+Allan Posted December 9, 2014 Share Posted December 9, 2014 I just uploaded a bunch of Atari cassette WAV files up on the Atarimania ftp that I recorded with Audacity and was wondering if one of you CAS gurus could try to convert them to CAS images. If any of them seem off and need to be re-recorder louder, softer, etc., let me know and I can re-record them. I have a couple of more I need to make and will post soon. untitled folder: Antic_Software_Library_Games_#1.wav APX_Caterpiggle.wav APX_Instedit.wav APX_Math_Mission.wav APX_Outlaw_Howitzer2.wav APX_Quarxon2.wav APX_Sketchpad.wav APX_Terminal_Emulator_2_0B.wav Danger_in_Drindisti.wav Home_Financial_Management_Thorn_EMI_Side_2.wav Allan 3 Quote Link to comment Share on other sites More sharing options...
Kr0tki Posted December 13, 2014 Share Posted December 13, 2014 (edited) The Sketchpad WAV file is missing! The rest got converted successfully; attached below. - Antic_Software_Library_Games_#1.wav contains 3 BASIC CLOAD programs: 1. Alien Craft! - the program is identical to the one already available on Atarimania (I mean the program listings are identical; the .BAS files differ, that's just the nature of saving to .BAS). 2. Play It Again - again, identical to the one already on Atarimania (again, identical listing, different .BAS files). BTW.: The listing contains comments that you may use to fill the blanks in the metadata: "Atari User Group of Dallas Program Library Volume 2 Name: AGAIN.B01 02/05/81" 3. Othello - not yet on Atarimania. Author and date unknown. - APX_Caterpiggle.wav - the tape contains some leftover of APX's "Babel" at the end - apparently the tape originally contained "Babel" and was later overwritten with "Caterpiggle". I've decided to not include this "Babel" part from the resulting CAS. - APX_Quarxon2.wav - the same situation, a leftover of some other program at the end. Again, not included in the CAS. - APX_Terminal_Emulator_2_0.wav - the program says it's actually version 3.1. Danger_in_Drindisti.wav - couldn't test it, as it requires a cassette version of Hellfire Warrior, which we don't have yet. - Home_Financial_Management_Thorn_EMI_Side_2.wav - this is odd. It requires OS rev. B and doesn't work on any other version, but the program was published by a British company, and OS rev. B existed only in NTSC 400/800 computers - in PAL we had rev. A only. But the program doesn't work with rev. A. I've seen other software by Thorn EMI on Atarimania, that shares this odd characteristic. Atarimania-CAS-2014-12-09.zip Edited December 13, 2014 by Kr0tki 4 Quote Link to comment Share on other sites More sharing options...
_The Doctor__ Posted December 13, 2014 Share Posted December 13, 2014 It is version 3.1 someone taped over 2.0 Quote Link to comment Share on other sites More sharing options...
+Allan Posted December 13, 2014 Author Share Posted December 13, 2014 Thanks Kr0tki, Thanks for converting these. I just posted: APX_Holiday_Programs_Side_1_Cas_1.wav APX_Holiday_Programs_Side_1_Cas_2.wav APX_Holiday_Programs_Side_2_Cas_1.wav APX_Holiday_Programs_Side_2_Cas_2.wav APX_Sketchpad.wav Vocabulary_Builder_1_Side_A.wav Vocabulary_Builder_1_Side_B.wav The four Holiday wavs are front and backs of two copies of the same APX cassette. I did both just in case one copy was difficult to convert. They are all original tapes. Some of these I guess could have been copied over but I know that APX_Quarxon2.wav for example I bought in the original sealed packaging. I guess APX was trying to be conservative with their supplies. Allan 2 Quote Link to comment Share on other sites More sharing options...
_The Doctor__ Posted December 13, 2014 Share Posted December 13, 2014 (edited) might want to look at some of these scans to complete things, the loose leaf loading instruction that are missing might be in these. Could be different pdf scans as well... http://www.mediafire.com/?p0mbcrck1y9jm Edited December 13, 2014 by _The Doctor__ 1 Quote Link to comment Share on other sites More sharing options...
+Allan Posted December 13, 2014 Author Share Posted December 13, 2014 These will all be posted on Atarimania in the next couple of days (I posted some already) with scans of instructions if I have them or at least a note on how to load them. Allan Quote Link to comment Share on other sites More sharing options...
