T+a dac 8 dsd

[EuroDriver]

Makes sense to me, and I would add 'high frequency' before 'analogue signal'. It would explain how people are getting great results with chip-less implementations too.

 

As I've often said, DSD really is analogue barring the zero coding which appears odd, and needs simply be output on a pin for the 'D/A' to occur.

 

Lothar is doing a lot more than simply outputting to an pin :-)

 

I need to speak to him next Friday about how much he is willing to disclose publicly (if at all)

 

In private with Jussi, I am sure they will be exchanging to lot of information as Jussi is never going to be a competitor

 

Turns out Lothar has been experimenting with one bit conversion for about 25 years, and has formulated his elegant solutions to the key issues

[orgel]

I hope they try and make a Linux driver; there are lots of linux playback devices around now, and a shame for their owners not to be able to use the DACs.

 

Could not agree more. If this DAC were Linux compatible and a personal audition were to yield a similar reaction to what @Eurodriver et al. have been describing, this DAC would be just what I've been looking for.

 

--David

[YashN]

Lothar is doing a lot more than simply outputting to an pin :-)

 

Well if you mean using a low-pass filter after that, yes, it's a necessity.

 

I'd keep this as simple as possible but no simpler here personally.

 

So far, all my dabbling in SS circuits, and especially the new SET Tube amp for audio purposes provide distinct auditory advantages the less transistors and the less opamps and the less complex circuitry is there.

 

Seeing the reports of excellent sound in their implementation, I have no doubt they already know this and hence use a discrete implementation which is optimised much more than just taking several chips and stringing them along.

[EuroDriver]

Well if you mean using a low-pass filter after that, yes, it's a necessity.

 

I'd keep this as simple as possible but no simpler here personally.

 

So far, all my dabbling in SS circuits, and especially the new SET Tube amp for audio purposes provide distinct auditory advantages the less transistors and the less opamps and the less complex circuitry is there.

 

Seeing the reports of excellent sound in their implementation, I have no doubt they already know this and hence use a discrete implementation which is optimised much more than just taking several chips and stringing them along.

 

Good speculation on your part, part of the dam clever implementation has to do with a reduced parts count !

[YashN]

Good speculation on your part, part of the dam clever implementation has to do with a reduced parts count !

 

Yay to that!

[EuroDriver]

I found a review of the PDP-3000HV on a German website and I have edited the Google translation. It tells part of the story

 

T + A built the 1-bit converter itself - and not by specifying an externally manufactured converter chip. Instead, the resourceful Westphalian company created a discrete, series-connected shift register (flip-flop circuits or 1-bit memory), with each register location having its output current (binary value) weighted by resistors and then summed.

This sounds and is also technically considered relatively simple, but implementation in practice is an enormous challenge. This primarly is due to increasing jitter problems at the higher sampling rates, which requires an extremely accurate interconnection routing. In an integrated chip, this would be something hardly feasible. In addition, the 1-bit converter of T + A processes extremely high clock frequencies up to about 23 megahertz, which corresponds to 8-fold DSD with amazing 512 times oversampling

 

So the question is why are the current values of each register location being weighted by resistors and then being summed ?

 

:-)

[juanitox]

if it internally oversampling to DSD512 why many of you use HQplayer to do that task ? is there a big benefit ?

[YashN]

So the question is why are the current values of each register location being weighted by resistors and then being summed ?

 

:-)

 

Because it's one of the simplest ways of doing the output.

 

I was doing a similar thing with PWM output on the Arduino to get finer resolutions at the analogue output than a single pin would provide.

[YashN]

if it internally oversampling to DSD512 why many of you use HQplayer to do that task ? is there a big benefit ?

 

The algorithms near the DAC are resource-constrained and can affect the actual D/A if the DAC is not thoroughly isolated - a hard thing to do as I understand it.

 

On the server computer, you can run much more sophisticated algorithms/filters/modulators, and then isolate the DAC from the computer noise.

 

Hence, the processed stream is then sent to the DAC for just the D/A part.

[hifial]

if it internally oversampling to DSD512 why many of you use HQplayer to do that task ? is there a big benefit ?

 

As far as I know it is not up-sampling.

 

It can process up to 8 x single rate DSD (64) aka DSD512. So I believe it is referring to that it can handle DSD512. It takes what ever DSD rate you give it and processes it at it rate. So DSD64 is at DSD64 and DSD128 is at DSD128 and so on. EuroDriver can confirm for me.

 

However, it is best to NOT feed it DSD64 without up-sampling to get the best out of that rate.

You can feed it DSD64 straight but it is best to up-sample DSD64 to at least DSDF128, or as high as your DAC and Player can handle (up to DSD512).

