Direct-to-Disc

[morricab]

It has to capture it on a relative basis.

The problem I have with Tascam is that it loses the room ambience

Yes, to some degree you are right but if the losses are too signficant I would argue it will bias a relative discussion as well.

I don't find the Tascam I have loses much of the room ambience at all...if anything it seems too sensitive to background noise often times and makes it seem a bit worse on the recording than in real life.

 

[morricab]

So what is it about the impacts of stereo reproduction (as a technique) that is then especially relevant to the recording chain be it direct to disc or otherwise. You often refer to the stereo approach as being an illusion like some kind of end statement (and did so again) but I still hold that all of musical reproduction is inherently an illusion so why differentiate? It is all building to an experience and all of it is inherently synthetic whether it is stereo, mono or multi-channel but that is just the way of it. The true art is when the synthetic seems natural... but that is just a summative assessment for those who feel that is valuable. Direct to disc could simply feel less synthetic because it could be one simple step less removed from the truth.

It is the same approach I now take with reproduction...to minimize the steps. My DAC has only 1 active element and my amp has only 2 active elements for a grand total of 3 active stages from DA conversion to speakers...and each active stage is a single tube...not an opamp (that has a whole bunch of active elements) for example or a "compound" element (two or more triodes or transistors to making the active element). Also, the path is strictly one directional, meaning there is zero negative feedback through the whole chain...not even local feedback. IMO, this is key to stripping away signatures that can be perceived as "synthetic". Cleaning the power also has a positive effect in removing artifacts that impact soundstage and 3d imaging. That is for my main rig.

For my experimental horn rig, it is similar where the tweeter DAC has only one active stage (it is a "compound" element though) and the tweeter amp is the simplest PP tube amp possible with three active elements. The mid/bass uses a DAC that is direct off the chip that feeds a SEP with only two active stages and no negative feedback.

 

[morricab]

Exactly my point. We can learn a lot about it if we do understand the technicalities and the psychoacoustic why's of the stereo system. BTW, stereo sound reproduction is much more than just separation.

Back to the main subject, many people refer they prefer direct to disc to tape sourced vinyl. Surely there are reasons behind this preference, this is an interesting thread.

I am convinced it has to do with reduction of losses that comes from multiple transfer steps. Why direct to disc analog still sounds more dynamic than direct to digital is probably worth further discussion. It could have a lot to do with the analog circuitry that feeds the AD converters more than the AD itself. For cutting of records directly, this is often done with tube amps to drive the cutter. At least the small rig I saw was driven with a tube amp.

Analog tape also involves amplifcation circuitry in addtion to things going on with tape saturation, bias etc. I think you can capture larger dynamics without saturation with direct to disc (grooves just get larger and larger modulation). Digital you always have to be conscious of the hard clip and tape the saturation as you overdrive it. It is VERY hard without adding compression or external limiting to capture large dynamic swings in either analog tape or digital...not sure what the headroom with direct to disc is like but I guess as long as you don't overdrive the cutting amp you can probably handle a very wide dyanmic range...although I could be wrong about this...

 

[bonzo75]

Here goes yesterday

 

[Alrainbow]

May I ask what model tascam ?

 

[morricab]

May I ask what model tascam ?

DR-100MKIII

 

[microstrip]

It is the same approach I now take with reproduction...to minimize the steps. My DAC has only 1 active element (...)

The DAC chip is a not a passive element - it includes hundreds of active ones. IMHO this talk about minimization after the signal has been digitized and re-built by such complex device is like telling people to use AWG 1 power cables to minimize losses in power mains. There are very good and valid reasons for your choice, but not the minimization of steps. Otherwise you would have only one power tube after the DAC chip!

Two decades ago there was a project of a digital speaker without analog steps - the bits value were weighted by the speaker elements area. The developers used the same type of minimization talk to promote it, but reports said it sounded terrible. But yes, my Edison phonograph and EMI gramophone have zero active elements, no gain at all!

 

[microstrip]

I am convinced it has to do with reduction of losses that comes from multiple transfer steps. Why direct to disc analog still sounds more dynamic than direct to digital is probably worth further discussion. It could have a lot to do with the analog circuitry that feeds the AD converters more than the AD itself. For cutting of records directly, this is often done with tube amps to drive the cutter. At least the small rig I saw was driven with a tube amp.

