Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Can DJs get away with 8-bit audio? (Read 12643 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Can DJs get away with 8-bit audio?

Reply #25 –
Bahahahaha! This topic makes me laugh, speaking as someone who has played a great many tracks made with 8-bit audio while out DJing.

Can DJs get away with 8-bit audio? Yes. I do.  I've also played 100+ people parties using decoded 100kbps Opus files...

And now back to your regularly scheduled Serious Business Technical Talk...

Can DJs get away with 8-bit audio?

Reply #26 –
@Canar:

Define "played" and define "people". Also, year would be interesting to know. (loudness war plays in favour of less bits...)

To put some more things into the mix. In the (early to mid) 90's electronic music was done with computer trackers (ok, not as a rule, but at that time, it was mostly sample-based, with either software or hardware sequencers), and samples used to be 8bits. (One could also see 12-in-16bits)
But also, soundcards used to have 60 to 70db of actual SNR.

I am not objecting to the fact that 8bit can be enough in some contexts, but I also know the added distortion (more so if that 8 bit file was postprocessed later), and I am trying to make a clear distinction between "perception" versus "caring about".

"Caring about" is subjective. Perception is not (although it might require attention)



The addition of using Opus at 100kbps is not interesting in the context of this discussion. MP3 was already mentioned. (you can see posts 2, 3, and 9).

Can DJs get away with 8-bit audio?

Reply #27 –
Tracker music, Amiga style. 8-bit chiptune (yeah, I know, you're talking PCM), whether NES or whatever. Years '88 to '13. Live DJ performance, a la Tiesto, Skrillex, etc. People being a crowd of friends gathered to dance.

I was just being a cheeky jerk more than anything, given the topic title. Relax.

Can DJs get away with 8-bit audio?

Reply #28 –
Amiga style... hehe.. That says it all.

(not that it should be bad in any way, but its characteristics, with the arpeggios, cut-samples, note retrigs and whatnot would made the 8bit side of the equation less worrysome).

Can DJs get away with 8-bit audio?

Reply #29 –
Quote from: pdq on
So if you take 20 kHz of white noise and compress it into 2 kHz it gets ten times bigger?


Something like that. Take 20 bricks, lay them out side by side. There's the noise energy, spread out across a 20 KHz bandwidth. Now pile the bricks into a stack 2 wide, 10 high. There's the same total noise energy in a 2 KHz wide bandwidth.
Regards,
   Don Hills
"People hear what they see." - Doris Day

Can DJs get away with 8-bit audio?

Reply #30 –
But even if you filter the dither with a high pass filter that suppresses frequencies below a certain cutoff point, you cannot increase the signal to noise ratio below that cutoff indefinitely.

So in the bricks analogy you start with two or more rows of bricks (depending on the number of bits used for the dither) and can never take away bricks from the lower row.
"I hear it when I see it."

Can DJs get away with 8-bit audio?

Reply #31 –
Quote from: xnor on
But even if you filter the dither with a high pass filter that suppresses frequencies below a certain cutoff point, you cannot increase the signal to noise ratio below that cutoff indefinitely.

So in the bricks analogy you start with two or more rows of bricks (depending on the number of bits used for the dither) and can never take away bricks from the lower row.

No, it doesn't work like that. You don't filter the dither. You don't even need dither for noise shaping. You put a filter in a negative feedback loop around the quantiser, which shapes the frequency of the quantisation error (and dither, if you use dither).

You can't reduce the overall power noise, but you can push it around spectrally. The narrower the spectral band you push it into, the larger the peak values of the resulting waveform. Within the constraints of digital full scale, stable filters, friable tweeters, and maintaining the overall spectral noise power, you can make the noise at any given frequency as low as you like.

See here:
http://www.hydrogenaudio.org/forums/index....st&p=580394

Cheers,
David.

Can DJs get away with 8-bit audio?

Reply #32 –
Thanks for that link, David. It really clarifies things for me.

Can DJs get away with 8-bit audio?

Reply #33 –
@David: Make that quantization noise or error instead of "dither". Still, I don't think you can make the noise as low as you like. Maybe you have an example that pushes the noise with 8 bit quantization really low at the frequency (range) of your choice?

Btw: there are algorithms that shape dither...
"I hear it when I see it."
 

Can DJs get away with 8-bit audio?

Reply #34 –
He did state the ifs and buts, which presumably result in the sort of noise shaping related weighted SNR gains we are seeing in practice.

8 bit is handy for demonstrating the effects that dithering etc. have, without having to turn the volume way up (at 16 bits, the same things are happening ~48 dB lower). It doesn't take a golden ear to hear how lousy it sounds undithered, how things much improve when fully dithered and how noise is reduced significantly when making use of shaped dither.

It has to be noted that shaped dither only comes out first as long as you have a high enough sample rate to hide the noise somewhere. Try the same at 8 kHz, and there are no more convenient frequency regions of reduced hearing sensitivity - unshaped dither tends to work best then.