant Systems, Inc. San Jose
408/433-5588