Analog tape also involves amplifcation circuitry in addtion to things going on with tape saturation, bias etc. I think you can capture larger dynamics without saturation with direct to disc (grooves just get larger and larger modulation). Digital you always have to be conscious of the hard clip and tape the saturation as you overdrive it. It is VERY hard without adding compression or external limiting to capture large dynamic swings in either analog tape or digital...not sure what the headroom with direct to disc is like but I guess as long as you don't overdrive the cutting amp you can probably handle a very wide dyanmic range...although I could be wrong about this...

Sorry, but it seems to me these are arguments of the 90's on average equipment. Currently used analog amplifier IC's have 140 dB signal to noise ratio and ADCs have true 116 dB dynamic range. The existing problem is just the opposite - the digital equipment captures too much information and many professionals did not know how to handle it - and the audiophile systems were not also able to reproduce it adequately. Fortunately the situation improved significantly professionals learned a lot - and the way people listen to stereo music in top systems also changed, we can't forget it!
BTW, I would love to know what are the specific D to D recordings you are exactly thinking about when making these comments on the D to D technique.

 

[morricab]

The DAC chip is a not a passive element - it includes hundreds of active ones. IMHO this talk about minimization after the signal has been digitized and re-built by such complex device is like telling people to use AWG 1 power cables to minimize losses in power mains. There are very good and valid reasons for your choice, but not the minimization of steps. Otherwise you would have only one power tube after the DAC chip!

Two decades ago there was a project of a digital speaker without analog steps - the bits value were weighted by the speaker elements area. The developers used the same type of minimization talk to promote it, but reports said it sounded terrible. But yes, my Edison phonograph and EMI gramophone have zero active elements, no gain at all!

A resistor ladder DAC has mostly...resistors...and I have only 3 tubes after the DAC to the speakers. IV conversion is done with a transformer. There are many technical reasons why only one tube would not work from DAC chip to speakers. 3 is probably the practical minimum (maybe two but not less I think). I have found that minimizing the circuit complexity and maximizing the power supply quality has resulted in steadily better sound. You can doubt but until you have tried you cannot say it is not important.

Do your Gramophones sound bad? In some important ways they can sound very good. There is a maxim about oversimplification...from the sound I get I would not say I have gone too far.

 

[morricab]

Sorry, but it seems to me these are arguments of the 90's on average equipment. Currently used analog amplifier IC's have 140 dB signal to noise ratio and ADCs have true 116 dB dynamic range. The existing problem is just the opposite - the digital equipment captures too much information and many professionals did not know how to handle it - and the audiophile systems were not also able to reproduce it adequately. Fortunately the situation improved significantly professionals learned a lot - and the way people listen to stereo music in top systems also changed, we can't forget it!
BTW, I would love to know what are the specific D to D recordings you are exactly thinking about when making these comments on the D to D technique.

And the analog IC in many important ways STILL sounds worse than a single properly designed triode stage.

 

[microstrip]

A resistor ladder DAC has mostly...resistors.. (...)

No, it includes current sources and analog switches to commute them using a digital signal that in reality leaks through the switch and are not perfect - ideal electronics only exist in elementary courses. But yes, some people like the noise and non linearity intrinsic to these techniques - choice is part of our preferences. Most others prefer other types of defects.

and I have only 3 tubes after the DAC to the speakers. IV conversion is done with a transformer. There are many technical reasons why only one tube would not work from DAC chip to speakers. 3 is probably the practical minimum (maybe two but not less I think). I have found that minimizing the circuit complexity and maximizing the power supply quality has resulted in steadily better sound. You can doubt but until you have tried you cannot say it is not important.

Yes, you have found that it pleases your individual preference and some people share your findings .We should be happy with it, but it is not enough to validate a general theory, since there is no technical support at all in it.
BTW, it can be easily done with two tubes, one tube only needs some expertise and an high-efficiency speaker. Did you ever go in tube DIY?

Do your Gramophones sound bad? In some important ways they can sound very good. There is a maxim about oversimplification...from the sound I get I would not say I have gone too far.

My main complain about my gramophone was that it sounded too loud - when I finished refurbishing it late in the night I still lived in a condo and I could not play it for long fearing to disturb the neighbors.

 

[microstrip]

And the analog IC in many important ways STILL sounds worse than a single properly designed triode stage.

