Open Apple Vol 6 No 09 OCT 1990

A2-Central 2 A journal and exchange of Apple If discoveries formerly Open-Apple Es October 1990 Vol.6, No.9 ISSN 0885-4017 newstand price: $2.50 photocopy charge per page: $0.15 Miscellanea IBM has launched an assault on the home market Apple has long ignored. As Tom Weishaar mentioned in “Personnel moves rock Apple” (March 1990), Apple has repeatedly said it believes that there is a business segment and an education segment to the personal computer market, but that there is no distinct home segment. Apple believes that home computer sales fall out of those education and business sales. This month IBM’s new model PS/! started appearing in stores and reports started appearing in the press. It is hardly state of the art: a 1OMHz 80286 processor in a system that includes a 1.44 megabyte floppy disk, 2400 baud modem and mouse. The basic system is cur- rently sold in three configurations; the floppy-based monochrome sys- tem ($1000), the same system plus a 30 megabyte hard disk, and a color system with hard disk ($2000). In deference to the first-time buyer, 1BM has left the configuration decisions for the basic system simple: decide if you want a hard disk and decide if you want a color display. Those are the only choices you have to make. The hardware is not imposing, but the implication is: Bill Machrone of PC Magazine reports in his August editorial that IBM believes there are 66 million households that can afford this type of system. That kind of potential has apparently dragged IBM salivating into a market- ing strategy aimed directly where Apple isn’t: presenting systems as attractive for a home market that is not directly tied to education or business. Toward that end, IBM has tailored the PS/I to be easy to use. You bring the system home in one box; the setup manual is packaged already opened to the page telling you how to set up the system. You attach the monitor, mouse, and keyboard, then run the power cord to the wall and the modem phone line to a phone jack. Then turn it on; the system prompts you from there, and help is always avaitable. One of the options the machine automatically gives you is support via the Prodigy information service (which will be expanded nationally). Another is a limited MouseTextlike version of MicroSoft Works, which is a clone, right down to the name, of AppleWorks. The version bun- died with the PS/1 includes a word processor, a spreadsheet, a database, and a communications module. These make the system immediately useful. {BM has even found ways to get around callous dealers interested in moving product but not helping entry-level customers. The units will be sold primarily through Sears retail outlets; IBM is putting this machine in the path of potential customers who woutd never enter a computer store, Rather than clepend on dealer salespeople, IBM also offers PS/! customers an 800 number available for basic technical questions. We played with the units at a local Sears outlet; the system design prevents the user from ever seeing the intimidating MS-DOS com- mand-line mode by accident (you can get there on purpose if you like), The system demo thal was running on one of the systems did an admirable job of presenting attractive features of the system for home. users; of course, much was made of the fact that IBM was bringing its office environment expertise “home”. That IBM logo carries weight that will sway customers hesitant to buy an “off brand” MS-DOS com- patible. But the sad thing is that an Apple llgs setting next to the PS/I would be very competitive. The animations running on the PS/I looked stilted. The sound quality, even with the optional audio card, doesn’t even get off the ground compared to the llgs. The screen has an annoying flicker when switching modes. The price is less than a comparable IIgs system, but then the ligs pricing is set to allow for dealer discounting (it's unlikely Sears’ pricing structure allows for dickering). And the pricing gap is close when you consider the capa- bility and expandability built into the IIgs that costs extra on the PS/1 (expansion slots, sound capability, and so on). Apple could respond to the competition by bundling AppleWorks or AppleWorks GS (or both) with the Ilgs, particularly now that Claris is back under Apple's total control. Appie could also have used Apple- Link Personal Edition as its on-line support vehicle, though it's too late for that now. Apple continues to believe technology alone can cary the company; yet seems to self-destruct every opportunity to make it palatable to the public The big disadvantages the Apple IIgs has compared to the PS/1 are distribution and software availability; both are closely tied to the myopic way in which Apple has marketed the Apple Il line since the introduction of the Mac. Despite the fact that Apple's customers have clearly stated their desires and despite a four-year lead time, IBM has beat Apple to the market that the IIgs is perfect for. | hope Apple finds the results truly embarrassing. Apple has announced a price drop on the Mac Ilci. Price reduc- tions of up to 20 per cent were announced for various configurations. That leaves the least expensive Mac Ilci system price at $5,969. Meanwhile, rumors of the new “low cost" Mac systems to be introduced in October seem to be converging on three models. The Mac Classic will be a replacement for the Mac Plus and Mac SE coming in at about $1,600 retail. The Mac LC will be a 68020-based system that accepts a color monitor and a future Apple {1 emulator card. And the Mac ilsi will be a 68030-based 16 MHz color system predicted to come in at about $2800, We'll see for sure in October. This has been the month of Iigs teasers. Coming on the heels “WE FIGURE THE EQUIPMENT MUST HAVE SCARED HER AWAY. A FEW DAYS AFIER ‘SETTING UP, LITTLE SNOWBALL’ JUST DISAPPEARED,” 6.66 A2-Central of demonstrations at the A2-Central Summer Conference, we have seen several new programs written by a nucleus of true IIgs gurus, In many cases the programs are “under construction’, but show that the ligs is as (or more) powerful than many of us have believed. First to be seen was Modulae, a demonstration of three dimension- al animation from the FTA (Free Tools Association), a group.of IIgs hackers based in Dijon, France. FTA accrued fame via their previous Nucleus and “Space Harrier” demos as well as the shareware disk util- ity, Photonix. Like their earlier productions (and somewhat to our cha- grin), Modulae eschews use of the Ilgs toolbox to the point where it is incompatible with GS/OS; it must be booted from its own disk and ~ requires booting the system again to return to GS/OS. On the positive side, both the ROM 01 and ROM 03 ligs versions are supported. Modulae cycles through an introductory screen and demonstrations of star field effects, wire frame objects, objects composed of spheres, star flight effects, and use of the Iigs graphics “fill mode” to draw solid objects and in calculated animations. In the final sequence, your vantage point is from the perspective of flying through a maze with colored walls; suddenly you are teleported outside the “maze” to dis- cover thal you are flying around and through the FTA logo, which draws back and presents itself in a horizontal roll. Catchy music is flowing from the IIgs during all of the demos, and (by turning off the “automatic” display mode) you are allowed to play with the rotation and sizing of the objects. Yol. 6, No. 9 Two other striking demos are from a new company still look- ing for financing called DreamWorld. (Contact Jason Anderson at 319-354-7959.) The first shows DreamGraphics, a GS/OS paint pro- gram. The second demonstration, DreamVision, is not GS/OS based and boots from its own disk. Both of these are only demos currently. DreamGraphics is designed to create 16, 256, or 3200 color images. The 256-color {with some placement restrictions) mode is possible in the ligs's 320x200 Super High Resolution graphics by using “dithering,” a technique in which colors are mixed within a pal- tern to give the impression that the pattern represents a conglomera- tion of the two colors. Dithering is analogous to the use of small, closely placed colored dots to generate color newspaper photographs. The simultaneous display of 3200 color images is possible in the Ilgs. but it isn’t easy. The llgs SHR display includes a series of 16 palette values that define available colors, and screen control bytes that determine which of these palette values are used for each of the 200 lines of the display screen. By doing some carefully timed fancy footwork as the IIgs video hardware draws the display screen to your monitor, the 16 palette values can be changed as the screen is being drawn so that each line can have a distinct palette. 