CHANGSTAR: Audiophile Headphone Reviews and Early 90s Style BBS

  • December 31, 2015, 02:13:16 PM
  • Welcome, Guest
Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 ... 5 6 7 8 9 [10] 11 12

Author Topic: Mytek Stereo 192-DSD DAC Review - Impressions  (Read 30585 times)

0 Members and 1 Guest are viewing this topic.

Hroðulf

  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +262/-1
  • Offline Offline
  • Posts: 538
  • MOT: Sonarworks
Re: Mytek Stereo192-DSD DAC "Review" and Un-Appreciation Thread
« Reply #90 on: March 31, 2013, 09:42:23 PM »
I don't know if this should be in some different thread in the DIY section like "The hurdles of correctly implementing the ES9018", but nonetheless I'll share some info I've stumbled upon on certain Russian forums.

Now there is this one electronics designer Dimitri Adronnikov whom I deeply respect as one of the very few members of the DIY community who does truly high-end audio designs. And by high-end I mean complex cost-no-object circuits that have their performance limited by the parts used. Unfortunately I lack the technical expertise required to translate some of his papers but I'd like to share some ideas from his latest ES9018 based design. Basically all of these thesis are his considerations on why it is so goddamn hard to make the ES9018 work well.

1. The ES9018 IC has tremendous potential, however it is evident that it requires a certain method to unlock it.
2. Many of the companies that have tackled the task of employing the ES9018 have done so with wildly varying results.
3. Current DIY designs that are based on this IC are barely scratching the surface of what is possible with the said DAC chip. He mentions the Buffalo circuit as one of the simpler implementations that utilizes only the most basic functions of the ES9018. The attempts of the Chinese "peasants" are not even worth analyzing as he hasn't heard one that didn't sound all kinds of terrible.
4. The ESS Technologies information policy doesn't help much either, to acquire the datasheet one must present credentials of himself as a manufacturer. Even more- during the process of working with the ES9018 with the datasheet provided Mr. Andronnikov has noticed some downright misleading info given by the ESS Tech people. The points he mentions as misleading or erroneous are info regarding ES9018 IC command registers and info about what clock frequencies should be used at certain sample frequencies. I have noticed that the last point corroborates with the findings of Ian and Qusp at Diyaudio. Andronnikov states that the ES9018 is unable to work with 192Khz sample rate when paired with manufacturer recommended 75.115Mhz master clock frequency. This problem was partially alleviated with upping the clock frequency to 80Mhz, however this presented another problem- decoding of incoming data was very unstable with frequent peaks in distortion. Getting the DAC to work with the said sample rate was only possible by master clock frequencies higher than 85Mhz.
5. Only by working in current output mode can one hope to attain high performance with the ES9018. Andronnikov dismisses voltage output mode as being fit only for simple test purposes.
6. The ES9018 requires a special IV stage because in stereo mode by summing up 4 channels one gets 16mA of current. As I said- my technical prowess isn't nearly high enough to understand what this means to a circuit designer however it is evident that this fact has given Andronnikov some trouble along the way of designing the final circuit.
7. And lastly- it shouldn't be a surprise to many, the ES9018 requires extremely clean power to work at its best.

That's all! I hope you are able to decipher something of use to you regarding the Mytek incident and other implementations of the ES9018. If not- feel free to delete or move this post.

P.S. Here's the link to the paper I'm referring- http://www.lynxaudio.narod.ru/articles/Lynx_D48.pdf Even if you don't understand Russian the circuit may be interesting to someone.
Logged
If it keeps on ringin', levee's goin' to break..

MoNelly

  • Powder Monkey
  • *
  • Brownie Points: +3/-0
  • Offline Offline
  • Posts: 13
  • hipster doofus wannabe
Re: Mytek Stereo192-DSD DAC "Review" and Un-Appreciation Thread
« Reply #91 on: March 31, 2013, 11:36:44 PM »
Quote from: RudeWolf on March 31, 2013, 09:42:23 PM
I don't know if this should be in some different thread in the DIY section like "The hurdles of correctly implementing the ES9018", but nonetheless I'll share some info I've stumbled upon on certain Russian forums.


