Preamps with DSs

[Spannko]

Just a couple of things, guys.

As well as not using the ESS DAC, Linn use a "volume control module .... placed at the end of the digital signal path, just before the DAC". The VCM operates at 35 bits/384 kHz and the DAC at 24 bits/384 Khz. So the Linn DAC doesn't adjust volume like the ESS DAC.

I've watched a video of the presentation the ESS slides came from. At the end, the designer talks a bit more about what he means by "Exquisitely" designed pre-amps. He was referring to an absolute state of the art laboratory grade design maintained at absolute zero. He went on to say that domestic designs will never be able to reach such low levels of distortion due to thermal noise.

Also, I'm listening to a KDS/2 directly into a power amp as I type this. I can assure you all it's definitely not "unbearable"! My experience of the DVC is nothing like Buggi666's - the sound's not "flat, dark " etc. I've experimented by setting the DS volume at 20 and the pre-amp volume at 100. Given what some people have said about DVC's, I expected the music to be buried in digital noise, however I can't detect any distortion at all.

Please note that I'm not commenting on whether my system sounds better with or without a pre-amp, just presenting information that will help us to keep discussing Linn's new developments from a neutral stance. I appreciate this is very hard to do (some might say impossible), but something certainly worth striving for.

As Einstein said, "I have no special talents, I am only passionately curious".

At the moment, I'm still very much a fence sitter with regard to Exakt, SPACE and DVC. I honestly don't believe that Linn have changed the way they evaluate their products - that would be financial suicide. There's no way they can loose their traditional buyers and compete with the likes of B&O. They don't have the products and their traditional dealers would drop them like a hot potato. I don't know of one Linn dealer who would be happy to sell under performing Linn hifi - there are too many musical alternatives now. However, like many, I remain to be fully convinced, even though I've heard some positive (musical) aspects of all three new technologies.

[matss]

Just a couple of things, guys.

As well as not using the ESS DAC, Linn use a "volume control module .... placed at the end of the digital signal path, just before the DAC". The VCM operates at 35 bits/384 kHz and the DAC at 24 bits/384 Khz. So the Linn DAC doesn't adjust volume like the ESS DAC.

I've watched a video of the presentation the ESS slides came from. At the end, the designer talks a bit more about what he means by "Exquisitely" designed pre-amps. He was referring to an absolute state of the art laboratory grade design maintained at absolute zero. He went on to say that domestic designs will never be able to reach such low levels of distortion due to thermal noise.

Also, I'm listening to a KDS/2 directly into a power amp as I type this. I can assure you all it's definitely not "unbearable"! My experience of the DVC is nothing like Buggi666's - the sound's not "flat, dark " etc. I've experimented by setting the DS volume at 20 and the pre-amp volume at 100. Given what some people have said about DVC's, I expected the music to be buried in digital noise, however I can't detect any distortion at all.

Please note that I'm not commenting on whether my system sounds better with or without a pre-amp, just presenting information that will help us to keep discussing Linn's new developments from a neutral stance. I appreciate this is very hard to do (some might say impossible), but something certainly worth striving for.

As Einstein said, "I have no special talents, I am only passionately curious".

At the moment, I'm still very much a fence sitter with regard to Exakt, SPACE and DVC. I honestly don't believe that Linn have changed the way they evaluate their products - that would be financial suicide. There's no way they can loose their traditional buyers and compete with the likes of B&O. They don't have the products and their traditional dealers would drop them like a hot potato. I don't know of one Linn dealer who would be happy to sell under performing Linn hifi - there are too many musical alternatives now. However, like many, I remain to be fully convinced, even though I've heard some positive (musical) aspects of all three new technologies.

+1. Well put. As I believe Einstein also said, "Everything should be made as simple as possible, but not simpler".

/mats

[donuk]

And while we are on Einstein quotes:

Here is one that is universally misquoted - what he actually said was -
"God does not play DVCs with the universe".

Don, rainy downtown York, which is why I am writing this rubbish.......

[Spannko]

Let's not forget,

"Not everything that can be measured matters, and not everything that matters can be measured"!

[matss]

Let's not forget,

"Not everything that can be measured matters, and not everything that matters can be measured"!

Absolutely true. As Linn and other sound people states - "measurements can never tell us what sounds better, only help us to better understand why".

/mats

[stefan]

....
Have a dem of the volume on an ADS2 then decide whether you need a preamp, it is much better than the ADS1...

Before I've heard ADS/1 vs ADS/2 directly to poweramp. Now I've compared ADS/1 and the new ADSM through a KK. If the improvements would be in the internal volume then the gap between them should have been reduced. But I feel it isn't.

Then, at the ADS/1->2 demo, I didn't really understand why a preamp wasn't used (this was a short time before their announcement of ending them). Now I do.