200 lines times 16 colors {with palette switching) available for each line gives 3200 avail- able colors. However, changing the palettes “on the fly’ like this consumes almost all of the Ilgs’s processor time just to display the picture; The ligs promise The latest issue of A2-Central prompted me to write and add my bit of confusion to the Ilgs/Mac controversy. | am an engineering man- ager for a large aerospace corporation and nat- urally we depend heavily on computers, ranging from mainframes to “personal computers”. Over the past four years Macs have been creep- ing into the engineering organization, due partly to my urging and example. I have had an Apple Il since about 1980 which I use to provide home analysis and inde- pendent review of the work provided by my engineers. For engineering applications, 1 need the ability to handle large arrays of data, and a good math package. | have been “making do” with Applesoft by virtue of the various compil- ers which have slowly come on the market. My magnus opus was an industrial grade perfor- mance program aggregating 40,000 lines of Applesoft in eight separate but related pro- grams, one of which had 198 variables. Last year when the time came to replace my trusty Apple Ile, I carefully considered the Macs (which I could purchase through the company for a significant discount}, the Iigs, and those “blue” things. | finally bought a Ilgs for one rea- son and one promise. - The reason is that I like the TimeOut enhanced AppleWorks word processor better than any word processor that | have used, including those on the Mac. It does not have all of the bells and whistles but it does about 90% of the things | want to do, The promise was that of ORCA/C. | have been looking for a good compiled language for the Ile/llgs family and this seemed to be it. 1 therefore elected to purchase a IIgs and soup it up with a hard drive, a TransWarp GS card, and alot of memory. Just as | had made the decision to purchase the IIgs Apple announced System 5.0, so | was able to wait a-week and buy one of the new models. | now find that | have the best of both the text-based and graphics-based worlds. I use the text-based approach for word processing and technical programming. I use the graphical interface for desktop publishing, | nearly have my dream machine. The one additional capability that 1 would like to have on the Ilgs system is a good techni- ca! plotting program. Graph-lti and TimeOut Graph won't hack it. A not-widely distributed DOS 3.3 program cailed Scientific Plotting Package by CMI-Cascade comes close but the company is out of business and the package is no longer supported. The only other contender is the plotting capability of SuperCatc 3a, but tt is somewhat cumbersome to use, and needs a patch which I don’t have to make it work on the IIgs. 1 took with longing at KaleidaGraph and DeftaGraph on the Mac and wish that we had approximately the same capability on the II family. Now in your previous issue you mentioned that someone demonstrated a 640x400 pixel screen at the A2-Central Summer Confer- ence. How do we get the message out to Apple that this is needed!? This Mac-like display capa- bility should be the impetus for some good pro- grammer to do the plotting packages [ have been needing. ‘Nother subject. I would like to send kudos to Mike Westerfield and his group of Byte Work- ers for the continuing development of ORCA/C. I have been treated most courteously when [ bring up what I consider a problem to them in spite of being an amateur when it comes to C. The new version 1,1 release of ORCA/C is a vast improvement over the past one. And speaking of kudos, A2-Central deserves ‘one in my book. I always look forward to the artival of the monthly issues. Please keep up the good work. Sidney A. Powers Valencia, Calif. The problem with Apple’s repressive Apple It marketing “strategy” is that they have forced most Apple If software companies to market almost entirely toward educational institutions. For many companies, this is the only venue they advertise in, Educational companies such as Scholastic do almost all of their sales via direct mail; that is, many of their products are not sold through national marketing agencies and may not show up “on the rack” at your local software store, Those with specialized needs like yourself should wander through a directory like the MENU Information Directo- ty to check into possible software sources. Meanwhile, we can supply you a copy of the SuperCalc patch to allow it to print on an ImageWriter from the 'lgs.—DJD More SuperCaic memory If anyone has a patch to allow SuperCalc 3a to recognize additional memory in the igs, it would help me very much. SuperCalc 3a recog- nizes up to 512K on an Apple Ile auxiliary slot memory expansion card, but won't recognize my IIgs memory expansion card. Paul T. Hogan San Diego, Calif. This is one that we don’t have an answer for curently.—DJD ProDOS 8 bad blocks Will ProDOS 8 write information on bad blocks of a hard disk drive? I am concerned about a program writing information to a bad block and losing important information. Douglas Bailor Spokane, Wash. The occurence of a bad block and its affect on your file conients is timing-critical, from the time ProDOS has decided which block it is October 1990 some of that time has to be stolen back to allow your paint program to function. It appears from the demonstration shots in the Dream- Graphics demo that the program compromises by letting you see and modify a subset (a horizontal band) of the image in 3200 mode, allowing you to gradually work your way through the entire image. It’s a reasonable solution, and we hope the 3200 color image format and display techniques can be standardized (so far, not all images and dis- play utilities are compatible). DreamVision is more mysterious. The demo boots into a screen simulation of a video monitor and what appears to he a video editing controller (a device to allow you to cut, paste, and copy video sequences}, The monitor then erupts into a brief sequence of a video recording of a tap dancer with with synchronized sound; the effect is spectacular, as if the IIgs had become a digital video tape recorder. Then an information screen is presented announcing that DreamVi- sion is coming, and the demonstration loops. Some people like to say the Apple II is “ten year old technology”; put this demo next to 1980's AppleVision, with its stick-man dancing to asimple tune on a simulat- ed TV screen, and you'll see what a difference ten years has made. The last demonstration can also be launched from GS/OS; it’s called GS Experience and displays a very impressive three-dimension- al rotating “2” that author Tim Meekins created (“rendered”) using a superminicomputer and then transferred to the llgs. The object appears to leap out of the screen at you, even reproducing the A2-Central 6.67 changes in reflected light that were recorded during its creation. American Printitg House for the Blind has announced a llgs version of TEXTALKER. The {\gs version can run in the background, permitting its use with “friendly” copy-protected programs as well as many non-copy protected programs such as AppleWorks, Copy II Plus, and the Apple System Utilities, The program installs as a ProDOS 8 application, but can remain system resident to work with other operating systems such as DOS 3.3 and GS/OS text applications. It requires a Ilgs with at least 512K bytes of RAM, a 3.5 drive, and an Echo (or DoubieTalk, in Echo mode) speech synthesizer. The cost is $54.50 {catalog number D- 8957 1-00}, We gleaned this tidbit from the Raised Dot Newstetter from Raised Dot Computing, 4088, Baldwin, Madison, Wisc. 53703, 608-257-9595 or FAX 608-241-2498, a newsletter covering microcomputer products for the visually impaired, Subscriptions are $18/year in large print, or $20 