ksbolduan@amherst.bitnet (11/06/90)
In article <5932@munnari.oz.au>, iand@mullian.ee.mu.OZ.AU (Ian Robert DOBSON) writes: > Has anybody out there started using the Classic ROM disk as the active system > folder? IMHO, it all sounds fairly useless - if the system folder is in ROM, Is this ROM disk just in the Classic or in all the machines with the 512k ROMs? And, BTW, which machines have those ROMs? Just the Classic, IIsi, and LC ? And, if it is in the other machines, how does one go about accessing it as the Startup "disk"? Kevin Bolduan KSBOLDUAN@AMHERST
hp48sx@wuarchive.wustl.edu (HP48SX Archive Maintainer) (11/09/90)
I have personally tried to boot from the ROM disk on a Classic, a si and a LC, and it only worked on the Classic. The system is almost unusable, as there is only one font in the system. -- ******************************************************* Povl H. Pedersen hp48sx@wuarchive.wustl.edu HP48sx archive maintainer
wnn@ornl.gov (Wolfgang N. Naegeli) (11/09/90)
In article <11074.2736d042@amherst.bitnet> ksbolduan@amherst.bitnet writes: > Is this ROM disk just in the Classic or in all the machines with the 512k ROMs? I have only seen it on the Classic, however an Apple insider has hinted that it is just better hidden in the LC and the IIsi. It's purpose is simply to boot the machine for use as a diskless workstation, then switch to a System (7.0 or higher) on an AppleShare server. AppleShare 3.0 (or whatever it's going to be called) is supposed to provide good support for that. The local diskette drive, if present, will be available for data files, without a need for disk swapping, and you can have dozens of fonts available, as many as are installed on the server. I think this really is an excellent idea for teaching labs, etc. This will make it much easier to guarantee that all Macs use the same configuration. I hope there will be a mechanism that a network administrator can optionally set to prevent any applications to be run from or copied to or from a local diskette. This would help prevent the spreading of viruses and with enforcement of copyright laws. ************************************************************** Wolfgang N. Naegeli President, MacClique--East Tennessee Macintosh Users Group Internet: wnn@ornl.gov Bitnet: wnn@ornlstc Phone: 615-574-6143 Fax: 615-574-6141 (MacFax) QuickMail (QM-QM): Wolfgang Naegeli @ 615-574-4510 Snail: Oak Ridge National Laboratory, Oak Ridge, TN 37831-6206 **************************************************************
kanefsky@cs.umn.edu (Steve Kanefsky) (11/09/90)
In article <1990Nov9.032405.13315@cs.utk.edu> wnn@ornl.gov (Wolfgang N. Naegeli) writes: >In article <11074.2736d042@amherst.bitnet> ksbolduan@amherst.bitnet writes: >> Is this ROM disk just in the Classic or in all the machines with the >512k ROMs? > >I have only seen it on the Classic, however an Apple insider has hinted >that it is just better hidden in the LC and the IIsi. That seems unlikely, since the LC and the IIsi need the extra ROM for things like Color Quickdraw, right? >It's purpose is simply to boot the machine for use as a diskless >workstation, then switch to a System (7.0 or higher) on an AppleShare >server. AppleShare 3.0 (or whatever it's going to be called) is supposed >to provide good support for that. The local diskette drive, if present, >will be available for data files, without a need for disk swapping, and >you can have dozens of fonts available, as many as are installed on the >server. Hmm, I wonder if there's some PRAM or something for AppleShare Prep info. >I think this really is an excellent idea for teaching labs, etc. This will >make it much easier to guarantee that all Macs use the same configuration. Yep. Right now, I'm trying VolumeImage, which makes sure that each hard disk is an exact image of a partition on the file server. Sometimes it runs into problems, but when it works it's great. Come to think of it, I can't imagine how slow it would be to have a lab full of Classics running diskless over LocalTalk (or PhoneNet or whatever). Perhaps with some extra RAM and a healthy RAM cache... >I hope there will be a mechanism that a network administrator can >optionally set to prevent any applications to be run from or copied to or >from a local diskette. This would help prevent the spreading of viruses >and