I've inquired before, so sorry for the repeat: Any opinions as to how well Oppo implemented the 9018 in the BDP 105? It's their second go around with that chip (also in the BDP 95), and they've been using ESS chips for even longer than that (BDP 83SE used 9016).

I'm on the fence about this player. I have a small SACD collection, but the 105's USB DAC capability is what makes it pretty tempting. My current DAC is the Centrance DACmini. Lots of fanboy raves for the 105 on AVS Forum, and a little on HF. I believe there was one at the most recent Bay Area meet (which I sadly missed). Any knowledgeable opinions here?
Logged

Marvey

  • The Man For His Time And Place
  • Master
  • Pirate
  • *****
  • Brownie Points: +555/-33
  • Offline Offline
  • Posts: 6698
  • Captain Plankton and MOT: Eddie Current
Re: Mytek Stereo192-DSD DAC "Review" and Un-Appreciation Thread
« Reply #92 on: April 01, 2013, 01:10:32 AM »
Quote from: RudeWolf on March 31, 2013, 09:42:23 PM
I don't know if this should be in some different thread in the DIY section like "The hurdles of correctly implementing the ES9018", but nonetheless I'll share some info I've stumbled upon on certain Russian forums...



It's quite obvious Mr. Bond...
« Last Edit: April 01, 2013, 08:36:31 PM by purrin »
Logged

omegakitty

  • Guest
Re: Mytek Stereo192-DSD DAC "Review" and Un-Appreciation Thread
« Reply #93 on: April 01, 2013, 08:18:16 PM »
Quote from: RudeWolf on March 31, 2013, 09:42:23 PM
I don't know if this should be in some different thread in the DIY section like "The hurdles of correctly implementing the ES9018", but nonetheless I'll share some info I've stumbled upon on certain Russian forums.

Now there is this one electronics designer Dimitri Adronnikov whom I deeply respect as one of the very few members of the DIY community who does truly high-end audio designs. And by high-end I mean complex cost-no-object circuits that have their performance limited by the parts used. Unfortunately I lack the technical expertise required to translate some of his papers but I'd like to share some ideas from his latest ES9018 based design. Basically all of these thesis are his considerations on why it is so goddamn hard to make the ES9018 work well.

1. The ES9018 IC has tremendous potential, however it is evident that it requires a certain method to unlock it.
2. Many of the companies that have tackled the task of employing the ES9018 have done so with wildly varying results.
3. Current DIY designs that are based on this IC are barely scratching the surface of what is possible with the said DAC chip. He mentions the Buffalo circuit as one of the simpler implementations that utilizes only the most basic functions of the ES9018. The attempts of the Chinese "peasants" are not even worth analyzing as he hasn't heard one that didn't sound all kinds of terrible.
4. The ESS Technologies information policy doesn't help much either, to acquire the datasheet one must present credentials of himself as a manufacturer. Even more- during the process of working with the ES9018 with the datasheet provided Mr. Andronnikov has noticed some downright misleading info given by the ESS Tech people. The points he mentions as misleading or erroneous are info regarding ES9018 IC command registers and info about what clock frequencies should be used at certain sample frequencies. I have noticed that the last point corroborates with the findings of Ian and Qusp at Diyaudio. Andronnikov states that the ES9018 is unable to work with 192Khz sample rate when paired with manufacturer recommended 75.115Mhz master clock frequency. This problem was partially alleviated with upping the clock frequency to 80Mhz, however this presented another problem- decoding of incoming data was very unstable with frequent peaks in distortion. Getting the DAC to work with the said sample rate was only possible by master clock frequencies higher than 85Mhz.
5. Only by working in current output mode can one hope to attain high performance with the ES9018. Andronnikov dismisses voltage output mode as being fit only for simple test purposes.
6. The ES9018 requires a special IV stage because in stereo mode by summing up 4 channels one gets 16mA of current. As I said- my technical prowess isn't nearly high enough to understand what this means to a circuit designer however it is evident that this fact has given Andronnikov some trouble along the way of designing the final circuit.
7. And lastly- it shouldn't be a surprise to many, the ES9018 requires extremely clean power to work at its best.

That's all! I hope you are able to decipher something of use to you regarding the Mytek incident and other implementations of the ES9018. If not- feel free to delete or move this post.

