[sajunky]

Comparing the Holo May KTE is not fair from price perspective. Holo MAY wins this game without doubt. Even Level 2. Holo May Level 1 would be interesting comparison though!

Not sure. Holo May is comparing well with good DS implementation like Chord Dave here, but I didn't see any trusty source comparing with top end R2R designs. It looks like it carries sound characteristic different to the true R2R DACs. What you chose it depends on what you are looking for.

 

[gk2013]

Not sure. Holo May is comparing well with good DS implementation like Chord Dave here, but I didn't see any trusty source comparing with top end R2R designs. It looks like it carries sound characteristic different to the true R2R DACs. What you chose it depends on what you are looking for.

What is not true R2R about the Holo May in your opinion ?
Denafrips is not a top end R2R? Agree, it always depends what you are looking for….. ..You haven’t heard it, that’s totally clear. If you are happy with your current dac keep it!

 

[sajunky]

What is not true R2R about the Holo May in your opinion ?
Denafrips is not a top end R2R? Agree, it always depends what you are looking for….. ..You haven’t heard it, that’s totally clear. If you are happy with your current dac keep it!

True, I didn't. It is what I get from the reviewers. Comparison with R2R DAC's is avoided.

From the design point of view it is not a true R2R ladder, but a segmented architecture. Similar to the chip-DACs used in Shiit Bifrost 2 and up. TDA1541 is an example, it could never compete against a true R2R which was PCM63 or the latest 1702/4. Segmented architecture is not so well defined for a dynamic response, but it measure better. Holo Audio makes it even worse, as for the marketting purpose they added a compensation ladder in parallel. Once again, purely for better measurements. And a Stereophile tests show that there is some bit manipulation in software which affects linearity of a sinewave output in NOS mode. Really strange machine. Here is: Fig.2 HoloAudio May, NOS mode, waveform of 1kHz sinewave at 0dBFS, 16-bit data.

It looks like a six MB bits are randomised, creating such stepping behaviour of a single sweep shot.

 

[DACLadder]

The “stair steps” on the May’s sine wave graph is in NOS mode with no anti-aliasing filter. You see absolutely no smoothing between each 44.1khz sample. ”However, without such a filter, the DAC's output consists of the unsmoothed voltages represented by each digital word. The DAC's sample-and-hold mechanism results in a "stair-step" waveform”. The Stereophile reviewer concludes though the Holo May L2 kte was the best measuring DAC ever as of last summer.
https://www.stereophile.com/content/holoaudio-may-level-3-da-processor-measurements

A better comparison with the L2 May might be the Magna tweaked R7 HE Mk2. I would certainly like to give the May a try. Free of course.

 

[gk2013]

True, I didn't. It is what I get from the reviewers. Comparison with R2R DAC's is avoided.

From the design point of view it is not a true R2R ladder, but a segmented architecture. Similar to the chip-DACs used in Shiit Bifrost 2 and up. TDA1541 is an example, it could never compete against a true R2R which was PCM63 or the latest 1702/4. Segmented architecture is not so well defined for a dynamic response, but it measure better. Holo Audio makes it even worse, as for the marketting purpose they added a compensation ladder in parallel. Once again, purely for better measurements. And a Stereophile tests show that there is some bit manipulation in software which affects linearity of a sinewave output in NOS mode. Really strange machine. Here is: Fig.2 HoloAudio May, NOS mode, waveform of 1kHz sinewave at 0dBFS, 16-bit data.

It looks like a six MB bits are randomised, creating such stepping behaviour of a single sweep shot.

Thank you for sharing your opinion.
Apparently the TDA1541 architecture can compete very well with other R2R designs I remember Audio-GD also likes the TDA-1541 sound (simulation) on their R2R dac’s. The compensation ladder has always been part of HoloAudio R2R dac’s, the first spring included. Every R2R has some sort of correction done to manipulate the output. This is not for marketing purposes. Hahahaha, good joke though, like it .

 

[Crow123]

Did you have a hard time in high school and now you come to audio threads to make yourself feel better?

Agree or disagree with the facts - even share an opinion but this school yard taunting only makes one person look bad

 

[sajunky]

The “stair steps” on the May’s sine wave graph is in NOS mode with no anti-aliasing filter. You see absolutely no smoothing between each 44.1khz sample. ”However, without such a filter, the DAC's output consists of the unsmoothed voltages represented by each digital word. The DAC's sample-and-hold mechanism results in a "stair-step" waveform”. The Stereophile reviewer concludes though the Holo May L2 kte was the best measuring DAC ever as of last summer.
https://www.stereophile.com/content/holoaudio-may-level-3-da-processor-measurements

Conclusion is wrong. Man, I know what Attkinson wrote, but it I don't agree with. First things there a lie. He measured it after a low pass filter, see the next paragraph. The same, I also disagree with explanation to a figure 1. In my opinion it is linked to the randomisation of MS bits which is shown on a figure 2.

