Just downloaded Mr. QuestionMan (thanks!) and now I have a bit of a quandry. I looked at one album in my collection with different utils, and got the following:
• EncSpot Pro: 8 x Blade, x 2 Gogo (before 3.0)
• AudioIdentifier: 10 x FhG
• Mr. QuestionMan: 10 x FhG
• dbPowerAmp: 1 x Xing (old), 3 x Blade, 6 x Unknown
The album was not my rip so I don't know what was actually used. As I understand, Mr Q and AudioID are variations of the same program or something, so no surprise that they show the same results.
I know that encoder identification is a bit of a guessing game, but I'm surprised at the variations. Is one of these utils generally thought to be more accurate than the others?
Full Version: Encoder ID Utils
If the mp3 contains identifying information such as Lame header, or FHG VBRI header then the encoder is know for certain. The rest is just guessing by all programs, the latest dbpoweramp beta will show if it is guessing or knows for sure.
QUOTE (spoon @ Jul 10 2005, 08:27 AM)
If the mp3 contains identifying information such as Lame header, or FHG VBRI header then the encoder is know for certain. The rest is just guessing by all programs, the latest dbpoweramp beta will show if it is guessing or knows for sure.
is that guessing engine based on some other code or did you come up with a complete new alogorithm? Is it any better than EncSpot or MQM?
I looked at various implementations and wrote my own.
AudioIdentifier is a fork from MrQM 0.6.
It seems that album is encoded with FhG: there is no reason to encode different tracks with different yet similar encoders.
It seems that album is encoded with FhG: there is no reason to encode different tracks with different yet similar encoders.
QUOTE (Mo0zOoH @ Jul 11 2005, 02:35 AM)
It seems that album is encoded with FhG: there is no reason to encode different tracks with different yet similar encoders.
...that would assume that the album WAS encoded collectively.
if you didn't do it, and you don't know the person who did, then it's a toss up...
what does mp3guessenc have to say about it? (look at the output info)
later
QUOTE (xmixahlx @ Jul 11 2005, 07:38 PM)
...that would assume that the album WAS encoded collectively.
if you didn't do it, and you don't know the person who did, then it's a toss up...
what does mp3guessenc have to say about it? (look at the output info)
if you didn't do it, and you don't know the person who did, then it's a toss up...
what does mp3guessenc have to say about it? (look at the output info)
Precisely. I don't know if the poster ripped the album or downloaded the tracks from some P2P app. I'll have a look at the tracks with mp3guessenc and see what it says. It'll probably tell me they're Qdesign or Gogo or something...
Well, that wasn't exactly helpful. MP3guessenc gave the same response for all 10 tracks:
"Maybe this file is encoded by dist10 encoder or other encoder"
I guess it will have to remain a mystery.
"Maybe this file is encoded by dist10 encoder or other encoder"
I guess it will have to remain a mystery.
IMHO, It's not a very realistic scenario, anyway.
One thing is clear: there must be a kind of standard not to guess the encoder used and get ridiculous results, but to know it for sure.
One thing is clear: there must be a kind of standard not to guess the encoder used and get ridiculous results, but to know it for sure.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.