Topic: Tascam DV-RA1000 DVD Recorder Surprise

[L Ron Hoover]

I'm in WI, and the initials FC are baffling me.

[BWolf]

alright, having taken many systems and signals classes and a lot of digital signal processing, here is my guess as to why it will sound better:

when a PCM wave is recorded, the 24 bits (or 16) from the AD are stored somewhere.  as the recording progresses, each sample is recorded at a distinct time (k*f where k>=0 and f is the sampling rate).  This means that when it goes to play it back, there has to be some interperlation.  this means that it places the sample next to the previous sample and connects them with a line (or best guess curve using previous values).  then it goes to the next sample and repeats the process, forever guessing what is exactly between the sample.  i know most people think that when sampling at such a high rate it wouldn't make much difference, but it will.

now DSD uses a 1-bit-delta scheme.  this means that it doesn't interpolate, but just add or subtract one unit (guessing its somewhere in the range of a millionth of of Hz) from the previous value.  this means that it will always be smoother then PCM because it never "guesses".  this would also account for the reason it must sample so fast.  if the sound went from 0 dB to -20 dB, it needs to be able to complete that transition very fast.  if it lags (that is, doesn't get down to the desired value fast enough), the sound will suffer.

please correct me if i'm wrong, but i think this may help some people understand the difference between the two

[Digital Quality]

It would also be interesting to see what D/A your old playback had. The D/A in your new setup has a lot to do with the new sound IMO. It's the whole reason we have $5K blue book players. You can only make the transport and laser so well.

[MattH]

ok, full compass.
comparing to: Apogee, HHB, and Denon D/A's
using: Meyer 500 amp (600wpc), Meyer 500 speakers, Denon and Gemini pre's, Infinity Kappa 5.1, Denon AVR-85, Sennheiser HD-280

BWolf has it right. But I'm not sure how upsampling really works.

[BWolf]

i think the most important point to understand is that digital systems are discrete time, that is each sample is taken at a time k*T where k is some positive integer.  when using the DSD encoding scheme, they effectively make it a continuous time system by just using 1 bit delta encoding and by making the sampling rate extrememly high.  like other digital components, switching time is usually the most importand factor.  because DSD can switch the output faster and more accurately then PCM, it will in essence sound better

i could be way off ....i guess this is a good test to see how good my schooling was

[MattD]

BWolf, I'm following, but a little confused as to how amplitude changes are represented. One bit isn't a lot of numbers to work with to represent amplitude. How does that work?

[BWolf]

alright, here are the two scenarios.  first PCM take discrete time sample and actually records each sample independently, then interpolates between them.  DSD uses whats called one bit delta encoding.  i'm not sure exactly how they encode, but lets just say a digital "1" means to go up one unit and a digital "0" means go down one unit.  DSD uses whats called a differential scheme, where the next state depends on both the current state and the input.  so lets just say that the "current state" is 5 volts and the input (DSD encoded number) is a one.  the next state would then be "5 V + 1 unit".  if the DSD bit was a zero, then the next state would be "5 V - 1 unit"

edit to add: the unit that is added or subtracted MUST be a predetermined value.  none of this would work if it was just a random amount of change!!

does that make any sense?

[Jammin72]

If you want to know why Sony thinks that it's such a good idea check out this document on the technology behind the digital amplifiers.

DSD is esentially the same type of idea  Skip the filters sample 1.92 million times per second to re-create the sine wave!  Much smaller stairs there for sure.

Note that it metions the analog input is upsampled to DSD for the purposes of sending it to the reveiver, our experience was that the analog inputs were still a limitiation due to the input circuits used before the upsampling occurred.  I'm not sure if PCM is upsampled in the Digital path or not? 

http://www.docs.sony.com/release/STRDA9000ES_TWP.pdf

[MattH]

I would think the digital in upsampling to DSD on the Sony receivers is better than the anolog in upsampling to DSD.

[mhibbs]

sounds like i'm going to have to invite myself over to hear this thing in person

 

mitch

[dklein]

thanks for the link to that sony document.

I ain't no digital engineer but as far as upsampling to DSD, the only benefit that I can see would be jitter reduction (if you believe everything they wrote).  And there are other ways to go at jitter. 
Anyone else have insights?

DSD recording does sound interesting, but there were far too many references to "digital transmission content protection".  We all know how Sony likes to lock everything down.

There were a few things in there that made me skeptical about some of the theory.  I like the way they talked about using switch mode power supplies but then for the top model, move back to a toroidal transformer just like other high end gear.  All in all, it read like propaganda.  I suppose I should shut up and go listen to one now.

[MattH]

No question about it. DSD completely rules. So much higher resolution than 16 bit and 24 bit. It really is that much better and I don't think it has anything to do with jitter .

[dklein]

No question about it. DSD completely rules. So much higher resolution than 16 bit and 24 bit. It really is that much better and I don't think it has anything to do with jitter .

ok but inquiring minds want to know why dsd upsampling a regular old CD would improve sound.

Analogy 1: I can take a picture in photoshop and increase the DPI to whatever I want but the actual resolution will be limited to the original.  Any extrapolations of accuracy will still be based on the original resolution.

Analogy 2: I can take an audio file, convert to 24/192 but it will remain no more accurate than the original.

Not saying it doesn't work...I just want to know why!

[Tim]

I run an upsampling DAC in my hi-fi system (Bel Canto DAC 2... 24/192 upsampling) and though I don't really grasp the science behind it, nor do I really care to, my ears do not lie... it definitely sounds better

here are a few links...
http://www.aslgroup.com/dcs/dacsandupsamplers.htm
http://www.audioholics.com/techtips/specsformats/upsamplingvsoversampling2.php
http://www.aslgroup.com/dcs/upandover.htm

[Juan Carlos Medina]

Discwelder has new SACD authoring software out to take DSD-Raw to SACD. $3,000 now but probably $300 in a year.

So we WILL one day be able to create SACD's from these raw DSDIFF files? If that's true, that just means we simply have to wait until it is easy to do so.

I've heard two different sides as to whether or not SACD will be in our future; that's one thing I'd really like to know.

I see that this unit has balanced XLR inputs. Is that what you're feeding it? You could just throw a V2 in front of this bad boy and you have one slick setup...