Mark Levinson and Premium Audio - Notes from an Audiophile
03-18-14, 04:50 PM
#76
You have a valid argument, especially if you factor in the equipment and environment. In-car listening is probably just as good with 320 kbit/s MP3s. BUT - do not think that there is no difference between MPS / FLAC / CD / DVD-A / etc. It's important to understand what the differences are and how the effect the sound.
Using high-end hi-fi equipment it is easy to hear the difference between MP3 and FLAC. I've been known to quickly run back to my computer to skip an MP3 if it is randomly selected.
Here's some information to digest:
MP3
8-bit / 44.1 kHz
320 kbit/s
CD quality is
16-bit / 44.1 khz
1,411.2 kbit/s
DVD-A
24-bit / 192 khz
4,608.0 kbit/s
The biggest difference between "high-quality" MP3s and FLAC / CD is the bit depth. This directly effects two things. 1. the noise, there will be a higher chance and higher level of background noise. 2. the dynamic range - this quality is greatly overlooked, but when a drum kicks or cymbal crashes I want it to be LOUD! It also 'blends' the range between instruments (there are only so many 'levels' to specify). Unfortunately most recording do not utilize the available range or bit depth and some are 'over' engineered to comply with general use requirements.
All of that being said. I exclusively use my Blackberry to steam FLAC files to my GS, but only 'cause I already have them loaded as such. I would not hold it against anyone to use high-quality MP3s in their car (my home system is another matter).
Using high-end hi-fi equipment it is easy to hear the difference between MP3 and FLAC. I've been known to quickly run back to my computer to skip an MP3 if it is randomly selected.
Here's some information to digest:
MP3
8-bit / 44.1 kHz
320 kbit/s
CD quality is
16-bit / 44.1 khz
1,411.2 kbit/s
DVD-A
24-bit / 192 khz
4,608.0 kbit/s
The biggest difference between "high-quality" MP3s and FLAC / CD is the bit depth. This directly effects two things. 1. the noise, there will be a higher chance and higher level of background noise. 2. the dynamic range - this quality is greatly overlooked, but when a drum kicks or cymbal crashes I want it to be LOUD! It also 'blends' the range between instruments (there are only so many 'levels' to specify). Unfortunately most recording do not utilize the available range or bit depth and some are 'over' engineered to comply with general use requirements.
All of that being said. I exclusively use my Blackberry to steam FLAC files to my GS, but only 'cause I already have them loaded as such. I would not hold it against anyone to use high-quality MP3s in their car (my home system is another matter).
I think you said it best yourself. Unfortunately, most recordings do not utilize the range--almost all music is recorded at 16-bit/44.1khz. it is upsampled to 192khz for DVD audio, which really makes no difference considering it is the same audio source and 99% of the population can't hear the difference.
If you can pass a blind test of a 320kpbs MP3 and FLAC file then you have some super human hearing
03-18-14, 07:54 PM
#77
Uh, not quite. I'd like to meet ANYONE over 30 years old that can hear ANYTHING at 19khz (20khz-5%). NOT bloody likely, my friend. The "mosquito" used to chase away loitering teenagers from 7-11s is an annoying 17khz drone heard ONLY by 13-25 year-olds. To hit everyone, the drone must be lowered to a frequency of 8khz. You need to look at a few more Fletcher-Munsons before making generalizations about human hearing capabilities.
03-18-14, 08:19 PM
#78
Lexus Test Driver
Uh, not quite. I'd like to meet ANYONE over 30 years old that can hear ANYTHING at 19khz (20khz-5%). NOT bloody likely, my friend. The "mosquito" used to chase away loitering teenagers from 7-11s is an annoying 17khz drone heard ONLY by 13-25 year-olds. To hit everyone, the drone must be lowered to a frequency of 8khz. You need to look at a few more Fletcher-Munsons before making generalizations about human hearing capabilities.
The following users liked this post:
bone215 (12-09-20)
03-19-14, 04:35 PM
#80
Rookie
Join Date: Mar 2013
Location: California
Posts: 27
Likes: 0
Received 0 Likes
on
0 Posts
Just because you can't hear above or below a particular frequency doesn't mean you can't hear the difference between lossy and lossless. With lossy, there is information missing, not just high and low frequencies. I'm in my 50s and I have done some "blind listening" and I could hear the difference. I did test as well as a 20 something YO on my last hearing test, which is amazing since I have blasted tunes for over 40 years, but I think many over 40 people could hear the difference between lossy and lossless. It takes a really high end audio system (which I have), but you probably couldn't hear a difference in most car audio systems. I use lossless in my car system, but I doubt I could hear a difference. I do it because it's easy to do, I've got the storage and it 's possible that it might sound better. In other words, it doesn't hurt.
03-19-14, 05:37 PM
#81
Uh, not quite. I'd like to meet ANYONE over 30 years old that can hear ANYTHING at 19khz (20khz-5%). NOT bloody likely, my friend. The "mosquito" used to chase away loitering teenagers from 7-11s is an annoying 17khz drone heard ONLY by 13-25 year-olds. To hit everyone, the drone must be lowered to a frequency of 8khz. You need to look at a few more Fletcher-Munsons before making generalizations about human hearing capabilities.
03-19-14, 06:32 PM
#82
Lexus Test Driver
Here's some files for you fellow audio nerds to compare. I suggest comparing them via whatever device in your car vs. headphones (not iPhone headphones) vs. a good home stereo setup. Linn is pretty well known for being serious about quality.
http://www.linnrecords.com/linn-down...testfiles.aspx
Ultimately, most modern music doesn't take advantage of the ML system. You really have to go for some obsessive-compulsively recorded operatic and symphonic pieces to really hear it sizzle, so I think Linn's test files are a good start.
