We have the technology...

[Mallette]

...to bridge the digital/analog divide if someone will just do it. The music server thread's swerve towards digitizing vinyl got me to thinking on this. I'd mentioned how convenient it was to use a Korg MR-1 DSD recorder to get LP's recorded at such accuracy that only a very highly trained ear would ever be able to accurately A/B.

In fact, it still isn't all that convenient. Many have written about the superiority of applying RIAA in the digital domain as opposed to the analog. I've never heard this, but the logic is sound. Combine that with the "universal" nature of the DSD recording process that yields a recording that, at least to these ears, can be transcoded transparently to any other digital format and you have a product waiting to be born.

The DSD analog to digital phono stage. Listen direct from the TT and/or store the results as a recording. One might even have a 60 minute or so buffer so that once you'd played a side, you could go back easily to listen to a previous cut a la PVR.

Seems to me that since the Korg MR-1 with DSD capability is only a few hundred dollars and provides such a marvelous quality that such a dedicated phono DSD AD stage could be made even more profitably (no moving parts required) at around the same price.

I guarantee I'd be first in line to purchase.

Dave

[Daddy Dee]

Dave,

Most interesting. It would be absolutely wonderful to get vinyl with digital gear.

Pardon my ignorance about this, is the end product a cd or dvd? That would seem like a no brainer for the recording industry.

or is the playback available only from the dsd device?

[Mallette]

The end product of the device built for basic purposes would be a line level analog signal. What I described is intended to be a 21rst century phono stage, not fill an existing purpose. If you could get the DSD license to allow it, you could also supply that signal as a file (to make CD's, playback from a server, whatever...same as ripping a digital disc today) as well as to provide a playback buffer (as per Tivo or PVR). If you wanted to make a "deluxe" model you could add internal transcoding to all PCM bit/sample rates for the same purposes. The "super deluxe" could add on-board digital domain click and pop suppression, dynamic range enhancement, even pitch change and such. The RIAA code could be enhanced to allow for "tailoring" deemphasis for folks like me that like to play old 78's that have non-RIAA pre emphasis. Heck for that matter, using the computer or an internal hard drive buffer (we are talking "super deluxe" here) you could listen to a 78 on a 331/3 TT simply by manipulatiing it in the digital domain. Of course, you'd need a change of stylus or cart, but that's a lot cheaper than a high quality turntable with 78 as an option. Since we are talking the digital domain, ANYTHING is possible here...

But basically I am describing a high resolution digital phono stage similar, but with much better performance, to the USB turntables and USB stages made for making CD's from vinyl.

However, this device would have the potential to rival the best analog equipment at a fraction of the cost, as well as provide considerable flexibility.

In fact, these would be ideal for turntables at any price range to provide "plug and play" out of the box functionality with greatly improved audible performance and functionality that the current rather "stuck in the 20th century" peformance. Many "borderline" audiophiles who just have a bunch of old vinyl they'd like to hear either do nothing because they immediately get confused about needing a phono stage (which their HT systems probably don't have) or wind up with pretty poor vinyl performance purchasing the USB plug and play ones for a 100 bucks at WalMart.

I'll be Music Hall could put one of these out with fine performance at 400.00 or less ready to go with a CD burner right in the base and a USB port for MP-3/Ipod storage. If my mom were alive, I'd buy one for her...she made cassettes of her LP's right up to her death.

Dave

[Daddy Dee]

Dave, thanks bro.

[Edgar]

It's a fine idea, except that it's extremely difficult to apply EQ directly to the DSD bitstream -- it has to be converted to PCM first. Once in the PCM domain, techniques have already been implemented to match analog filters extremely well, either as magnitude-only (See posts by Robert Orban -- yes, that Robert Orban -- at http://groups.google.com/group/comp.dsp/browse_thread/thread/59232b93211c041a/1786c9d17474f221. Even though he discusses de-emphasis filters there, the same technique can be applied to EQ.), or as magnitude and phase (See posts by Greg Berchin -- that's me -- at http://groups.google.com/group/comp.dsp/browse_thread/thread/ce5cc4bfe193070/b05e9312fa36a7be.). So as long as you aren't adamant about keeping the data in the DSD format, yes, we have the technology.

