SHULMAN@RED.RUTGERS.EDU (Jeffrey Shulman) (11/20/86)
Delphi Mac Digest Thursday, 20 November 1986 Volume 2 : Issue 60 Today's Topics: RE: Hard disk drives (Re: Msg 14839) (4 messages) Hyper Drawers warning RE: Booting SCSI & HD20 Hard disk RE: human touch "One Touch" board (alter (Re: Msg 14845) (2 messages) RE: Apple fellows & "Application already open" message? RE: Expanding your system heap (3 messages) power supply parts New Apple Sales Promotion? (3 messages) Programmers at Work (2 messages) DataFrame 40 XP is _Fast_ RE: Wanted: the MacBinary specification RE: Computerworld Focus 11/12 RE: IBM-PC <-> VAX <-> MAC network? RE: APL for the Mac (2 messages) RE: ZoomWindows...Help Wanted RE: Mac fan comparison Re: mac video - drawing into ScreenBits PCPC's HFS Backup (3 messages) RE: Font questions RE: HD vs. Floppy: $/Kb (3 messages) SCSI pinouts HFS Blues Revisited RE: mac video - drawing into ScreenBits fast? ----------------------------------------------------------------------- From: HSTARR (14851) Subject: RE: Hard disk drives (Re: Msg 14839) Date: 15-NOV 01:17 Hardware & Peripherals Levco make an internal 20mb drive called the Overdrive-20, and it couples with the 2Mx board, and the Prodigy 4 (I have both). They are also releasing a 40mb Overdrive early in December (based on the CP340 I think). I personally like the Overdrive (even if it is not the speediest, at the moment) because of the way it mounts internally, and the reasonably quiet, but very efficient fan (no more heat problems). On the Prodigy at least, the Overdrive's power supply runs the Disk and the upgrade, so there is absolutely no extra load on the damned pple supply. I have only had the Prodigy/Overdrive for 2 months, but (cross all appendages) nothing has gone wrong, YET! The Overdrive retails for about $1000, so it is not the cheapest. ------------------------------ From: BRECHER (14920) Subject: RE: Hard disk drives (Re: Msg 14907) Date: 17-NOV 02:22 Hardware & Peripherals The allocation block size on HFS volumes is 0.5K for volumes up to 32MB in size, 1.0K for 32-64MB, 1.5K for 64-96MB, etc. The reason for this is that a volume can have at most 65,535 allocation blocks. ------------------------------ From: DWB (14958) Subject: RE: Hard disk drives (Re: Msg 14920) Date: 18-NOV 04:32 Hardware & Peripherals Shouldn't that be the "minimum allocation block size is" I believe that you can change that on a per volume basis by twiddleing a number in the volume header block or whatever it's called. Another reason for having larger allocation block sizes is to reduce the amount of fragmentation a disk undergoes. You ideallly want to tailor the size of the allocation block to the average and mean sizes of the files on the volumes. That tailoring process ain't easy though. An ideal number for most users should actually be about 2k though... David ------------------------------ From: BRECHER (14959) Subject: RE: Hard disk drives (Re: Msg 14958) Date: 18-NOV 10:24 Hardware & Peripherals Yes, I was referring to the minimum and default allocation block size. It can be changed prior to initializing the disk as described in the Disk Initialization Package chapter in IM vIV. Any change after initialization would have to be done before any files (even Desktop) were created, and would require changing several fields in the master directory block and the copy of the virgin MDB at the end of the volume. ------------------------------ From: HSTARR (14852) Subject: Hyper Drawers warning Date: 15-NOV 12:28 Hardware & Peripherals Be aware that the stupid GCC Drawers accessory fails to UnLock itself when it quits -- thus leaving a large LOCKED, unrelocatable block in your application heap. If anyone has a REAL line of communication to GCC, maybe YOU can get thru to them. This causes real PAIN when using Lightspeed Pascal. ------------------------------ From: LOGICHACK (14863) Subject: RE: Booting SCSI & HD20 Hard disk Date: 16-NOV 03:17 Network Digests > From: <CLAK100%BGUNOS.BITNET@WISCVM.WISC.EDU> > Subject: Booting SCSI & HD20 Hard disk One way (not super elegant) way to make a SCSI drive boot before the HD-20 is to kill the boot block signature off of the HD-20 with Fedit. Just change Larry Kenyan's initials in the first to bytes of a volume