with enforcement of copyright laws. Using LaunchBreak, one can enforce licensing agreements while still allowing free copying of the software (the copies are no good outside the lab) and running of the software from floppies, hard disks, or file servers. With VolumeImage, a one-step operation automatically restores any modified files with clean copies, and optionally closes opened folders and deletes superfluous files (or moves them to a lost&found folder). I just make system disks with VolumeImage as the startup application, and lab attendants are instructed to just restart with these disks whenever problems occur that may be due to missing or misplaced files, viruses, etc. I've had a few problems with it, though, especially when trying to restore multiple machines at the same time. It also seems to want to replace a lot of files that probably don't need to be replaced, but when there aren't many to be replaced it works very quickly. -- Steve Kanefsky kanefsky@cs.umn.edu
doner@henri.ucsb.edu (John Doner) (11/10/90)
In article <1990Nov9.044036.7385@cs.umn.edu> kanefsky@cs.umn.edu (Steve Kanefsky) writes: >Come to think of it, I can't imagine how slow it would be to have a lab >full of Classics running diskless over LocalTalk (or PhoneNet or whatever). >Perhaps with some extra RAM and a healthy RAM cache... Yes, I've often thought about the time it takes to set up a roomfull of Macs over Appletalk. One thing we tried a few years ago was to provide each with a floppy configured to set up a ramdisk upon startup, and then copy various programs and files over from a server. This worked reasonably well, but was slow when every student in the room was simultaneously trying to copy the same stuff over the net. But it would seem that an Appletalk broadcast mode would solve the problem in such a situation. Normally, a packet is sent from one sender to one receiver, and is ignored by all other nodes. I see no hardware reason why several stations couldn't simultaneously receive the same packets. There would be problems of sychronization in getting started, but in principle you ought to be able to boot all the machines at once this way. However, I don't think Appletalk software offers this sort of broadcast mode; am I right? John E. Doner | "The beginner...should not be discouraged if...he Mathematics, UCSB | finds that he does not have the prerequisites for Santa Barbara, CA 93106| reading the prerequisites." doner@henri.ucsb.edu | --Paul Halmos, Measure Theory
wnn@ornl.gov (Wolfgang N. Naegeli) (11/10/90)
In article <1990Nov9.044036.7385@cs.umn.edu> kanefsky@cs.umn.edu (Steve Kanefsky) writes: > Come to think of it, I can't imagine how slow it would be to have a lab > full of Classics running diskless over LocalTalk (or PhoneNet or whatever). > Perhaps with some extra RAM and a healthy RAM cache... Many schools really are on a shoestring budget and willing to sacrifice performance for having the basic capability at all. I know of a couple of labs that boot off a floppy, but run all their applications off the server. Once an application is running, little access is made to the System Folder. If you select your applications carefully, you can get pretty decent performance even with some 20 machines off a single AppleShare server. WriteNow, for example, preforms admirably in such situations. Such environments and low end machines were one of the prime considerations in its development. It also happens to be very easy to learn and to offer everything that is needed for standard business correspondence as well as for writing a dissertation. It's feature list may not be as long as that of some other word processors, but almost all of its features are truly useful and not just gimmicks of little practical value. You get a lot of bang for the buck, not only in terms of software costs but also in terms of what you have to spend for hardware to make good use of it. ************************************************************** Wolfgang N. Naegeli Internet: wnn@ornl.gov Bitnet: wnn@ornlstc Phone: 615-574-6143 Fax: 615-574-6141 (MacFax) QuickMail (QM-QM): Wolfgang Naegeli @ 615-574-4510 Snail: Oak Ridge National Laboratory, Oak Ridge, TN 37831-6206 **************************************************************
dorner@pequod.cso.uiuc.edu (Steve Dorner) (11/10/90)