Kr0tki Posted December 13, 2014 Share Posted December 13, 2014 Btw. I've noticed that Instedit works correctly only with BASIC rev. A. Quote Link to comment Share on other sites More sharing options...
Kr0tki Posted December 14, 2014 Share Posted December 14, 2014 Allan, please re-record both Vocabulary Builder tapes at higher volume. These two tapes (and Sketchpad, too) are recorded so quietly than the signal drops off to silence at many points. I'm not saying these WAVs are completely worthless - fully correct CAS files can be created from them, I've done it with Sketchpad - but the automatic process fails and a lot of manual intervention is required. So rather than spending a few hours creating the CAS files, I'd like to see if re-recorded WAVs would be any better. (You definitely should verify signal levels of all WAV files you're sharing, open them in a sound editor and make sure that the signal fills about 50-80% of the full range.) Atarimania-CAS-2014-12-13.zip Quote Link to comment Share on other sites More sharing options...
+Allan Posted December 15, 2014 Author Share Posted December 15, 2014 I've been recording them with Audacity. I'm a total nubby with this stuff so bare with me. I will re-record vocabulary builder. Thank you again. Allan Quote Link to comment Share on other sites More sharing options...
+Allan Posted December 15, 2014 Author Share Posted December 15, 2014 In the mean time...thanks to Kr0tki..... http://www.atarimania.com/game-atari-400-800-xl-xe-three-holiday-games_12540.html Merry Christmas. and here is Sketchpad. A brand new ancient program. http://www.atarimania.com/utility-atari-400-800-xl-xe-sketchpad_19614.html Allan Quote Link to comment Share on other sites More sharing options...
Kr0tki Posted January 6, 2015 Share Posted January 6, 2015 BUMP! So, any progress with Vocabulary Builder? Quote Link to comment Share on other sites More sharing options...
+Allan Posted January 6, 2015 Author Share Posted January 6, 2015 BUMP! So, any progress with Vocabulary Builder? I was having trouble loading stuff up on Atarimania FTP section. I'll give it another try today. Allan Quote Link to comment Share on other sites More sharing options...
thorfdbg Posted January 6, 2015 Share Posted January 6, 2015 (edited) - APX_Terminal_Emulator_2_0.wav - the program says it's actually version 3.1. Danger_in_Drindisti.wav - couldn't test it, as it requires a cassette version of Hellfire Warrior, which we don't have yet. Thanks, I checked myself the above file (and some others). While some of them just have the problem that almost all of the signal is lost, the above one is special: The tape seemed to have been in a pretty bad shape, causing high variations in the recording speed. The variations are easily audible when playing the file. In principle, this wouldn't be a problem because such variations can be recognized in the block header and can be compensated, but unfortunately, the speed breaks down right in the middle of the block, causing the serial transmission to run out of sync. Anyhow, I would be curious to learn how you made this tape accessible. Edited January 6, 2015 by thorfdbg Quote Link to comment Share on other sites More sharing options...
+Allan Posted January 6, 2015 Author Share Posted January 6, 2015 Thanks, I checked myself the above file (and some others). While some of them just have the problem that almost all of the signal is lost, the above one is special: The tape seemed to have been in a pretty bad shape, causing high variations in the recording speed. The variations are easily audible when playing the file. In principle, this wouldn't be a problem because such variations can be recognized in the block header and can be compensated, but unfortunately, the speed breaks down right in the middle of the block, causing the serial transmission to run out of sync. Anyhow, I would be curious to learn how you made this tape accessible. How I recorded it? With a regular stereo cassette deck hooked up to my Imac audio input and recorderd with Audicity. Originally I had the sound levels in Audacity to low. Allan Quote Link to comment Share on other sites More sharing options...
thorfdbg Posted January 6, 2015 Share Posted January 6, 2015 How I recorded it? With a regular stereo cassette deck hooked up to my Imac audio input and recorderd with Audicity. Originally I had the sound levels in Audacity to low. Allan Thanks - though I afraid there is possibly nothing you can do about it. The problem here is not the audio level (volume) but rather the pitch of the beep. If you listen carefully, you'll notice that the frequency of the beep goes down by about a third intermittedly from records 28 and on, probably at around 1:30 into the file. That's a typical effect of an aged tape, or a tape teck that has problems to keep up the speed of the tape transport. Actually, my question was less how to record it in a better way - after all, you got what you got - but which kinds of tricks have been applied to reconstruct from there. Following the usual Pokey serial decoder algorithm just shows that this frequency deviation is already large enough to make the serial decoder run out of sync. Quote Link to comment Share on other sites More sharing options...