to anything else and you got. Paul :) ------------------------------ From: LOGICHACK (14864) Subject: RE: human touch "One Touch" board (alter (Re: Msg 14845) Date: 16-NOV 03:41 Hardware & Peripherals Doug: I have a Human Touch 3 to 1 Touchboard and have been using it since July. I'm extremely pleased with its performance and compatibility. It has, in my experience, been compatible with all software that also works on other 2MB upgrades. The only exception to that I have found is Pheonix 3D. I might note that the 321 board speeds up everything on the Mac to 12.5mhz including the Apple ROMs. I don't think the HyperDrive 2000 (a similar product) does that. Note that I'm not completely unbiased since I wrote most of the ROM code for the 321 board. The bad news is that Human Touch has apparently disappeared for the time being. Their phone number 'has been temporarily disconnected due to the customer's request' according to a phone company recording. One major complaint of the upgrade was its ugly external power supply. My solution was to mount the power supply internally. Not too tough to do, even for a hardware klutz like me. I'm not sure if you should by one (if you can even find one) due to the unknow status of the company. One last thing, I'm using a SuperMac DataPort to hook up a DataFrame to it. Sucker works great. Welcome to ICONtact. Paul :) ------------------------------ From: DDUNHAM (14869) Subject: RE: human touch "One Touch" board (alter (Re: Msg 14845) Date: 16-NOV 03:54 Hardware & Peripherals A friend of mine bought a One Touch board, but there were some problems with it, so he had to send it back for a new one (regretfully, since he loved the speed). And then Human Touch went chapter 11. I don't know what his dealer was able to do. ------------------------------ From: DDUNHAM (14870) Subject: RE: Apple fellows & "Application already open" message? Date: 16-NOV 03:54 Network Digests > From: shimono@tkov58.dec.com (Takao Shimono - the TongKing boy) > I know Dr. Alan Kay is one of three Apple Fellows. Bill Atkinson is another. > From: mcb@styx.UUCP (Michael C. Berch) > Subject: "Application already open" message? I think I've seen that happen when I've moved an application from one folder to another. Apparently Finder doesn't update its APPL resource (in which it keeps track of what applications are where). One way around this was to open the folder containing the application. David Dunham "If voting could change the system, it would be illegal. If Maitreya Design not voting could change the system, it would be be illegal." ------------------------------ From: PEABO (14882) Subject: RE: Expanding your system heap Date: 16-NOV 17:19 Network Digests >From: howard@amdahl.UUCP (Howard C. Simonson) >Subject: Expanding your system heap >Date: 11 Nov 86 22:53:17 GMT >Organization: Amdahl Corp, UTS Products Group >Thought I'd make a quick request to the net about the proper way to go >about expanding the system heap. In addition to fooling with the field in Boot Blocks containing the size of the System Heap, make sure you also increment the Boot Blocks version number. The 128K ROMs look at the version number first before deciding whether to use the settings in the Boot Blocks or whether to default to settings from the ROM. Last I heard, it was not advisable to set the size of the System Heap larger than 64K. Among other things, system patches have in the past been referred to the base of the System Heap using a 16-bit offset, and if this is still true, it might cause problems. peter ------------------------------ From: DDUNHAM (14957) Subject: RE: Expanding your system heap Date: 18-NOV 04:12 Network Digests While testing some prerelease software that gave misleading error messages, I cranked my system heap up to 100K with no apparent ill effects. I do have a Mac+; I think the 64K limit is much more of a problem with 64K ROMs. ------------------------------ From: PEABO (14963) Subject: RE: Expanding your system heap Date: 18-NOV 13:28 Network Digests Could be ... there was something to do with the coding of the trap vector during initialization, which was compressed in the 64K ROMs but not in the 128K ROMs. (Now where is my master concordance of Inside Mac where I can look this up???) peter ------------------------------ From: CCWOLFE (14902) Subject: power supply parts Date: 16-NOV 21:57 Hardware & Peripherals Does anyone know where I can obtain parts for the power supply mines almost shot ------------------------------ From: DSACHS (14929) Subject: New Apple Sales Promotion? Date: 17-NOV 20:16 SIG Business Does anyone know why Apple has been sending out large empty envelope? After reading someone else's report, I recevied one today. ------------------------------ From: MOUSEKETEER (14931) Subject: RE: New Apple Sales Promotion? (Re: Msg 14929) Date: 17-NOV 20:38 SIG Business That's the new press kit for the new *** due out next January. The way I look at it, if they got the postage right, it's a plus (in several terms of the word). Then again, maybe it's a sneaky way of bringing up the topic of one slot. (stop me! stop me!) Alf ------------------------------ From: LOFTUSBECKER (14939) Subject: RE: New Apple Sales Promotion? (Re: Msg 14929) Date: 17-NOV 21:15 SIG Business They hope you will put some money in them and return them... For what it's worth, I got an envelope (nonempty) with a mailing pushing APDA today. They paid 56 cents postage for it. That might be what you aren't getting. Lofty ------------------------------ From: TRAINBRAIN (14947) Subject: Programmers at Work Date: 18-NOV 01:57 Programming While killing some time in a Computerland recently, I spotted a title "Programmers at Work" on the bookshelf. The title was enough for a mental yuk, but since the car was not going to be ready for an hour, I ended up thumbing through it. Turned out to be one of the most delightful books I have ever read. It's a series of interviews of the well known programmers of our day (Gates, Kildall, Hertzfeld and 16 others). They express feelings, philosophies on a variety of subjects, programming techniques, where computers are going, the effect of age on programmi ng skills (scary, I'm 46), how they got started, etc. The techniques and hints are useful, but the book is so much more than that. I identified with many of the feelings, dreams, frustrations, and triumphs expressed. It gave me a better appreciation of what it is to be a programmer. I can now put into word s why I enjoy it so much. It gives the profession some dignity. Next time I fill out a form with a box marked 'occupation' I will say 'programmer' without the mental prefix 'just a'. It's published by Microsoft Press ($14.95). The editor is Susan Lammers. I have also seen it in Crown Books. The other big chains should be carrying it also. Steve Seidensticker, Author of PROLOG/m P.S. I have no connection with Microsoft or anyone mentioned. This is strictly an unsolicited testimonial. ------------------------------ From: PEABO (14960) Subject: RE: Programmers at Work (Re: Msg 14947) Date: 18-NOV 13:19 Programming I'll second the testimonial. It's a great book and I stretched it out over a week or two instead of gulping it down at one sitting. peter ------------------------------ From: MACINTOUCH (14964) Subject: DataFrame 40 XP is _Fast_ Date: 18-NOV 14:24 Hardware & Peripherals I've just completed benchmarks on the DataFrame 40 XP, and I've been using it for real work. It's very fast. Our benchmarks show about a 20% increase over the run-of-the-mill, and the speed increase is quite noticable in use. Ric ------------------------------ From: MACINTOUCH (14965) Subject: RE: Wanted: the MacBinary specification Date: 18-NOV 17:57 Network Digests To:<INFOEARN%HLERUL5.BITNET@WISCVM.WISC.EDU> Subject: Wanted: the MacBinary specification The MacBinary specification should be available on CompuServe. It's Dennis Brothers' baby, and he's a Sysop there. Ric Ford ------------------------------ From: MACINTOUCH (14970) Subject: RE: Computerworld Focus 11/12 Date: 18-NOV 20:30 Network Digests To: jacobs@utah-cs.UUCP (Steven R. Jacobs) Subject: Re: Computerworld Focus 11/12 I've made something of a specialty of editors on computers from IBM MVS systems to VAX/VMS, VAX/Unix, IBM/PC and Mac systems, and I find mouse-based editing far more efficient for me than any command-driven editors I've seen. There are functions missing from Mac editors, but they aren't the fault of the mouse. I agree that it is nice to have a keyboard option for mouse commands -- the Standard File dialogs are a good example (on the Mac Plus), but take a look at the standardization of Command keys (there isn't any) versus the standardization of mouse-based functions (menus, double-clicking). I thank Apple daily for forcing some standards on the application developers, and find it amazing and wonderful that the real non-standard, non-Macish applications have almost entirely gone "down the tube." Ric Ford ------------------------------ From: MACINTOUCH (14971) Subject: RE: IBM-PC <-> VAX <-> MAC network? Date: 18-NOV 20:32 Network Digests To: berger@datacube.UUCP Subject: Re: IBM-PC <-> VAX <-> MAC network? Bill Joy of Sun recently spoke in Boston and said that U Michigan is porting NFS to the Mac... Ric Ford ------------------------------ From: MACINTOUCH (14972) Subject: RE: APL for the Mac Date: 18-NOV 20:41 Network Digests To: dave@bucket.UUCP (Dave Munroe) Subject: APL for the Mac Although I'm not an APL hacker, Portable Software's APL for the Mac seemed to me to be well done. They are at 60 Aberdeen Ave., Cambridge, MA 02138; 617-547-2918 (as of May 1986). Ric Ford "MacInTouch" newsletter ------------------------------ From: DDUNHAM (14982) Subject: RE: APL for the Mac Date: 19-NOV 03:49 Network Digests > From: dave@bucket.UUCP (Dave Munroe) > Subject: APL for the Mac Quote Quad is the newsletter of the APL SIG of the ACM. David Dunham "A mind is like a parachute. It only functions when open." Maitreya Design ------------------------------ From: DDUNHAM (14983) Subject: RE: ZoomWindows...Help Wanted Date: 19-NOV 03:50 Network Digests > From: jdm@ut-ngp.UUCP (Jim Meiss) > Subject: ZoomWindows...Help Wanted > Is it possible to set the size that the window zooms to? Yes. I recommend setting it to the smaller of the paper size or the screen. This way if you have a FPD or Megascreen or E-machine, you won't get obscenely large windows. if (((WindowPeek)window)->dataHandle != NIL) { pr = &(**(WStateData **)((WindowPeek)window)->dataHandle).stdState; pr->right = pr->left + WIDTH; b = pr->top + HEIGHT; if (b < pr->bottom) pr->bottom = b; } BTW, both your ZoomWindow calls use inZoomOut... David Dunham "A mind is like a parachute. It only functions when open." Maitreya Design ------------------------------ From: MACINTOUCH (14989) Subject: RE: Mac fan comparison Date: 19-NOV 13:49 Network Digests to: tod4@sphinx.UChicago.UUCP (phillip a todd) Subject: Mac fan comparison Thanks to the information you posted about Mac fans, I bought a System Saver Mac from MacConnection (for $65). I'm very pleased with it, finding it quiet and very well designed. The extra outlets and switches are useful (and gave me a spare Mac cord back). Ric Ford "MacInTouch" newsletter ------------------------------ From: BRECHER (14984) Subject: Re: mac video - drawing into ScreenBits Date: 19-NOV 04:28 MUGS Online To: alen@ttidca.UUCP (Alen Shapiro) Subject: mac video - drawing into ScreenBits fast? > Is there a way to retarget the video display mechanism to read from > other areas of memory [besides the alternate screen buffer]? No. > ... my own very fast 3 line 68000 code memory-to-memorymap copy function... Have you tried unravelling the loop, i.e., generically speaking, replacing Move #longs,Dn @0 Move.L (An)+,(Ay)+ Dbra Dn,@0 with Move #longs/x,Dn @0 Move.L (An)+,(Ay)+ ;1 Move.L (An)+,(Ay)+ ;2 ... Move.L (An)+,(Ay)+ ;x Dbra Dn,@0 Even with a fully unravelled loop -- which would take more than 40K of code -- you might not make 60 frames/sec. Move.L (An)+,(Ay)+ is 20 cycles nominal, and the Mac's effective CPU rate is somewhere around 6MHz. ------------------------------ From: NANOCHIP (14991) Subject: PCPC's HFS Backup Date: 19-NOV 18:33 Hardware & Peripherals Dave> Is the updated version of HFS Backup available yet? I've got <1800 files on my Micah20AT and am no longer able to use version 1.1B5. I had talked to someone (Tammy?) at PCPC a few weeks ago and they told me the new version would have a file limit of ~5k (V1.1 has a limit ~1.4k files)). Is this the correct poop? When I first got the "Directory Full, File 'Foo' could not be copied" ALRT I assumed "Foo" was a bad file and deleted it, thinking it had caused my problems ( I didn't realize there was a FIle # limit ). I found the ALRT and the