85 replies to this topic

[Marius1976]

@krotki there is one thing you seem to overlook what ivop isn't

You ignore the fact that .cas is not perfect. It is an interpretation and not a 100% copy of the original.

I love to use real tapes. I love the loading sound. When I recreate a tape from a cas file there are three possibilities:

1. The created WAV file does not want to load on real equipment. I did not succeed to create a working tape from eastern front and some other games alrhough the cas file appeared to work properly.

2. The recreated WAV file does sound different than the original. I am blessed/cursed with a perfect pitched pair of ears and I am a professional musician. Those waves created by cas2wav are sounding completely different than the original. You say in fact : who cares, it is the data that has to be preserved. I disagree. The original sounding has to be recorded and saved. I am happy I have the record of original Glenn Gould Bach playing. Your statement is that with the music score and a 100% description of what mr Glenn Gould played his recording could be reproduced. That is a very poor theory and is absolutely wrong.

3. The last thing that can happen is that the conversion DOES actually work. This is the case in most situations.

[Marius1976]

What i tried to say with point 2 in my post above is, that for a real preservation project it counts that you safe todays experience for another generation of people.

I mean the complete experience!

If you listen carefully to your tapes while they are loading, you will notice that one tape or another has a different 'color' of sound. Some have a louder mechanical noise, while other tapes let us hear more the sound of the ATARI itself (the lower pitched sound). Also when you listen carfully you'll notice that sometimes the sound has some strange distortion.

All this is for me a part of the total Atari tape experience. This might sound weird to you, but I'm real into audio and that kind of things.

I still miss my modem calling out (now I have DSL connection for years)

Also with my Vinyl recordings. Even that sometimes a scratch or dirt is in the tracks of the disc, I love those side-affects. They are a part of the experience.

As soon as you 'only' have this .CAS files as archive, I don't see the benefit anymore of just using .XEX files.

It is not only the DATA that counts.

What you need is this:
First a first quality tape player and recording setup.
After you converted real tape to WAVE file, you need a setup that is able to TEST this WAVE file. I heard fox-1 telling something about emulators that accept a raw .WAV file? Right? This is a good first test.

Secondly
You need an interface (like the one in the link from krotki) where you can test that WAVE file on real equipment.

If it works. It works, and will always work. As long as the wave will not get corrupeted ofcourse.

If you are sure your hardware is ok, the tape is running at right speed, the audio is not disturbed (due to too much db or other reasons). Your WAVE file is a very good file format to store this 'forever'.

Next to that you could ALSO save a .CAS file, and even a .WAV created from the .CAS file.

But for me (and that was what I tried to explain in point 2 in the post above) is a WAV created from the .CAS file, is NOT as good as the original.

[Kr0tki]

Marius1976, on Thu Jan 12, 2012 1:04 AM, said:

@krotki there is one thing you seem to overlook what ivop isn't

From what I understand, your following points reflect F#READY's stance more than ivop's. As it can be deduced from my response to him, I am deliberately ignoring it, not overlooking. Keep the recording of the data track if you want, it's not my problem.

Marius1976, on Thu Jan 12, 2012 1:04 AM, said:

1. The created WAV file does not want to load on real equipment. I did not succeed to create a working tape from eastern front and some other games alrhough the cas file appeared to work properly.

That could be a failure in the recording procedure. I've seen various reports about problems with loading tapes recreated from CAS files and also from original WAVs. When recording a tape from a PC, one has to correctly adjust volume level, ensure that the data track gets recorded on the right channel, etc. This issue applies regardless of whether we record the tape from an original WAV or from WAV created from CAS.

Edited by Kr0tki, Thu Jan 12, 2012 8:15 AM.

[Marius1976]

Kr0tki, on Thu Jan 12, 2012 8:13 AM, said:

That could be a failure in the recording procedure. I've seen various reports about problems with loading tapes recreated from CAS files and also from original WAVs. When recording a tape from a PC, one has to correctly adjust volume level, ensure that the data track gets recorded on the right channel, etc. This issue applies regardless of whether we record the tape from an original WAV or from WAV created from CAS.

No it is not a record failure. It is a failure in the .cas system. Last year I have recorderd over 40 tapes. Every tape succeeded, and I have retried this Eastern Front more than 10 times and kept failing. The original tape I have works. I recorderded the WAV file... this WAV file works too. Creating a .CAS file from it works. Re-creating a .WAV file from it fails. This means: this recreated .WAV file does NOT work on real equipment.