+Allan Posted January 6, 2015 Author Share Posted January 6, 2015 Actually, my question was less how to record it in a better way - after all, you got what you got - but which kinds of tricks have been applied to reconstruct from there. Following the usual Pokey serial decoder algorithm just shows that this frequency deviation is already large enough to make the serial decoder run out of sync. You'd have to ask Kr0tki. He's more of the expert. I've just made the recordings and he's turned them into .CAS images. Kr0tki, I'm still getting timing out issues with the atarimania ftp. I'm not sure if my upload speed is to slow for the larger file and it's timing out or what. I'm upgrading my Internet provider speed soon so I'll have to wait and try it after I upgrade. In the mean time I'll try find another place to upload it. Allan Quote Link to comment Share on other sites More sharing options...
Kr0tki Posted January 8, 2015 Share Posted January 8, 2015 (edited) Actually, my question was less how to record it in a better way - after all, you got what you got - but which kinds of tricks have been applied to reconstruct from there. Following the usual Pokey serial decoder algorithm just shows that this frequency deviation is already large enough to make the serial decoder run out of sync.I'm gonna dissapoint you. A8CAS is not able to decode these frequency deviations properly. I've spent an hour or two examining the WAV file visually (in Audacity) and manually decoding the parts on which the automatic process had failed. It would be possible to improve the decoding algorithm to support these kinds of errors, but I think that it's not worth the effort - the WAV's that are that much problematic are few and far between, so I find the task of fixing the files manually less time-consuming than fixing the algorithm itself. You'd have to ask Kr0tki. He's more of the expert. I've just made the recordings and he's turned them into .CAS images.I believe the question was targeted at me. Kr0tki, I'm still getting timing out issues with the atarimania ftp.It might help to use an FTP client that supports file transfer resuming. I would recommend WinSCP, if you're on Windows. Edited January 8, 2015 by Kr0tki Quote Link to comment Share on other sites More sharing options...
thorfdbg Posted January 9, 2015 Share Posted January 9, 2015 I'm gonna dissapoint you. A8CAS is not able to decode these frequency deviations properly. I've spent an hour or two examining the WAV file visually (in Audacity) and manually decoding the parts on which the automatic process had failed. It would be possible to improve the decoding algorithm to support these kinds of errors, but I think that it's not worth the effort - the WAV's that are that much problematic are few and far between, so I find the task of fixing the files manually less time-consuming than fixing the algorithm itself. Wow, thank you for the tremedous work then. Actually, I'm more interested in the signal processing part of it than to really make the files accessible and I though you had probably used some kind of preprocessing that could be used to improve the current quality. Apparently, no. That said, I would have a couple of ideas for a better algorithm, but it's all quite an efford and a complete change of the (still rather simple) serial decoder that mimics the hardware. At this point, I'm already supporting speed variations below the block level, but what I have seen in this specific file was too dramatic and happened too quick to allow automatic correction. Quote Link to comment Share on other sites More sharing options...
Kr0tki Posted January 10, 2015 Share Posted January 10, 2015 Wow, thank you for the tremedous work then.It's worth it, when it comes to preservation of historical artifacts. (That's why when Allan said in another thread that he doesn't see the point in preserving disk copy-protection intact, my heart stopped - dude, we do not simply spread the software, we practice archeology!) Actually, I'm more interested in the signal processing part of it than to really make the files accessible and I though you had probably used some kind of preprocessing that could be used to improve the current quality. Apparently, no. Right, no complicated preprocessing involved. I simply use two matched filters - one for each frequency - and compare their outputs to determine the signal. (Same algorithm as in Altirra.) As the filters are matched to the two specific frequencies, I can't adjust them to support frequency changes due to tape speed variations. I'd have to throw the matched filters away and start to measure the signal's frequency, comparing it against the nominal frequencies of mark and space, and adjust the signal's "speed" according to that. Doable, but not worth it IMHO. I'm already supporting speed variations below the block level, but what I have seen in this specific file was too dramatic and happened too quick to allow automatic correction.A8CAS does that too - it adjusts the baudrate a little bit after each decoded byte, according to duration of that byte. But that obviously can't help when the incoming signal is incorrect due to frequency shift. Quote Link to comment Share on other sites More sharing options...