per year on audio tape {add $20/year air mail outside North America). Checkmate has resurfaced under a new name: Micro Memory. The new address is 7655 East Gelding, #B1, Scottsdale, Ariz. 85260, 602-998-0227. Former Checkmate president, Earl North is still presi- dent. going to write to until the data is. actually writ- ten to the disk. Before ProDOS can write a block to the disk, it has to locate the block. It does this by read- ing the block header (an “address” on the disk that identifies the block) as it moves the disk drive read/write head over the disk. If the header is unreadable and can’t be found, Pro- DOS wil! report an error. Once ProDOS has found the header, it tries to write the block. Looking at the code that does this for a 5.25 drive, it doesn’t appear that the driver verifies that the data was actual- ly retained by the disk. However, assuming that the disk had been previously formatied and the header was stil! readable, the odds are very good that the data is there. ProDOS repeats this for every block in the file it is trying to write. If no unmanageable errors occur, your file is on the disk intact. tf not, ProDOS will return an error and your pro- gram should warn you that the file was not saved. However, that does not mean that the file will necessarily be intact the next time you read the disk. Magnetic fields or physical dam- age to the disk, among other things, can cause the data to fade. There may simply be a “weak spot” in the disk’s magnetic coating; not weak enough to cause the write to fail, but too weak to retain the data written there between uses. You can lessen the chances of some of these problems by keeping the disks away from hazardous environments, and (for fop- pies) by noticing which brands of disks seem to be the most reliable for you and sticking to that brand where possible. This won't remove absolutely 100% of the risk (there’s always the sinall chance that the disk will decide to forget what it Knows immediately after you write to it), but you can get pretty darn close. Other than that, the only way to make sure you don’t lose data Is to make sure you have a backup of every file you consider to be impor tant. If you fail to follow that simple rule, there is no disk recovery program, virus detector, or other utility on the market that can help you. Period.—DJD Wrong number [have problems in using the new ProDOS 8 v1.9 in conjunction with GS/OS 5.0.2. 1 renamed the PRODOS file to “P8” and copied it into the System folder. When | then launched a ProDOS 8 application with my ProSei selector, | got the message “incorrect ProDOS version”. Can you: help, or is the new ProDOS not com- patible with GS/OS 5.0.27 Gunther Unger Enimench, West Germany There seems to be confusion in the concept of ligs System Software, GS/OS, and ProDOS 8. The gs System Software is the entire pack- age of GS/OS; ProDOS &, the tools, drivers, and So on. The most recent version of this is 5.0.2, GS/OS is the 16-bit native operating system of the IIgs. GS/OS itseif is responsible for com- municating with devices while the ligs is in native (16-bit) mode and for managing the rest of the components of the system software (loading tools, for example}. The version of GS/OS supplied with 5.0.2 fs version 3, but there are several minor components al! with their own version numbers, For example, in the change from System Software 5.0 to 5.0.2 the version number of the supplied ProDOS FST changed from 3.0 to 3.01. ProDOS 8 is responsible for communicating with disk devices from the 8-bit “Ne emulation” mode of the ligs. ProDOS 8 is also a compo- nent of GS/OS; GS/OS launches ProDOS 8 in order to run an &-bit program. It is here that the version number of the ProDOS 8 “P8" file comes to matter to GS/OS; it needs to check the version number to verify that a version-spe- cific patch can be made to P8’s quit code so that it can return contro! to GS/OS . if you change the version of P8, it is no longer recog- nized as the “correct” version by GS/OS. I've always recommended using the latest version of the system software for your system (barring contrary advice from Apple, which has happened only rarely in the last few years). [ should now clarify that in saying “system soft- ware* | am referring to the complete system software environment; individual components shouldn‘t be modified without Apple’s recom- mendation, In other words, figs System Software 5.0.2 is the current release version for the ligs, and you shouldn't modify any component files except per approved updates, This can be a modification to the operating system as released by Apple (the Apple High-Speed SCSI card comes with a minor revision to the Iigs system software used to support new features of the card) or a driver supplied by a third party in accordance with Apple's documented methods of enhancing the system software. However, changing files arbitrarily (even with Apple’s own files) may not work to your advan- tage. The new ProDOS 8 system disk works fine by itself. But don't use it to “update” the ligs System Software: Apple has been releasing those updates independently.—DJD Giving it the boot | have an Apple Ilgs, | would like to know how to configure a program disk so that, when Tun, it will bypass the finder. | am trying to run an application from the desktop, but } would also like to know how to do this (from the beginning). I have at least two applications that | launch from the desktop. To get there, | must use my system disk. Immediately after arriving at the desktop, | remove the system disk and insert, say, PaintWorks Plus. | then double-click on the PaintWorks Plus application icon and the pro- gram loads. During this process, | am continual- ly asked to swap disks from the system disk to the PaintWorks Plus one. | know that I have only one drive, but | would like to know why all this swapping is necessary and if there is a way to avoid it? John Guggenheim Bryn Mawr, Penn. There is no real magic to avoiding disk Swaps. Unless the operating system software (GS/OS} and the program itself wili fit into the 800K of disk space on current ilgs 3.5 drives, you must swap disks once or twice while 6.68 A2-Central starting a GS/OS application on a single-drive system. The ligs is not unique in this; both the Mac and MS-DOS machines are in a similar quandary. To see if the program might fit on a single disk, try the following. Make a copy of your usual startup disk, and use the ligs Installer (on the /System.Tools disk) to configure the disk with the minimum features you need to run the program at issue. There are no hard rules as to what items may not be needed, but common sense should dictate some; for exam- ple, a game that doesn’t allow you to access the printer should indicate that you won't need a printer installed. When you've stripped every- thing that you know isn’t needed, boot your new startup disk, Now insert the disk with your program file and launch the program. Finder will “unload” itself and launch the new program. The pro- gram may require some of the ligs tools that are not currently in memory, requiring the first Swap back to the system startup disk. Then the program may have additional portions of itseif to load, requiring a switch back to the program disk. In most cases, those two swaps should be adequate to get the program going. If the program does launch and run, it indi- cates that the program may not require files to be added to the System folder of your startup disk, which greatly simplifies the detective work in finding out what files need to be added to your startup disk. Quit the program (more disk swaps) and boot your normal start- up disk. Insert your program disk and make note of the files that appear in the root (volume) direc- tory of the disk that do not appear to be among the files found in the directory of the startup disk you created to use with that pro- gram (PRODOS, BASIC.Launcher, BASIC.Sys- tem, System, AppleTalk, and Icons). Insert the new Startup disk you've made for your program and, unless they appear on the program disk, drag the files BASIC. System and BASIC.Launcher into the trash. Now open the System/ folder and drag the file named START (which is actually Finder) into the trash can. Empty the trash, and note how much free disk space you have left. Eject the startup disk (don’t