Strange eh?

[Marius1976]

The only thing I want to clear with this is that it is at least doubtful whether .CAS is a good choice for preservation project or not.

When I create a .CAS from an original tape, and the .CAS works.
And the RE-creation of a .WAV file from this .CAS file fails, this means that it is a wrong idea that your .CAS archive is a safe way of preserve your so beloved atari tapes.

I know what I'm talking about. It is not meant to have a stupid no-go discussion.

[Kr0tki]

Marius1976, on Thu Jan 12, 2012 6:06 AM, said:

Also with my Vinyl recordings. Even that sometimes a scratch or dirt is in the tracks of the disc, I love those side-affects. They are a part of the experience.

Now I'm considering adding a feature to the emulator, which would emulate physical degradation of tape images during their use, including creating crooks in a tape when the Play button is left pressed for prolonged time while the motor is stopped. That's what I'd call complete experience.

Marius1976, on Thu Jan 12, 2012 8:26 AM, said:

No it is not a record failure. It is a failure in the .cas system.

There are many success stories of working tapes recreated from CAS files. So when a person comes and says it doesn't work for him/her, it's not necessarily the "system" that is to blame, you know.

Take one of your original WAV's and compare it with a CAS-recreated WAV file using some sound editor, Audacity for example. Adjust the loudness of the recreated WAV to match that of the original. If the CAS-created WAV is mono, add an empty left track. Then it will be more likely to work.

You may view the need of adjustment as a "failure", but, it's actually required part of the process of recreating a WAV form a CAS. It might change in the future, by updating the CAS software to allow to adjust volume level during conversion, but it's not that important to implement since it can be done manually. (And then, the volume level still will have to be adjusted manually, because the actual volume level of the WAV-playing computer is still a variable that's unknown when creating an image.)

Edited by Kr0tki, Thu Jan 12, 2012 8:51 AM.

[Marius1976]

@Krotki

You sound a bit 'p*ssed off' ... you do not seem to take responses very serious here.

The Atari hobby is all about experience. You don't have to feel the same experience, or think the same about things, but you make a fool of others (mine) responses, and I ask you politely to stop with that.

My daily job is for 50% audio editing, and I am 100% sure that the re-created WAV file is not having problems because of Loudness issues.
I don't feel the need further to continue with you on this.

My statement stays that .cas is unusable as a preservation format. Period.

[ijor]

ivop, on Wed Jan 11, 2012 5:35 PM, said:

I am not saying CAS files have no use, they definitely have. But for long time preservation one needs something better, too, IMHO...Sure, for the short term, just copying the damn file will suffice, but what'll happen in hundred years when we're all dead? This is where redundancy and error-correction codes and the like come in.

I have to agree with Kr0tki, at least on this. This is starting to be ridiculous.

Yes, of course, redundancy is important. Not just important, it is imperative. So let's have redundancy, you don't need to save WAV files for this purpose (it might be good, or not, for other purposes). There are much better and more efficient ways to provide redundancy. Make tens, hundreds, or even thousands of copies. Produce and save checksums, crc, hashes using SHA-1, SHA-2, MD5, etc. Do you want specific error correction besides full redundancy (which is obvious, an error correction in itself), then use par files, use Reed–Solomon codes.

Claiming that you need to use audio files for the purposes of redundancy, is akin to say that for preserving some text, instead of saving the ASCII code, you should print it, scan it, and save the scan of the hardcopy using something like a PNG (or TIFF) uncompressed format.

Quote

And, paraphrasing Freddy here, having a copy as near to the original as possible leaves open the possibility to reprocess the original in different ways in the future.

This is a completely different issue, I fully agree that this might make some sense. But claiming that the CAS (or similar) format is bad because it doesn't provide redundancy, IMHO, makes no sense at all.

Marius1976, on Thu Jan 12, 2012 1:04 AM, said:

@krotki there is one thing you seem to overlook what ivop isn't

As krotki is saying, you are mixing up the issues, between what was raised by Freddy and by Ivo. Not that I'm trying to make this personal in anyway. But let's not mixup the two debates. The two issues might be somewhat related, but they are different and separated.

Quote

1. The created WAV file does not want to load on real equipment. I did not succeed to create a working tape from eastern front and some other games alrhough the cas file appeared to work properly ... No it is not a record failure. It is a failure in the .cas system.

This certainly should be investigated. But instead of debating if this is a recording or a cas system failure, why don't you post the culprit WAV (and CAS) files?