In a conversion there is of course stepping (20-bit resolution is acceptable), but in this case there are ~32steps - far to big! When you overlay one sreenshot over the other like it is seen on the multisweep osciloscope screen, you will see that each clock period carries different values. Screenshots of the Spings made on the Russian website show that 6 MS bits are randomised using (at least) 16th discrete values. There are two sets of sreenshots, one is taken from the DAC output, like Attkinson did, the other is taken directly from the ladder output, it gives bettter clarity how many randomisation steps are involved. An example of 4kHz:

Number of discrete randomisation values is changing with a frequency, as randomisation algorithm is synchronised with a specific fixed frequency, close to Nyquist.

 

[gk2013]

Did you have a hard time in high school and now you come to audio threads to make yourself feel better?

Agree or disagree with the facts - even share an opinion but this school yard taunting only makes one person look bad

Sorry, but I found his response very childish as did mine. You could also find all sorts of things about the Audio-GD DAC, from a technical point of view and also from a measurement point of view. I expressly do not include that.
Holo MAY is an R2R DAC and measures very well. Audio-GD R7HE is also very good but does not excel in measurements. I do not care. I'm mainly concerned with how something sounds. Someone asked how the two DACs compare and I answered.

 

[ymlee]

There is a website put his scoring among different dacs he had listened, Audio-gd R7He should be same league as Denafrips Terminator (not Plus version). Anyway, just his personal opinion.

Post in thread 'Best DACs comparison test result [[My Subjective Opinion]]' https://www.whatsbestforum.com/thre...esult-my-subjective-opinion.25584/post-525489

 

[sajunky]

There is a website put his scoring among different dacs he had listened, Audio-gd R7He should be same league as Denafrips Terminator (not Plus version). Anyway, just his personal opinion.

R-7HE is upgradeable, T+ is the end-game (it is why the name). You can upgrade R-7HE with a quality external clock which is far better than an internal OCXO clock in T+. This upgrade path is important.

 

[Jackula]

R-7HE is upgradeable, T+ is the end-game (it is why the name). You can upgrade R-7HE with a quality external clock which is far better than an internal OCXO clock in T+. This upgrade path is important.

The weak link in the Denafrips line up is the Gaia IMO. I haven't tried but honestly I think the T+ will sound better with a DI-20HE. Because the T+ is obviously better than the R7HE, but R7HE /w DI-20HE is about the same as T+ /w Gaia.

 

[FredA]

The external clock makes a clear difference for sure. With the best units available out there, you can kiss the digital glare goodbye. This was my main reason to buy another audio-gd dac. I have no regret. I am sure the May is great. But with the external clock, my di20he/r7he mkii combo is just awesone, just perfectly analalog. It competes with the best out there, no doubt.

 

[mark44]

What are the differences between Schiit Yggdrasil 2 (the newest) and Audio GD R7 HE MK2? What's the best?

 

[DACLadder]

The Schiit Yggdrasil uses two 20-bit Analog Devices A/D chips in a balanced array where they claim 21-bit resolution. R7 uses discreet resistor DACs at least 24-bit in a balance config.
The Schitt does not support DSD direct.
The R7 HE has better power supplies. The R7 HE is 2/3rds power supply.
The R7s supports 10M clocking.
The R7s have ACSS outputs.

I had a Iggy 1 and used it exclusively for 6 months. It sounded thin and lean compared to the old Master 7. The Iggy 1 rendered vocals well but did not move my soul. Schiit claims you don’t need more than 20-bit resolution but the Iggy 1t was dry in its presentation (not as much echo, reverb, depth). A lot of people love the Iggy though. Hope the 2nd gen is better.

 

[mark44]

The Schiit Yggdrasil uses two 20-bit Analog Devices A/D chips in a balanced array where they claim 21-bit resolution. R7 uses discreet resistor DACs at least 24-bit in a balance config.
The Schitt does not support DSD direct.
The R7 HE has better power supplies. The R7 HE is 2/3rds power supply.
The R7s supports 10M clocking.
The R7s have ACSS outputs.

I had a Iggy 1 and used it exclusively for 6 months. It sounded thin and lean compared to the old Master 7. The Iggy 1 rendered vocals well but did not move my soul. Schiit claims you don’t need more than 20-bit resolution but the Iggy 1 was dry in its presentation (not as much echo, reverb, depth). A lot of people love the Iggy though. Hope the 2nd gen is better.

So you haven't compared the Yggdrasil 2 and the R7 HE MK2. It's correct? Who should I ask for this comparison?