 

All other DSD rates will do better, (not just because they are higher DSD) but they also will benefit greatly by using HQ Player up-sampling all to DSD512 into the DAC 8 DSD.

 

But let me say that most of the time you can benefit greatly by feeding a DAC the highest capable rate it can handle instead of allowing the DAC to use its internal up-sampling. Assuming that you use a quality up-sampler like HQ Player.

[juanitox]

The algorithms near the DAC are resource-constrained and can affect the actual D/A if the DAC is not thoroughly isolated - a hard thing to do as I understand it.

 

On the server computer, you can run much more sophisticated algorithms/filters/modulators, and then isolate the DAC from the computer noise.

 

Hence, the processed stream is then sent to the DAC for just the D/A part.

 

 

the bad thing is that we need a special turbo charged computer to fo that , exit my mac mini ..

[hifial]

the bad thing is that we need a special turbo charged computer to fo that , exit my mac mini ..

 

 

Well, you do not need a "special turbo charged" computer to do it. Yes, one needs a computer that has the processing power that is needed. But as was said in an earlier post on this thread, one only needs a bare bones PC with the right Processor. Cost is approx $700-800. Not expensive. Yes, not cheap. But my Mac Mini was way over that in 2012-2013 dollars. If some one wants better sound quality they could spend more on the PC build. Just like anything else in audio.

 

What some of us heard at lmitche house with the DAC 8 DSD was a bare bones PC with the right Processor.

 

Hope that helps.

[firedog]

Wrote T+A and asked if there was any possibility of a Linux driver for DSD 256 and 512. Answer: no.

[YashN]

Wrote T+A and asked if there was any possibility of a Linux driver for DSD 256 and 512. Answer: no.

 

One will have to count on the community to deliver this.

 

Do they use the XMOS interface?

[hifial]

Wrote T+A and asked if there was any possibility of a Linux driver for DSD 256 and 512. Answer: no.

 

Yea, I got the same response a few days ago. But if enough of us ask/bug them maybe they will give us a present one day.

 

I know some people are trying to explain to them why it is worth their time and effort. Hopefully some day soon. Then again...

[juanitox]

Wrote T+A and asked if there was any possibility of a Linux driver for DSD 256 and 512. Answer: no.

to bad it will not work in native DSD with the microrendu too..

[Miska]

Do they use the XMOS interface?

 

Someone said it is Amanero...

[kravi4ka]

It is indeed using the Amanero with one of the older stable firmware versions. I asked them too if it is possible to update it but got no straight answer, hopefully that can be done in the future, it would be magnificent to use it at DSD512 with the microRendu...

[lmitche]

It is indeed using the Amanero with one of the older stable firmware versions. I asked them too if it is possible to update it but got no straight answer, hopefully that can be done in the future, it would be magnificent to use it at DSD512 with the microRendu...

Agreed, I would like to see DSD512 with Linux as well.

[YashN]

Someone said it is Amanero...

 

There is some activity this year on the Github pages to adapt things to the new firmware versions, so I would say there is hope:

 

[30-mar-16]

Updated patch for Amanero Combo384. Requires at least firmware 1099rc2

 

[04-mar-16]

Added experimental support for the Amanero Combo384. Firmware 1099rc1 is needed (see Amanero site). [update: waiting for firmware update, rc1 does not work correctly]

[YashN]

It is indeed using the Amanero with one of the older stable firmware versions.

 

As an end-user/owner of the DAC, can one update the firmware?

[EuroDriver]

It is indeed using the Amanero with one of the older stable firmware versions. I asked them too if it is possible to update it but got no straight answer, hopefully that can be done in the future, it would be magnificent to use it at DSD512 with the microRendu...

 

There is a reason why "one of the older stable firmware versions" is being used :-)

 

A note to the Linux fans, Lothar has a very deep and extensive knowledge of Linux. I don't think there is anything anyone can teach Lothar about Linux.

[kravi4ka]

Indeed, it is not just the firmware, the whole way that the amanero solution is implemented is different from what you can buy as DIY, I have a pretty good knowledge about DACs, at least theoretically and that little device is an engineering marvel. Cannot wait to hear it properly.

 

Let's hope that native DSD512 comes to the DAC 8 though, would be nice

[Miska]

There is a reason why "one of the older stable firmware versions" is being used :-)

 

I think the spec Amanero released contains needed info also how to make the older firmware work on Linux. I was planning to try it out at some point when I have time. Doesn't take much time or effort...

 

For me personally, Linux support is very important because Linux is my primary development platform.

[Miska]

There is some activity this year on the Github pages to adapt things to the new firmware versions, so I would say there is hope:

 

I flashed one of my Amanero boards with the 1099rc2 firmware and swapped it on my DSC1. I will try it out hopefully tomorrow to see if/how it works.