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: MP3 Listening Test at 128 kbps (Read 26417 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

MP3 Listening Test at 128 kbps

Reply #25 –
I don't post here much now, but I just noticed this thread... and think I would have some interesting samples to post.
Is this thread still alive?

MP3 Listening Test at 128 kbps

Reply #27 –
OK, here are samples than I think are interesting. Some of them come from lossy sources but I believe this does not matter. Most of the samples can be shortened if they are too long for the test.
-
shokk, airscape - pre-echo killer samples
starburst - the high frequency "chirping" gets smeared
hissing - causes "crumpling" low frequency noises with some encoders (Fhg, Lame)
peaceful_world - warbling of lead and bass sounds, loss of treble
trippy - this is all-killer sample mainly for CBR, I'm not sure about this one


I hope all the attachements are added, if not I can upload them on external server.

-J.M.

[attachment=4612:attachment][attachment=4613:attachment][attachment=4614:attachment][attachment=4615
:attachment][attachment=4616:attachment]
[attachment=4617:attachment]

MP3 Listening Test at 128 kbps

Reply #29 –
Quote from: Sebastian Mares on
What do you mean with "come from lossy sources"? The FLAC files come were transcoded from MP3 / Vorbis / etc.? You should post samples that come from lossless sources, please.

Yes some were mp3 converted to Wav, cut out and converted to flac. If it's a big problem, I can tell you which ones come from lossy sources (you should see this also on spectrograms): "Starburst", "hissing" and "trippy". I especially liked the hissing sample as it caused really noticable problems.

MP3 Listening Test at 128 kbps

Reply #30 –
Quote from: jmartis on
Quote from: Sebastian Mares on

What do you mean with "come from lossy sources"? The FLAC files come were transcoded from MP3 / Vorbis / etc.? You should post samples that come from lossless sources, please.

Yes some were mp3 converted to Wav, cut out and converted to flac. If it's a big problem, I can tell you which ones come from lossy sources (you should see this also on spectrograms): "Starburst", "hissing" and "trippy". I especially liked the hissing sample as it caused really noticable problems.


You should only post samples from lossless sources.  Otherwise it will be impossible to ABX anything unless we happen to have the lossless source that you have.  You see, the problems that we might hear in your hissing sample might not be problems at all unless we reference them to the lossless source.

MP3 Listening Test at 128 kbps

Reply #31 –
Quote from: Martel on
Quote from: /mnt on

...

That artifical infrabass on the Vangelis Main Titles is pretty brutal! 

Most of the track was done on a Yamaha CS-80 synthesizer.

Also I have another sample, which might be usefull.
"I never thought I'd see this much candy in one mission!"
 

MP3 Listening Test at 128 kbps

Reply #33 –
As one can read in my post earlier today in other forum I've ABXed Pet Shop Boys - In the Night from their album Disco (The Remix Album). As it isn't hard to ABX at around 128 kbps and Lame, Ogg and Nero produce different kind of distortions I thought it would be a nice sample for including in test.

[attachment=4636:attachment]

MP3 Listening Test at 128 kbps

Reply #34 –
11/16 = failed with snares on Pet Shop Boys (with lame 3.98 -V 2)
6/6 = easy on -V 4
PANIC: CPU 1: Cache Error (unrecoverable - dcache data) Eframe = 0x90000000208cf3b8
NOTICE - cpu 0 didn't dump TLB, may be hung

MP3 Listening Test at 128 kbps

Reply #35 –
Quote from: smok3 on
11/16 = failed with snares on Pet Shop Boys (with lame 3.98 -V 2)
6/6 = easy on -V 4

Could you try lame 3.98 -V 0? I could ABX it but couldn't at -V 2, I wonder if it's me...

MP3 Listening Test at 128 kbps

Reply #36 –
yeah, something is wrong with -V0, i can abx it easily just based on volume change, compared to original, will try with RG enabled as well... (edit: can't abx with RG applied)

edit2: -v0 comes out as 264 Kbps
edit3: it appears i can abx 0.02 dB level difference, something must be wrong with my setup.
PANIC: CPU 1: Cache Error (unrecoverable - dcache data) Eframe = 0x90000000208cf3b8
NOTICE - cpu 0 didn't dump TLB, may be hung

MP3 Listening Test at 128 kbps

Reply #37 –
There was report about possible regression on V0 and 320 kbit/s.
I tried very hard samples (castanets and eig). 3.98 was horrible at 320 kbit/s comparing it to 3.97 and 3.90.3
I can post ABX results lately.

MP3 Listening Test at 128 kbps

Reply #38 –
Quote from: IgorC on
There was report about possible regression on V0 and 320 kbit/s.
I tried very hard samples (castanets and eig). 3.98 was horrible at 320 kbit/s comparing it to 3.97 and 3.90.3
I can post ABX results lately.

My experience with 3.98 -V0 on eig is that it compares very favorably against 3.97 -V0 --vbr-new.
lame3995o -Q1.7 --lowpass 17