Quote

What you need is this: First a first quality tape player and recording setup. After you converted real tape to WAVE file, you need a setup that is able to TEST this WAVE file.

Verification is very important. I agree. And ideally, I agree that we should try to verify on real hardware (not just under emulation). You could verify if the recreated audio from CAS works with real hardware or not. If it doesn't work, then you know that there is some kind of problem, and you could save the WAV file in these particular cases (disregarding if the recreated audio fails because of a CAS system issue, or because the original tape was damaged and it could be repaired, or any other reason). It doesn't necessarily means that you should save the AUDIO on every single tape, at least not for this specific reason.

Quote

I am blessed/cursed with a perfect pitched pair of ears and I am a professional musician...If you listen carefully to your tapes while they are loading, you will notice that one tape or another has a different 'color' of sound. Some have a louder mechanical noise, while other tapes let us hear more the sound of the ATARI itself (the lower pitched sound). Also when you listen carfully you'll notice that sometimes the sound has some strange distortion.

Well, that's very interesting. But you might want to elaborate about the technicals of the difference. Otherwise, "for the rest of us", it would be difficult to appreciate how much this is important or not. I guess that two major differences could be mono vs stereo recording (and in the latter case, if there was something in the audio track), and the recorded volume. The exact frequencies used might be different as well, but this should depend not only on the recording but also on the reproduction? And the absolute actual volume should depend on how you recorded the audio file as well, correct?

[Marius1976]

Quote

Claiming that you need to use audio files for the purposes of redundancy, is akin to say that for preserving some text, instead of saving the ASCII code, you should print it, scan it, and save the scan of the hardcopy using something like a PNG (or TIFF) uncompressed format.

The 'problem' is (what krotki already admitted) is what your explanation is of the word "preservation" ...

By calling something "some text" you exactly give the difference in opinion about things. When I talk about preservation of text I am ALSO interested in the original book (or at least the looks of the book), the type of ink, the size and weigh of the paper etc. Not in every case, but when it comes to preservation: YES. Indeed. I find those things interesting/important. I have all the letters I got from my wife when we were in the very beginning of our relationship years ago. Ofcourse I could preserve all the text (like you said) but I would be MORE happy with 100% hardcopies (or better the originals) in stead of just plain text of the content of those letters.

Do you feel the difference?

Quote

This is a completely different issue, I fully agree that this might make some sense. But claiming that the CAS (or similar) format is bad because it doesn't provide redundancy, IMHO, makes no sense at all.

At least I did not say, that is the reason why .CAS format is bad. I even did not say it is bad. I stated (and I still do) that it is not the right format for preservation. It is a very cool format: yes it is. And it has a lot of functionality, but simply not for this goal. It is a DATA-extractor. That is what it does. And it does that VERY nicely.

Quote

1. The created WAV file does not want to load on real equipment. I did not succeed to create a working tape from eastern front and some other games alrhough the cas file appeared to work properly ... No it is not a record failure. It is a failure in the .cas system.

Quote

This certainly should be investigated. But instead of debating if this is a recording or a cas system failure, why don't you post the culprit WAV (and CAS) files?

You are right. I will have to look things up for that. I will have them posted as soon as I have everything installed back for that.

Quote

Verification is very important. I agree. And ideally, I agree that we should try to verify on real hardware (not just under emulation).

I think that is the biggest problem today. Most active people seem to do almost anything on emulator, cross assemblers whatsoever. I only use the real thing. And I am only interested in the real thing. .CAS is cool, but for me personally it is another PC/Emulation related invention. And my experience with the creating REAL tapes from .CAS archives was that a lot of games actually did work on the real hardware, but also a lot did not work (and that were not only the copy protected tapes)

Quote

You could verify if the recreated audio from CAS works with real hardware or not. If it doesn't work, then you know that there is some kind of problem, and you could save the WAV file in these particular cases (disregarding if the recreated audio fails because of a CAS system issue, or because the original tape was damaged and it could be repaired, or any other reason). It doesn't necessarily means that you should save the AUDIO on every single tape, at least not for this specific reason.

Well... the most important thing is that EVERY file that you store away, and you THINK it is safe, should be double checked. When you really need to restore a tape (because the original is lost forever), and then you find out the preserved file isn't functional, it is too late. And that is what my experience is from what I was doing with real tapes: the .cas archive is not a reliable source for the purpose of creating REAL tapes, to run on REAL hardware. And again: the volume level was perfect for atari tape.

