SRS.TXT - Text file accompanying the SRS program This file can and should be modified by any and all users with comments they wish to pass along to others who might receive this program. Licensing This program is distributed as "advertisement", and consists of an executable program, the external "play" and "record" batch files, and their accompanying text files. As advertising, this matter may be freely distributed provided that all supplied files reach their destination in their received format, except for the file "srs.txt" (this file) which may have any comments (warnings) added to it. Miscellaneous Notes The use of "marks" is, we believe, a superior method for selecting a range from a sound buffer, as opposed to that requested by mouse-users who desire a visual interface; remember, this is a "sound" program, and your primary goal is to manipulate what you "hear". Trying to ' find exactly what you want to hear by looking is, well, rather counter-intuitive, is it not? Couple this with the ability to play back at a far slower samplerate, and precise cuts can be made easily, by simply pressing the 'm' key at the desired points, stat'ing the buffer and noting the mark numbers, and "cut mark M upto mark N"! Notice that the program will prompt for a channel if the source and target buffers differ in channel numbers and a "left", "right" or "both" designator is not present. Pressing ESC will cancel the process, but the "into" buffer will not be removed (bug). The reason for providing "shadow" is to look at a single buffer in a number of different ways. Using shadows, you can refer to different parts of one buffer using other buffer table entries, in contrast to using loads, which copy the sound buffer data into new files for ecah buffer entry. So modifications made to the source are then available in the shadowed buffers, immediately. Another, highly overlooked, feature is the ability to generate exact waveforms, if you are so include to do. In no other program is this abilitym available, not even in SRS! The secret lies in being able to put together other SHARK tools in series so as to obtain the desired result. Indeed, allowing the use to "free-form" ideas in way unimaginable by either the programmers or the Ancients, this lies at the heart of SHARK programming. An example is a the generation of a sawtooth, square, or sinusoidal waveform. Prior to the introduction of SRS and the SHARK utilities, this could only be either approximated using analog devices or built directly using sophisticated (expensive) digital signal processing equipment. With srs, you can take any wave and add or subtract any other wave you can possibly think of, and immediately hear the results. You can do this thanks to both the SHARK ability to convert output to any other format and the program YOU used to generate the data! Concatenate your sound files into a single sound library file. The name chunk serves to identify your sound using a 16-character key. Do this: fe i ( *.wav ) ? mv $i $i:r.snd ? srs save from $i:r.snd into $i:u ? cat -b $i >> sndlib1.slb ? rm $i $i:r.snd ? end "fe" will process each .wav file, "mv" renames "FILE1.WAV" to "FILE1.snd", "srs" saves "FILE1.SND" into "file1.wav". This merely includes the filename in the .wav file so that it can be referred to by that name in the resulting library file. "cat" concatenates the file1.wav file onto "sndlib1.slb". "rm" removes the .wav and .snd files. srs load sndlib1.slb ; play *.wav This will load the .slb file, and then play the original .wav files. Note that this example requires the programs fe, cat and rm. You can accomplish the same thing in a rather roundabout way using DOS, but since the DOS "for" does not provide the ability to modify the "substituted" string and SRS will not overwrite existing files, you need to create a big .bat file: srs save file1.snd into file1.wav srs save file2.snd into file2.wav (get the SHARK utilities and simplify your DOS file operations) This method is recommended for the following reasons: storing "file" information is space-consuming 5 buffers can be stored in the space of 1 file the program is written to handle 16 files simultaneously, but up to 512 buffers. 512 buffers sounds a little crazy; maybe I should alter that balance. Unfortunately, some programs fail to process .wav files when an unrecognized "chunk" is present. This is bunk. Fortunately, the Windows Wavedit program does not have this problem. If your program has this problem, then use the -N option (put it into the SRS environment variable "set srsopts=-N"). Don't Tell Anyone But, the latest idea is to obtain a CD-ROM dictionary, and then record the pronunciation key found therein. With a little scripting, this should provide the ability for the computer to speak any word in the dictionary. Then, the idea is to add a very-low frequency wave to the generated sound wave based on the punctuation of each sentence, or phrase. After writing this thing for the past few months, I am really looking forward to actually using it! If you want a list of all prime numbers up to 10,000,000, just ask. If you want the program that uses this list to find new primes, ask. Now I am thinking about using the "dadd" and "dsub" functions for searching for prime and perfect number. Hmm, a 360000-digit hexadecimal number... I wonder what its factors are? Actually, to consider a "sampling of a wave" as a "number" (and