>It's purpose is simply to boot the machine for use as a diskless >workstation, then switch to a System (7.0 or higher) on an AppleShare >server. AppleShare 3.0 (or whatever it's going to be called) is supposed >to provide good support for that. It's kind of ironic that the current AppleShare REFUSES to switch-launch to a network volume. I do wish sometimes that Apple wasn't so very interested in protecting us from ourselves. -- Steve Dorner, U of Illinois Computing Services Office Internet: s-dorner@uiuc.edu UUCP: uunet!uiucuxc!uiuc.edu!s-dorner
hpoppe@ncar.ucar.edu (Herb Poppe) (11/10/90)
In article <1990Nov9.190042.28795@ux1.cso.uiuc.edu> dorner@pequod.cso.uiuc.edu (Steve Dorner) writes: > It's kind of ironic that the current AppleShare REFUSES to switch-launch > to a network volume. I do wish sometimes that Apple wasn't so very > interested in protecting us from ourselves. Nonsense - Apple is trying to protect itself from you (us). :-) Herb Poppe hpoppe@ncar.ucar.edu NCAR (303) 497-1296 1850 Table Mesa Dr. Boulder, CO 80307-3000
KPURCELL@LIVERPOOL.AC.UK (11/10/90)
Some questions about the ROM disk: 1. Is there any way that you can get it to execute just a single init (well where would you put it?) like INITshare to pull over inits 2. Does anybody have a use for it in this mode? 3. Why do they put in macsbug as one of the files? 4. Are there any other short-cuts (that may make it useable) that Apple haven't yet documented. and finaly In article <1990Nov8.200002.22937@wuarchive.wustl.edu>, hp48sx@wuarchive.wustl.edu (HP48SX Archive Maintainer) says: > >I have personally tried to boot from the ROM disk on a Classic, a si and >a LC, and it only worked on the Classic. > >The system is almost unusable, as there is only one font in the system. Don't you need a minimum of 3 fonts for a system (Chicago 12, Geneva 9 and 12)? or do they just put in Chicago 12? Hmmm, the question is WHY? Whats it for? Kevin Purcell | kpurcell@liverpool.ac.uk -or- kgp@cxa.dl.ac.uk Surface Science, | Liverpool University | There is now a *Twin Peaks* email discussion list for Liverpool L69 3BX | UK Peak Freaks. Mail me for details ....
chapman@ug.cs.dal.ca (Paul Chapman) (11/11/90)
In article <1990Nov9.163823.29938@cs.utk.edu> wnn@ornl.gov (Wolfgang N. Naegeli) writes: >In article <1990Nov9.044036.7385@cs.umn.edu> kanefsky@cs.umn.edu (Steve >Kanefsky) writes: >> Come to think of it, I can't imagine how slow it would be to have a lab >> full of Classics running diskless over LocalTalk (or PhoneNet or >whatever). >> Perhaps with some extra RAM and a healthy RAM cache... > >Many schools really are on a shoestring budget and willing to sacrifice >performance for having the basic capability at all. I know of a couple of >labs that boot off a floppy, but run all their applications off the server. Here at the Dal UCIS Computer Centre (or whatever it's called), they have Mac Pluses set up so that they boot from an internal floppy and then run from the server like that. The network is MacJanet and it is guaranteed to drive you bezerk before you can get anything done on the machines. I have never used AppleShare, so I can't comment on it but if you want an idea of how slow I'm talking, think of trying to connect to a server over a modem (maybe a bit of an exaggeration but the setup makes floppies look speedy). >Once an application is running, little access is made to the System >Folder. If you select your applications carefully, you can get pretty >decent performance even with some 20 machines off a single AppleShare >server. WriteNow, for example, preforms admirably in such situations. Ever tried to run PageMaker? I'm serious, they've got that on the network! It literally takes 5 minutes to load. WriteNow's a lot better, but Macintosh programs usually only load a part of themselves when you start up, so there's a big delay when you click on anything, pull down a menu, etc etc. Remember, there's 20 other people competing for the network, as well as several people printing things over it as well. > [nice things said of WriteNow] > >************************************************************** >Wolfgang N. Naegeli >Internet: wnn@ornl.gov Bitnet: wnn@ornlstc >Phone: 615-574-6143 Fax: 615-574-6141 (MacFax) >QuickMail (QM-QM): Wolfgang Naegeli @ 615-574-4510 >Snail: Oak Ridge National Laboratory, Oak Ridge, TN 37831-6206 >************************************************************** Although I suppose MacJanet is better than nothing, I am currently confined to using the MS-DOS computers in the next room. Much though I would like to use the Macs, they're just too slooow. A friend of mine who uses an Amiga got an unfairly bad impression of Macintoshes after he spent a good afternoon trying to get something productive out of them. Just throwing in my 2 cents' worth (about 1.6 cents U.S. :) - Paul -- -= Paul Chapman -=- chapman@ug.cs.dal.ca -=- pc@ac.dal.ca (BitNet) =- -=-=-=-=-=-=-= Put Humourous Quote of Your Choice Here =-=-=-=-=-=-=-