drag it into the trash; use open-apple-E or the drive’s eject button so that the dimmed image of the disk remains on the desktop. Insert the program disk, select the files and folders that don’t appear on your startup disk. To determine whether the files will fit, you can use the “Icon Info” item of the Finder's “Spe- cial” menu (for folders, the “Calculator” icon in the info box will sum the sizes of the files the folder contains). If there appears to be enough space, drag them onto the dimmed startup disk image. Follow the commands to swap disks until Finder has copied the program disk files onto the new startup disk. Now check the startup disk to see if the pro- gram file’s name ends in *.SYS16*. If not, you'll need to rename it; for example, you may have to rename the file “MyProgram’ to “MyProg.Sys 16”. Once the file is renamed, try booting the startup disk. Either it will work, or it won't, If it doesn’t work, and you received a “vol- ume full” error message during the copy, then there isn’t enough room on the 800K disk for GS/OS and the program. All you can try to do is remove files you believe the program doesn’t need. Once you've removed as many elements as you can manage with Installer, you can try deleting other “flexible” compo nents of the startup disk’s system software. Here are the minimum file requirements need- ed to boot system software version 5,0.2 (ref erence: Apple Il GS/OS Technical Note #1: Contents of System. Disk and System.Tools): Prefs Systex/ ‘Start.65.05 68.08.Dev Exror. sq PSts/ PrP? Ghar FSt Drivers/ Appledisk3.5 required for Apple 3.5 drives Console.Driver required for ProQ0S disks System, Setup/ (DEV. Init ——requized for Control Panel XDA Took. Setvp 182 RM OL tool setup 183 204 03 tool setup Resource. Mgt Sys.Resources Cevs/ 8 per your system configuration Desk. Aocs/ CelPanel 3A Sppressload Tools Ponts Fastfont speeds up Shaston 8 % required for FroQ0S 8 BASIC.Systen required for AppleSoft BASIC BASIC. launcher required by Pinder for AppleSoft. Comments indicate special considerations. Italicized items should be included unless infeasible. Here are some further comments. The Control Panel NDA program is contained in the CtlPanel.NDA file and uses the CDev (Control Panel Devices) files as data; CDev.INIT loads the CDevs during boot initialization. If CDev.INIT is included the CDEVs folder must be present. if you plan to delete the Control Panel NDA (and. consider that this is the only way to select certain functions of the system; the Alphabet, printer, and AppleShare selec- tion protocols are not available in the Control Panel CDA) you should delete CDev.INIT, the CDevs folder, and the CtiPanel.NDA. If you are running a program that does not present you with an Apple menu where you can access the Control Panel NDA, these files are expendable. ExpressLoad is a system utility that makes it possible to joad files in a special compatible format much faster than normally; it should be retained if at all possible. Without it, the files will still load, but more slowly. The use of files within the Fonts and Tools folders are determined by the program. The best rule of thumb is to see whether the pro- gram you are using comes on a bootable disk; if so, open these folders on the program disk and see what files are present. Leave those files in place in the corresponding folders on your startup disk, FastFont is a special form of the default system font (Shaston 8) that allows faster drawing of the font on screen; it should be retained if possible even if your program disk’s Fonts folder is empty. The remaining folders of interest are Desk.Accs, and System.Setup. if you've started with your “minimalized” startup disk, the con- tents of these folders should approximate the minimum configuration suggested above. Vol. 6, No. 9 Apple’s tech note does not recommend it for disks developers are preparing (which need to be compatible with expected system configura- tions), but if you are creating a disk that will be booted only on your system you can omit the tool setup file not specific to your ligs ROM version. Specifically, if you are a ROM O1 Iigs user you Can delete T33; for a ROM 03 ligs you can delete TS2. Once you have done this, the disk will not boot on the “other” system, These are all stop-gap measures; at some time; the Ilgs’s 65816 can theoretically address 16 megabytes as easily as the 6502 in earlier Apple if models can address 64K (the igs is restricted to using less than 16 megabytes of RAM by system design con- straints). Just as Apple il programs such as AppleWorks and Pascal outgrew a single 140K 5.25 disk, ligs programs are fast outgrowing the use of a single 3.5 disk. The solution is to add more drives, especially a high-capacity drive such as a hard disk. For the current crop of 16-bit (and larger) machines from al! manu- facturers a hard disk is becoming a near necessity.—DJD Postscript downloads I am able to create a PostScript file from Publish-t! 3.0 and get it over to a Mac with a LaserWriter NT to successfully print text and Publish-It! line art, However, | get a PostScript error when the PostScript contains any graphic images. Also, the printouts are such that | can’t get anything to print in an area one-half inch from each paper edge. Is this a limitation in the LaserWriter printer or do | have to do something extraordinary? Michael Gooding Chesterfield, Mo. We talked to Publish-It! guru Bruce Rosen- blum and he clarified some of the “features” of the PostScript file generation. Most programs that print via “PostScript” actually download a PostScript program called LaserPrep that contains subroutines (written in PostScript) for common opera- tions. For example, there is no inherent PostScript command to draw a rectangle on the virtual page, but a subroutine named “rc” contained in the LaserPrep file will draw the figure. When the computer wants to draw a rectangle as part of the printer output, it can then use the “rc” command. One problem that can crop up Is trying to Install incompatible versions of LaserPrep. Publish-It! 3.0 ships with a version of Laser- Prep that is compatible with the version used by the ligs System Software 5.0 and Mac Sys- tem Software 6.0. When creating your PostScript document in Publish-It! 3.0 by printing to a disk file, elect to include Laser- Prep in the disk file when you know the Laser- Writer has not been initialized (that Is, no one else has printed to it since the LaserWriter was powered upj, and don’t include LaserPrep when you are going to be downloading the file to a LaserWriter that has been initialized. If you are going to print a file to the Laser- Writer via downloading, make sure you have at least one-half inch borders; the PostScript output won't support printing closer to the edge of the page than that (the “Larger Print October 1990 Area’ command in the Mac’s Page Setup won't result in a larger print area for your “preformat- ted’ document file). The absolute limit is about one-quarter inch in any case. If your original document is formatted for an ImageWriter, you may find the page width defined at 8 (rather than 8.5) inches with one- quarter inch borders; to prepare the page for the LaserWriter, widen the page width to 8.5 inches and center the print area within one- half inch borders. You should be aware that EPS (Encapsulated PostScript) files are different than the down- loadable files that Publish-It! creates, so you can’t print a file to disk and then import it as an EPS file. Another problem you may run into at a ser- vice bureau is in getting them to handle an Apple i! disk when all they have is Macintosh- es. That's relatively easy, just be sure to bring your PostScript file on an 3.5 ProDOS disk and ask for a Mac that has Apple File Exchange installed. You can use Apple File Exchange to import the file from the ProDOS disk to the Mac via the text translation option. All that having been said, we have run into some instances where a Publish-It! file con- taining an SHR graphic would generate a PostScript file that the LaserWriter gagged on. We've sent an example to TimeWorks to look at.—DJD Yummy? Ackk! My daughter says the gum on the envelope should be apple flavored. Jerry Giaccai Canton, MA Given the typical response we've gotten to the “yummy” glue, if you made her lick the envelope we‘re surprised she’s still speaking to you.