P.S. Here's the link to the paper I'm referring- http://www.lynxaudio.narod.ru/articles/Lynx_D48.pdf Even if you don't understand Russian the circuit may be interesting to someone.

I have an engi neering friend that thinks the ES9018 might be the most difficult DAC to properly implement. Certainly nothing DIY is even coming close; though I don't think he knows of the stuff the Eastern Europeans are working on. At the very least companies will need university level measurement instruments to get the most out of them.
Logged

Boda

  • Guest
Re: Mytek Stereo192-DSD DAC "Review" and Un-Appreciation Thread
« Reply #94 on: April 02, 2013, 12:47:34 PM »
Quote from: Questhate on March 30, 2013, 01:16:35 AM
Just thought I should pop in here since I got to spend the afternoon with the Mytek today.

The firmware on the unit that arrived was version 1.7.1, which is the most current firmware according to the Mytek site so rest assured Purrin, Anax and LFF all heard the latest version.

I downloaded the latest drivers from the Mytek site as well and let it warm up for a couple of hours.

Honestly, I was a bit relieved at first because the treble wasn't as bad as I thought. Keep in mind that my sensitivity for treble nasties is much lower given all of my years of Grado and Ultrasone damage. But as I turned the volume to analytical levels and as listened more, the glare became much more apparent. There was extra hardness in the treble, and vocals had a brittle and grainy quality to them. And it's ever-present too as music plays -- sorta like a rock skipping upon water. It's not quite as obvious as LFF's initial reaction leads you to believe, but keep in mind that LFF is a highly trained acoustic engineer. I can see people overlooking the treble considering how many people are happily living with their Objective 2's and Benchmarks.

I had more issue with the bass of the unit. The timbre of the bass on a lot of my test tracks was odd. Hard to put my finger on an exact description, but the timbre was off enough to be jarring to me. It sounds very synthetic. I'd also agree with criticisms of the Mytek's lean-ness (err, I think "limp dick" is the technical term). I'm used to the explosiveness and dynamics of the Gungnir. I found myself turning the volume up to get the same impact, but that exacerbates the digititus in the treble.

I think ultimately I'm much more forgiving of the Mytek than the other guys here, but those issues were enough to give me no regrets about selling it especially considering a prefer the Gungnir by a good amount. There's just too many choices in this price range.

I have a question kind sir. Did you happen to use any DSD? And what headphones and amp did you use? Also, Purrin has implied that firewire from Apple may sound better than USB2 from Windows? Did you use Windows / USB2? Thanks.
Logged

Questhate

  • Stops to get gas, buys some stax.
  • Mate
  • Pirate
  • ****
  • Brownie Points: +83/-1
  • Offline Offline
  • Posts: 916
  • Banned for putting ice cubes in fine Scotch
Re: Mytek Stereo192-DSD DAC "Review" and Un-Appreciation Thread
« Reply #95 on: April 02, 2013, 04:11:37 PM »
Quote from: Boda on April 02, 2013, 12:47:34 PM

I have a question kind sir. Did you happen to use any DSD? And what headphones and amp did you use? Also, Purrin has implied that firewire from Apple may sound better than USB2 from Windows? Did you use Windows / USB2? Thanks.

Oops, sorry I was in a rush typing this before heading out Friday night so I wasn't very thorough.

Headphones used were HD800 and M50. I briefly tried the UERM, but the outputs of the Mytek were so hot that there was too much gain on the amps for them (even with my lowered-gain Leckerton). I had to use an impedance adapter to get decent usable volume range with them, which skewed the FR so I opted not to use them much.

As far as amps, I ran to the Super 7 mostly, but also ran the DAC to a Leckerton UHA-6Smk2 and I also tried the headphone output on the Mytek (which was pretty crap).

I only tried the USB2 input since I don't have any Apple products to even pair it with. It may very well be the Firewire is the better input.

I did NOT get to try any DSD (which I know is a big appeal of this unit). I ran a couple of Hi-Res files, but mostly used my FLAC test tracks.
Logged

Boda

  • Guest
Re: Mytek Stereo 192-DSD DAC Review - Impressions
« Reply #96 on: April 02, 2013, 06:15:00 PM »
Questhate, thank you for answering my question.