Quote

Well, that's very interesting. But you might want to elaborate about the technicals of the difference. Otherwise, "for the rest of us", it would be difficult to appreciate how much this is important or not. I guess that two major differences could be mono vs stereo recording (and in the latter case, if there was something in the audio track), and the recorded volume. The exact frequencies used might be different as well, but this should depend not only on the recording but also on the reproduction? And the absolute actual volume should depend on how you recorded the audio file as well, correct?

The reason I brought that 'ear thing' up was to prove the fact that I can judge whether the by cas2wav created WAV file differs from the original. The fact that the NEW created WAV file is different than the ORIGINAL WAV file, is for me a reason not to chose .CAS as a preservation file format.

I do not want to sound stubborn, or j*rky or whatever. I want to explain with serious motivations, why .cas is not the first choice when you seriously consider a preservation project for Atari tapes.

The .cas 'lovers' seem only to be interested in the preservation of the DATA of the original tape. I think that is not the right motivation.

[ijor]

Marius1976, on Thu Jan 12, 2012 2:10 PM, said:

Quote

Claiming that you need to use audio files for the purposes of redundancy, is akin to say that for preserving some text, instead of saving the ASCII code, you should print it, scan it, and save the scan of the hardcopy using something like a PNG (or TIFF) uncompressed format.

The 'problem' is (what krotki already admitted) is what your explanation is of the word "preservation" ...

By calling something "some text" you exactly give the difference in opinion about things. When I talk about preservation of text I am ALSO interested in the original book (or at least the looks of the book), the type of ink, the size and weigh of the paper etc. Not in every case, but when it comes to preservation: YES. Indeed. I find those things interesting/important. I have all the letters I got from my wife when we were in the very beginning of our relationship years ago. Ofcourse I could preserve all the text (like you said) but I would be MORE happy with 100% hardcopies (or better the originals) in stead of just plain text of the content of those letters.

Do you feel the difference?

No, no, no. Please Marius, I asked you to not mixup the two issues, and you are once again mixing them up. I was talking about oranges, and you are replying about apples.

This paragraph you are quoting and replying has nothing to do with what is the meaning of preservation, and what you should preserve. It was only about the redundancy issue raised by Ivo.

Ivo is claiming that CAS format is bad for preservation purposes because it has no redundancy. Note once again, this is beyond, and has nothing to do with the original issue raised by Freddy (and that you continue), that CAS doesn't preserve all the information we want to preserve. Ivo is claiming that we need to use a format that provides redundancy (such as an audio dump), because otherwise a format like CAS would eventually degradate or corrupt and we'll lost everything.

It was about that that I was replying in the text you quoted me above. I think it was more than clear than (the quoted text) was in reply to Ivo, not to you, not to Freedy. And I think it was more than clear that I was addressing there, only the redundancy issue. So please, do not reply me with a preservation concept when we are talking about a pure technical redundancy concept.

And when I gave the example of text, I didn't mean about an originally printed text. I meant just about any text that you want to preserve. So let me rephrase. Consider some text that was written say, by Bill Gates on Notepad, or by Steve Jobs (in whatever is the equivalent of notepad on the MAC), or some text written by Linus Torvalds on ED (Linux basic editor). There was never an original hardcopy. Text existed only virtually on pure ASCII format. Do you still believe that we should make a hardcopy, scan it, and save the picture of the scan, just for the purposes of having some redundancy? Because that's (more or less) what Ivo is claiming.

Also, I don't think your analogies are fair to the debate (we all agree that we should save a picture of say, a manuscript hand written by Shakespeare, and not just the text).

Quote

And my experience with the creating REAL tapes from .CAS archives was that a lot of games actually did work on the real hardware, but also a lot did not work (and that were not only the copy protected tapes)

I'm afraid I don't have that much experiencie with CAS dumps to judge. But again, there is no point debating this without samples. Please do post some of that "lot that did not work".

Quote

When you really need to restore a tape (because the original is lost forever), and then you find out the preserved file isn't functional, it is too late.

But what you are saying (if I understand you correctly), is that they aren't exactly non functional. They are non functional on real tapes with real hardware. That's quite different than a dump (on whatever format) that doesn't work at all. In this last case, then chances that the information was indeed completely lost. But if the CAS works, then the information is likely recoverable.

