• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Review and Measurements of Soekris dac1421 Multibit DAC

gvl

Major Contributor
Joined
Mar 16, 2018
Messages
3,427
Likes
3,982
Location
SoCal
I don't know which one is better, but from my experience DS always sounds a tad cleaner. R2R sounds less clean, especially in the highs, and sometimes more "emotional" overall, whatever it means, likely due to the higher harmonic content. That's the explanation I settled on, I think it correlates well with what we've seen here.
 

pwjazz

Addicted to Fun and Learning
Forum Donor
Joined
Sep 21, 2018
Messages
507
Likes
746
then afterwards its a job for the human hearing to finish the design

I find this part fascinating, it's sort of like where art meets science meets industrial engineering. Can you please talk more about your process?
  • Do you create several prototypes and then ABX test them? Or do you actively tweak parameters on a single unit while listening?
  • What parameters do you vary to come up with your test candidates and before you vary them do you have a working hypothesis on what impact they'll have on the listening test, or do you take more of a shotgun approach of randomly adjusting a bunch of parameters and then just seeing what sounds best?
  • Do you conduct your testing in the morning with fresh ears or in the evening after a hard day of work?
  • Do you have a standard music playlist that you use our do you listen to test signals?
  • Do you rely just on your own ears or do you have employees who perform testing as well?
  • Once you've dialed in a prototype that sounds good to your ears, how do you verify that production units meet the same specification? Since the subtle performance benefits of an optimized R2R design aren't measurable with an audio analyzer, do you have to perform a listening test with each production unit?
 

g29

Addicted to Fun and Learning
Joined
May 1, 2019
Messages
520
Likes
318
...

Conclusions
Up until now, I had only tested multibit DACs from Schiit. Unfortunately all seem to have glaring design flaws, especially when it came to linearity. Thankfully, there is no such problem with Soekris dac1421. It has excellent linearity, and near absence of inharmonic distortion that plagues Schiit multibit DACs. This tells me the design is thought through and verified.
...

Would be interesting to see how the 5-figure ($$$) MSB multi-bit/ladder DACs stack up in comparison.
 
Last edited:
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,506
Location
Seattle Area
Would be interesting to see how the 5-digit MSB multi-bit/ladder DACs stack up in comparison.
Hopefully one of those owners gets motivated to have theirs measured. The one person I know with one doesn't want it measured.
 
  • Like
Reactions: g29

Calexico

Senior Member
Joined
May 21, 2019
Messages
358
Likes
72
I am one of your customers and sorry but I can't confirm that. When accurately level matched and blind tested it sounds fine, but not fundamentally better than a good DS dac.

Unfortunately there is no real evidence to justify your words there.
You cannot infirm that too.
It sounds just fine to you doesn't mean much. Maybe to others it will sound more to their taste.
Also having different technologies and designs in dac is good.
I wouldn't like all designs to be the same.
It's not marketing to make r2r dac i think it's courageous to design these dac that measure a little worse that modern dacs.
Even if ds were proven to be the best (to me it's quite not proven) it's good to keep different designs and to have people who want to try new designs.

Having all dacs designed with classic Delta sigma and aop output stage would be very boring.

5 years ago chinese dac were a lot cheaper.
I think now that new chips have extravagant specs designers want to use it and sell it at higher cost.

They design in order to have good specs.
They don't care about the sound or musicality.

So again it's good to have designers who care of the ear feeling and who want to try other designs.
They have the right to choose less good specs while trying to find more musicality.
And if people ear this you must respect it.
.
 
Last edited:

g29

Addicted to Fun and Learning
Joined
May 1, 2019
Messages
520
Likes
318
I am one of your customers and sorry but I can't confirm that. When accurately level matched and blind tested it sounds fine, but not fundamentally better than a good DS dac.

Unfortunately there is no real evidence to justify your words there.

