[charbs]

Hello,

I have installed a 2GB sd chip and am in the process of filling it up with music.
The methods I play the music are through the factory headphones (HDW-13019-001) and the JBL On-tour portable music box (Frequency Response 100Hz-20kHz).
I understand there are much better techniques for listening to music! However, if these TWO methods were what you listened through -what formats do you find preferable?
In the past, I have encoded aac files @192kbps -and find them ~ 1Mb smaller than comperable .mp3s. This is really a waste of space concerning the above setup. Sample rates of 22050 / 24000Hz seem to do the trick with as low as 64kbps (5min songs ~ 2.47 Mb).
What kbps do most of you encode with? -again I understand the Pearl is not the ideal music platform -just trying to get an idea of what kbps / sample rates other audio people find optimal.

Thanks for your reply, I look forward to the discussion!

charbs

[ImportSport]

I'm in the same general range as you. I willingly sacrifice space for quality. I won't go below 192kbps. It's difficult to predict what you are sensitive to in terms of audio quality. I would take a familiar song and encode a few different versions while tweaking some settings. Find your prefered balance and just stick with what sounds best to your ears.

[JoelDF]

Originally Posted by charbs

...Sample rates of 22050 / 24000Hz seem to do the trick with as low as 64kbps (5min songs ~ 2.47 Mb).

Sample rate of 22/24kHz???

That gives you a frequency response of up to only 11-12kHz... tops. Talk about cutting off the top end and muffling the sound.

Sorry to say, but the "audiophile" label, IMO, is something you give up using any form of lossy compression on music. The CD format of 16bit/44.1kHz sampling rate PCM is minimum (uncompressed wave gives you that). 24 or 32 bit is better as is 88.2, 96, or 192kHz sampling rates. Of course, analog is optimum .

If you gotta compress, don't change the sampling rate, leave it at least 44.1. Even though most compression techniques trim off the top end progressively lower as use a lower kbps - 192 cuts off the frequency response at about 18kHz during strong signals, 128 cuts off at about 16kHz, 96 cuts down to 14kHz or lower. Other sounds are "removed" as well, but you're not supposed to notice because you're not supposed to care (I do). If I have to compress for whatever reason, I don't use less than 192kbps.

[stealthx32]

Honestly, I doubt the Pearl is fitted with any decent decoder.

[charbs]

Thanks Joel,

good points for me to consider. I see 96kbs is the minimum bit-rate for an .mp3 @ 44KHz. Any advice on a process or song that illustrates the differences best?

thanks again,
sc

[TheBigNewt]

Some time ago I encoded that same song 128mp3 (both CBR and VBR), 128 wma (supposedly better like AAC claims), and both formats/ways at 160kbps. Loaded them and passed the player around to people doing the blindfold test. Nobody could tell the difference. Personally I now use mp3 at 192kbps CBR. I think it sounds better than 128, but I haven't done the blindfold test again. I know I can tell that 96kbps is inferior.

[charbs]

I find encoding with the .aac saves about 1 Mb / minute @ 192kbps -but don't really hear or see much of a difference below 92kbps. It seems 192 is a bit overkill for the Pearl and the factory headphones.
I'm going to try a test with three styles of music (classical, rock, and some ambient-pad styles) to test dynamic ranges tonight. I plan on encoding the files from .cda to pure .wav at a few sample rates -still want to see if I can hear a difference between the 22KHz & 24KHz sample rates. I know mathematically there is a difference. For some rock songs -I couldn't tell the difference between 92kbps and 192kbps through the supplied headphones!

Do you know if there is a way to calculate optimal quality if you know a spakers output?
i.e. The JBLs have the following:
Nominal Power Out: 6W (I understand nominal in the economic sense -but not really here -unless we have to adjust for inflation I'll google this.
Response Bandwidth: 100-20000Hz
Input Impedence: 10Kohm
Signal to Noise Ratio: 80dB
Audio Amplifier: Integrated
Connectivity Technology: Wired
Voltage Input: 100-240 Vac, 50-60Hz
Unweighted Pink Noise: 95dB

and a review that said the Output dropped off below 90Hz -I'll have to Google this as well -not really clear on this right now.

[TheBigNewt]

If you're using the supplied headphones use anything you want and you can't tell the difference. OEM phones/buds stink. Anything that plugs straight into the 8100 stinks. Use an adapter and decent cans. If you use headphones that cost $60-100 you can tell 96kbps stinks. 128-192 depends on your ears.

[stealthx32]

Regardless, I doubt the headphone amplifier is worth a damn.

Actually. I did an unscientific test. I plugged both the computer and the Pearl w/ the Verizon headphone adapter into my Onkyo home theater. Comparing to my laptop IBM T43 w/ an Analog Devices AD1981B chipset (which touts a rather plebian 90 dB S/N ratio...iPods are rated ~98 dB, my Onkyo receiver is rated at 100 dB), and the laptop sounds leagues better than the Pearl.

Both are reading the same 320 kbps MP3 file. Maybe the iTunes software decoder is a lot better than the Pearl one? Who knows. The end result is it's really not worth your money to spend a lot of good drivers when your source is noisy.