mp3 artifacts used as effets ? ( two examples inside )
- KVRAF
- 6504 posts since 25 May, 2002 from Bobo-dioulasso\BF__Geneva/CH
Out of curiosity
I'm amazed by one typical artifact resulting of mp3 encoding,,,,
( and as much i can judge in other analysis/resynthesis methods )
...that seems to smooth the attack of percussions, much in the highest frequencies through their harmonic content, without adding any coloration
Especially on cymbals, this is very spectacular
Seems to be an interesting effect in itself and i'm wondering if their are in the actual huge plug-in arsenal many effects able to recreate it ?
I'm amazed by one typical artifact resulting of mp3 encoding,,,,
( and as much i can judge in other analysis/resynthesis methods )
...that seems to smooth the attack of percussions, much in the highest frequencies through their harmonic content, without adding any coloration
Especially on cymbals, this is very spectacular
Seems to be an interesting effect in itself and i'm wondering if their are in the actual huge plug-in arsenal many effects able to recreate it ?
Last edited by Krakatau on Sun Sep 24, 2006 9:33 am, edited 2 times in total.
-
- KVRist
- 204 posts since 10 Mar, 2005
not directly that i know of but you could try and encode at a shit rate and then use voxengos' MSED plug to bring out the crapness.
-
- KVRAF
- 3964 posts since 31 Aug, 2003 from In a foreign town, in a foreign land
You could try and use KTgranulator to get similar effects. Experiment with settings and grain count.
You can get the warbly effect of poorly encoded mp3s by setting the transpose to two values.
Groet, Erik
You can get the warbly effect of poorly encoded mp3s by setting the transpose to two values.
Groet, Erik
Pop music delenda est.


- KVRAF
- 3944 posts since 7 May, 2004 from behind his workbench
- KVRAF
- Topic Starter
- 6504 posts since 25 May, 2002 from Bobo-dioulasso\BF__Geneva/CH
thanks for the input folks, i think in particular that this artifact that smoothes the impacts very interesting and would deserves to be turned to an effect as itself, especially considering it transparency in regard to the timbral aspect that strictly remains the same
Did you notice this ?
Did you notice this ?
-
- KVRAF
- 3404 posts since 15 Sep, 2002
How about encoding something with mp3 and then decoding it, then subtracting the processed sound from the original?
- KVRAF
- 3944 posts since 7 May, 2004 from behind his workbench
-
- KVRAF
- 3511 posts since 27 Dec, 2002 from North East England
Most encoders tend to low pass the signal before encoding when using low bit rates. Typically 16kHz for 128k. I'd hazard a guess that what you're hearing is related to this, since most (all?) analysis/resynthesis procedures cope with the high end much better than the low. Cymbals do indeed have prominent partials in this area, so perhaps what you're hearing is the low pass taking the 'edge' off the sound.
-
- KVRAF
- 4692 posts since 28 Jan, 2003 from In these very interwebs
- KVRAF
- Topic Starter
- 6504 posts since 25 May, 2002 from Bobo-dioulasso\BF__Geneva/CH
yeah right, that's what i suspect about spectral effects,
at another dregree,but more extreme Spectral monkeyage seems to have a somehow similar rendering
Some time stretch algorithms too :
on Amazing slow downer , one of the algo has a similar artifact ( even without stretching an audiofile )
at another dregree,but more extreme Spectral monkeyage seems to have a somehow similar rendering
Some time stretch algorithms too :
on Amazing slow downer , one of the algo has a similar artifact ( even without stretching an audiofile )
- KVRAF
- Topic Starter
- 6504 posts since 25 May, 2002 from Bobo-dioulasso\BF__Geneva/CH
By listening carefully, i would descriped it as the instantaneous, impact of a cymbal especially ,is turned to a kind whipped sound from it lowest to highest frequencies
I can figure out that a dynamic low-pass filter would do something approaching on a single cymbal shot , or a cymbal at a certain degree, but not in a mix...
I' ll try to create some significant examples later if i can find the time to do a serious job
I can figure out that a dynamic low-pass filter would do something approaching on a single cymbal shot , or a cymbal at a certain degree, but not in a mix...
I' ll try to create some significant examples later if i can find the time to do a serious job
- KVRAF
- Topic Starter
- 6504 posts since 25 May, 2002 from Bobo-dioulasso\BF__Geneva/CH
ok
here i go :
2 examples, each with an original 16bits/44.1Hz wave files and its 128 kbps mp3 encoded counterpart
at first :
www.anak-krakatoa.net/~downl/misc/artifacts/1.wav
www.anak-krakatoa.net/~downl/misc/artifacts/1.mp3
In this first example comparing both files, just notice, as other instruments and arrangements remains the same how, how the hi-hat is changing character
The "flutes" are also affected, but this is not such spectacular because of the natural smoothness of their attack
in second :
www.anak-krakatoa.net/~downl/misc/artifacts/2.wav
www.anak-krakatoa.net/~downl/misc/artifacts/2.mp3
Here i just want you to notice on how the sustained synthetic sound on the last measures is behaving after encoding
Something that reminds a heavy chorus or a detuning effect created with a pitch shift ( like mda detune ) and others organ-like sustained harmonies seems also slightly affected
( other percussives sounds are also quite affected, most of them because they where themselves heavily threated in a previous time, ...maybe revealed byt the encoding limitation ? )
I' m just wondering if this artifacts is revealed only inside such an overall mix or if it would sound the same applied on these tracks individually, apart of the whole orchestration ?
here i go :
2 examples, each with an original 16bits/44.1Hz wave files and its 128 kbps mp3 encoded counterpart
at first :
www.anak-krakatoa.net/~downl/misc/artifacts/1.wav
www.anak-krakatoa.net/~downl/misc/artifacts/1.mp3
In this first example comparing both files, just notice, as other instruments and arrangements remains the same how, how the hi-hat is changing character
The "flutes" are also affected, but this is not such spectacular because of the natural smoothness of their attack
in second :
www.anak-krakatoa.net/~downl/misc/artifacts/2.wav
www.anak-krakatoa.net/~downl/misc/artifacts/2.mp3
Here i just want you to notice on how the sustained synthetic sound on the last measures is behaving after encoding
Something that reminds a heavy chorus or a detuning effect created with a pitch shift ( like mda detune ) and others organ-like sustained harmonies seems also slightly affected
( other percussives sounds are also quite affected, most of them because they where themselves heavily threated in a previous time, ...maybe revealed byt the encoding limitation ? )
I' m just wondering if this artifacts is revealed only inside such an overall mix or if it would sound the same applied on these tracks individually, apart of the whole orchestration ?
- KVRAF
- Topic Starter
- 6504 posts since 25 May, 2002 from Bobo-dioulasso\BF__Geneva/CH
fred-hal wrote:connection timed out..
...in other words ?