jtt@cunixd.cc.columbia.edu (James T. Tanis) (11/12/90)
Appletalk certainly does implement a broadcast mode. the destination node, I believe, is set to 255 on the packet.It is supported at the ATP level, and who knows if any existing software implements it. -JT
philip@pescadero.Stanford.EDU (Philip Machanick) (11/12/90)
In article <1990Nov11.012533.8672@cs.dal.ca>, chapman@ug.cs.dal.ca (Paul Chapman) writes: |> In article <1990Nov9.163823.29938@cs.utk.edu> wnn@ornl.gov (Wolfgang N. Naegeli) writes: |> >In article <1990Nov9.044036.7385@cs.umn.edu> kanefsky@cs.umn.edu (Steve |> >Kanefsky) writes: |> >> Come to think of it, I can't imagine how slow it would be to have a lab |> >> full of Classics running diskless over LocalTalk (or PhoneNet or |> >whatever). |> >> Perhaps with some extra RAM and a healthy RAM cache... |> > |> >Many schools really are on a shoestring budget and willing to sacrifice |> >performance for having the basic capability at all. I know of a couple of |> >labs that boot off a floppy, but run all their applications off the server. [Sluggish performance of MacJanet] |> >Once an application is running, little access is made to the System |> >Folder. If you select your applications carefully, you can get pretty |> >decent performance even with some 20 machines off a single AppleShare |> >server. WriteNow, for example, preforms admirably in such situations. [More sluggish performance of MacJanet] |> Although I suppose MacJanet is better than nothing, I am currently |> confined to using the MS-DOS computers in the next room. Much though I |> would like to use the Macs, they're just too slooow. A friend of mine who |> uses an Amiga got an unfairly bad impression of Macintoshes after he spent a |> good afternoon trying to get something productive out of them. My experience of running off a server (both MacServe and later AppleShare) is you absolutely have to have plenty of RAM and big RAM caches to get respectable performance. The caches need to be on the user machines; more RAM on the server doesn't help nearly as much. The exact size of cache needed depends on your application. If you have things carefully set up, the performance should come close to local diskette (I haven't tried ethernet; if you can afford this, you can probably also afford local hard disks). I don't know how your MacJanet machines are configured. I suggest you look in the Control Panel and see if you can bump the RAM cache up a bit. -- Philip Machanick philip@pescadero.stanford.edu
chapman@ug.cs.dal.ca (Paul Chapman) (11/12/90)
In article <1990Nov11.194615.15478@Neon.Stanford.EDU> philip@pescadero.stanford.edu writes: [lots of old quotes] >My experience of running off a server (both MacServe and later AppleShare) is >you absolutely have to have plenty of RAM and big RAM caches to get respectable >performance. The caches need to be on the user machines; more RAM on the server >doesn't help nearly as much. The exact size of cache needed depends on your >application. If you have things carefully set up, the performance should come >close to local diskette (I haven't tried ethernet; if you can afford this, >you can probably also afford local hard disks). I don't know how your MacJanet >machines are configured. I suggest you look in the Control Panel and see if >you can bump the RAM cache up a bit. Of course, the mandarins at the UCIS computer centre decided that the Control Panel would be just too dangerous for mere mortals to handle, so it's not on the boot disks. Unfortunately, the boot disks are permanently confined to the internal drive by means of a piece of paper stuck over the slot. I am going to complain loudly to them though. The