[Music Lover]

I admit to hesitating for a long time before I posted that link. Why? Because its focus is entirely technical and my focus is entirely musical.

I was a bit puzzled myself actually.

Regardless it was a good read, even if conclusions from technical measurements should be avoided as you can almost "prove" anything, depending of your agenda.
As the LP12 analogy clearly describes.

[Charlie1]

I expected the music to be buried in digital noise, however I can't detect any distortion at all.

I wonder how the distortion will manifest itself though. I recall that the harmonic distortion of early CD led to a brittle harsh sound, quite different to how analogue distorted. Or is harmonic distortion something different?

I like the Einstein quote btw.

[Spannko]

I expected the music to be buried in digital noise, however I can't detect any distortion at all.

I wonder how the distortion will manifest itself though. I recall that the harmonic distortion of early CD led to a brittle harsh sound, quite different to how analogue distorted. Or is harmonic distortion something different?

I like the Einstein quote btw.

I agree that inaudible distortion could well produce other unwanted artifacts, particularly a reduction in a systems tune playing ability. When the DS has run in I'll listen to the DS's DVC at different levels and see how it affects the tune.

Although I couldn't detect any distortion at 20 on the DS and 100 on the pre-amp, I had to get within 150mm of the speaker to hear anything - it was that quiet. I wanted to push the DS into producing as much distortion as possible and wasn't in tunedem mode at the time.

I've also planned a comparison of ADS/1 v ADS/2 (with and without a pre-amp) and the KDS/2 with and without a KK. Hopefully, I'll then know more about the DVC's strengths and weaknesses.

Re: Einstein, I think my favourite quote is:

"We cannot solve our problems with the same thinking we used to create them"

[matss]

When the DS has run in I'll listen to the DS's DVC at different levels and see how it affects the tune.

Although I couldn't detect any distortion at 20 on the DS and 100 on the pre-amp, I had to get within 150mm of the speaker to hear anything - it was that quiet. I wanted to push the DS into producing as much distortion as possible and wasn't in tunedem mode at the time.

Interesting comparison.

Quoted from the Linn forum and from Linn staff:

Murrays,

Does the signal pass through the DVC even when the DS's volume control is switched off in Konfig, or does the signal bye-pass the DVC?

Thanks

Yes. Even when the volume control is switched off in Konfig, the volume control module remains in the signal path (equiv. to volume on and set to 80). The volume control module contains the dithering algorithm that matches the internal 35-bit word width to the 24-bits required by the DAC chip.

That should make it easy for more of us to see what DVC and AVC does to musicality.

BR, Mats

[Spannko]

I found murrays reply really interesting, even though I don't really know what the implications are!

It suggests that we're listening to the DVC at all times. I think it's fair to say that when the DVC is switched off (i.e. set to unity gain [80]) everyone is happy with its performance. It will be interesting to see if there is any difference in the performance between KDS/2 with DVC switched on (@80)/off into a pre-amp. In theory there shouldn't be any difference - but we'll see!

I can only think of 2 other possible factors influencing the deterioration in musicality people have heard when using the DVC directly into a power amp:

1. Somehow the DAC doesn't work as well with lower volume inputs. I don't understand the workings of DAC's so I'll leave that for others to comment on.

2. The change of input/output impedances (by removing the pre-amp) is unsettling the fine balances needed for optimum tune playing.

There may (probably!) be other explanations too!

[lejonklou]

It will be interesting to see if there is any difference in the performance between KDS/2 with DVC switched on (@80)/off into a pre-amp. In theory there shouldn't be any difference - but we'll see!

When I have switched the DVC off there's been a clear improvement in sound quality. I have assumed this is due to the dithering being switched off.

[matss]

It will be interesting to see if there is any difference in the performance between KDS/2 with DVC switched on (@80)/off into a pre-amp. In theory there shouldn't be any difference - but we'll see!

When I have switched the DVC off there's been a clear improvement in sound quality. I have assumed this is due to the dithering being switched off.

If I understand Murrays correct that is exaktly the reason it is still in the loop even when it is switched to off in Konfig - "The volume control module contains the dithering algorithm that matches the internal 35-bit word width to the 24-bits requred by the DAC chip". The DS would not work correctly without the DVC module in the loop. I guess something else must be going on, for you to experience a clear improvement in sound qualtiy.

BR, mats

[lejonklou]

It will be interesting to see if there is any difference in the performance between KDS/2 with DVC switched on (@80)/off into a pre-amp. In theory there shouldn't be any difference - but we'll see!

When I have switched the DVC off there's been a clear improvement in sound quality. I have assumed this is due to the dithering being switched off.

If I understand Murrays correct that is exaktly the reason it is still in the loop even when it is switched to off in Konfig - "The volume control module contains the dithering algorithm that matches the internal 35-bit word width to the 24-bits requred by the DAC chip". The DS would not work correctly without the DVC module in the loop. I guess something else must be going on, for you to experience a clear improvement in sound qualtiy.