May be there are some bugs in the utility to write back tapes from CAS files? May be there are some limitations on the CAS files, and may be the CAS format should be extended (and I understand there is already some work on that). This doesn't necessarily means that a "CAS like" format is not good enough for preservation. But again, we'll know better when you'll post those WAVs and CAS files.

Quote

The reason I brought that 'ear thing' up was to prove the fact that I can judge whether the by cas2wav created WAV file differs from the original. The fact that the NEW created WAV file is different than the ORIGINAL WAV file, is for me a reason not to chose .CAS as a preservation file format.

I don't think that's a reason good enough. In first place, because it might be just an issue of enhancing CAS2WAV. In second place, because multiple copies of the same original tape might sound different. And unless you want to preserve every single copy, then you could ignore these differences.

You said that you were feeling a different experience. This is something, to me, important. But if you don't elaborate about that, then it would be very difficult to judge.

Quote

The .cas 'lovers' seem only to be interested ...

And now you are being despective. Personally I don't mind because I am not really a "CAS" fan. As I said, I'm not even too familiar with "CAS". But once you start being despective, then don't be surpised if somebody else feels pissed off.

Edited by ijor, Thu Jan 12, 2012 3:35 PM.

[Marius1976]

@ijor

I do not mix up things for sure! Read the whole thread. My first response was a bit inclear because I used the wrong name indeed. My response is at the first post of krotki where he writes this:

Quote

I hope you are thinking only about tapes with audio tracks - for other tapes keeping a CAS image is enough, as its contains all data needed to recreate such tape.

My posts are against that sentence. It makes no sense what he wrote there (why, read my previous posts again please).

As F#Ready also wrote. It is perhaps a bit hard to communicate in foreign language. (I'm Dutch)
I placed to be sure Quotation Marks/Signs (here: i don't know the word again). in the word 'LOVER' (.CAS 'LOVER')

If you read the complete thread again, you will find that krotki is rather defending the .cas system. I summarize this by writing about .cas 'lovers'.

If you want .cas files that do not work, simply browse to:
http://cas-archive.pigwa.net/cas1.htm

Again, to be clear:
the ONLY reason I'm writing about this, is to explain, why it is better to preserve WAVE files for a preservation project in stead of .cas
I'm not mixing up things. This is the only thing I'm trying to contribute in this thread.

Btw. it was already clear to krotki I mixed up the NAMES (ivop vs F#ready) (not the subjects).

[Marius1976]

Since I can not find my original Eastern Front 1941 tape so soon. I ordered a new one on ebay.

I will record this tape on my MBP and test the file by record it back to a fresh tape. That record will work (if the tape I bought on ebay is ok of course).

Then I will create a .CAS file from that WAV file. That .CAS file will probably work too. I will post both original WAV and CAS files here.

Then I will create a WAV file from this .CAS file... and record that one back to tape. And you will see: that one will fail. (Yes audio settings are ok ofcourse)

I will post everything.
But in the meantime: you can have same experience, when you try some of these projects yourself. Also the fact that there circulate (enough) not working .cas files must be a sign I guess.

[ijor]

Marius1976, on Thu Jan 12, 2012 3:46 PM, said:

If you want .cas files that do not work, simply browse to: http://cas-archive.pigwa.net/cas1.htm
...
But in the meantime: you can have same experience, when you try some of these projects yourself. Also the fact that there circulate (enough) not working .cas files must be a sign I guess.

What I would do with non working CAS files? Bad CAS files are completely useless and the fact that they circulate don't mean anything at all.

A working CAS that doesn't work on real hardware is interesting. A working WAV that doesn't work when converted to CAS is interesting. A non working CAS alone is not interesting at all, why it would be?

The rest of the issues, I decided it would be better to move to PM. I sent you a PM before that I assume you already received.

[Marius1976]

@IJOR,

Thanks for your PM. I will come back to it later.

I will provide you enough interesting material as soon as I have my tape-setup for this back up running AND when I have received my Eastern Front tape. Since my movement I have put those tapes in a box, and it is one of the boxes somewhere I don't know which one it is. So that is why I ordered a new one yesterday.

You will hear from me soon (I hope)

[charliecron]

Eastern Front is an interesting tape I think because the tape will run on a 16K machine, I believe I read it in the manual, or maybe it was 8K machine I can't remember, but I labeled my conversion 16K so it's probably 16K. And I think at the time I tested a couple file versions, cart version on a 16K machine and they wouldn't work. So the tape really worth preserving.

Anyway here are links to my conversion if you would like to play with while your waiting for your new tape. This is from an original APX tape.