other problem is that these computers only have 1 meg, so it'll probably start squeezing the application area if the cache is too big. Over the summer, I set up a network of computers in Physics which had their own hard drives. I almost can't believe how fantastically much better they are than the ones run by UCIS (and I own a Mac!). They don't have the right software for what I need though, and they also don't connect to the Sun 4 that I'm writing this on. Oh, well. Thanks for the info, tho'. - Paul -- -= Paul Chapman -=- chapman@ug.cs.dal.ca -=- pc@ac.dal.ca (BitNet) =- -=-=-=-=-=-=-= Put Humourous Quote of Your Choice Here =-=-=-=-=-=-=-
gillies@m.cs.uiuc.edu (11/12/90)
Re: running diskless over appletalk We a labs here full of Appletalked Mac Plus 1Mb machines, that run with RAM DISKs to get things done. Insert system disk into floppy, boot, copy system disk to RAM disk, insert user disk into floppy, get real work done. What a pain it is to have on ~300-500K for your user program. To run without the RAM disk would be a great boon. So
jmunkki@hila.hut.fi (Juri Munkki) (11/14/90)
In article <1990Nov11.181457.24416@cunixf.cc.columbia.edu> jtt@cunixd.cc.columbia.edu (James T. Tanis) writes: >Appletalk certainly does implement a broadcast mode. the destination node, >I believe, is set to 255 on the packet.It is supported at the ATP level, >and who knows if any existing software implements it. I have used appletalk broadcasts on the DDP level and it does work, if you don't mind a few problems. The first problem is that every machine has to have the DDP listener registered to the same socket. Normally sockets are allocated dynamically, so there is no way to guarantee that every machine has the same socket. The solution to this is to pick a socket from the experimental socket number range and hope that someone else is not already using it. The other problem is that broadcasts do not seem to be able to cross physical bridges. If you have one machine on the other side of a Kinetics fastpath and the others on the other, broadcast packets will not be able to cross. There might be a way to fix this, but I can't remember the specifics well enough to be able to know for sure. Bottom line: unless you are writing a game or you are at Apple writing system software, I don't recommend using using broadcasts. What do broadcasts have to do with ROMdisks? ____________________________________________________________________________ / Juri Munkki / Helsinki University of Technology / Wind / Project / / jmunkki@hut.fi / Computing Center Macintosh Support / Surf / STORM / ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
phil@and.cs.liv.ac.uk (Phil Jimmieson) (11/15/90)
In article <1990Nov9.044036.7385@cs.umn.edu>, kanefsky@cs.umn.edu (Steve Kanefsky) writes: > In article <1990Nov9.032405.13315@cs.utk.edu> wnn@ornl.gov (Wolfgang N. Naegeli) writes: [with reference to the ROM disk stuff that's built into the Classic...] >>It's purpose is simply to boot the machine for use as a diskless >>workstation, then switch to a System (7.0 or higher) on an AppleShare >>server. AppleShare 3.0 (or whatever it's going to be called) is supposed >>to provide good support for that. The local diskette drive, if present, >>will be available for data files, without a need for disk swapping, and >>you can have dozens of fonts available, as many as are installed on the >>server. > > Hmm, I wonder if there's some PRAM or something for AppleShare Prep info. > Yes, that's what appears to happen. I had a Classic for a couple of days that a friend of mine had bought - and I read about the ROM disk stuff and so tried it out. The files that are in the system folder include the Appleshare and Appleshare prep ones so I tried logging in to a server and getting it to save my name and password. The next time I started the Classic up (having set the startup disk to be the ROM one), it logged into the server for me!! Phil Jimmieson, *************************************************** Computer Science Dept., * JANET : phil@uk.ac.liv.cs.and * Liverpool University, * INTERNET : phil@and.cs.liv.ac.uk * PO Box 147 *************************************************** Liverpool L69 3BX "This message contains wit, sophistication, class, (UK) 051-794-3689 style, and 3 other words I don't understand as well..."