M3DVQ wrote:the 40 pin connector is IDE, for data... will this contain the original audio data?
if the drive has a digital-audio out (a 3 or 4 pin connector) isn't that what you want?
The only way that my drives can output audio is via the EIDE connector to the motherboard which has the needed D-As etc. They work suprising well considering their cost.
The digital audio-out is in a serial format, (not that my drives have it anyway, they're kinda old) it's a headache to de-serialize as with the standards proposed the 16-bit word needs to be processed in a non-standard parallel manner.
I also don't want to use the on-board D-A, I want access to the raw 16-bit parallel data on the EIDE connector. That in itself is not a problem, I'm just having trouble finding how the data stream is organised. The hardware (pin-outs etc.) are not an issue.
What I am trying to to is see if it's possible (there is no reason why not) that external D-As could be DC-coupled and each chunk of the 16-bit data be diverted into either the monochrome channel, or the appropriate colour decoder, or processed in some other way.
It's also conceptually possible to use the lower 8-bits of the right channel for more video data. The right channel D-A would simply ignore these, OK, we end up with 8-bit audio, but as with all TV audio is the Cinderella of the industry, and few care too much about it.
But try watching the news with the sound off!
Steve A.
OK, I'm learning yet again. It seems that the data is organised into 'frames' of 2352 bytes each which represents 1/75th of a second, = 176,400 bytes/sec. Divided by two for two channels, then divided by two for 16-bit audio = the famous 44100Hz sample rate. Someone please check my maths!
This is for audio CDs, it appears that data CD frames are of 2048 bytes each, which would seem sensible.
So my next task is to find out how these frames are composed and what each byte represents.
The interesting thing is that 1/75th of a second is exactly one sixth our frame rate of 1/12.5 sec.