Master1.wav is the original recording.
Master2.wav is the original recording put through a sox filter, and what I would have used to make the .cas.

Pretty sure these files are too big to attach here..

http://www.local-spi...ont_APX_16k.cas
http://www.local-spi...cas/Master1.wav
http://www.local-spi...cas/Master2.hex
http://www.local-spi...cas/Master2.wav

[Marius1976]

Hi There,

Thanks Charliecron for providing this. I created a fresh .WAV file by using your .CAS file and krotki's tool. I recorded this back to tape, and indeed this works perfectly.

It's a pity I can't find my Eastern Front tape, I was using.
I'm also going to look for the other games I have had problems with.

If now comes out that there was a problem with the software I have been using for this, and it is always possible to create a fully functional WAV file (compatible with real hardware) I will change my mind, and apologize ofcourse for my statements. But first I want to reproduce the problems I had in the past. Perhaps I used an outdated version of the conversion tools.

One more thing to keep in mind. If the goal is to preserve atari tapes for a Looooooong time. Make sure that the .CAS -> WAV converter will still work in a long time on the newest computers that exist THEN. The chance that .WAV files will be playable in 100 years is perhaps bigger than the possibility to still being able to run those .CAS related tools.

[charliecron]

Great, success.

One thing to keep in mind. A couple years ago I was doing some conversions and I was having a terrible time with a tape that worked fine on real hardware. Someone here at atariage suggested it may be related to system load, the system could have trouble keeping up when recording, and you can miss samples. I believe I was recording in a virtual machine, and sure enough when I used Audcacity natively on the host, I succeded the first time!

I suppose maybe the reverse could also be true? When playing a wav file, maybe sound cards can sometime miss samples on playback. Who knows. One thing some folks who are really into tapes do is rather than make the recording from a sound card source, they will record it on to CD ROM first, and then make the tape from that. They don't trust the DAC. I guess nice audio CD players have some higher resolution DAC's (24bit vs 16bit? I really don't know). I haven't done this, but who knows maybe there is something to it.

If your into audio even a little bit, I suppose you know how to bias a tape, but if not definatley get a tape deck with adjustable bias and even adjusting by ear will produce superior recordings.

[ivop]

If I seemed to imply that redundancy is the sole purpose of using wave files for preservation, I'm sorry about that. It's just something extra on top of being closer to the original.
For day to day usage I'd use a .cas file, too.
@marius, I hadn't even thought about the timbre of the sounds. Thanks for pointing that out.
As for whether the arcane skill of copying data to preserve it will be lost, I don't think so, but it can be forgotten (i.e. the act of copying). For reference, think about all the celluloid films from the early 20th century that are already lost because they were not copied soon enough and those that will be lost in the coming years because there is not enough time and manpower to preserve them.
BTW There are people that actually make hardcopies of digital text to prolong preservation in case making a digital copy in time is forgotten. Quite recently somebody on here wrote he made hardcopies of all Atari related documents he downloaded from the internet. A decent pigmented ink on quality paper and stored under proper conditions will be good for hundreds of years.
Yes, this might be a bit overkill, but who knows....

Edited by ivop, Sat Jan 14, 2012 6:19 AM.

[ijor]

ivop, on Sat Jan 14, 2012 6:18 AM, said:

BTW There are people that actually make hardcopies of digital text to prolong preservation in case making a digital copy in time is forgotten. Quite recently somebody on here wrote he made hardcopies of all Atari related documents he downloaded from the internet. A decent pigmented ink on quality paper and stored under proper conditions will be good for hundreds of years.

But that's NOT what you are suggesting, is it? Saving a physical hardcopy makes a lot of sense (that's quite different than saving a digital scan of the hardcopy). Using a physically different media can certainly help. And if we could save everything in something like the gold record used in the Voyager probe, then that would be great (but then, lots of such records, might mean lot of pollution? )

But you weren't talking about a different media, did you? You were talking about using the same weak media ...

Quote

As for whether the arcane skill of copying data to preserve it will be lost, I don't think so, but it can be forgotten (i.e. the act of copying).

When and if would be forgotten, then the actual encoding (CAS or WAV) wouldn't make much of a difference. Data on weak media will be eventually lost no matter how you encode the information. Redundancy would help, but as I said already, use the methods most efficient for redundancy. And even redunancy won't help too much in the very long term.

[Kr0tki]

ijor, on Thu Jan 12, 2012 3:26 PM, said:

I'm afraid I don't have that much experiencie with CAS dumps to judge. But again, there is no point debating this without samples. Please do post some of that "lot that did not work".

