Gustard DAC-A26 - AK4499EX DAC :: Impressions Thread
Apr 22, 2024 at 6:22 PM Post #977 of 1,166
I thought the whole idea was to make computer audio to sound or get close to analogue sound. With oversampling and upscaling you are doing the opposite - getting away from the analogue sound.
If the goal is accurate reconstruction, then there are theories that tell us how to achieve this. Of course there are debates on how best to get there, but that is ultimately the goal of some (but not all) of the upscaling algorithms available in commercial products. To my ears, PGGB does the best job of this and is vastly superior to native playback on my Gustard x26pro.

The following is from the PGGB website and it explains why it upscaling can be beneficial:
1713824254515.jpeg
 
Apr 22, 2024 at 6:37 PM Post #978 of 1,166
You appear to be trying to draw conclusions from this that the author didn’t intend. He was advocating for DSD to be the format for the music we purchase and download. “This makes DSD a serious contender in the choices of high resolution audio formats” is a quote from this.

He wasn’t advocating for converting PCM to DSD, which is what you’ve been advocating for. He was advocating avoiding conversion entirely: “As science progressed as well as our experience with digital audio, we started to realize that the algorithms for the DSD-to-PCM and PCM-to-DSD conversions can have a profound impact on the sonic performance when they are developed according to classic formulas… Hence the motivation by the engineering teams of Sony and Philips to remove these steps altogether from the conversions between analog and digital.” That’s what the illustration you posted was trying to convey. It shows that this is only achieved if conversion steps are actually skipped. Thus it only applies to recordings that were captured in DSD with no conversion to PCM at any point. That represents a very small number of recordings.
 
Apr 22, 2024 at 7:13 PM Post #979 of 1,166
You appear to be trying to draw conclusions from this that the author didn’t intend. He was advocating for DSD to be the format for the music we purchase and download. “This makes DSD a serious contender in the choices of high resolution audio formats” is a quote from this.

He wasn’t advocating for converting PCM to DSD, which is what you’ve been advocating for. He was advocating avoiding conversion entirely: “As science progressed as well as our experience with digital audio, we started to realize that the algorithms for the DSD-to-PCM and PCM-to-DSD conversions can have a profound impact on the sonic performance when they are developed according to classic formulas… Hence the motivation by the engineering teams of Sony and Philips to remove these steps altogether from the conversions between analog and digital.” That’s what the illustration you posted was trying to convey. It shows that this is only achieved if conversion steps are actually skipped. Thus it only applies to recordings that were captured in DSD with no conversion to PCM at any point. That represents a very small number of recordings.
Not in the least.
I posted that diagram to show that the very first format analog is converted to is dsd. If the data is then converted to PCM we can then at least convert back to DSD outside of the dac and then avoid all that processing that the internal dac must do to convert the pcm signal to dsd (or a form of it).
We can apply way better modulators and noise shapers than the dac can as well.
I do agree that native dsd recording not converted to pcm in the first place, kept in dsd all the way to the analog reconstruction filter will sound the closest to the original recording.
Just like the diagram also illustrates.
 
Apr 22, 2024 at 7:21 PM Post #980 of 1,166
You appear to be trying to draw conclusions from this that the author didn’t intend. He was advocating for DSD to be the format for the music we purchase and download. “This makes DSD a serious contender in the choices of high resolution audio formats” is a quote from this.

He wasn’t advocating for converting PCM to DSD, which is what you’ve been advocating for. He was advocating avoiding conversion entirely: “As science progressed as well as our experience with digital audio, we started to realize that the algorithms for the DSD-to-PCM and PCM-to-DSD conversions can have a profound impact on the sonic performance when they are developed according to classic formulas… Hence the motivation by the engineering teams of Sony and Philips to remove these steps altogether from the conversions between analog and digital.” That’s what the illustration you posted was trying to convey. It shows that this is only achieved if conversion steps are actually skipped. Thus it only applies to recordings that were captured in DSD with no conversion to PCM at any point. That represents a very small number of recordings.
Unfortunately with the Ess chip in your x26
you cannot bypass any of the internal oversampling and modulation.
All of your PGGB is well… being undone. You need an actual NOS dac to keep the effects of PGGB.
With the A26 AKM you can see how dsd has a path that can bypass all of the internal processes and go right to analog reconstruction.
 

Attachments

  • IMG_9286.jpeg
    IMG_9286.jpeg
    45 KB · Views: 0
  • IMG_9285.png
    IMG_9285.png
    115.6 KB · Views: 0
Apr 22, 2024 at 7:52 PM Post #981 of 1,166
Unfortunately with the Ess chip in your x26
you cannot bypass any of the internal oversampling and modulation.
All of your PGGB is well… being undone.
Yes there is still processing, but the benefits from PGGB can still be clearly heard. It’s best not to comment on what you haven’t heard for yourself.
 