thorfdbg Posted January 10, 2015 Share Posted January 10, 2015 I'd have to throw the matched filters away and start to measure the signal's frequency, comparing it against the nominal frequencies of mark and space, and adjust the signal's "speed" according to that. Doable, but not worth it IMHO. A8CAS does that too - it adjusts the baudrate a little bit after each decoded byte, according to duration of that byte. But that obviously can't help when the incoming signal is incorrect due to frequency shift. Actually, my wav2cas has adjustable frequency filters, i.e. it has about 20 filter pairs that are selected from for optimal separation. That part works pretty well, and the problem is really not the analog filter - the output is pretty robust and continues to work for this particular file. The problem is really the emulation of the serial input. If you follow the file along the decoding, the speed variation is so drastic that the decoder runs out of sync, i.e. the bits become so long that the decoder samples some bits as two bits, and that *despite* that it monitors the distance between start and stop bit continuously and keeps the baud-rate adjusted. If the tape speed variation would be a bit smoother, it wouldn't be a problem at all, but here the speed breaks in more or less within a block. Quote Link to comment Share on other sites More sharing options...
Kr0tki Posted January 10, 2015 Share Posted January 10, 2015 Actually, my wav2cas has adjustable frequency filters, i.e. it has about 20 filter pairs that are selected from for optimal separation.Interesting idea. How do you choose the particular filter pair being used? If you follow the file along the decoding, the speed variation is so drastic that the decoder runs out of sync, i.e. the bits become so long that the decoder samples some bits as two bits, and that *despite* that it monitors the distance between start and stop bit continuously and keeps the baud-rate adjusted.To cancel the effect of varying tape speed, baudrate would have to be adjusted not by measuring decoded byte durations, but by continuously monitoring frequency fluctuations in the input signal. I hope you get the idea. Quote Link to comment Share on other sites More sharing options...
thorfdbg Posted January 10, 2015 Share Posted January 10, 2015 Interesting idea. How do you choose the particular filter pair being used? Highest amplitude that is consistent with the required interpretation. We know that the signal must be mark in the pauses, we know what stop bits are marks too, and start bits are spaces. Hence, we know already a bit which filter is which, and which filter pairs to exclude. To cancel the effect of varying tape speed, baudrate would have to be adjusted not by measuring decoded byte durations, but by continuously monitoring frequency fluctuations in the input signal. I hope you get the idea. Yes, but how? You can only sample the duration of entire bytes because you have two defined signals, the start bit and the stop bit. You do not know the duration of a bit - if the signal level is "mark", it can be the stop bit, one bit and the stop bit and so on. Thus, one can only guess. - Well, actually, one can do a bit more, one can try to estimate the probability of the number of bits, or correlate the bit length with the filter frequency. But that's all quite a bit more complicated than the current algorithm. It's probably a nice exercise in signal processing, unfortunately, my vacation is ending... Quote Link to comment Share on other sites More sharing options...
Kr0tki Posted March 21, 2015 Share Posted March 21, 2015 BUMP RETURNS! So, Allan, what about that Vocabulary Builder? Quote Link to comment Share on other sites More sharing options...
+Allan Posted March 21, 2015 Author Share Posted March 21, 2015 BUMP RETURNS! So, Allan, what about that Vocabulary Builder? We got a another way of moving big files so in the next couple of days I will send you a PM with a link to download Vocabulary Builder. Allan Quote Link to comment Share on other sites More sharing options...
tschak909 Posted March 22, 2015 Share Posted March 22, 2015 How is the quest for finding all the Dorsett tapes, coming? Do you think somebody actually has all of them? -Thom Quote Link to comment Share on other sites More sharing options...
Like us on Facebook
Follow us on Twitter
AtariAge on Instagram
Watch us on YouTube
Chat on Discord
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.