MP3's

[Craig Maier]

Most of us do not like Microsofts products. This is especially true for those of us who have to develop programs to run on their platform.

But that is not the point. We do not want to cut off our noses to spite our faces. Micrsoft is not going anywhere in all probability. So, if I want to have an archive of compressed audio, I am going to want to be confident that I will still be able to play those files 20 years from now.

Wil ape for flac still be supported then?

[Craig Maier]

I thought that it would be interesting to take GB's data and convert it to % of the original file size. Here is GB's data presented in that way:

.flac: 31.1 %

.wma: 30.2 %

.ape: 29.5 %

320 bit mp3: 22.7 %

Maximum Variance between non-lossy formats: 1.6 %

Maximum Variance between all formats: 8.4 %

My Conclusion: The non-lossy formats produce essentially the same data reduction. The 320 bit mp3 is probably not worth the additional 8.4 % data reduction savings over non-lossy compression considering that information is tossed out.

[Geebster]

Like I said earlier, technology doesn't stand still; so there is no need for the product we use today to last indefinitely. It is a natural progression. Aside from us hobbyists, most of the general public today do not own turntables. Does that prevent them from listening to the BEATLES "White Album"? No. Because most people have moved on to iPods, CD or other media, and our restorations we do today will ultimately have to be transferred to some other format.

So, I have no problems using a computer program (of any kind) that may not be around long term. If the program is good and does what I want it to do, I will use it.

I, in no means think OVERALL MS is a bad company. I just can't abide by their media division's tactics. So, I will use .ape or .flac until something better comes along.

GB

[Geebster]

Sorry, Craig!

I could have put the percentages down! I didn't know if anybody would be interested in that, but thanks for putting it down, so everybody else could see it.

GB

[Craig Maier]

I am kind of conservative with these sorts of things (formats, etc). I strictly use:

1. .wav

2. Red Book CD Audio

I have no mp3's, .flacks, .ape's or anything else like that. I do not even own an i-pod or anything similar. Guess that I am just getting old.

[Geebster]

I am kind of conservative with these sorts of things (formats, etc). I strictly use:

1. .wav

2. Red Book CD Audio

I have no mp3's, .flacks, .ape's or anything else like that. I do not even own an i-pod or anything similar. Guess that I am just getting old.

Ha Ha... I bet you and I are close to the same age .... I may be older; when somebody asks my age I usually reply: I am older than dirt!

I do have some MP3's but after performing my tests, I don't even want to use them.

GB.

[cromag]

Quite a few mp3 players will play .wav files, and you can find them built around hard disks up to 60GB (or larger), so don't let dislike of mp3s hold you back.

I mostly use my mp3 player, an "old" (2002) Creative Nomad Jukebox 2 with a modest 10GB disc, when I go for long walks with my dog. The music is important, but so is the ability to hear oncoming traffic, hostile dogs, etc., so I use a good pair of "open" headphones -- Koss SportaPros. Most of my mp3s are at 128 kbps -- I can hear the difference in quality between these and the source in the house, but on the street they sound great.

I think the user base will keep mp3s around for a long time. There will probably be a way to listen to .WMAs, but Microsoft has a history of being without loyalty when they decide that a change in technology will be of benefit to them, and new encoders would have to be licensed from them, so I wouldn't bet the farm. Last I've heard, mp4 may be the next big file format.

[cromag]

And don't worry about getting old, guys. It's a lot better than having your age carved in stone, if ya catch my drift!

[Geebster]

Quite a few mp3 players will play .wav files, and you can find them built around hard disks up to 60GB (or larger), so don't let dislike of mp3s hold you back.

I mostly use my mp3 player, an "old" (2002) Creative Nomad Jukebox 2 with a modest 10GB disc, when I go for long walks with my dog.

Last I've heard, mp4 may be the next big file format.

The media player I use will play ANY media format; audio or video, so that is why I am trying to stay away from the MP3 format. It supports among all others: .wav, .flac and .ape. File size for listening really isn't a issue, it's only for long term storage that size becomes important for me. Having 5 or more full length albums on 1 CD is nice when you are trying to locate a specific album or track.

Call me old fashioned but I don't use a portable device of any kind. I have the luxury in living in a secluded cul de sac next to a large woods and when I take long walks, I prefer to listen to the sounds of nature. Birds chirping, insects buzzing, chipmunks chatting are the sounds I like to hear when I'm out in nature. Now if I lived in a city atmosphere, I probably would use a portable just to block out the noise pollution.

MP4 is a variant of QuickTime and has been around since around '98 or '99. It has the advantage of being a "compartment" format, so it can handle a lot of differing types of media. Right now it is popular because of the popularity of the iPod series of products. I see it being offered as a 'choice' among formats, but rarely the first choice. I think we are just in the beginning stages of format development. In the next few years, I see many new formats emerging.

GB

[Craig Maier]

Given that there is no free lunch, I wonder what the data error rate of the non-lossy compression techniques are compared to non-compressed files. I have been told that non-lossy compression works on the principle of storing the incremental value of a sample rather than the entire value. Since the delta value from one sample to the next is usually less than the sum of each value, compression is achieved.

I once heard an accountant say that the road to hell is paved by the technique of incremental accounting. This sounds like a very similar situation. Hmmmm.

In other words, the errors build up over time. One error carries forward forever or until the next error correction frame.

[Geebster]

In other words, the errors build up over time. One error carries forward forever or until the next error correction frame.

Yes, I wondered about that too. I remember when the ZIP programs first came out and they often corrupted the original files in the process.

Given that this scheme has the potential of creating errors, would they be reduced (as much as possible of course) if you compressed a given file only once. If you had to restore a file you could from the original compressed, but never compress a file that was once compressed?

I don't know if that makes any sense... It seems to me that errors would build up to the point of no return if you kept compressing...decompressing...recompressing the same file.

But, since I would be using these files as an emergency back-up; in theory I shouldn't have to touch them much.

It is very interesting though.

GB

[Geebster]

Oh, speaking of which Craig...

That is another thing I did on my experiment, that I didn't document in my post. After compressing all of the files, I decompressed them to see if there was any change in the file size after the process.

I did this to satisfy myself more than anything. I can remember in the instance of the ZIP programs, that sometimes they used to alter the file size by a byte or two, back in the day when they first came out.

That wasn't the case in regards to the audio compression schemes. They ALL left the original file the exact size after compression/decompression.

FYI

GB

[Craig Maier]

Quoting GB:

"They ALL left the original file the exact size after compression/decompression."

-----------------------------------------

That is an excellent sign; I am not knowledgable about non-lossy audio compression file formats so that is why I posed the general question about that facet of things.

[Geebster]

Yeah. In my own un-scientific mind, I thought that was promising too.

I'm sure they all do this to some degree, but I thought I would share what the author's of Monkey's Audio say about errors and their product specifically.

"Monkey?s Audio incorporates redundant CRC?s to ensure proper decompression of data (errors never go unnoticed)"

GB

[Dan McDonald]

I've stored some very large files in shorten and flacs formats and have always gotten the same filesize when I uncompressed also. 45 minutes or so of music per file.