consp11@bingsunm.cc.binghamton.edu (Brett Kessler) (03/13/90)
> [...] > Does AmyOs multitask as well as Multifinder? Are most Amy programs > pretty reliable when multi-tasking? > [...] This is the only question of the bunch that I can answer of the few you asked, but I must say, quite simply, that Apple's MultiFinder is a joke, basically because the hardware itself is still not capable of true multitasking. What the MultiFinder does is basically the same thing as the Mac cDEV called Switcher. It lets you swap out one program and swap in another, but ONLY ONE IS RUNNING AT ANY GIVEN TIME. The other one is put to "sleep" until it is switched back in. The Amiga, on the other hand, has the hardware support for true multitasking. There are custom chips for I/O, sound, graphics, etc., freeing up the CPU to do more, such as run more than one program at a time. As an example, do the following on both a Mac and an Amiga: Recalculate a large spreadsheet WHILE printing out a document WHILE downloading software via modem WHILE unarchiving some other software that was downloaded earlier WHILE editing a text file. MultiFinder can't do it. In fact, the printing alone will kill it (though a public-domain ImageWriter spooler init or cdev can get around that). And the software D/Ling kills it alone, too - if you switch away from the terminal program, it stops executing (ie: sending checksum information, etc.), making your D/L a useless, incomplete hunk of nothing. Not to mention the fact that the spreadsheet is no further along in its calculations since you switched out of it.... As for reliability while multitasking, I routinely do the above (minus the spreadsheet - I don't own one!) nearly every day, and haven't had a system crash in well over three months. Ah, well. Hope it helped. +------///-+------------------| BRETT KESSLER |------------------+-\\\------+ | /// | consp11@bingvaxu.cc.binghamton.edu | \\\ | | \\\/// | consp11@bingvaxa.BITNET | \\\/// | | \XX/ | (PeopleLink) B.KESSLER | \XX/ | +----------+-----------------------------------------------------+----------+
dlcogswe@vela.acs.oakland.edu (Dan Cogswell) (03/13/90)
In article <3137@bingvaxu.cc.binghamton.edu> consp11@bingsunm.cc.binghamton.edu (Brett Kessler) writes: >but I must say, quite simply, that Apple's MultiFinder is a joke, basically >because the hardware itself is still not capable of true multitasking. True, but not for the following reason... >What the MultiFinder does is basically the same thing as the Mac cDEV called >Switcher. It lets you swap out one program and swap in another, but ONLY ONE >IS RUNNING AT ANY GIVEN TIME. The other one is put to "sleep" until it is >switched back in. The Amiga, on the other hand, has the hardware support for >true multitasking. There are custom chips for I/O, sound, graphics, etc., >freeing up the CPU to do more, such as run more than one program at a time. You're talking about different levels of multitasking here. In effect, the Exec and Multifinder do the same thing: they schedule processes for the CPU by putting the current process to sleep, then, based on some algorithm, wake another. ONLY ONE PROCESS IS RUNNING ON THE CPU OF THE AMIGA *OR* THE MAC AT ANY GIVEN INSTANT. PERIOD. The main difference between the two involves the WAY in which the scheduler is allowed to make the context switch. The Amiga Exec is tied into the hardware clock interrupt, which cannot be disabled by software. In other words, a process cannot take hold of the CPU forever (as long as correct programming practices are observed). Multifinder, on the other hand, is not tied to any hardware interrupt, which means a process can grab the CPU and keep it until it decides to let it go. Also, (and I may be wrong on this one), Mac software can busy-wait at times. A big no-no in a multitasking system. The CPU is spinning doing no work. >+------///-+------------------| BRETT KESSLER |------------------+-\\\------+ >| /// | consp11@bingvaxu.cc.binghamton.edu | \\\ | >| \\\/// | consp11@bingvaxa.BITNET | \\\/// | >| \XX/ | (PeopleLink) B.KESSLER | \XX/ | >+----------+-----------------------------------------------------+----------+ -- Dan Cogswell | If *ONE MORE* | Disclaimer: (313)625-3234 | person makes | Oakland University INET: cogswell@vela.acs.oakland.edu | a joke about | doesn't HAVE a cogswell@unix.secs.oakland.edu | "Cogswell Cogs!!"| position on this.