—DJD Gliding mouse Having to clean my mouse’s ball once a week is a real pain. Do you folks know of any- one that makes an ADB compatible optical mouse that works reliably for the Apple [lgs? For anyone else having the same problem, | find the best way to freshen a dirty ball is to do the following: 1. Open the mouse, remove the bali and clean it with alcohol, per the user's manual instructions, 2. Clean your mouse pad with alcohol. 3. Pour a small amount of alcohol on your mouse pad and play your favorite arcade-style game for about 15-20 minutes. 4. Wipe any excess alcohol from your mouse pad and mouse. This may sound strange, but it works fairly well. For the faint at heart, fear not; you won't get shocked or ruin your mouse doing this, Alcohol and Freon are commonly used to clean elec- tronic equipment. Bill Materse San Jose, Calif. A ligs version of the A+ Mouse is being advertised by several mail-order companies. Its the only Apple i! optical mouse we know of—DJD Mixed-up backup Enclosed is a bulletin put out by Applied Ingenuity that says that the Backup 1 program (I believe v1.1) will not handle a hard drive with GS/OS 5.0 or higher. Also enclosed is an article from the figs Buy- ers Guide about the Apple II High-Speed SCSI Card that says that a new version of Backup I (and this is v2.0) will handle 5.0 and higher. However, the manual for the new High-Speed SCSI Card states that the utilities thereon, one of which is the new Backup 1, will not work on earlier SCSI cards. As the possessor of an older Apple Il SCSI card purchased last March, this leaves me up a tree. What backup program can be used with earlier SCSI cards and System 5.0.2? Several programs are in my collection such as Easy Drive and ProSel, but which are suitable? Does Apple expect one to purchase another SCSI card every time they bring out a new operating system? Incidentally, the documentation for David Hill's new version 3.0.2 of JumpStart says that the copy program included can correctly copy GS/0S files with any resource forks. Robert Allison Punta Gorda, Fla. Two available incremental GS/OS-based backup utilities that will handle System 5.0.2 “extended” files are Salvation Guardian (for- merly just Salvation) from Vitesse and ProSel- 16's backup (see “Backup only”, p. 6.14, March 1990). Most ProDOS 8 utilities will not work Correct- ly with extended files, and many will not even report an eitor when dealing with them {an “unknown storage type” error is feasible since extended files do return a unique storage type identifier when the file information is request- ed). Utilities that do not work on a file-by-file basis can be used; for example, the ProSel-8 BACKUP utility that copies the entire used por- ‘tion of a ProDOS volume block-by-block to a series of floppies or a file. The only 8-bit incremental backup program we know of that works with extended files is Apple’s Backup If v2.0. it is supplied with the new High-Speed SCSI card and is also supplied as part of the Apple Ii High-Speed SCSI Card Utilities from APDA ($30, part #A02422LL/A; currently classified as “beta” since it includes the beta versions of the Apple Scanner and Apple Tape Backup 40SC drivers for the Higs). Although the manual cautions against. using the new SCSI utilities on the old cards, Matt Deatherage at Apple DTS verified that Backup I (itself not actually a "SCSI" utility) specifically works with any ProDOS (but not AppleShare) volume, regardless of interface, The manual’s warning is to prevent users from doing things such as using removable- media drives (which the new drivers will recog- nize) with the older Rey. C SCSI card, which doesn’t work properly with some aspects of the new drivers. For example, | can eject and swap volumes on a SyQuest drive attached to the High-Speed SCSI card without problems; the Rev. C card becomes confused. Our expe- rience is that a warning not to eject a disk with the Rev. C card may not be followed by a cus- tomer, so Apple probably decided not to sup- Port the use of the new drivers with the older card rather than deal with liability issues. In faimess, and in response to complaints about the new card's price, it is only $129 including the new software. The Rev. C card is not functionally obsolete; the DMA version just adds some capabilities. while maintaining the same price (You don’t get to upgrade cars A2-Central 6.69 when the new model comes out, though it would be nice.}—DJD BASIC, C? 1 ordered from your organization and received the Addison-Wesley Technical Library Set. It is most impressive. However, it is for someone much more advanced in computer techniques than | am. I have owned an Apple I! Plus, Ile, and now the ligs. Since the first Apple 1 have tried to learn the inner workings and have tried to learn (on my own) machine Ian- guage programming—unsuccessfully. So when | got the library set | was rather frustrated to find that the textbooks gave examples on how to use the text in machine language and C. Is there any way that the ToolBox can be accessed via BASIC? The instructions in the Pro- grammer's Introduction are rather convoluted and make little sense to me. Is it possible to get a translation into simple and understandable English of this *highevel explanation’? I cite “Calling from a high-level language” on page 65: “The interface libraries that allow C program- mers to access the Apple IIgs ToolBox are included in APW C. Those libraries contain the function definitions for the tool... “1. Make the routine accessible by using an #include statement that includes the appropri- ate file... “2. Invoke the call by entering its name and supplying the correct parameters...” Etc.. What does it mean? Could | please get some help as to how to access the powerful ToolBox from BASIC? Peter H. Smith South Dartmouth, Mass. Let's start with the concept of the ToolBox routines in general, then move on to the lan- guage-specific considerations. The ToolBox is a library of routines supplied by Apple with the figs and Mac system soft- ware. The specific routines for the two comput- ers correlate reasonably well, making some of what you learn about the tools on one machine useful for programming the other. The purpose of the ToolBox (for either machine) is to provide standardized ways of accessing system features. Each routine is a “black box” that performs a specific function; ail your program has to do to use the routine is to set up a list of information (or data struc- ture) that the routine can use to perform its function, and then to call the desired routine. An example of the ToolBox call Read- AsciiTime to read the lIgs clock is presented in “Time to look In the toolbox", pp. 3.21-22, April 1987. Using these standardized routines provides two definite benefits, The more you use the ToolBox, the more your program is indepen- dent of the specific hardware features of the computer it is running on. This frees Apple engineers to make improvements in the com- puter hardware without having to worry about existing programs “breaking”. By only requiring a list of data to be provided to the routine, the workings of the routine itself can also be “hid- den” from your program, so that the ToolBox routines themselves can be improved without your program having to be re-written. Occa- sionally, completely new routines may be added that a program can take advantage of 6.70 A2-Central (The Apple lgs ToolBox Reference, Volume M1, is full of these), but the important factor is that existing programs don’t have to (indeed, shouldn't, if properly written) fail to work as the system is upgraded. The third benefit also has a dark (well, slightly gray} side: by using standardized rou- tines, most programs will have a similar “look and feel". Programmers should feel encour- aged to use the existing tools for the reasons in the previous paragraph, and because they are there in the system waiting. By encouraging programmers to do this, Apple gets the benefit of making their computers appear easier to use by “unifying” the look of programs that run on the system. Sometimes there is too much uniformity, which can be a bit boring, so the recent upgiades in ilgs System Software 5.0.2 include some new functionality to allow pio- grammers to design more “independent” ele- ments for the user