I'm totally with you on Les Paul - it's hard to imagine what modern music would have been like without him.

Not so convinced by the audio argument, however. The Nyquist Theorem just says that, to represent an audio signal digitally, you need to use double your highest audio frequency as the sample rate. So if you have amazing ears (I assume that you are an 8-year-old boy, right?) that can still hear to 20kHz, you need a 40kHz sample rate. So the standard CD rate of 44.1kHz should cover it, assuming an imperfect (non-brick-wall) filter at the top end. This is backed up by heaps of empirical testing; I've never seen well-executed research that could prove the merits of higher sampling rates.

Besides, most popular mics don't output past 20kHz, so you're wasting space recording data that isn't even there. At best, you're recording noise artifacts.

One redeeming feature about some (but not most) A-D converters that use higher sampling rates is reduced jitter, and there is some evidence that some listeners can hear the difference. But this just speaks to the merits of a better clock, as the reduced jitter is usually still there at lower sampling rates (when using the same equipment).

24 bit certainly makes sense, as you can back off the preamps, record more cleanly and with a lot more headroom, and have room to adjust things later in the mixing/mastering stage (where you'll almost always go back to 16 bit anyway, hopefully with some good dithering).

...Sorry for the Professor Audio lecture. I'll leave it there, and you can agree or not. Time to rock the evening away in a tribute to Les Paul. Where to begin?

Oh no! You guys started talking about the Nyquist theorem!! Now I'll have to say something about it!!!

The whole thing has a strong theoretical justification. If your signal is "brick wall" filtered, as you said, you can sample and then use this perfect filter to recreate the same exact analog signal, with no loss at all.

The problem is that filters, as you said again, are not always perfect. So it's always good to have some extra band. If you sample something that has too much power above the Nyquist frequency, you get aliasing. That means you introduce a noise in the signal, you add to the perfect signal some other fake signal. You can actually hear this aliasing noise sometimes. So it's not that the experts have good ears, and listen above the nyquist frequency, they are just listening to this audible noise generated by the aliasing because of the original inaudible signal.

And then there is the processing issue. As you said (one more time), when you start processing stuff these extra bits of precision may help preserving the quality of the signal you are producing. So, even if you are producing 16 bit audio, it's good to process it in 24. But I'm not sure actually recording stuff in more than 16 bits is possible, because it's very hard to get not only a decent microphone, but to build a room that quiet.

But the thing is: just as it is good to make your processing in 24 bits, it is also good to oversample it, because any filter which is not strictly linear will add some high frequencies to your signal, that would become aliasing noise. For strictly linear filters, trough, it makes very little sense. It might help a bit with rounding noise, but again, it might not be an issue.

Myself, I have lost almost a good octave from my hearing, I can't go much over 14kHz. Also, I usually listen to music in the streets anyway, so I've been recoding all my music collection with 16000 samples per second (8kHz limit). This is the double of the usual band of a telephone call... Goes with my crappy headphones, and save a lot of space in my SD cards.

I wish I knew his own music better. But if you want one of his customers, I would definitely start with the Zep...

There is a great Brazilian band called Os Mutantes whose lead singer Arnaldo Batista considers that only Gibson guitars can be used to make good music. He's passionate about it! (Actually, he's not quite sane ). His brother, Serginho, plays Fender, and together they embody the great guitar schism of the 20th century...

I can point you guys to my masters dissertation on electric guitars and distortions, it has some great graphics!

Even at age 48 I can hear WAY up and out there. That includes those cell phone rings designed for teens and the frequencies designed to scare them off. It's a mixed blessing/curse. Bats are jealous.

But I just like having overhead upstream. Still, I rarely sample over 48 anyway. And I have yet to hear an MP3 that matches what can be done with 24 bit WAV. They always sound brittle to me.

Oh, and I installed a new soundcard tonight so I feel much closer to true electronic musicianship...lol. It's an Audiotrak Prodigy 7.1 HiFi and MAN does it blow away the old Creative X-Fi gamer card I had... and a bargain at $130.

Hey TenSpeed-- did you hear the song I just put up in my "Hillbilly" thread? The new vocalist working with me is awesome.

Most of my work involves rendering midi direct to WAV. No external room required.

Is it wrong of me to wonder if my autographed Les Paul t-shirt will be worth anything?

Tim

In your situation? Heck no. It could put food on the table man.

Recently I did some research to reconstruct the intervals in the scales of the Byzantine music (the old Byzantine music, not the modern Greek ecclesiastical music, although it turned out the scales are pretty much close). In order to satisfy my curiosity, when I completed my theory, I decided to play something using zynaddsubfx. And despite my poor musical abilities many friends asked me to give them the mp3s. This music is so much different from anything we hear today!

I played some modern ecclesiastical interpretation of the Byzantine music but in order to imagine the sounding of the real Byzantine music, you should know that the Byzantines considered the organ their national instrument. Two organs played simultaneously in the court of the Emperor. These organs however did not have a keyboard, so the playing style must have been similar to what is demonstrated in the the following clip on youtube: Persian piano solo.

The files are here:

http://lml.bas.bg/~anton/Byzantine/pachymeres-1.mp3
http://lml.bas.bg/~anton/Byzantine/pachymeres-2.mp3
http://lml.bas.bg/~anton/Byzantine/pachymeres-3.mp3
http://lml.bas.bg/~anton/Byzantine/pachymeres-5.mp3
http://lml.bas.bg/~anton/Byzantine/pachymeres-8.mp3

The file names are because the calculation of the intervals in the scales is based on information provided by the Byzantine scholar George Pachymeres (1242 – c.1310).

Damn, I wish we could give Thanks in this forum.

Thanks, nwerneck! I'll have to look up your thesis - maybe I can make the students in my Research Methods course find it (I teach, part-time, at a faculty of music). I completely agree with you, especially when it comes to quiet recording spaces. I've actually been adding some carefully selected noise to recordings lately, to mask little problems in the audio chain. The best advice I've heard is "do an amazing performence", in which case no one will notice the recording quality!

Thanks, Texrat! I haven't listened to the new song, but will do so today. I certainly agree with you on the sound of MP3s - I bought new studio monitor speakers last year (Genelec), and now most of my music collection sounds like crap... unless I play the original, non-MP3 files.

And Thanks, Fontus! I love singing early music, especially trying to stay true to the original tuning. If you're interested in alternative scales and tuning systems, check out Scala <http://www.huygens-fokker.org/scala/>. Powerful stuff!

I love that there are so many skilled, capable musicians involved in Maemo. It's a (barely) hidden strength of the project.

I'll have to give the Byzantine music a listen in a few minutes, when this MASSIVE thunderstorm goes past. Same for the new hillbilly music - can't miss that! (hmm, that's an interesting mix...)

I've been working on some multitracked vocal music lately (just like Qole, I'm a choir guy), but it's far from done. Too busy with my other recording project, creating audio lectures for work (in Cultural Enterprise, or the business of culture). Anyway, I'll post the vocal stuff this autumn, if/when I find time to get it done.

Actually the last song I posted isn't remotely hillbilly...lol. But the overall porject is country & western oriented.

I'm working with the new singer on a style I'm calling "Western Soul"...

Oh, and I am always looking for more vocalists, especially for Eagles-style harmony work. If anyone is interested by all means let me know!

That is seriously weird to my western ears. I'm familiar with middle eastern intervals, and those ones sound all "off-key" to my ears -- mostly a little flat, some a little sharp...

But I see what you mean about sounding similar to modern Eastern Orthodox music...

I notice that there's a drone constantly playing. Was this organ that they used something like the bagpipes?