The No.1 Website for Pro Audio
24th April 2005
#1
Here for the gear

๐ง 15 years
DSD Refresher...

Evening all, been reading about DSD and wouldn't mind if anyone could help me get my head around the concept of the amplitude being represented by 1-bit instead of 16 or 24.

As I understand it, with PCM, after the amplitude level is read by the system it is then decimated so that it can be quantised to one of 16,777,216 steps provided by a 24-bit for example. DSD does no decimation but rather keeps the original value of the amplitude which is represents as a 1 bit word. Is this correct?

Also, if this is the case, was the initial purpose of the decimation in PCM because there would have been an infinite amount if amplitude values unless it was quantised?

RG

btw. didnt want to post this in the recent DSD thread as it was too heated for a dumb question like this!! Ive also seen the DSD docs from Sony etc. but their terms aren't as simple as I'd like them.
24th April 2005
#2
Gear Guru

1 Review written
๐ง 15 years

Think of the bit stream as a PWM (pulse width modulator). The 1 or 0 really only indicates if the signal is louder or softer than it was the last sample.

Remember that at 2.8MHz, the bit rate is much faster than the fastest transient in the recorded signal. To go from zero to the negative rail and then to the positive rail, you'd need something like 50 zeros followed by 100 ones, and it would still be too high a frequency to hear.

See:

http://www.digit-life.com/articles/sacd/

-tINY

24th April 2005 | Show parent
#3
Here for the gear

๐ง 15 years
aah, so instead of each amplitude value being given a value as such, it is just either higher or lower than the previous one?
24th April 2005 | Show parent
#4
Here for the gear

๐ง 15 years
Thanks for the info. Another point is, what is the min and max amplitude for a DSD system? Actually I should rephrase that. Is there a resolution for the number of amplitude values or is this idea defunct as it has no steps as such?
24th April 2005 | Show parent
#5
Motown legend

๐ง 15 years
Quote:
Originally Posted by Sounds Great
Doesn't DSD more closely mimic the action of an analog tape recorder? In other words, isn't this true analog tape modeling?
Not really. It just lets you apply very different filters that a lot of people think sound better.

The good news is that the threat of DSD sounding better forced the PCM chipmakers to clean up their acts.

I'll return to my recuring argument which is that implimentation seems to make a much bigger difference than format.

## Welcome to the Gearslutz Pro Audio Community!

###### Registration benefits include:
• The ability to reply to and create new discussions
• Interact with VIP industry experts in our guest Q&As
• Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD \$20/year