Latest release date: 2/13/88 version 1.5 ------------------------------------------------------------------------------- Changes making version 1.20 IAU.EXE 1. Cursor is not displayed now for a nicer display during the read-format- write operation. 2. Some of the source was modified for Turbo-C compatibility vs. C86 C. 3. A potential bug with large hard disks was fixed. 4. Time displays fixed (if you change interleave twice, the old "finished" time remained on the screen, giving misleading information). 5. 'C' source code has been passed through the XENIX "cb" (C beautifier) program, causing the structure to be easier to follow. 6. Some of the comments in the .DOC files were slightly changed. 7. The .DOC files may now be in either your current (default) directory, or in any directory in your "PATH=" environment string. Note: If you want to change the text of the .DOC files, use any text editor (or WardStar in the Non-Document mode). The program will reformat your text during run-time to fit in the pop-up window. This is true for all earlier versions, too. Note: Some of the screen subroutines and other subroutines are from the Hammer library (O.E.S. Systems). I use at least one (and usually several) of their routines in just about every 'C' program I write. I can recommend it highly. Changes making version 1.21 IAU.EXE 11/17/87 1. When the cylinder count goes above 655, the percent complete is now correct. 2. There are two versions of the executable; the 8088 version is IAU.EXE, and the 186/286/386 version is called IAU286.EXE. It runs a little faster on these processors. The only difference is a few compiler switches - not in the implementation. ------------------------------------------------------------------------------- Changes making version 1.3 IAU.EXE 12/11/87 1. A new displayed value is shown -- the current interleave value is measured and displayed. This works for most STANDARD configurations, except for the following special situations: A) If you are using disk cacheing, results will be wrong. B) If your hard disk isn't 17-sector, no result will be shown. C) If your current interleave is too small for your system (such as 1 on most systems), you will get the get the "effective" interleave value. This means that if your interleave is set for 1, and your controller can't support it, then the calculated value will be 18. This is the value which will be displayed, and this is the performance (penalty) that your other software will experience. Thanks to Ed Eggleston for this suggestion for a new feature. ------------------------------------------------------------------------------- Changes making version 1.4 IAU.EXE 1/6/88 1. Version 1.3 hangs when used on an 8088 or 8086 machine. I linked in the wrong library for that version. As far as I can tell, it can't ruin any data on your hard disk -- it just hangs. Version 1.4 is correct. Thanks to George Nail for his time in reporting it. 2. I deleted the 286 version for quicker upload/download times. 3. Now the program measures the performance on your system to calculate the best interleave for current configuration. 4. The "current interleave" measurement was fine tuned. 5. A dial-up phone number for questions/answers/suggestions is now displayed on the screen. This is a unix system and you can mail to me. When you connect, answer the Login prompt with "hello" and press return (without the quotes). You can use uucp from your unix system to that number, also: mail concord!dad or via USENET ...wang7!concord!dad If you do, let me know how to get back to you (phone number, hours to call, system id, etc.) You may also use amateur packet radio if you'd like: kz1o @ wb1dsw ------------------------------------------------------------------------------- Changes making version 1.5 IAU.EXE 2/13/88 1. The sectors-per-track was fixed at 17. While this accomodates most hard disk drives, some of the newer (Maxtor, etc.) drives use RLL encoding, which gives more sectors per track. The 1.5 version allows this value to be read and the corresponding table is built for this parameter. 2. The source code was removed from the archive file since most users weren't using it anyway. The source code is still avail-able for anyone who wants it, either via modem or mail to the author. Because of restrictions on some of the Hammer Library routines (copyrights) you must obtain these on your own (or write them). 3. Most users weren't aware that the .DOC files are read at run time, and can be changed by anyone who wants to. They are read and formatted for the little window on the screen, so if you do edit them, you needn't worry about margins, etc. Just be sure that you don't use a text editor which uses hidden characters (for example, WordStar). My algorithm counts and displays all of the characters, and I'm not sure what would happen if some of them were non-printing types. 4. The phone number given for help is CORRECT; however, because of various problems, it doesn't always answer. The best time to try is late at night and on weekends, but I try to keep it up all the time. In some cases you might get an answer by voice instead. If you have a pressing problem, you could call the regular (voice) number at (603) 228-4683. Please make it during reasonable east-coast times, since I have a family. Oh, yes -- my CompuServe ID is 70441,2456. 5. The display of the "Best Value For You" value is not always correct. The value displayed as the "best so far" is the correct number, but a typo caused the wrong variable to be passed to the printf statement. Now the program displays the correct value. ------------------------------------------------------------------------------- I am always interested in change and improvement suggestions that you may have. Most of the updates were made because someone had an idea about how to make the program better. As a result, it is a much better program than when originally written. I didn't want to say this before, but now I will so that it's out in the open: Some people have called with questions about the program, and indicated that they have had problems of one kind or another. Due to the nature of this program's operation, people usually run it once and shelf it. Because of human nature, it seems that I am more eager to solve problems for people who have sent me pictures of dead presidents ($$$). I didn't write this program for money, but it helps to pay the phone bill, etc. any case, I'll do what I can to help you.