Greg

[Mallette]

As I tell my programmers "Are you telling me it can't be done or that you don't know how to do it?"

I'd want to keep it in the DSD domain. It's digital, and that means there is a way to handle it digitally. I suspect the main problem is largely Sony's insistence on killing it by limiting it's use. PCM isn't remotely as universal and only 24/192 comes close to DSD. I suspect a transcode to 24/192 and back wouldn't be a showstopper, but it strikes me as an unnecessary kludge.

I think a good coder could work it out.

Dave

[Edgar]

As I tell my programmers "Are you telling me it can't be done or that you don't know how to do it?"

A little bit of both. DSD is a 1-bit format (the commercial version; the "Pro" version is multibit). Filtering involves scaling and adding multiple samples together. Do that to 1-bit samples and you get multibit samples -- there's no way around it (well, no simple way ... afterward you can re-encode to 1-bit, but that's not quite the same as staying in the 1-bit domain).

In addition, trying to design filters that operate below 100 kHz, when the sampling rate is 2.8224 MHz, inherently causes numerical precision problems. Not insurmountable, but much easier to deal with when the ratio of the sampling rate to the frequency of interest is lower, within Nyquist constraints.

I think a good coder could work it out.

It's not a coding problem. It's a mathematical problem.

Greg

[Mallette]

It's not a coding problem. It's a mathematical problem.

Thanks, Edgar. You told me some things I didn't know at all and while hardly a digital genious I can see the issues. I write algorithms, and let the coders make them work. However, I suspect a good mathematician could work this out.

If not, I'll bet I one could write an algorithm could be developed using some lateral digitial transforms that would do the trick with little if any artifacts.

I have yet to see any limits to what can be done in the digital domain by real black belt nerds who won't take no for an answer...

Dave

[artarama]

In fact, these would be ideal for turntables at any price range to provide "plug and play" out of the box functionality with greatly improved audible performance and functionality that the current rather "stuck in the 20th century" peformance. Many "borderline" audiophiles who just have a bunch of old vinyl they'd like to hear either do nothing because they immediately get confused about needing a phono stage (which their HT systems probably don't have) or wind up with pretty poor vinyl performance purchasing the USB plug and play ones for a 100 bucks at WalMart.

Dave,

I just want to get that "bunch of old vinyl" I have into digital storage that will archive them without degrading the sound quality. I would like easier access for locating and playback from this storage than my shelves and bins are now ie. a search function. I would like the new device to work with my system not add another TT (turntable) into it. I don't want to have to learn, research and understand all the intricacies of how this happens ie. it has to be intuitive to use. Honestly right now I don't know what DSD, PVR & PCM refer to so I can't have any input on that. This kind of un-referenced jargon drives me crazy!

I am definately a target audience for a device like you envision sans the TT. I am really suprised it does not already exist, there may be some serious legal/artist rights issues. Remember the stink about read/write CD's? Keep up the brainstorming!

[Edgar]

However, I suspect a good mathematician could work this out.

If not, I'll bet I one could write an algorithm could be developed using some lateral digitial transforms that would do the trick with little if any artifacts.

I have yet to see any limits to what can be done in the digital domain by real black belt nerds who won't take no for an answer...

People have definitely been working on it. I know of at least one patent for processing 1-bit signals directly (5,990,818; "Method and apparatus for processing sigma-delta modulated signals"), but it is not nearly as simple as processing linear PCM.

Also, the infamous Lipshitz and Vanderkooy paper is worth reading, not because it points out the flaws in the format, but because in section 3 it shows how linear PCM can easily be made to outperform DSD, while retaining the inherent ability to be filtered.

