Support and General Use > Audio Playback, Database and Playlists
Rockbox Audio Quality versus Original Firmware (RMAA comparison)
omendelovitz:
Saratoga,
While I appreciate your enthusiasm, Platitudes don't help. I'm hoping you can be a little more specific with your statements - e.g. how does stereo crosstalk effect perceived spaciousness of the soundstage? How does it affect imaging? exacly what is 16 bit? How would it affect instrument separation?
oh, and let me correct myself - I've never found an audio hobbyist whose been able to measure the above types of audio cues; I've also never had a chance to speak/interact with someone who works with these issues professionally to teach me such nuances in audio reproductions - if you have, please fill me in or at least refer me to places where I can learn all these types of goodies for myself! Yumm
Febs:
--- Quote from: omendelovitz on August 27, 2006, 03:52:28 PM ---Here you go:
...
Dixie Chicks - Landslide
--- End quote ---
Omendelovitz, since you hear a difference between the Apple FW and Rockbox, I would be greatly appreciative if you could do a test for me. I have posted two clips of Landslide (one of your reference tracks) for you to compare:
Wave files:
Clip 1
Clip 2
WavPack Files
Clip 1
Clip 2
It doesn't matter which set you choose; they are identical except that obviously the WavPack files are smaller. To prepare the clips, I ripped Landslide from CD using EAC. The .wave file was placed on my iPod, and then played back using both the Apple firmware and Rockbox. In each case, the clips were recorded to .wav file, using the iPod's line out and a pocket dock. In each firmware, the iPod's volume was set on maximum (i.e., +6dB in Rockbox; a "full blue line" in the Apple FW). No EQ or other processing was enabled. I did no post-processing other than to (1) peak normalize both files so that there would be no volume differences to skew the results, and (2) trim the starting and ending points of each clip so that as near as humanly possible, they were exactly the same length.
Please compare the two clips using an ABX program such as the one built into Foobar2000. You should do 12 trials. Turn off the option allowing you to view the results of each trial as the test is in progress. When you have finished 12 trials, exit the ABX program (being sure to save the results) and post the results here.
Thanks! Please let me know if you have any questions.
saratoga:
--- Quote from: omendelovitz on August 27, 2006, 04:00:45 PM ---Saratoga,
While I appreciate your enthusiasm, Platitudes don't help.
--- End quote ---
Platitudes are statements such as "life is great" that don't actually mean anything. I posted definitions of those terms, as commonly used.
--- Quote from: omendelovitz on August 27, 2006, 04:00:45 PM ---I'm hoping you can be a little more specific with your statements - e.g. how does stereo crosstalk effect perceived spaciousness of the soundstage? How does it affect imaging?
--- End quote ---
Stereo cross talk collapses the sound stage. At 0 dB you have mono. At negative infinity you have perfect channel speration. In this case though, since the noise floor is roughly the same as the cross talk, its a safe assumption that the channels are not audiably coupled.
Regarding imaging verses sound stage, they really are the same thing in this context. Heres the first random site I found googling:
http://www.electronixwarehouse.com/education/glossary/S.htm
"
Soundstage
...
(See also Imaging above)
"
So it seems its not just me who uses them interchangably.
--- Quote from: omendelovitz on August 27, 2006, 04:00:45 PM --- exacly what is 16 bit? How would it affect instrument separation?
--- End quote ---
The resolution of the PCM audio is 16 bits in Rockbox and Apple firmwares. It doesn't effect imaging, it determines how well the system can resolve amplitude changes in reconstructed waveforms. This in turn determines the dynamic range of the system. 16 bit systems typically have a maxium of 96 dB SNR.
--- Quote from: omendelovitz on August 27, 2006, 04:00:45 PM ---oh, and let me correct myself - I've never found an audio hobbyist whose been able to measure the above types of audio cues; I've also never had a chance to speak/interact with someone who works with these issues professionally to teach me such nuances in audio reproductions - if you have, please fill me in or at least refer me to places where I can learn all these types of goodies for myself! Yumm
--- End quote ---
A lot of audio terminology is simply made up by hobbyists as the go along. Its very common on the internet for people with no real understanding of audio to hide behind the equipment they use and the nebulous terms they throw around (I'm not singleing you out). Its best to just avoid fluffy language like that entirely and say what you mean. At least thats my preference.
But yeah, take the ABX test and post your results. Either way it'll be interesting.
saratoga:
Hey Febs, I think you mean
http://jazzexplosion.com/Febs/Rockbox/Landslide1.wv
http://jazzexplosion.com/Febs/Rockbox/Landslide2.wv
are the wavepack files.
Anyway, thats a difficult test. I tried but could not ABX it. Though it could be that I'm not sure what to look for, which doesn't help.
Febs:
Thanks, I fixed the links.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version