Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: dist10 encoders (Read 8487 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

dist10 encoders

Out of historical interest, I tested out a few famous (and famously crappy  ) dist10 based encoders. SoloH, 8Hz, Blade and Plugger.

All of them except for Plugger produced the same result (with Blade being faster than the rest), but Plugger produces a sound that is different altogether. It is much "softer" in the fatboy sample, but even more metallic and smeared than the other dist10 encoders in real music, I can upload samples if you want.

Why is this? Is Plugger really dist10, or some other beast entirely?

dist10 encoders

Reply #1 –
[a href='index.php?showtopic=4967']This thread (click)[/a] may be of some interest.

dist10 encoders

Reply #2 –
Thanks, but I've already read that thread before. I just checked my version of Plugger, and yes, it is the "ultimate" plus version  . Does it really lack any psychoacoustic model, or just the iso one? I am not sure if mp3s can even be made without psychoacoustics, to me it seems like dehydrated water, or non-poisionous lead. On the other hand, considering it sounds even worse than Shine I wouldn't be surprised if this was in fact true.

Anyone has the original Plugger or knows where to get it to compare? I tried Google, webarchive version of Really Rare Wares, and webarchive version of author's website, but all files seem to be missing.

dist10 encoders

Reply #3 –
Also, did the l3enc FhG encoder use any dist10 code, being the first software mp3 encoder?

dist10 encoders

Reply #4 –
I've also seen claims here that WMA has no psymodel, and the Shine mp3 encoder certainly has none as well... how can lossy encoding work without a psymodel? Aren't psychoacoustics essential to MDCT lossy encoding? Do encoders w/o a psymodel just "force" and trunicate the file to fit an mp3/wma/whatever bitstream?

I understand most people are probably more concerned about more "practical" issues of music encoding, but I always found it interesting to know how exactly does it all work. And yes, I did extensive searches here already before, but many questions are still unanswered in my mind.

dist10 encoders

Reply #5 –
Quote from: Neuron on
I've also seen claims here that WMA has no psymodel, and the Shine mp3 encoder certainly has none as well... how can lossy encoding work without a psymodel? Aren't psychoacoustics essential to MDCT lossy encoding? Do encoders w/o a psymodel just "force" and trunicate the file to fit an mp3/wma/whatever bitstream?

I understand most people are probably more concerned about more "practical" issues of music encoding, but I always found it interesting to know how exactly does it all work. And yes, I did extensive searches here already before, but many questions are still unanswered in my mind.


Here's how all these transform codecs work, essentially.  This glosses over a lot of details but gets the point across.
With a time-domain signal (raw PCM data), if you reduce the bits of precision, you effect the entire frequency spectrum.  However, if the time-domain signal is converted into a frequency-domain signal, you can reduce the bits of precision of some areas (frequency bins) of the signal without effecting others.  The psychoacoustic model decides which frequencies should be reduced in precision and which should not.  However, you could have a very simple psymodel which always leaves the loudest frequency bins at full precision and reduces all others.  When we say "without a psymodel" it really means "with a very simple psymodel".

dist10 encoders

Reply #6 –
Thanks. Anyone knows the details of wma by the way? I personally think it has to have a psymodel (mp3 without psymodel sounds like shit hitting the fan, while wma is not my preferred codec but is nowhere near Shine terrible [although even Blade beats it in one problem sample I've encountered]), but I'm not sure.

dist10 encoders

Reply #7 –
Quote from: Neuron on
Thanks. Anyone knows the details of wma by the way?


I typed up a description of how it works for the wikipedia article some years ago.

Quote from: Neuron on
I personally think it has to have a psymodel (mp3 without psymodel sounds like shit hitting the fan, while wma is not my preferred codec but is nowhere near Shine terrible [although even Blade beats it in one problem sample I've encountered]), but I'm not sure.


It has a psymodel.  Don't believe random speculation you read on the internet. 
 

dist10 encoders

Reply #8 –
Thanks for the answers,

Also, Plugger+ supposedly has no psymodel, but why it sounds very different to Shine? I also see both use no short blocks, but Shine's sound seems a lot worse (not that Plugger+ doesn't sound shit), why?