BR, mats

The dithering should of course be switched off when no DVC calculations are performed. I take for granted this is the case. That it passes the module is another matter.

[Spannko]

Is dithering used when converting from 35 bits to 24 bits?

[lejonklou]

Is dithering used when converting from 35 bits to 24 bits?

I can't answer for Linn's DS players, but if you use the full signal and don't manipulate it (no DVC or Space), the least significant bits will all be zero and there is no need for dithering.

[matss]

The dithering should of course be switched off when no DVC calculations are performed. I take for granted this is the case. That it passes the module is another matter.

That is not how I interpret what Murrays from Linn is writing:

Murrays,

Does the signal pass through the DVC even when the DS's volume control is switched off in Konfig, or does the signal bye-pass the DVC?

Thanks

Yes. Even when the volume control is switched off in Konfig, the volume control module remains in the signal path (equiv. to volume on and set to 80). The volume control module contains the dithering algorithm that matches the internal 35-bit word width to the 24-bits required by the DAC chip.

Is dithering used when converting from 35 bits to 24 bits?

I can't answer for Linn's DS players, but if you use the full signal and don't manipulate it (no DVC or Space), the least significant bits will all be zero and there is no need for dithering.

I'm not the digital expert, so maybe I best shut up before I end up in too deep water. But as far as I understand what Linn is trying to show and explain in their white paper, as well as what ESS does in theirs - yes, dithering is used to get rid of quantization noise/distortion in transforming 35 bit information into the required DAC format of 24 bit. I also believe that the digital process going on in the DS is a little bit more complicated than just deleting the unnecessary zeros in the 35 bit word length to reach optimal performance. But, like I said, I'm not the designer.

BR, mats

[lejonklou]

What I'm saying is that adding dither when no recalculation of the digital signal is being performed will degrade performance. Therefore it appears unlikely that they would leave it on when the digital volume control is switched off.

[per]

In my understanding, all signals in a DSM is converted to 35 bit information. Incoming material can be 16 or 24 bit. The conversion to 35 bit 384 kHz always involves calculations. Later on, the 35 bit signal is converted to 24 bit, 192 kHz to match the DAC. This step always involves calculations and diethering.

In case you use the volume control feture, an additional step of multiplication is applied in the 35 bit area of signal processing. Turning the volume control "off" does not disable the 35 bit to 24 bit calculation that uses diethering as a step.

[lejonklou]

Thanks Per! I didn't know 35 bits was used other than for the DVC step. I thought all signals were upsampled to 24 bit and left at that word length until the DAC.

My experience in this field has so far indicated that its beneficial keeping digital audio signals in their original format. Every reclocking, resampling and dithering is problematic. Even lossless manipulations affect sound quality.

[matss]

In my understanding, all signals in a DSM is converted to 35 bit information. Incoming material can be 16 or 24 bit. The conversion to 35 bit 384 kHz always involves calculations. Later on, the 35 bit signal is converted to 24 bit, 192 kHz to match the DAC. This step always involves calculations and dithering.

In case you use the volume control feature, an additional step of multiplication is applied in the 35 bit area of signal processing. Turning the volume control "off" does not disable the 35 bit to 24 bit calculation that uses dithering as a step.

This is all displayed and explained in the white paper I was referring to, and further elaborated on in the adjacent webinar held by Linn. The paper also hints at some of the techniques Linn say they use to minimise possible negative impact of reclocking, resampling, dithering and other digital signal processing. I can fully understand all this affects sound quality if not done properly.

BR, mats

[lejonklou]

I can fully understand all this affects sound quality if not done properly.

Even if done "properly", all of that still affects sound quality.

[matss]

Even if done "properly", all of that still affects sound quality.

I meant properly, not "properly". ;-)

I fully understand most of us don't fully understand everything electrical, digital, mechanical, magical that affects sound quality through our devices. I hope we all remain curious to learn and find out more on why things work, and not let our prejudice block us in assumptions of what should work. We only know what we know, not what we don't know. I try to remind myself of this as often as possible.

BR, mats

[Spannko]

In response to my question about what happens inside the Volume Control Module when it is switched on/off and the volume is adjusted, murrays posted:

"The only thing that changes is the volume setting (from 80 to whatever the start-up volume is). Dithering is always on and the algorithm does not change with volume setting."

[matss]

In response to my question about what happens inside the Volume Control Module when it is switched on/off and the volume is adjusted, murrays posted:

"The only thing that changes is the volume setting (from 80 to whatever the start-up volume is). Dithering is always on and the algorithm does not change with volume setting."

Great. Thanks for clarifying this from Linn. Now we know.

So, what else is going on to explain the clear improvement in sound quality some of us experience when the DS volume control is set to off in Konfig, besides seeing the switch toggle on the PC screen?