bmacintyre@watdragon.waterloo.edu (Blair MacIntyre) (03/13/90)
consp11@bingsunm.cc.binghamton.edu (Brett Kessler) writes: >> [...] >> Does AmyOs multitask as well as Multifinder? Are most Amy programs >> pretty reliable when multi-tasking? >> [...] > >This is the only question of the bunch that I can answer of the few you asked, >but I must say, quite simply, that Apple's MultiFinder is a joke, basically >because the hardware itself is still not capable of true multitasking. Say what? Yes, MultiFinder doesn't multitask, it "switches" between applications. _BUT_, where in the world did you come up with a conclusion that the *hardware* can't multitask?!?! The OS, perhaps, but _not_ the hardware. The MacII harware is _at_least_ as capable of multitasking as the Amiga hardware, considering the processor, etc. -- -- Blair MacIntyre, Professional Leech on Society ( aka CS Graduate Student ) -- bmacintyre@{watcgl, watdragon, violet}.{waterloo.edu, UWaterloo.ca} -- Date, verb: prearranged socializing with intent.
cmm1@cunixa.cc.columbia.edu (Christopher M Mauritz) (03/13/90)
Hehe, OK guys you can stop filling my mailbox with Multifinder sucks pud type messages. :-) I get the message. My main use for Mac emulation will be to continue work (from time to time) that I've begun on my Mac IIcx at work. Currently, I do have a Mac IIcx at home, but it has to go back to work in a month or two. (Rats!) I am going to buy the Amy 030 box as soon as I gather the cash. Thanks for all the helpful replies. Chris ------------------------------+--------------------------- Chris Mauritz |Where there's a BEER, cmm1@cunixa.cc.columbia.edu |there's a plan. (c)All rights reserved. | Send flames to /dev/null |Need I say more? ------------------------------+---------------------------
daveh@cbmvax.commodore.com (Dave Haynie) (03/14/90)
In article <21904@watdragon.waterloo.edu> bmacintyre@watdragon.waterloo.edu (Blair MacIntyre) writes: >consp11@bingsunm.cc.binghamton.edu (Brett Kessler) writes: >>> [...] >>> Does AmyOs multitask as well as Multifinder? Are most Amy programs >>> pretty reliable when multi-tasking? >>> [...] >>This is the only question of the bunch that I can answer of the few you asked, >>but I must say, quite simply, that Apple's MultiFinder is a joke, basically >>because the hardware itself is still not capable of true multitasking. >_BUT_, where in the world did you come up with a conclusion that the >*hardware* can't multitask?!?! The OS, perhaps, but _not_ the hardware. The Amiga OS is better designed for multitasking than the Mac OS, basically because the Mac OS was never designed for multitasking in the first place. The basic fact that Mac people find the performance of Multifinder acceptible in the first place is a credit to the original Mac OS. You certainly wouldn't find as fluid an add-on multitasking system under MS-DOS if it weren't for the 640K physical limitations and hardware support for multiple virtual 8086s under the management of a '386. The best multitasking OS is one designed to do so, but there are varying degrees of How Good Is This Multitasking; it's not a 1 or 0 situation. As for the second comment, I imagine that Brett was thinking of the Amiga's support hardware, rather than the CPU. All 68030s multitask equally well at the base level, though the OS that they're running has alot to do with efficiency. The Amiga's use of preemptive task switching, interrupts, DMA slots, and bus DMA make it more efficient in a multitasking environment. However, to a degree, this would also help out in a non-multitasking environment -- in terms of raw transfer, for instance, the Amiga hard disk interface is about 4x the speed of most Macs. In real life, disk speed is the bottleneck, you really won't notice much difference in disk transfer. However, with a multitasking system, the CPU will get all the cycles the hard disk controller can't use on the Amiga, so your throughput is better -- the Mac would simply wait for the transfer to complete. Folks who don't