interface. Thal’s the ToolBox philosophy. The imple- meniation of accessing the ToolBox within a specific language varies, but basically it con- sists of building the data structure that the ToolBox routine requires and then calling the routine. Some languages, like Pascaf and C, allow calling a function (or procedure} and passing data to the function as it is called. Most BASIC imptementations, like Applesoft, don't do this and a method of passing the data must be contrived. Some extended BASIC implementations, like Micol Advanced BASIC, do have a function-calling mechanism that allows passing data items. In any case, the eventual form of the call is broken down into machine language and executed. The ToolBox manuals present the machine language and C forms of the calls because these are the hightevel languages supported by the APW tools, The first line of each defini- tion in the manual gives the function number and tool set number in the form of a four-byte hex word (“$0F05" for function “SOF of too! set “$03’) followed by Apple-specified name for the call ("ReadAsciiTime”). A description of the function follows. Below that is a “Parameters” Section that shows what values are passed on the Hgs stack, and what values are returned. Next is an “Errors” section that reports what error values may be returned. Next comes the °C” version of the call. The € calls are actually converted to a Pascal call- ing convention (which is why you continually see the declaration of the “C” function as a “pascal” cail in the ToolBox References). igs hightevel languages usually are enhanced with some way of executing a “generic’ ToolBox call, and then a series of interface routines are built upon the call. The APW, Metlin. Complete (formerly TML), ORCA, and some other languages for the ligs include such Interfaces as part of the language package. Let's use ReadASCiiTime as an example, since Tom Weishaar has already described the machine language form (p. 3.22). In C, the form of the call is given as (page 14-14 of The Apple ligs ToolBox Reference, Volume I): exter pascal void ReadASCIITine (puffer?) pointer buffer?tr; The first line defines a C function *ReadASCiiTime” that accepts the data item bufferPtr (a pointer to memory that you have reserved to accept the ASCII data) and returns nothing ("void" in current C parlance}. “extern” means that the function is defined somewhere besides the current C file {the one you're using the function in), and “pascal* refers to the fact that “bufferPtr’ is to be passed to the ToolBox using Pascal, notC, conventions. The second line tells us that “bufferPtr” is a variable of type “pointer”; this tells the C com- pilér how Jarge the data item is so the compiler can treat it properly. By knowing the type, the compiler can also do some checking for errors in our program (for example, if you try to assign the pointer value to an integer variable, the compiler can warn you). Here’s a trivial program using ReadAsciiTime in ORCA/C; dinclede <stdio.D include <aisctoc] b> int mainivoid) [ ar Buffer[21]; /* allocate 21 bytes */ pointer GuiufPtr =a; /* set pointer to bufver */ Buffer[20] = "\0'; /* set last byte to “moll” */ Readisciitine(OurBufetz); /* cet time/date string */ puts (QurBufPtr); {* and display it #/ return 0; /* return “no error” to shell */ } For simplicity, this program runs in text mode from the ORCA shell; all we're trying to do is took at an isolated example of one too! call. The reason you probably didnt grasp “#include” is that it doesn’t have an equivalent in Applesoft. The C compiler includes a prepro- cessor that looks for certain patterns in the source text and performs certain actions before the actual program compilation takes place. “#inchide" accepts a filename and caus- es the preprocessor to locate and load the named file. if the file is enclosed in “<>” char- acters, the preprocessor looks in a specific defined directory; if the name is only enclosed in quote marks, the preprocessor looks in the current directory. “#include <stdio.h>” fells the preprocessor to include the conlents of the file “stdio.h", which contains definitions used for standard input/output operations (we need this to access the display screen). “stdio.h” is a part of the standard C language library, and details on lis use are included in any book on C. “#include <misctool.h>" is the lgs-specific operation that the Programmer's Introduc- tion was referring to in item 1. The ToolBox function ReadAsciiTime is not inherent to C, so we need to have it defined as a C function before we can use It in our C program. That definition {similar to the example in the Toal- Box Reference) is in the file ’misctool.h” that is included with both current Apple ligs C compil- ers (APW C and ORCA/C), We use “#include” to copy that file into our program source so that we have access to use ReadAsciiTime tater in our program. if you know C, you can figure out the rest; if not, there are several good books that can help you with the overall program structure (among them The C Programming Language by Kernighan and Ritchie, and either of two books by Kelley and Pohl A Book on C or C by Dissection). Our program contains one function, main, that creates a buffer for the ASCH data returned by ReadAsciiTime, sets OurBufferPtr to the buffer address, and stores a mull byte (‘\0° in C parlance) into the last byte of the array (C expects sirings to be null- terminated in most cases). Next we pass OurBufferPtr to the ReadAsci- Vol. 6, No. 9 iTime function defined in misctool.h, This cor- responds to item 2 of high-level language access: calling the named function with the appropriate parameters. Finally we use the C puts() function from Stdio.h to print the string to standard output (in this case, the Iigs text screen “console”), “return 0” sends a zero to the ORCA/C shell as the error return code; we've defined main as returning a value of type “int” (integer), which the shell expects. This is about as brief an introduction to C as I can manage in an attempt to show what the Apple llgs ToolBox Reference definitions mean, What I hope it conveys is how the defini- Uon in the manual translates into the syntax of a real program. Now let’s try translating ReadAsciiTime into a BASIC dialect, Here's a similar program in Micol Advanced BASIC for the ligs: FROGRAM PrATine DIM Batter (20) PROC ReadAsci iTima [OurBuiPtr] Yow = TNT (OurBusPtr/ (2"26)) ASH = 1S TSW = OurmafPtr ~ (Mee (2*16)) {get ackiress into 198} ‘LSW = LW TOOLECR (3,15: HSWE, LSW¥) BORO PROC puts[OurBufPtr] LocalPtr = OurbufPtr RET CHAR = PERK (LocalPtr} CHARS = CHRS (CHAR) ‘TF CHAR = 0 THEN CHARS = CHRS(13) ‘PRINT CEARS; ‘LocalPtr = LocalPtr +1 UNTIL CHAR = 0 BPR ourdufPtr = ADDR (Buffer) PORE OurlbufPtr420,0 SOSUB Readhscii ime [OusBufPtz) GOSUB pete [CorBuftr] FT {get bank into SK} ED Micol BASIC requires that we start our pro- gram with an identifying name, so we used PrA- Time (print ASCH time}, We allocate a 21-byte buffer imaginatively named “Buffer”) for the ASCH data. Then we define two procedures (bracketed by “PROC” and “ENDPROC*) that we'll use Iater. The main program starts by using Micol’s “ADDR” function to return the starting address of the array Buffer, assigning it to OurBufFtr, and setting the last byte of the array to zero, Then we call the procedure ReadAsciiTime, passing it the value of OurBufPtr. Micol BASIC doesn’t include interface Hles for the ToolBox, so we have written our own interface function. ReadAsciiTime accepts a pointer to a buffer, and we break it into two words (two-byte integers} representing the bank number and the address of the array within the bank. We pass these to the ToolBox via Micol’s TOOLBOX function, which accepts the tool number (“3° for the Miscellaneous Toolset), tool function number (“15° for ReadasciiTime within the Miscellaneous Tools), and (following a colon) the parameters (a word October 1990 at a time} in the order that they need to be pushed onto the 65816 stack to be passed to the tool, In our ReadAscifTime call, we pass the pointer as the most significant word (MSW%) followed by the least significant word {LSW%) of the four-byte OurPointer value. It is here that we need to know how the ToolBox accepts the data in assembly language format; that knowledge Is required to develop the form in which the