DX200=Affordable High End Audio. Dual ES9028Pro dacs. AMP1, AMP3, AMP5, AMP7 & AMP8 ***Firmware support now up for AMP9***
Jul 11, 2018 at 3:53 PM Post #14,731 of 22,021
a brief PCM refresher is key first. Hope this helped. Please don't tell me you already knew all this and I just :deadhorse: : )

No, this was immensely helpful, many thanks! And thanks to @Whitigir too :)
However, it does leave me with a few questions.

1) I believe Whitigir said that CD's can be converted into DSD?
My understanding from what you and he wrote is that DSD of any type has a far higher sample rate (and hence quantity of information, even at 1 bit) than a 16/44 CD (or FLAC file).
So how can a CD be converted into a DSD file of a higher quality? Or is this also just a form of 'fill in the blanks' upscaling?

2) From which sources can one obtain music in DSD format without any upscaling?

3) I'm still a bit confused about the difference/relationship between DSF and DSD?
Perhaps it would help me if you could describe using practical examples involving playback on a DAP? As in, what format do I ultimately need it to be in so that I can just stick the file on a micro-SD card (or external HDD) and play it on the DX200?

Because that's ultimately my goal here lol :p
 
Jul 11, 2018 at 4:04 PM Post #14,732 of 22,021
It's pretty much the same thing what chord Dave is doing with Blu mk2 [16k pounds a pair I believe] (16x CD on the fly upscaling). No way though I'm gonna do that with a mobile setup and waste TBs of storage and time for that matter :p
 
Jul 11, 2018 at 4:09 PM Post #14,733 of 22,021
No, this was immensely helpful, many thanks! And thanks to @Whitigir too :)
However, it does leave me with a few questions.

1) I believe Whitigir said that CD's can be converted into DSD?
My understanding from what you and he wrote is that DSD of any type has a far higher sample rate (and hence quantity of information, even at 1 bit) than a 16/44 CD (or FLAC file).
So how can a CD be converted into a DSD file of a higher quality? Or is this also just a form of 'fill in the blanks' upscaling?

2) From which sources can one obtain music in DSD format without any upscaling?

3) I'm still a bit confused about the difference/relationship between DSF and DSD?
Perhaps it would help me if you could describe using practical examples involving playback on a DAP? As in, what format do I ultimately need it to be in so that I can just stick the file on a micro-SD card (or external HDD) and play it on the DX200?

Because that's ultimately my goal here lol :p


There is no fill in the blank. Any sigma-delta DAC will Upsample 2x or 8x, or 16x before it got digital filtering applied and then going into capacitor array to put out music. DSD64X is simply an offline conversion of this very Sigma-Delta Modulation processing at 64X, and so is DSD512 is just 512X.

This is not Upscaling conversion, it is a Processing conversion. Your DAC chip is capable of applying and processing signals on the fly but it is in real time, and so the supportive systems are very dependent on clock systems, power linearity and so on. Instead of putting all the load on your DAC processing, you simply use a computer to process it because the computer can take forever to process at the utmost precision, the final outcome is only depend on the algorithm and DSP filters processing. Again, this process is very complicated.

The majority of recordings are all in the PCM format for eases of editing purposes and storage, time processing .....so on....and then it get Processing conversion into DSD. The majority of DSD music you can find are all being done this way, but by the Studio, and whatever program, DSP and algorithms they are using.

If you do not want any type of Oversampling DAC, you need to find those with true analog filters (-6Db analog filters) and true NOS (Non Over Sampling), and I am currently owning AudioGD R2R7. I can confirm that it is awesome with just your PCM 16/44.1.....but did I mention that I had to build a PC which cost near $4K with Linear power supply and upgraded OCXO clocking system ? The DAC by itself is also not Sigma Delta but is Resistor ladder.

It is your choice after all, listen to it for yourself. See If you can tell the differences ? Then decide what is to do
 
Last edited:
Jul 11, 2018 at 4:19 PM Post #14,734 of 22,021
No, this was immensely helpful, many thanks! And thanks to @Whitigir too :)
However, it does leave me with a few questions.

1) I believe Whitigir said that CD's can be converted into DSD?
My understanding from what you and he wrote is that DSD of any type has a far higher sample rate (and hence quantity of information, even at 1 bit) than a 16/44 CD (or FLAC file).
So how can a CD be converted into a DSD file of a higher quality? Or is this also just a form of 'fill in the blanks' upscaling?

2) From which sources can one obtain music in DSD format without any upscaling?

3) I'm still a bit confused about the difference/relationship between DSF and DSD?
Perhaps it would help me if you could describe using practical examples involving playback on a DAP? As in, what format do I ultimately need it to be in so that I can just stick the file on a micro-SD card (or external HDD) and play it on the DX200?

Because that's ultimately my goal here lol :p

1) converting to DSD does not mean it’s inherently better. It means the software is adding samples between the original samples in the file to the best of it’s ability to interpret what should go in-between. You can never extract more information from the already sampled data (digital music). Depending on the up-sampler / conversion software you may or may not prefer the sound, but it definately doesn’t have more fidelity to the sound (closest to the source). In fact, I prefer to let the DAC do the work and feed it bit-perfect files (not changed from the originally sampled format).