Apr 22, 2024 at 7:56 PM Post #982 of 1,166
Yes there is still processing, but the benefits from PGGB can still be clearly heard. It’s best not to comment on what you haven’t heard for yourself.
How do you know what I have heard and what I haven’t?
Explain how the benefits can be heard when they are being undone by your dac. X26 Pro is ESS based, and is never bit-perfect.
 
Apr 22, 2024 at 7:59 PM Post #983 of 1,166
I posted that diagram to show that the very first format analog is converted to is dsd. If the data is then converted to PCM we can then at least convert back to DSD outside of the dac and then avoid all that processing that the internal dac must do to convert the pcm signal to dsd (or a form of it).
We can apply way better modulators and noise shapers than the dac can as well.
Nice word salad.

What you actually wrote was:
Please look at the graph again.
What do you think is closer to analog sound?
That illustration doesn’t prove that point in the way you were suggesting. If you can’t see this, then I would encourage you to read the article he wrote that accompanies that illustration again.
 
Apr 22, 2024 at 8:05 PM Post #984 of 1,166
Nice word salad.

What you actually wrote was:

That illustration doesn’t prove that point in the way you were suggesting. If you can’t see this, then I would encourage you to read the article he wrote that accompanies that illustration again.
Word salad?
Look at first part of the diagram.
Analog is first turned to DSD in the ADC
So the closest format to analog is DSD
 

Attachments

  • IMG_9288.jpeg
    IMG_9288.jpeg
    90.5 KB · Views: 0
Apr 22, 2024 at 8:21 PM Post #986 of 1,166
How do you know what I have heard and what I haven’t?
Because if you actually had heard it, the opinion you expressed would actually sound informed. You would be actually able to comment in a way that reflected reality instead of just regurgitating the false narrative that’s been fed to you.

Explain how the benefits can be heard when they are being undone by your dac. X26 Pro is ESS based, and is never bit-perfect.

Thousands of listening comparisons have confirmed that ESS isn’t doing that much harm. I am one of the beta listeners who helped provide feedback on PGGB since before it was released to the public. My observations were nearly always consistent with others who own DACs that did less processing. The only time my observations weren’t consistent with others was in terms of the magnitude of improvement with PGGB upscaling to DSD. I hear all the same kinds of improvements just not to the same degree.

The improvement from PGGB can be clearly heard with the x26pro. It’s just not true that the ESS chip will undo the benefits of PGGB; lessen, yes, but not undo. Even HQPlayer can be clearly heard to be beneficial on the x26pro. More proof of that can be found in this post. He has posted a number of times about the improvements he heard from upscaling methods, both in terms of pros and cons. If you were actually speaking from experience it would be just as obvious as it was with him.

Upscaling is a very complicated topic that very few of us understand. Unfortunately there are a few so-called experts who spew a lot of nonsense just to discredit their competitors. One has to be careful to not be misinformed by them.
 
Last edited:
Apr 22, 2024 at 8:29 PM Post #987 of 1,166
Do you even understand what is happening within your dac?
All your PCM/PGGB is turned to DSD.
You are now trying to gaslight me so I won’t be responding to you again. I accomplished what I hoped to accomplish, which was to let others see how much your opinions on upscaling are worth. You helped me accomplish that.
 
Last edited:
Apr 22, 2024 at 8:50 PM Post #988 of 1,166
Because if you actually had heard it, the opinion you expressed would actually sound informed. You would be actually able to comment in a way that reflected reality instead of just regurgitating the false narrative that’s been fed to you.
What are you even talking about? What false narrative?
Upscaling is a very complicated topic that very few of us understand.
It is complicated and I understand it quite well. It is you who don’t seem to understand it at all.
You are now trying to gaslight me
Gaslight you? What are you talking about. Im asking you if you know how your dac works. It’s very clear now that you don’t.

Let me break it down for you.
Your PGGB PCM enters your dac. It gets processed by 8x FIR. Then the same sample is copied 8 times and is fed through 3rd order IIR before being copied again 8 times and entering the modulator which produces a 1-bit DSD stream. This is converted to 32 (2^5) or 64 (2^6) bits at a time and sent to analog through those 32/64 conversion elements.
Your PGGB is gone brother.
DSD with HQplayer and your dacs chips takes a different shorter path. It goes through the 3rd order IIR without any rate conversion and then goes through the modulator to the conversion. Better yes. Ideal no. Ess chips suck for external upsampling period.
Sorry you wasted your money on PGGB using that dac. Get a NOS dac and you can actually hear its effects.
I guess that only makes me the only one of us who has actually heard PGGB. 😂
 
Last edited:
Apr 22, 2024 at 8:55 PM Post #989 of 1,166
Your PGGB PCM enters your dac. It gets processed by 8x FIR. Then the same sample is copied 8 times and is fed through 3rd order IIR before being copied again 8 times and entering the modulator which produces 1-bit DSD stream. This is converted to 32 (2^5) or 64 (2^6) bits at a time and sent to analog through those 32/64 conversion elements.
Your PGGB is gone brother.
Actually I should continue responding to you as you keep helping my cause.

Clearly you aren’t familiar with the first law of holes.
 

Users who are viewing this thread

Back
Top