I was going to ask myself - as a developer of "the tool" I'm naturally the Person Most Interested in solving this problem. Marius1976, if you still have those CAS images, please post them here. Also, tell us what version ov CAS2WAV software did you use for conversion. As you yourself noticed yesterday, quality of CAS->WAV conversion has improved recently. I heard (not tested though) that some old version of CAS2WAV had its problems regarding that.

ijor, on Thu Jan 12, 2012 3:26 PM, said:

This doesn't necessarily means that a "CAS like" format is not good enough for preservation. But again, we'll know better when you'll post those WAVs and CAS files.

Obviously, the CAS format is NOT good enough for storing the audio track, or storing the data track as audio track. But it's well enough for preservation of the data on the data track. Today the CAS format can store all possible data from the track, meaning it can archive even copy-protected tapes with non-standard sequences of signals. That's why I was so defensive when you mentioned "a failure in the CAS system" - the format is by design able store all data from the data track, so if there are problems with recording CAS images to tape, it can be caused by a bad particular CAS file, or a broken software, but not by the format itself.

Marius1976, on Fri Jan 13, 2012 3:39 PM, said:

Thanks Charliecron for providing this. I created a fresh .WAV file by using your .CAS file and krotki's tool. I recorded this back to tape, and indeed this works perfectly.

Glad to hear this. Was it necessary to make any adjustments to the resulting WAV file before recording? As a professional in audio processing, do you expect any additional functionality in a8cas-convert regarding adjustment of the resulting WAV files?

Marius1976, on Fri Jan 13, 2012 3:39 PM, said:

The chance that .WAV files will be playable in 100 years is perhaps bigger than the possibility to still being able to run those .CAS related tools.

It is equal to the chance that in 100 years anybody will care for Atari tapes. Preservation of anything in digital format includes preservation of the software that operates that format, or at least documenting the format. In this case the documentation and the source code of the software is freely available for copying, you really can't hope for more.

charliecron, on Fri Jan 13, 2012 7:55 PM, said:

I suppose maybe the reverse could also be true? When playing a wav file, maybe sound cards can sometime miss samples on playback.

Not really an issue of a sound card, but again of a system being under heavy load. In such case it fails to keep sending the necessary data to the soundcard.

ivop, on Sat Jan 14, 2012 6:18 AM, said:

If I seemed to imply that redundancy is the sole purpose of using wave files for preservation, I'm sorry about that.

Yeah, that's how it sounded to me.

ivop, on Sat Jan 14, 2012 6:18 AM, said:

As for whether the arcane skill of copying data to preserve it will be lost, I don't think so, but it can be forgotten (i.e. the act of copying).

That would mean that the whole idea of preservation would be forgotten. Act of copying is the main point in preserving the data in digital format, specifically for the reason that the copies are 100% identical and it can be verified that they are, just to be sure. If the people have stopped copying the data once destined to endless preservation, then it would affect all of the data regardless of what file format it's stored in. No point in having a few redundant megabytes in a WAV file, when the mere existence of that file is already forgotten by the humanity.

ivop, on Sat Jan 14, 2012 6:18 AM, said:

BTW There are people that actually make hardcopies of digital text to prolong preservation in case making a digital copy in time is forgotten. Quite recently somebody on here wrote he made hardcopies of all Atari related documents he downloaded from the internet. A decent pigmented ink on quality paper and stored under proper conditions will be good for hundreds of years.
Yes, this might be a bit overkill, but who knows....

You can also print the CAS file you know. :-)

[Marius1976]

Kr0tki, on Sat Jan 14, 2012 8:46 AM, said:

Glad to hear this. Was it necessary to make any adjustments to the resulting WAV file before recording? As a professional in audio processing, do you expect any additional functionality in a8cas-convert regarding adjustment of the resulting WAV files?

I did not make any adjustments and I was really surprised by how good it worked. The WAV file created by your tool sounded very stable and the signal was strong and without distortion. when I record to tape I played the file back on my MacBookPro at 75% of the max volume. On my tapedeck I first adjusted the incoming level so that there was 1 bar of the vu meter 'red' colored, then I precisely turned the incoming level back so that red bar only flashed up sometimes. Not too loud, but definately a strong signal.

