I'm not sure if I'm putting this question in the right section, but here it is:
So 8-bit color would be 2*2*2*2*2*2*2*2 or 2^8, so 256, which means there are 256 colors in 8-bit color. But what exactly is 8-bit music? I know what the sound of 8-bit music is, but what defines "8-bit" music?

THANKS GUYS!!!

Christian

 

I'm not sure if I'm putting this question in the right section, but here it is:
So 8-bit color would be 2*2*2*2*2*2*2*2 or 2^8, so 256, which means there are 256 colors in 8-bit color. But what exactly is 8-bit music? I know what the sound of 8-bit music is, but what defines "8-bit" music?

THANKS GUYS!!!

Christian

Regarding the video, it is common to have True Color, which is 24 bits for 256 shades of red, 256 shades of blue, and 256 shades of green, i.e (256x256x256)=16,777,216 color variations. My video is 32 bit, which is 24 bit color and 8 bits for an alpha channel (opacity).

The term "8 bit music" often refers to music that is generated by 8 bit computers, such as the venerable Commodore 64 or games like the Nintendo. The music usually consists of sine waves, triangle waves, sawtooth waves, and maybe white noise for percussion.

 

I had never heard of this, but then it is something that I recall as being mostly "trendy" at the time and those kinds of things don't attract me.

Here is an interesting Wikipedia: Chiptune.

If it hadn't been for mlog's response, mine would have been that the same idea applies. That 8-bit color implies, as you said, 256 different colors (more about that in a bit) and that 8-bit music implies 256 different values in the waveform. I believe most telecommunications (telephone and cellphone) use 8-bit audio. Music, on the other hand, sounds pretty poor at this resolution. If I recall, something on the order of 12 bits is needed before most people consider it "good" and by the time you get to 16 bits (as on a CD) only a tiny fraction of people claim to be able to tell the difference (say between 15-bit and 16-bit).

Back to that 8-bit color. As mlog said, 24-bit color provides 8 bits each of red, green, and blue (or perhaps three other primary colors). So you can directly work with 256 shades of each. 8-bit color almost always is palettized, meaning that instead of a continuous set of shades of various colors, you can define up to 256 different colors (typically 24-bit colors) and associate them with the index numbers 0 to 255 just like a paint-by-number set. You then set each pixel in the picture to one of the 256 numbers and, when rendered, that pixel is actually colored according to the 24-bit color associated with that index.

 

Telephones do use 8-bit audio but it is a non-linear conversion (companding) to increase the dynamic range and improve the signal-to-noise ratio over a straight 8-bit linear conversion.

 

Hello,

In the old days there where a lot of MOD files on the BBS-systems.
These could be played with a MOD-player, using an interface consisting of a couple of resistors on the parallel port of the PC.
See this tale on the internet:
http://sonarmods.com/sonarmods/doc/ducktales.php

Bertus

 

Hello,

In the old days there where a lot of MOD files on the BBS-systems.
These could be played with a MOD-player, using an interface consisting of a couple of resistors on the parallel port of the PC.
See this tale on the internet:
http://sonarmods.com/sonarmods/doc/ducktales.php

Bertus

The parallel port is deprecated + no longer supported by Microsoft.

 

Telephones do use 8-bit audio but it is a non-linear conversion (companding) to increase the dynamic range and improve the signal-to-noise ratio over a straight 8-bit linear conversion.

Yep. It's called companding (compressing and expanding). I believe the basic compander simply uses a logarithmic model and thus, as far as the response of the human ear is concerned, achieves a performance comparable to about 12 bits if a linear model were used.

With modern cell technology, it gets a lot more interesting than that. As I understand it, they basically take short snippets (20ms, maybe?) of the signal and encode it as parameters for a synthesizer and transmit the parameters. I think they actually do some predictive modelling so that the receiver can predict what the next sample set will be based on the prior sets and then the transmitter (which knows what that prediction will be) is able to encode just the error between the actual signal and what the receiver's prediction will be.

While it's amazing that it works at all, not to mention the bandwidth savings they are able to squeeze out of it, I still hate having a conversation when either party is using a cell phone.

 

In Europe digital telephone systems use a-Law companding (http://en.wikipedia.org/wiki/Alaw) while in the US they use u-Law (http://en.wikipedia.org/wiki/%CE%9C-law_algorithm) - very similar but slightly different!

The companded bytes can be translated back to 16-bit samples easily and quickly with a 256 entry lookup table, but going the other way is a bit more tricky...