Back to the wishlist, I think if you're going to provide a system like this in a car, it should be able to decode damn near anything including FLAC, ALAC, etc. Like an earlier poster stated, Blu-ray support for surround audio would be nice since DVD audio is dead, but I'd much rather have a good surround format on an ipod.
Maybe I will stop being lazy and convert one of my 5.1 DVD audio discs to WMA surround and see if the ML handles it properly.
http://www.linnrecords.com/linn-down...testfiles.aspx
Ultimately, most modern music doesn't take advantage of the ML system. You really have to go for some obsessive-compulsively recorded operatic and symphonic pieces to really hear it sizzle, so I think Linn's test files are a good start.
Back to the wishlist, I think if you're going to provide a system like this in a car, it should be able to decode damn near anything including FLAC, ALAC, etc. Like an earlier poster stated, Blu-ray support for surround audio would be nice since DVD audio is dead, but I'd much rather have a good surround format on an ipod.
Maybe I will stop being lazy and convert one of my 5.1 DVD audio discs to WMA surround and see if the ML handles it properly.
03-19-14, 08:42 PM
#83
Agree on decoding everything. ASIC decoders are cheap and available. No excuse not to use them.
I hope you will do the WMA surround conversion and playback. I, for one, would be very interested in your impressions. To this day, I pine for my old multichannel SACD setup - what I wouldn't give to hear Herbie Hancock's Chameleon again in all its glory. Shame hirez MC bit the dust.
03-20-14, 04:59 AM
#84
Pole Position
I think I'm starting to understand the confusion. The 192khz specification is not the bandwidth, it's not the highest frequency. It is the sampling frequency, the number of times per second the track it sampled, really you would want it to be infinite so the track is constantly sampled, but that's not possible.
03-24-14, 07:10 AM
#85
I think I'm starting to understand the confusion. The 192khz specification is not the bandwidth, it's not the highest frequency. It is the sampling frequency, the number of times per second the track it sampled, really you would want it to be infinite so the track is constantly sampled, but that's not possible.
03-24-14, 10:32 AM
#86
Pole Position
Yes, as a simple rule, the max frequency is half the sample rate. The point of a higher sample rate is not to increase the max frequency, that's really more of a byproduct. The point is to sample a more accurate waveform (specifically for non-sinusoidal waveforms).
03-24-14, 04:48 PM
#87
Lexus Test Driver
Thanks for file link.
Agree on decoding everything. ASIC decoders are cheap and available. No excuse not to use them.
I hope you will do the WMA surround conversion and playback. I, for one, would be very interested in your impressions. To this day, I pine for my old multichannel SACD setup - what I wouldn't give to hear Herbie Hancock's Chameleon again in all its glory. Shame hirez MC bit the dust.
Agree on decoding everything. ASIC decoders are cheap and available. No excuse not to use them.
I hope you will do the WMA surround conversion and playback. I, for one, would be very interested in your impressions. To this day, I pine for my old multichannel SACD setup - what I wouldn't give to hear Herbie Hancock's Chameleon again in all its glory. Shame hirez MC bit the dust.
The head unit showed the files as WMA but scanned them quickly and skipped through each one without playing. I think it ended up deciding none of the files would play and said something to the effect of no supported files found on the screen.
Back to the drawing board, but I'm not giving up. I emailed ML who told me to talk to their automotive division, who then told me to talk to Lexus. It's ridiculous that ML won't just email me tech specs of what this head unit can handle so I get to play mad scientist with formats until I get it right.
03-24-14, 04:49 PM
#88
http://xiph.org/video/vid2.shtml
03-24-14, 06:26 PM
#89
For fun, I took some m4a files I ripped from DVD-A to ALAC (apple's lossless codec) and used dbpoweramp to convert them to 5.1 WMA 10 Pro files. Copied them to a USB thumb drive, put it in the car...
The head unit showed the files as WMA but scanned them quickly and skipped through each one without playing. I think it ended up deciding none of the files would play and said something to the effect of no supported files found on the screen.
Back to the drawing board, but I'm not giving up. I emailed ML who told me to talk to their automotive division, who then told me to talk to Lexus. It's ridiculous that ML won't just email me tech specs of what this head unit can handle so I get to play mad scientist with formats until I get it right.
The head unit showed the files as WMA but scanned them quickly and skipped through each one without playing. I think it ended up deciding none of the files would play and said something to the effect of no supported files found on the screen.
Back to the drawing board, but I'm not giving up. I emailed ML who told me to talk to their automotive division, who then told me to talk to Lexus. It's ridiculous that ML won't just email me tech specs of what this head unit can handle so I get to play mad scientist with formats until I get it right.
03-25-14, 01:57 PM
#90
Pole Position
I am by no means an expert in any of this, but I found this video pretty educational. I think it shows proof against what you're arguing, but I could be wrong.
http://xiph.org/video/vid2.shtml
http://xiph.org/video/vid2.shtml
Now that you have watched it as well, look up am image of a typical audio waveform and imagine the points placed on both axis, you can see the advantage to more points, especially bit depth. It allows a more accurate sample of the waveform, it 'forces' less adherence to one of the points.
It's much easier to reproduce the waveform when it's repetitive, but what about when there is less then half a wave or the wave changes during the cycle?
http://goodrob13.com/wp-content/uplo...tereo-wave.gif
http://www.mathworks.com/help/daq/c1_introdaq13.gif
http://zone.ni.com/images/reference/...c_fnyquist.gif
http://www.school-for-champions.com/...plex_sound.jpg
http://www.computeraudiophile.com/at...tracks-two.jpg