data needs to fit into the TOOL- BOX parameter list. When ReadAsciiTimef} returns, we call our procedure puts{) to print the string returned in our buffer. Via a REPEAT—UNTIL loop, puts() PEEKs successive bytes out of our buffer and prints each one, converting the terminating “0” value to a carriage return (ASCi! 13) and abort- ing (the UNTIL condition fails). It gets very tedious to sit down and write the interfacing code such as our ReadAsciiTime example each time you want to use a ToolBox function. That’s one reason | (personally) pre- fer C or Pascal; you have a high-level language that includes the necessary interface files, ready.to go. A third-party company, Codesmith Software, is working on a set of Micol BASIC procedures to aid in accessing the lgs ToolBox: Tthe Codemaster Collection ToolBox Library. Codesmith Software. Codesmith also has start- ed a Micol Advanced BASIC User Group, Membership Dept. #1316, 107 E. Vallette, Elmhurst, til. 60126, that shares source code and tips with members (this group is not affili- ated with Micol Systems). New members receive a library disk of ToolBox procedures, such as routines to allow using 16 dithered col- ors in 640 mode on the ligs SHR screen, And if you want to hack at ToolBox access from Applesoft, So What Software's CallBox (see "GS/OS Resources”, pp. 6.5-6, Feb, 1990) includes the tools to do it. Some of the data structures used by the ligs ToolBox get quite complex. The other advan- tage of C and Pascal is that each language allows description of more complex data struc- tures than typical BASIC. For example, here’s a window record definition as aC structure: Haclude <window b> [* where Paranlist structure is defined */ Paranlist RenoteWin = { /* new window record #/ 8, 7? paramlength */ Greta, J* wiraneBits */ title, it wfitle */ a, [* vhefton */ {0,0,0, 0} f* wloo #/ WL, & Color #/ ,0, J* wtOrigin,wiarigin +) 4,0, [* sata, watall */ 0,0, 1 Milan eal #/ 0,0, [+ weorollver, worel lor */ 0,0, {+ weagever, wagelior */ 4, 1 wInfoRefCon */ o, i* winfobeight */ NULL, (+ BremedefProc */ WL, /* winfobefProc */ WL, 1+ WeontDetProc */ {30, 20,190,260}, /* wosition */ (void *) -1L, it wlane */ WL I* wStozage */ ki The comments (bracketed by */*” and **/*) refer to the names of the “fields” within the record RemotelWin as defined in the C header file “window.h”. Accessing an Item requires only that you use a period to separate the name of the field from the name-of the record, so that “Remote Win,wPlane” if you prefer using simple arrays rather than abandoning BASIC, however, that option is open to you. You'll need to access the ele- ments of the array “manually” by determining the offset to the item you need; you could define the offsets In “standard” variables to help systematize things. Incidentally, several readers have also asked where they can get started learning C or Pascal on the llgs. The Byte Works (4700 Irving Blvd. NW, Suite 207, Albuquerque, N.M., 87114, 505-898-8183, GEnie BYTEWORKS, America Online MikeW50} has been offering on-line and correspondence courses in C and Pascal fea- turing their languages (ORCA/C and ORCA/Pascal) on the ligs. The lessons take you through the elements of the language up to the point where you can start concentrating on learning the lIgs ToolBox. From there, you should probably try to acquire as much Hgs Source code as possible (specifically the exam- ples provided by Apple DTS) and dissect how it works. Ihave a lapel button that reads “Calm down...it’s only ones and zeros*. The ToolBox is there to help; don’t fet it intimidate you.—DJD GOBASE Ill file translation Using the PC Transporter | would like to translate files from dBASE ‘jH+ to an AppleWorks database file. Can it be done? Jeffrey S. Bock Newbury Park, Calif. We've mentioned CrossWorks 2,0 as a method of moving files between and Apple I and an MS-DOS machine, but since the file conversions are actually done on the MS-DOS machine you could run the program with the PC Transporter (converting the files in the MS- DOS “partition“) and then copy the files over to ProDOS. This is faster than using the modem transfer method, but would also require you to correct the filetype of the converted files after you copy them fo the ProDOS side of your schizoid computer's world. There is a handy freeware utility called FAZ (File Attribute Zap- per) that will facilitate assigning the correct types. The other option would be to convert the BASE Illy daia to a text file format suitable to read into AppleWorks. You will lose some data in the transfer this way that CrossWorks will attempt to. convert, such as report formats.—DJD Pushing the envelope What effect would replacing my 7 Miz TransWarp GS's present 28 MHz crystal oscilla- tor with a 32 Milz one as shown in Applied Engi- neering’s advertisement have? Stephen Gant Manteca, Calif. We have never looked that closely at the advertisement, but... our suspicion is that it may {or may not) speed up your system slighi- ly, if the chip will operate reliably at the higher speed. It would probably also void your war- ranty, There have been instructions circulating on accelerating the TransWarp GS by replacing the 65816 processor on the card with a faster (higher MHz rating) chip, increasing the speed A2-Central 6.71 of the crystal oscillator, and (when necessary) installing faster static RAM to support the Speed increase. One of the more detailed doc- uments we have seen is credited to Andrew Hall of Western Design Center, and adequately explains the hazards of the “enhancement”. We haven‘t done the modification, but you should realize going in that it isn’t simple. The 65816 should operate at one-fourth the frequency of the crystal oscillator; that is, for a 7 MHz 65816, a 28 Miz oscillator is the cor- rect value. it might be that the 65816 may operate at a slightly higher speed than its rated yalue, but we couldn't presume to second- guess Applied Engineering or Western Design Center on this, According to Tony Vece at Zip Technology (Zip is also working on a Hgs accel- erator card), the static RAM cache used by the iigs accelerators should accept accesses at the crystal oscillator’s speed (four times the pro- cessor speed); ihe corresponding access rate for a 28 MHz oscillator is once every 35.7! nanoseconds. That means if you speed up the oscillator (and processor) too much, you'll need to unsolder and replace the static RAM chips with faster versions, Don’t put too much faith in advertising pho- tos; we remember the original Zip Chip promo- tional flyers where the chip was shown plugged into the MMU socket (rather than the processor socket) on a lle motherboard.—DJD More orphan help I sold my Il Plus and purchased a used enhanced Ile and now I need a patch which will enable me to use Checkmate’s MultiView 80/160 card with AppleWorks. | wrote Check- mate and they told me they no longer support their card. Also, | recently inherited a Qume Let- terPro 20S printer and am in need of a bidirec- tional forms tractor. Qume no longer supports that printer so [ am on my own for parts. If any- one can help me or put me in touch with a source [ will be forever grateful. Richard Washington P.O. Box 253. Fort Union, Va. 23055 Control Panel shortout Is there any way to disabie access (and later, of course, restore it) to the ligs CDA menu? For example, a game that selectively enables a TransWarp GS would be thrown off if it was manually set. John Morison Phitadelphia, Penn. it’s better to tell the user not to alter the Control Panel settings in your documentation that to lock them out of access; let the user accept responsibility for how the computer is used. The particular feature you don’t want them to alter is only one of many options the Control Panel controls —DJD Hard disk interleaves How can I determine the best interleave for my hard disk? Do I have to try all possible com- binations? | am using a Seagate ST-157N. Chang Yuh Kang Republic of Singapore You don’t have to try every combination, but you will have to try several in order to “frame” the best time. This may vary from drive modet 6.72 A2-Central to drive model as access characteristics change. You can, for example, start with the highest possible interleave setting, reduce it by five every time until you find the interleaves at which performance gets worse, and then try all the settings between the two where the “drop” in performance occurs. A2-Central’s Tom Vanderpool has done some timings on a Seagate ST-277N, with and without his TransWarp GS active, and these were his results (all timings in seconds): Interleave TWGS On TWGS OFF i 4 17.80 5:1 13.29 19.61 Be 10.69 18,09 9:1 Ban 15.66 phe 12.51 16.96 These timings were accomplished in the same way (move a large file onto a freshly for- matted HD and then load it while timing the load). Here are Tom's timings for an 80 meg ST296N: Interleave — TWS (nS OFF in nw 18.70 S21 4.10 20.00 ‘By 2.2 18.26 &:1 mu 18.58 91 1.4 17.8 10:1 11,36 17.36 bby not 16.65 You should pick an interleave at or slightly above where the timing seems to increase sig- nificantly: for example, for the ST296N with A2-Central’ a © Copyright 1990 by A2.