For people who like to hear the ATARI load sound (in stead of the high pitched recorded sound): record only the RIGHT channel. When you load this tape you will hear the lower pitched atari sound. I love this sound, it gives me some diagnostic information (by ear) of the stability of the loading. On the other side: when your TAPE loader has an I/O sound off instruction... you will hear nothing ofcourse.

When you want to create a copy as close as the original: don't do this.

@krotki I apologize in advance to you ... I think we both are really enthusiastic about the things we do. I have to admit on a certain point I got a bit too fanatic about this subject. I must admit that your .CAS tools do really a very fine job. IMO a very good task for those tools is creating a fresh-and-new WAV file. When you record this fresh-and-new WAV file to a brandnew tape of good quality you will have fabulous new tapes.

I did this to a lot of my tapes. It works great.

When I have found the non-working tapes. I will let you know in this thread. It is very interesting to know WHY things do not work.

One more thing: how is your experience with XEX2CAS (and then creating WAV files). I also had a lot of trouble doing that, while I really wanted to create tapes of certain XEX files.

When everything works so good as this, I really take back my words. I am still going to create my own audio-library of sampled Atari tapes, but I'm definately going to keep the .cas tapes as well.

[www.atarimania.com]

Something I don't get... You want to create "fake" tape images from executables / cracks yet insist on wanting to preserve and relive the true experience? That doesn't make sense.

--
Atari Frog
http://www.atarimania.com

[Marius1976]

@www.atarimani.com

No. I want to preserve ORIGINAL tapes, and indeed: using the original WAVs
When it comes to preservation, that is my preference.

But...
I also like to use tapes. Also playing games that never exist like tape is fun. So that's why I asked for XEX2CAS.

It's perhaps better not discussing THAT in this thread, since it is a preservation thread. I understand your confusion. I hope I cleared it up with this?

[www.atarimania.com]

Yes, I get the point now.

--
Atari Frog
http://www.atarimania.com

[Kr0tki]

Marius1976, on Sat Jan 14, 2012 9:09 AM, said:

For people who like to hear the ATARI load sound (in stead of the high pitched recorded sound): record only the RIGHT channel. When you load this tape you will hear the lower pitched atari sound. I love this sound, it gives me some diagnostic information (by ear) of the stability of the loading. On the other side: when your TAPE loader has an I/O sound off instruction... you will hear nothing ofcourse.

When you want to create a copy as close as the original: don't do this.

That's why I treat the sound of the data track as irrelevant: It was never meant to be heard. Properly recorded tapes (including the ones produced by our program recorders) have the data track recorded only on the right channel. However there exist commercial tapes (LK Avalon, for example) that had that track recorded on both channels, probably only because their proffessional audio equipment (Avalon had one that made twenty something copies at once) only recorded in mono.

Additionally, the variety of timbres of the data signal as noted by Marius1976 - it was never an intentional effect. It's always an effect of a particular characteristics of the audio recording equipment, of the characteristics of the tape itself, finally it's the effect of the signal being degraded from storing for 25 years.

Marius1976, on Sat Jan 14, 2012 9:09 AM, said:

@krotki I apologize in advance to you ...

I don't think you have any reason to apologise for.

Marius1976, on Sat Jan 14, 2012 9:09 AM, said:

When I have found the non-working tapes. I will let you know in this thread. It is very interesting to know WHY things do not work.

Great idea, I'm looking forward to it.

Marius1976, on Sat Jan 14, 2012 9:09 AM, said:

One more thing: how is your experience with XEX2CAS

XEX2CAS, the one from Schreurs' website, is bad on two levels.

One, it adds a crappy initialiser before the DOS file, the well-known "exclamation mark". It takes 6 blocks to load, and makes a few calls to the OS not using the vector table. It works on the XLs only by chance.

Two, the tool doesn't add longer gaps after blocks that write to INITAD. When a software initialises itself during loading, the process fails.

That being said, it works quite well for many files.

I have once hacked it up to fix the above issues, replacing the "exclamation mark" with a loader based on the one used in LK Avalon tapes (takes one block), and making it add a 10 second gap after each INITAD block. Still not the best solution - the initialiser should stop the motor after INITAD blocks. One day I might fix it.
Never published it though, so here's the source code for those that can compile it themselves. I don't say it works with all DOS files, but maybe it's better for some. I'd like to know with what files it fails. It definitely doesn't work with files that load below $0780 - the initialiser resides there.

Attached Files

•  xex2cas.cpp.zip   3.17K   14 downloads

Edited by Kr0tki, Sat Jan 14, 2012 1:56 PM.