Greg

[Mallette]

Artarama, tis folks like you such a phono stage would be targeted at. The basic level would require nothing more than be able to locate the inputs on one side and the outputs on the other, nothing more. However, it would provide a path to enormous flexibility.

Also, the infamous Lipshitz and Vanderkooy paper is worth reading, not because it points out the flaws in the format, but because in section 3 it shows how linear PCM can easily be made to outperform DSD, while retaining the inherent ability to be filtered.

I will comment on that statement first without having read the paper...which I will do. "....how linear PCM can be easily made to outperform DSD..." is a difficult to parse statement. While I found I could detect the difference between 24/96 and 24/192 in tests I did a number of years ago, that difference was very, very small and almost a "learned" thing of little consequence. At 24/192 the recordings were completely transparent as to the source and I find DSD to be the same. So I find it hard to imagine "greater performance" than that result unless we are talking about non-audible issues. Further, after more thinking, I suspect that internal transcoding...if that is truly necessary for various processing...and back to DSD would not likely produce audible artifacting except to the most highly trained ear in very demanding situations if at all.

The reason I insist on DSD for primary storage is it's universal transcoding capability...nothing else. File size and audbible performance of DSD and 24/192 PCM are essentially equal. However, unless one limits (which is also practicable) the signals in a PCM device to the evenly divisible you are going to have audible dithering issues when transcoding. The 16 or 24/48 that is standard for video becomes problematic here for those purposes.

I am not an engineer...but a pretty good idea guy.

In any event, I am glad I started this thread. There are a number of really advanced theorists on this forum I hope will weigh in. We might even wind up getting someones attention who can realize such a device.

Dave

[Edgar]

I will comment on that statement first without having read the paper...which I will do. "....how linear PCM can be easily made to outperform DSD..." is a difficult to parse statement. While I found I could detect the difference between 24/96 and 24/192 in tests I did a number of years ago, that difference was very, very small and almost a "learned" thing of little consequence. At 24/192 the recordings were completely transparent as to the source and I find DSD to be the same.

In the paper, L&V provide examples of linear PCM configurations in which the data rate is lower than DSD, the noise-shaper design is simpler and the noise-shaping less severe, and the signal can be properly dithered (a 1-bit DSD signal can not). That is all that I meant by "outperform DSD".

The reason I insist on DSD for primary storage is it's universal transcoding capability...nothing else. File size and audbible performance of DSD and 24/192 PCM are essentially equal. However, unless one limits (which is also practicable) the signals in a PCM device to the evenly divisible you are going to have audible dithering issues when transcoding. The 16 or 24/48 that is standard for video becomes problematic here for those purposes.

That is a very strong argument, indeed. From that standpoint, DSD is very attractive.

In any event, I am glad I started this thread. There are a number of really advanced theorists on this forum I hope will weigh in. We might even wind up getting someones attention who can realize such a device.

I can do it. I hate to be mercenary, but is there any money to be made from it?

Greg

[Mallette]

I can do it. I hate to be mercenary, but is there any money to be made from it?

I'd come up with a set of specs for the basic device...in this case, line and digital out...and float it here. In my opinion there is money to be made from the "fence sitting" digital types who would love to sample vinyl, but know little about phono stages and such and have no hangups about the digital domain.

Get rich? Probably not, but serious money...I think so. I'd want one for sure.

As to the mass market...maybe. Depends on the price. If you can go head to head with the cheapo USB TT's being marketed through the mail order and WalMart type places as well as claim "audiophile quality" there is significant coin to be made. Certainly if somebody would write the code to use Freedb database entries to make good guesses at the track breaks that would be a BIG plus. Even with the couple of programs out there to ease the pain of vinyl digitzing it's still a hassle. Of course, many LP's simply haven't been issued on CD, but even then using a good threshold algorithm to at least make good guesses that can be visiually confirmed would help a lot.

However, I mooted this idea because I think it's time for a high end phono ADAC that can rival, perhaps surpass, analog performance while greatly reducing cost. That's the bottom line...the rest is gravy.

Dave