FretfulDave wrote:I'll try to go through this succinctly...
There are a number of formats offered on archive. Two formats are what are referred to as "lossless" formats and the MP3s are "lossy" or formats that contain data loss.
All files that are compressed are decompressed as they are played back through your music player.
The files put up on archive.org are compressed from their original format. Two of the format types are "lossless" meaning that they are compressed to reduce the file size but the compression algorithm does not remove information from the original recording. When it is played back and decompressed, all information is restored and you listened to a duplicate of the original recording. These formats are SHN (Shorten) and FLAC (Free Lossless Audio Compression).
The other format is MP3, which as I have noted is a "lossy" format. What this really means is that when the original file is compressed, information (the bits) are thrown away to help get more compression and an even shorter file compared to what a lossless compression algorithm can produce. When you play back an MP3, the file is decompressed but you are not hearing all the information that was in the original recording, so it is either "interpreted" back into the data stream or in the case of silence, just replaced with timed silence, if you will.
So continuing the blather, the MP3 compression algorithm has numerous compression degrees that can be selected before the original file is compressed and this information is written into the MP3 file header so as to let the playback program, with the decompression algorithm, know how to appropriately decompress the information. The MP3 algorithm can compress at rates of 32, 40, 48, 56, 64, 80, 96, 112, 128, 160, 192, 224, 256, and 320 and also a mode called VBR. And guess what the lower the number the smaller the file and the more information that is thrown away from the original recording. I.e., it is a lower quality reproduction. OK, keep reading... don't fall asleep yet, the answer you were looking for is...
VBR stands for Variable Bit Rate compression. so VBR uses intelligence in the algorithm to do things like detect silence and compress that at 32 but when there is a full fabric of sound, only compress at 320 to produce the best possible compressed and lossy sound reproduction..
So, without a doubt VBR is FAR, FAR better than 64.
To my ear, 128 is about the lowest I will accept for MP3s. When I produce MP3s of my various jams that I record, that is what I compress at. I cannot tell much difference between VBR and 128, though one friend of mine say he can and strives to get only VBR MP3s.
For those of us older deadheads, 64 sounds kinda like 3rd, 4th or more, generation tapes to me. Another way to express it is 64 sounds a bit like listening to a recording over the old phone system (not cell phones. horrible audio). 32 sounds like talking through tin cans and string (or a cell phone...).
I never load anything but the FLACs or SHNs myself. Then I will produce my own MP3s, if I so desire and that allows ME to choose what level of lossy compression I will apply. I usually restore the FLACs or SHNs to audio CDs, which then are perfect replicas of the original recording. If you restore MP3s to CD audio format, you definitely produce a degraded result.
Granted to 99% of everybody, you will have a hard time telling a 128 stream or a VBR stream from an uncompressed recording. But for me, I believe I can detect a difference. Besides, there is the principle of the thing. I.e., retaining the music as it was originally recorded.
I know a lot of this information is already in various topics on the board, so apologies for consuming additional (and I hope compressed) bit space on the server.
Return to Tapers Section
Users browsing this forum: No registered users and 1 guest