always completely understand it say that the Amiga was designed for multitasking and the Mac wasn't. That's not true, since the capability for real multitasking lies in the CPU more than anything -- as proof, the Mac II machines run the same basic UNIX that the Amiga 25xx machines run. However, there's quite a bit of the Amiga system design that makes things much faster under multitasking by eliminating CPU waits in several different places. Macs currently don't do these kind of things, though other 680xx machines like Suns and NeXTs do, so I would expect it's only a matter of time before we see thing like peripheral DMA and video coprocessors in a Mac family machine. But expect to PAY for it. >-- Blair MacIntyre, Professional Leech on Society ( aka CS Graduate Student ) -- Dave Haynie Commodore-Amiga (Systems Engineering) "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: hazy BIX: hazy Too much of everything is just enough
bard@jessica.Stanford.EDU (David Hopper) (03/14/90)
In article <10143@cbmvax.commodore.com> daveh@cbmvax (Dave Haynie) writes: ...[stuff deleted]... >for multitasking and the Mac wasn't. That's not true, since the capability >for real multitasking lies in the CPU more than anything -- as proof, the Mac >II machines run the same basic UNIX that the Amiga 25xx machines run. However ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Oh, please, please, please say it's not so, Dave. All I hear around here is how A/UX *sucks*. Even people at Apple realise it's a joke; a hack whipped together to snag government contracts. I've been relying on the hope that AMIX will be a *truly* competitive UNIX package. If my comments are ignorant, forgive me. It just scares me to see AMIX compared to A/UX, after hearing the popular opinions of A/UX. Dave Hopper /// Yesterday, CS. | bard@jessica /// Today, Anthro/History. | .Stanford.EDU \\\/// | \XX/ Tomorrow... URBAN TERRORISM! | (Mel Blanc lives!)
daveh@cbmvax.commodore.com (Dave Haynie) (03/17/90)
In article <10155@portia.Stanford.EDU> bard@jessica.Stanford.EDU (David Hopper) writes: >In article <10143@cbmvax.commodore.com> daveh@cbmvax (Dave Haynie) writes: >>the Mac II machines run the same basic UNIX that the Amiga 25xx machines run. However >Oh, please, please, please say it's not so, Dave. All I hear around here is >how A/UX *sucks*. Even people at Apple realise it's a joke; a hack whipped >together to snag government contracts. I've been relying on the hope that >AMIX will be a *truly* competitive UNIX package. Perhaps I should clarify that, but it is technically correct. Both the Mac and the Amiga are running ports of UNIX System V. I think A/UX is Release 3.1, and the first released version of AMIX will be Release 4. But I would certainly expect Apple to upgrade A/UX to Release 4, and if they do, there's a real good chance that they'll be able to run the same executables. However, they are ports, and the porting process lets the porter make some changes. Apparently (based on my interpretations of probable the same popular opinions you've heard) Apple spent more time making A/UX have a very limited capability to run a single Mac program at a time, rather than making A/UX a very good UNIX implementation. Our UNIX people seem to have exactly the opposite philosophy -- AMIX should primarily be the best possile UNIX system running on Amiga hardware. So at that level, the two UNIX implementations are probably much different. >If my comments are ignorant, forgive me. It just scares me to see AMIX >compared to A/UX, after hearing the popular opinions of A/UX. The complaints I've heard about A/UX are specific to A/UX, not applicable to the general System V Release 3.1 UNIX or any other specific implementation of that OS. >Dave Hopper /// Yesterday, CS. | bard@jessica -- Dave Haynie Commodore-Amiga (Systems Engineering) "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: hazy BIX: hazy Too much of everything is just enough
mike@ames.arc.nasa.gov (Mike Smithwick) (03/20/90)
["moof! moof!"]