-Central Mes rghis resoved. Al pograns pubished m A2-Central 27 pubic domain and may be coped and dsirbuled without charge. Apple user groups ane saan ches may obtan pemisson to epi aries te me ty spent weten rapes Ecied by: Dennis Doms wits help trom: Tom Weishaar Sally Dwyer Dean Esmay JoyceHammond JetfNeuer Jay Jennings Tom Vanderpool Jean Weishaar A2Central—ied trough January, 1988—has boen pub- ished mony soe January 1985. Wortwide proes fn U.S. dlrs: armal elvan nee at no addcnal hare): $28 fo 1 year: $s for 2 yours: $78 or ‘3 years, Allback sss ae cureiyavalate or $2 och; bound. deed ec tors of ou re fou volumes are $12.95 each otumes end wih ine vanuary ‘sue an index bythe pro vole is ncder ne February seve. ‘The fl ext ofeach issue of A2-Centralis avaiable on 35 disks, along wt aseiacion ofthe best new puble domain and stareware les and pro- «grams. or $84 a year (newaleter and disk combined). Single asks are $10. lease sandal eoespondence a: A2-Central P.O. Box 11250 Overland Park, Kansas 66207 U.S.A. A2.Central is sid in an wnprolace tormal lr your convarisnce. You ze encouraged lo make Dack-up archival copies of e2sy-o-oad enlzted copies for yeu own use who charge. You may also copy A2-Centra ior ‘sintuson to others. The dstnbuton lees 15 cents per rage per copy ds- vbved. ‘WARRANTY AND LIMITATION OF LIABILITY. Iwaran:tnat ost of the inlormaton m A2-Central is uselul and conect,ahough drivel and mis- takes are incied Irom ‘ene 10 bme, usualy uninteionaly. Unsatsted subsoroers may cancel ther subsp at ary time and receive 2 lu refund of thaw Last subscnpton payment. The uniled potion of any pa Subscpion wil be relanded even to saished subscribers upon reques! MY LIABILITY FOR ERRORS AND OMSSIONS IS LIMITED TO THIS PUBLICATIONS PURCHASE PRICE. In no case shall or my coninbulos be lable fr any inc denial or cangequential damages. oor lor ANY can ages in excess cl the lees paid by a subscriber. (SSW 0885-4017 (GEnie malt: A2-CENTRAL Voice: 913-468-6502 Printed in the US.A, Fax; 919-4696507 the TransWarp GS “on” Tom picked the 9:1 interleave. However (as we've said), be aware that your specific configuration and usage determine the best timings; the interleave that facilitates the fastest time for writing a 5 megabyte file from BASIC may not be the one that is optimum for loading a huge AppleWorks database file. Chinook’s new SCSI utilities will include an option to “scan® access under ProDOS 8. and the best interleave to use. You may still want to verify the result with your own timings, how- ever.—DJD More on DMA SCSI As a-fellow DMA SCSI card owner, | must reply to Udo Huth's letter (“Don't hang up”, June 1990). I have a Jasmine 40 megabyte drive attached to the DMA card in slot 7 and | origi- nally thought the system was hanging if the drive was not switched on. In practice, it “times out” after 20-25 seconds and continues from the 3.5 disk, I have some comparative times for the old and new cards. My ‘ligs system has 3.25 megabytes of memory, a TransWarp GS, and my Jasmine's seek time is 28 milliseconds. The interleave is 1:1, which is better than 2:1 and the (unknown) original value, but I'm not sure if its the best! Both times use the new drivers (which are slightly faster for the old card as welll). A fascinating thing is that the times often vary up and down by several seconds under otherwise fixed conditions. The times listed are average timings: Rev. CD Boot to Finder 40 36.0 ~ 550K of DA's, ete Agplettorks 6 as ns ~ i, SS, Com ProGal volune stats ~ Linear zead 95 22 ~ randon read BS 315 = 0S overhead 4s Ma Digitized sound files ~ 200+ blocks 3.0 Lo ~ 600+ blocks 6.0 1.0 > 670+ blocks 65 15 ‘The sound files were what convinced me to buy the card. | can play through a directory of sounds in (almost) real time-- very impressive. The simple fact is that the card speeds up the transfer rate. If the drive spends its time seek- ing, as during a boot or loading a segmented file like AppleWorks GS, then the speed-up Is minor. Give it a large file to load and you are talking “greased lightening”. | am hoping Claris will provide an installation option to allow “true” preloading of AppleWorks GS so that people with enough memory and a DMA card will be up and running {all modules) in around 12 sec- onds! Peter Watson Box Hill North, Vic. More power to you I have a 512K IIgs and have noticed a slight problem, My system is configured as follows: Basic 512K ligs Apple RGB Monitor Kensington System Saver GS-RAM Plus with | megabyte Applied Engineering Datalink 2400 HyperStudio digitizer Vol. 6, No. 9 CMS 20 megabyte hard disk Apple 3.5 Apple UniDisk 3.5 Apple 5.25 I didnt notice this problem until 1 daisy- chained the UniDisk from the Apple 3.5. Howev- er, when | boot the system from the hard disk the read/write light on the UniDisk pulses! I took the UniDisk off of the system and then noticed that the 5.25 read/write tight did the same thing but wasn't as noticeable { then put the UniDisk back in the chain and noticed that the UniDisk and 5.25 drive were both pulsing at the same time! I stripped the whole system down to a basic machine with only the Apple 3.5 drive connect- ed and tried all versions of GS/OS (v2.0 through 5.0.2) and found that | had an audible pulse tone through the speaker, After disconnecting the speaker | noticed | could still hear an elec- trical click in the power supply. { took it to our local Apple “dealer” and he confirmed that the power supply is bad. Now my question: the Apple supply is $135 as opposed to the Applied Engineering supply which is $99. Which would you consider, forget- ting cost difference, as better for the ligs sys- tem? Walter Monhof Hephzibah, Ga. This sounds like a nightmare. Let's retrace a few steps. First, the pulse/click with the UniDisk 5.5 installed is normal. Some Hgs programs like the Finder have to monitor the drive to see if a disk has been changed: the UniDisk takes this “polling” rather hard. We've got ligs systems here that have survived this for years. Second, the click in the speaker may be nor- mal. My ROM 01 machine at home does this (with only an Apple 3.5 and 5.25 attached) when I turn the sound level up. It apparently is some sort of crosstalk reaching the sound cir- cuitry. Third, the “click” in the power supply may or may not be normal. The supply is a switch- ing type that emits a high-pitched whine in some people’s hearing range (t guess I'm one of the “fortunate” ones). As the supply has to “strain” a bit to send the poiling pulse to the disk port, I can hear the frequency of the whine change. My system has done this ever since I bought it. if you have had no other problems with the system other than the “click” from the power supply and speaker while the drives are polled (the pulsing of the drive lights is perfectly nor- mal) you may want to check a little more into how the dealer determined the supply is “bad”; that is, did he use some sort of diagnostic equipment or just recommend replacing the supply as a cure for the “problems” you've seen, If the latter, I'd suggest asking for a quar- antee that the new supply will eliminate the problems. If he’s willing to do it, paying the extra dollars may be worth it. If not, and you determine the supply is the problem, then it may be a toss-up. My feeling is that it is worth paying a little more for the Apple logo all other things being equal; for one thing, it eliminates any contention with the dealer about my sys- tem configuration. Others are quite welcome to disagree. —DJD