subsequent sorting of only the first 1400 files to be a little confusing. Also, ( as Raines Cohen once suggested) an option to back up files in a non-compressed format would be a nice feature. In the meantime... I once saw (Usenet Digest?) a PCPC public domain program which could extract individual files from HFS Backup disks. Will this be included on the update disk, or could you upload it here in the database? HFS Backup enjoys a strong following ( myself included :-) ) here on ICONtact and I'm sure it would come in handy for some members. Thanks! <Chip ------------------------------ From: DAVE99 (14993) Subject: RE: PCPC's HFS Backup (Re: Msg 14991) Date: 19-NOV 20:59 Hardware & Peripherals HFS Backup 2.0 is in the final stages of beta test, and should be released by 1 Dec. We've learned a lot in the past year, and much of that knowledge was put into the new version; I think you'll like it! Backing up in a non-compressed format is very contrary to our way of thinking these days... as a matter of fact, the new Backup puts one large BackupFile on each disk! I'm told there is a program called Mega Copy which does just that tho; haven't used it myself, but others can probably help you there. The new Backup also has a Recover option, which allows you to extract files even if the BackupDir and Copy of BackupDir have been lost, erased, run over, etc. It's not something you'd want to do all the time, but it's nice to know it's there. Thanks for the feedback, and the interest! -Dave W ------------------------------ From: MCOHEN (14998) Subject: RE: PCPC's HFS Backup (Re: Msg 14993) Date: 20-NOV 00:40 Hardware & Peripherals Will the new version be available to registered owners? Is it any faster than 1.1? I'm starting to get very frustrated with HFS Backup, almost to the point of switching to FlashBack (which my company uses; it seems much faster). Last night, it took me 40 minutes & 10 disks to backup a little over 8Meg of data from my HD20. One other slight thing I would like to see changed is the ability to do a 'changed files only' backup of selected folders only. When I do a backup of one folder (as I often do with my source code), doing a 'changes' backup will scan all folders and backup anything that isn't on the original backup set. - Mike ------------------------------ From: LOFTUSBECKER (14994) Subject: RE: Font questions Date: 19-NOV 21:12 Network Digests To: "Bob Soron" <Mly.G.Pogo%OZ.AI.MIT.EDU@XX.LCS.MIT.EDU> Subject: Font questions Yes, there is an FKEY like Other... that lets you temporarily use fonts not installed in the system. FONTSIE 1.51 works with many (but not all!) programs that have a "Font" menu, and with some ("sort of" with Word 1.05, very well with Acta and miniWriter) that don't. A beta-test version ("Special Fontsie"), somewhat less user-friendly, works with more programs, including MacDraw and Write/Now, but you have to make your additions before running the program. If Jeff wants to upload them from Delphi, he's welcome to (i.e. he can for this purpose break the restriction on further distribution of the beta-test version of Special Fontsie). Lofty [ Fontsie is available in the INFO-MAC archives. At this time I do not think it is appropriate to upload Special Fontsie. - Jeff ] ------------------------------ From: MACINTOUCH (14995) Subject: RE: HD vs. Floppy: $/Kb Date: 19-NOV 21:55 Network Digests To: harvard!ci-dandelion!carl@seismo.CSS.GOV (Carl A. Dunham) Subject: HD vs. Floppy: $/Kb It seems to me that this comparison would be a lot more valid if you included the price of the floppy _drive_ in your equation. Alternatively, you could compare the price of floppy disk media, per megabyte, with the price of Bernoulli Box media, per megabyte. Furthermore, there are 10MB Paradise drives selling for under $500, and Bernoulli Box 5MB units are listed at $499 from Icon Review. These approach the cost of an external floppy drive with a quantum improvement in capacity and performance. How about including in the equation the cost of the amount of RAM you're using for your RAM disk, if it were bought as an upgrade? Ric Ford "MacInTouch" ------------------------------ From: DDUNHAM (15003) Subject: RE: HD vs. Floppy: $/Kb Date: 20-NOV 03:38 Network Digests > From: harvard!ci-dandelion!carl@seismo.CSS.GOV (Carl A. Dunham) > Subject: HD vs. Floppy: $/Kb Yes, but cost/K of storage isn't the whole picture. How about multiplying access time (from Steve Brecher's DiskTimer II) to get $seconds per K? Some us may not be as patient as you are. ------------------------------ From: MACLAIRD (15004) Subject: RE: HD vs. Floppy: $/Kb Date: 20-NOV 04:17 Network Digests > Re: HD vs. Floppy: $/Kb There are two advantages to a hard drive. The first is convenience - that is, to both avoid the need to flip floppies and to keep the programs and maybe data that is frequently used in an always accessable location. The second is speed, which has some convenience value but really only has a true purpose in a production/commercial situation. Your point is very well taken. Hard disks are unsuitable as a repository of static information (as anyone who has had one go South can verify) and anyone who is buying a hard drive for that purpose is spending extra money for some- thing he really doesn't want. Laird ------------------------------ From: HSTARR (14997) Subject: SCSI pinouts Date: 19-NOV 22:58 Hardware & Peripherals Does any good person out there KNOW where the SCSI terminator gets its 5V rail from; I thought it was pin 26, but that just hangs my SCSI bus. Running without an energised terminator seems to be causing a lot of flaky's. I am using a lash up of drive + power supply + flat ribbon cables. Any help will be well appreciated. Harry ------------------------------ From: MACLAIRD (15005) Subject: HFS Blues Revisited Date: 20-NOV 04:19 Bugs & Features Just in case you though HFS funnies were confined to the Hard Disk 20 side of things... I noticed an 800K diskette all of a sudden grew to "779K used, 0K available" as indicated in the Finder, so I copied it onto a fresh one - which showed up with "778K used, 7K available". No crashes or intermediate giggles, but it was a work disk. Did you know that about a year ago, when a woman called WBCN with a talking Mac, Karlos 4K asked her "what size are your floppies?" It must have been one of those rare days when Josh was absent from the keyboard... Laird "A Waist is a terrible thing to mind" ------------------------------ From: MACLAIRD (15006) Subject: RE: mac video - drawing into ScreenBits fast? Date: 20-NOV 04:22 Network Digests To: alen@ttidca.UUCP (Alen Shapiro) Subject: mac video - drawing into ScreenBits fast? As usual, Steve BRECHER beat me to the punch, but I think I have a few comments to supplement his: I haven't tried drawing into the screen buffer (YVES has, for one..) but I can offhand think of a shorter and probably faster blit. [so did Steve] LoopLabel: MOVE.L (A0)+,(A1)+ DBF D0,LoopLabel The .L is there in the move to move four bytes per instruction; this should be better than the MOVE.B that I often see in print, as the 68000 will not have to fetch the instructions as often. You'll have to pre-calculate the counter DO as one-fourth of the number of bytes to move, and fudge for the excess, perhaps with FinLabel: CMPA.L A0,lastByte BGT.S EndLabel MOVE.B (A0)+,(A1)+ BRA.S FinLabel EndLabel: or something like that. Again, I haven't found the need to draw directly into the screen, but I have to believe that it's not so hard. BlockMove() runs out of ROM, which isn't interleaved with video memory, so you might trace through and see what his code is like. Is AppleTalk or a hard drive on your Mac? One more thing: check the hardware chapter of Inside Macintosh (Vol. 3, pp. 18-20). It describes the need to coordinate with the vertical blanking interrupt to achieve smooth animation of images. You can't do it all during vertical retrace, but you can apparently get enough of a head start to keep ahead of the raster. Otherwise, you might _appear_ to be achieving 30 Hz. Also, you might have to blit just changes to the screen. I have heard that the author of Vanlandingham tried to blit the images and to have digitized sound as well - but was unable to combine them. I wonder what the Macintosh 128K would have included if the 68000 had just a little bit more power; the hardware designers apparently squeezed it for just about all it was worth. Laird J. Heal (MACLAIRD) ------------------------------ End of Delphi Mac Digest ************************ -------