vice-versa) is tantalizing. Take a prime number. Take another prime number. Multiply the two. Now, the result will have a 0 modulus with respect to only the two original primes. Now, process using SRS. Is Public Key Crytography still safe? Obtain the orbital positions for the members of the Solar System. Play this data as a sound. Is that "the Music of the Spheres"? Just for the heck of it: set samplerate to 1000 load command.com into a play use the 's', 'S' and CTRL-S and f keys to modify the samplerate. Is that "b-i-i-ll gg-aa-tt-ss"? Reverse it. Did I just hear "ooohh-ee-ssss tt-oo-oo ii-zzzz dd-e-e-dd"? Just kidding about that, those sounds ARE NOT heard. Really! Believe me. Bart. --- In closing (please, release the code!) let me say once again that one phrase that sums our feelings towards computers: "... previously impossible, until the advent of modern-day computers ... " A fondest hope is to be a factor (!) in finding the next Perfect Number. --- Many know the meaning of a Prime Number, but what is a Perfect Number? A Perfect Number is a number that is the sum of its divisors: 6 is perfect because all of the numbers that divide it evenly (except itself, of course) add up to 6: 1 6/1 = 6 rem 0 + 1 = 1 2 6/2 = 3 rem 0 1 + 2 = 3 3 6/3 = 2 rem 0 3 + 3 = 6 The number 6 is Perfect because the sum if its factors equals 6. The next perfect numberis 28 (sum the even factors: 1+ 2+ 4+ 7+ 14 = 28). All told, less than a dozen Perfect Numbers are known! --- Yes, you TOO can LISTEN to your DATA FILES! We at SHARK deem regular "listening" of your data to be a sound (!) practice. Seriously, using sound to verify the integrity of your data is not an impossible circumstance! Since all SHARK tools work together, you really can "listen" to a data file, and have it actually mean something to you: take the "format" of each record multiply by the count of records subtract this "format" buffer from the actual "data" the result is the "regular" differences between YOUR data and the Default this will "sound" uniquely compare the bytes Imagine being the manager or supervisor of Some Information amd being able to walk into a situation and "hear" and so then "listen" to the current state of that data, thus enabling you to proceed with confidence knowing some of the state of data-integrity of over which you are responsible. Nowadays, as the manager or supervisor of this information, you would need relinquish this control to the MIS group. Imagine, being able to VERIFY (or contradict!) their assessment using "sound" (as opposed to staring at tens-of-pages of "account status" reports). I'm serious, folks. 21st century conceptions, happening NOW, at SHARK (we never sleep). This is what we mean by being "uniquely qualified" to handle problems of this sort. The standard MIS response to problems of this nature is: more reports! The games programmer leans to the extreme and responds with something off-the-wall: ordered files? you mean, with a regularity to them? Well, listening to them regularly will allow you to detect a problem, because any DISCREPANCIES will SOUND different! I know it sounds crazy, professional MIS contractors exhorting the benefits of playing accounting-data through the sound board in order to detect potential problems by "listening"... but, Multi-Media computing can mean more than business presentations. --- Command Line vs. User Interface Take this interaction one of us had with some accountants using a spreadsheet for the Macintosh: "S" What are you doing? "A" Taking this report from one computer and putting it into this spreadsheet... you use the "template" to define the format, highlight a range, and the computer modifies it! See how easy it is? "S" So what you are doing is stripping out the report headers, getting rid of a few columns, re-ordering the remaining columns, and then removing the dollar-signs and commas, right? "A" Right! It is so easy! "S" How long have you been doing this? "A" Oh, about 15 minutes. This is the third report I am reformatting. "S" So, about a half-hour per report? "A" Yes, these are large reports. "S" (chuckling) That's ridiculous! "A" Huh? "S" I could spend about 2 minutes THINKING about a command line that would strip out the headers, then another minute to come up with the command to strip out columns and re-order them, and then about 20 seconds to remove the punctuation. All told, less than five minutes doing nothing but THINK. Then, while the command is running, I get a cup of coffee, smoke a cigarrette, and THINK about what I have to actually DO with the resulting spreadsheet! "A" (dismayed) Really? "S" Really. But it would take YOU a little longer to think of the commands because you never used them before. But, after spending 15 minutes figuring out what to do the FIRST time, it will take you LESS time the next time you have to do this. Instead of doing the "accounting" thing and analyzing the data, you spend the SAME amount of time, EVERY time, performing mindless mouse-actions! I must laugh! What really happened here? The user confused "simple" with "easy". The Mac procedure was simple but the entire process was far from easy; the SHARK procedure is slightly more difficult, but the entire process is so much easier! We believe that energy spent THINKING is energy better spent! Let the computer perform the rote, repetitive, mindless actions; it doesn't care! At SHARK, we swallow problems whole.