Again feelings from the 80's. We are in 2020. Anyway most of our praised recordings went through tens of horrible sounding ICs after the tube microphone and sound excellent. Remember that Nick Doshi tape preamplfier uses an IC buffer in the input. But yes, anyone can assemble a decently sound buffer with a tube, a lot of expertise that transcends the DIY is needed to use the current top ICs.

 

[DonH50]

Just happened to see this thread. I'll skip the device debate (nobody ever convinces anybody else anyway) but can relate some direct-to-disc experience from the recording side. Yes, you can get a magical capture, and there were a few, but it is almost impossible to do as well as a "normal" recording. Maintaining focus through an entire side with no break and no margin for error puts intense pressure upon the performers. And if you do make a mistake, it is not cheap to fix, and you have to go through the entire performance again (very expensive in studio time!) whilst trying to maintain that level of focus, again. It felt to me more like a long audition with no breaks than a performance, much more stressful, and no way to fix any little blips during it. And to keep breaks between songs short you must "instantly" switch music and mindset to play again with no break (usually in a live performance there's a longer break to introduce the piece and give the musicians a breather). It is very fatiguing mentally and physically. And for the mixing/mastering engineer as well since s/he is riding the board and the cutter in real time. Tough job.

I suspect most musicians/singers would prefer to put their very best on the recording and that means being able to retake a piece, or a section, try different levels in the mix, tweak EQ, etc. to dial in the final product.

FWIWFM - Don

 

[bonzo75]

The best musicians always come to life during live shows

 

[DonH50]

The best musicians always come to life during live shows

And in the studio, too, but most (probably "all") live shows have little flaws they'd've liked to have fixed. I hate listening to recordings of myself playing; all I hear are my own flaws. That said, studio musicians (something I never was and never will be) are on a whole 'nuther level in terms of performing perfectly on the first (and every) take.

 

[microstrip]

The best musicians always come to life during live shows

I am sure that in a life show D-to-D they would overload the system, ruining the lacquers, after the first applause!

In this aspect current digital recording is great for life recordings. Several of the latest Savall recordings that I have too often praised are live recordings.

 

[morricab]

I am sure that in a life show D-to-D they would overload the system, ruining the lacquers, after the first applause!

In this aspect current digital recording is great for life recordings. Several of the latest Savall recordings that I have too often praised are live recordings.

D2D is live as it has to be done in one take. You wouldn't turn the amps way up like in a live amplified concert to make a recording...there would be no need to do so.

 

[morricab]

Again feelings from the 80's. We are in 2020. Anyway most of our praised recordings went through tens of horrible sounding ICs after the tube microphone and sound excellent. Remember that Nick Doshi tape preamplfier uses an IC buffer in the input. But yes, anyone can assemble a decently sound buffer with a tube, a lot of expertise that transcends the DIY is needed to use the current top ICs.

Yes, and tubes are more popular than ever and still sound better than ICs. Studies have been performed that show the overload potential for opamps is far lower than tubes...a high tolerance that is necessary for live recording to not sound distorted.

 

[cjfrbw]

Yes, and tubes are more popular than ever and still sound better than ICs. Studies have been performed that show the overload potential for opamps is far lower than tubes...a high tolerance that is necessary for live recording to not sound distorted.

Maybe why some of the most amazing recordings are still from 50's and 60's, blessedly preserved on some vinyl records. Toobs throughout the recording/mastering chain and onward and upward onto high ips tape.

 

[morricab]

Maybe why some of the most amazing recordings are still from 50's and 60's, blessedly preserved on some vinyl records. Toobs throughout the recording/mastering chain and onward and upward onto high ips tape.

There was an interesting study published in the mid-1970s where a recording engineer decided to investigate why he was hearing a deterioration in the sound at many recording studios. The major change he found was the switch to SS throughout the chain. They then looked at tube, discrete transistor and op amps and found that op amps (at that time) had poor tolerance for being overdriven, something that happens frequently in live music capture. Discrete transistors were a bit better and tubes much better in this regard. However, they also noted the tubes sounded more natural even below the overdrive thresholds. Maybe it applies less today...at least below overdrive...but it is well known that high feedback circuits , which op amps inherently are, don’t react well to being overloaded, so modern ones likely have similar issues.