In article <3137@bingvaxu.cc.binghamton.edu> consp11@bingsunm.cc.binghamton.edu (Brett Kessler) writes:
<> [...]
<> Does AmyOs multitask as well as Multifinder? Are most Amy programs
<> pretty reliable when multi-tasking?
<> [...]
<
<This is the only question of the bunch that I can answer of the few you asked,
<but I must say, quite simply, that Apple's MultiFinder is a joke, basically
<because the hardware itself is still not capable of true multitasking.
<
In a recent article in MacWorld magazine (I think), one of the editors said
"Multifinder is a hack".
*** mike smithwick ***
"E Pluribus Unix!"
[disclaimer : nope, I don't work for NASA, I take full blame for my ideas]lsr@Apple.COM (Larry Rosenstein) (03/20/90)
In article <10213@cbmvax.commodore.com> daveh@cbmvax.commodore.com (Dave Haynie) writes: > popular opinions you've heard) Apple spent more time making A/UX have a > very limited capability to run a single Mac program at a time, rather than That's no longer true with the introduction of A/UX 2.0. With A/UX 2.0 you can run several Macintosh programs at once, at the same time as X appliations and other UNIX applications. The ability to run standard Macintosh binaries is what sets A/UX apart from any other UNIX system. Larry Rosenstein, Apple Computer, Inc. Object Specialist Internet: lsr@Apple.com UUCP: {nsc, sun}!apple!lsr AppleLink: Rosenstein1
davidw@telxon.UUCP (David Wright) (03/20/90)
In article <7267@goofy.Apple.COM> lsr@Apple.COM (Larry Rosenstein) writes: >appliations and other UNIX applications. The ability to run standard >Macintosh binaries is what sets A/UX apart from any other UNIX system. Please, give us a break. You couldn't be expected to say anything else, but really. That's like IBM saying "The ability to run IBM software is what sets this computer apart from a Mac". I guess you never heard of VP/Ix by SCO, which lets you run multiple IBM applications at the same time under their Unix. At the same time as X. Who else besides Apple runs Mac software anyway?!?! Unless their is some new Mac clone I don't know about, that was able to produce compatible ROMs without Apple filing another rediculous lawsuit, no one else is ABLE to run Mac software at the same time as Unix. Don't compare apples (pun pun pun) with oranges. -- Telxon: | davewt@NCoast.ORG | Programs Plus presents: x4350 | uunet!hal!ncoast!davewt | CoreWar, CRobots, Empire telxon!davidw | uunet!cwjcc!ncoast!davewt | Amiga - The ONLY choice for personal computing
daveh@cbmvax.commodore.com (Dave Haynie) (03/21/90)
In article <7267@goofy.Apple.COM> lsr@Apple.COM (Larry Rosenstein) writes: >In article <10213@cbmvax.commodore.com> daveh@cbmvax.commodore.com (Dave >Haynie) writes: >> popular opinions you've heard) Apple spent more time making A/UX have a >> very limited capability to run a single Mac program at a time, rather >That's no longer true with the introduction of A/UX 2.0. With A/UX 2.0 >you can run several Macintosh programs at once, at the same time as X >appliations and other UNIX applications. The ability to run standard >Macintosh binaries is what sets A/UX apart from any other UNIX system. I just heard about this. It sounds pretty neat. It also is pretty much implying that Macintosh binaries will multitask better under UNIX than under Multifinder at present. Which, if you consider that under UNIX, the Mac binaries can have a virtual Mac to run on thanks to management of the system by a real multitasking kernel with memory management, rather than just being "well behaved" in software. And you should get virtual memory for the Mac applications today, rather than waiting for the new Mac OS. At least it would seem so, I don't suppose this is something that the user would necessarily be given control over. So, is there any remaining reason for a reasonably equipped Mac II to run the plain Mac OS, other than perhaps the cost of A/UX? >Larry Rosenstein, Apple Computer, Inc. -- Dave Haynie Commodore-Amiga (Systems Engineering) "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: hazy BIX: hazy Too much of everything is just enough