Purrin, I read your notes in the spreadsheet. Did you get a chance to use DSD on the Mytek?
Logged

Marvey

  • The Man For His Time And Place
  • Master
  • Pirate
  • *****
  • Brownie Points: +555/-33
  • Offline Offline
  • Posts: 6698
  • Captain Plankton and MOT: Eddie Current
Re: Mytek Stereo 192-DSD DAC Review - Impressions
« Reply #97 on: April 02, 2013, 06:53:11 PM »
Yes I have.

LOL, thank you for reading the important notes in the spreadsheet. I can't tell you how many people have asked me questions pertaining to the circumstances of the review but did not read the fine details.

To be honest, I only tried two DSD tracks from two albums. I didn't think they sounded any different from PCM (in terms of the treble issues). I also tried various tricks I've used in the past to smooth things out such as up-sampling PCM in JRiver to move the digital filter up way out of the audio band.

Quite frankly, I don't buy into the DSD dedicated DAC thing for computer audio - at least not yet. A native DSD DAC is nice, a DSD DAC can be a fun DIY project, but ultimately DSD can now be converted to hires PCM in JRiver or in Foobar? (Besides, 98% of my music is in PCM.) As I've said time and time again, I get SACDs only because of the availability of differently mastered recordings. Typically, SACD masters tend to be mastered or sometimes re-mixed better than their CD counterparts. (Then again, the MJ Thriller SACD was pretty badly messed up.)

Are there differences between PCM or DSD as they are handled by the chip and DAC? Yes, probably, but not significant enough to matter. I really don't masturbate about the fine details between 24/96 to 24/192 to DSD to DXD or whatever. The right mastering job of a recording will affect sound quality 100x more than the difference between 16/44 and DSD and 24/192.

God damn Sony. I sincerely hope they go bankrupt (pissed at them right now because the PS3 update bricked the box and Sony in it's infinite wisdom won't let you roll-back - I gotta pat $150 to get it fixed.)

This thought is continued here:
http://www.changstar.com/index.php/topic,924.msg22701.html#msg22701
« Last Edit: April 05, 2013, 05:40:44 PM by purrin »
Logged

sheya

  • Powder Monkey
  • *
  • Brownie Points: +3/-0
  • Offline Offline
  • Posts: 17
Re: Mytek Stereo 192-DSD DAC Review - Impressions
« Reply #98 on: April 02, 2013, 07:36:35 PM »
I saw this in a thread on audioasylum.  Thought it might be interesting to those who are using the Mytek.  Not sure if it is part of what was heard in the review, but perhaps worth considering.

Audio Asylum Thread on Digital Volume Control

http://www.audioasylum.com/cgi/vt.mpl?f=digital&m=167596

Tony Lauck:

"What I did was to adjust the internal jumpers (as per the Mytek manual). This had the effect of cutting the resistance of the IV converter in half, thereby halving the output voltage from the SABRE chip, which operates in current summing mode. Before I did this my average attenuation with the Mytek volume control had been about -24 dB and afterwards, about -18 dB. (All the needed resistors were on the circuit board.) There are a few recordings that require additional gain, typically live to DSD recordings, where the levels had been set low to prevent overmodulation and the original bits were distributed. There are some cases where I crank the gain up to -10 dB, but that's about it. There is enough headroom to run the system flat out on most recordings, because I am listening in nearfield with over a 1000 watts of power. Perhaps I could reduce the DAC output another 3 dB, but that's about as far as I would want to go, so as to not run out of gain should I encounter an even quieter recording.

The sideband peaks from jitter are turned down with the volume, since the effect of jitter is to create an error that is the product of the timing error times the rate of change of the signal. The same argument applies to dither noise on the input signal, which is also turned down.

What is not turned down is any dither noise injected by the DAC in the actual conversion process, since this happens after the volume reduction. Any analog noise also remains unchanged. "
Logged

Torpedo

  • Guest
Re: Mytek Stereo 192-DSD DAC Review - Impressions
« Reply #99 on: April 03, 2013, 05:26:22 PM »
I may try the use of the -6dB jumpers and report back. One thing that caught my attention is the DAC output levels being way higher than the usual 2V, for the compared loudness in this same system. TBH I haven't measured it.
Logged
Pages: 1 ... 5 6 7 8 9 [10] 11 12
« previous next »