2) There are a few sources for native DSD recordings others can recommend (I don’t really like DSD for a multitude of reasons).

3) DSF is just a form of DSD like FLAC is a form of PCM. You can search google for DSD format, or look at Wikipedia for more in depth information.
 
Jul 11, 2018 at 4:20 PM Post #14,735 of 22,021
There is no fill in the blank.

In the conversion there also must be ‘fill-in-the-blank’ interpretation happening. Most DSD converted in the studio uses high resolution PCM format for headroom to minimize losses / errors in the conversion.

You are correct that there is very little music natively recorded and distributed in DSD format as there can be no DSP on DSD recordings.
 
Last edited:
Jul 11, 2018 at 4:27 PM Post #14,736 of 22,021
In the conversion there also must be ‘fill-in-the-blank’ action happening.
Under this statement, every DAC chip will do “fill in the blank” by Applying Sigma-Delta modulations at 2-4-8-16X anyways. The differences is that it is much reliable on the clocks for real time communications where as offline processing can take it out of the equation, period.

You read a book once, then you read it the second time and realized howmuch you are missing out...are you a book reader ? This conversion is the same thing, the more time it sample, the more information it can achieve.

It is true that upsampling or more processing to not be able to create additional information, but it rather is that you are normally missing out information. Why ? Because

**** processing on the fly, a DAC need to rely on expensive clocking system, expensive power rail design, expensive low noises components ....etc....etc.....which your portable player never had, not even some desktop tier equipments have.

The more you sample the signals, the more information you can achieve, and the more noises of ultra sonic and other noises will also be achieved. This is where you need noise shaping, low-pass and high-pass filters applications. Otherwise the Ultra-Sonic will blow your brain out, or blow your tweeters. This is also where it matter the most and where different DAC brands will bring it own characteristics. For example, AKM is different house sound than ESS Sabres and so on...you get the thrills

Yes, one can argue that Studio DSD is using high quality PCM, then again...what is it ? 786Khz ? 192Khz ? It doesn’t matter, the standard verdict is out at 16-24 bits and 44.1 KHz is enough to contain all informations. The matter is that, could you have achieved all of these informations and not be missing out every little drop of it ? If every system can do it, then you can listen to your iPad and it cirrus logic DAC to end this debate. Others will be spending hundreds of thousands of dollar to do this, and then they would also demand an electric line straight from the power plant for their own Audiophile equipments LOL ....with the power lines burried underground, heavily shielded
 
Last edited:
Jul 11, 2018 at 4:47 PM Post #14,739 of 22,021
And Multibit from ex. Schiit? No guessing samples, no upsampling...
What ? Are you talking about Multibit in the name or in the format or what device are you talking about ?

Again, you are missing out information from my previous posts, a perfect example that the more sampling ‘ the more information achieved.

It is possible to stay Non-Over-Sampling, but you need a very beefy system and a good DAC to do so. Each one of these pieces will have to be heavily reliable on clocking systems, power supply designs, components (DSP interface...etc..) used and so on. I have a system , and no Schiit can not touch it LOL!

I am simply trying to squeeze out every last drop of performances the DX200 as a portable device can do, you can choose to do whatever you want :). However, there are real gains when you do DSD512 from your CD FLAC on your DX200. I believe all of you can tell the differences
CD33D73D-3060-44F6-8D82-629A8312F6A6.jpeg
 
Last edited:
Jul 11, 2018 at 5:00 PM Post #14,741 of 22,021
of the father, son and the holy Schitt

Showoff :D
It was convenient though :D I believe one day I may just sell them all and stick to Dx200 with 12Tb of DSD512 LOL!

Also another reason why I want a discrete Dx200 is because all these capacitor arrays to playback DSD512 can be....imagine the quality with juicy capacitors :D
 
Jul 11, 2018 at 5:28 PM Post #14,742 of 22,021
It was convenient though :D I believe one day I may just sell them all and stick to Dx200 with 12Tb of DSD512 LOL!
Before you do that, upgrade your AC outlet. It's the only thing I have on you LMAO....yet, no system here any more to suck from it's silver teets. I kept that, power conditioner and 1 AC cable...cuz you never know :D
...imagine the quality with juicy capacitors :D
Trying to but cannot fathom from a portable device...that's hugging random people in the street quality*

*Probably also broken face quality
 
Jul 11, 2018 at 5:59 PM Post #14,743 of 22,021
Are you talking about Multibit in the name or in the format or what device are you talking about ?

I think Tennessee was referring to the Bifrost, Gungnir and Yggdrasil multibit DACs from Schiit. If I'm wrong, ignore me, please. :)
 
Jul 11, 2018 at 6:50 PM Post #14,745 of 22,021
As long as they are using Sigma-Delta, there are over sampling and modulation somewhere. Unless they can technically do True-NOS (True Non Over Sampling)

While I'm a customer, I'm not an expert at that level of the technology for sure. I just now started messing around with XiRSC and DSD512 per your recommendation. As I understand it though, they offer both delta-sigma and multibit versions of some of their DACs. Others just multibit. They have NOS pass-thru for the higher sample rates. Beyond that, I don't know much. I will say though that the whole DSD conversation here has helped me better understand the options there. So thanks for that as well. :)
 

Users who are viewing this thread

Back
Top