Since measurements are established based on engineering norms/standards and the identified test equipment/environment has known published spec'd tolerances, capabilities and limitations, maybe subjective listener reviews should also be pre-qualified by publishing the results of their RECENT hearing tests (for both sides of the subjective fence). Maybe a correlation would emerge between individual's actual hearing levels and perceived sound performance.

I know my old ears are nowhere near as good as they were in my youth and I openly admit that. I also consult my wife's hearings acuity for her ability to hear higher frequencies that I no longer can. There were times in my younger years where I would hear a system and think, that is really harsh/shrill and the next guy would say, sounds great to me. Now when someone asks me if it sounds harsh/shrill, I tell them they are asking the wrong ears, ask my wife.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,506
Location
Seattle Area
Since measurements are established based on engineering norms/standards and the identified test equipment/environment has known published spec'd tolerances, capabilities and limitations, maybe subjective listener reviews should also be pre-qualified by publishing the results of their RECENT hearing tests (for both sides of the subjective fence). Maybe a correlation would emerge between individual's actual hearing levels and perceived sound performance.
Audiologist only check to 8 kHz so their testing is not useful for this application. Even if they tested higher frequencies, what matters is having critical listening skills to hear impairments, not just high frequencies. My high frequency range is shot but I can easily outperform many on finding audible distortions.
 

g29

Addicted to Fun and Learning
Joined
May 1, 2019
Messages
520
Likes
318
Audiologist only check to 8 kHz so their testing is not useful for this application. Even if they tested higher frequencies, what matters is having critical listening skills to hear impairments, not just high frequencies. My high frequency range is shot but I can easily outperform many on finding audible distortions.

Having a trained ear is the other half of the equation. Being able to hear and knowing what you are hearing are both prerequisites.
 

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
Since measurements are established based on engineering norms/standards and the identified test equipment/environment has known published spec'd tolerances, capabilities and limitations, maybe subjective listener reviews should also be pre-qualified by publishing the results of their RECENT hearing tests (for both sides of the subjective fence). Maybe a correlation would emerge between individual's actual hearing levels and perceived sound performance.

I know my old ears are nowhere near as good as they were in my youth and I openly admit that. I also consult my wife's hearings acuity for her ability to hear higher frequencies that I no longer can. There were times in my younger years where I would hear a system and think, that is really harsh/shrill and the next guy would say, sounds great to me. Now when someone asks me if it sounds harsh/shrill, I tell them they are asking the wrong ears, ask my wife.

I don't disagree with your statements there, they are entirely logical, however it is difficult for any individual to demonstrate this competence. BTW my hearing still extends to around 15/16 kHz., but as Amir says critical listening skills are the more important side of the equation.

One way to help mitigate this is to test with multiple subjects. I have done this with a (admittidly small) number of people with the Seokris dac and my own dac1. No one could reliably tell the difference when level matched and blind.
 

Calexico

Senior Member
Joined
May 21, 2019
Messages
358
Likes
72
I don't disagree with your statements there, they are entirely logical, however it is difficult for any individual to demonstrate this competence. BTW my hearing still extends to around 15/16 kHz., but as Amir says critical listening skills are the more important side of the equation.

One way to help mitigate this is to test with multiple subjects. I have done this with a (admittidly small) number of people with the Seokris dac and my own dac1. No one could reliably tell the difference when level matched and blind.
You should specify what songs you was listening to. To me lots of songs are badly mastered to my taste. I agree with some songs there are no differences between dacs. As the master sounds bad it's difficult to have a preference. Songs mastered with great transient with rich harmonics with some echo and complex effects rich timbres and no compressor will be more suited for such test and will make it easier to peak the differences.

Usually i found easier to pick differences with classical music. When there are finals with lot energy and big dynamic range it can be felt like the players are tired on some dacs and on other you feel woww. Also you can hear more easily some instruments.
 

Bds3151

Member
Joined
Sep 20, 2018
Messages
57
Likes
23
After reading this and several other test results, some that have listening reviews attached, no matter what the test says it's not audible. Buy on price alone with no other criteria. Is this the overall result being communicated? I buy audio equipment to listen to, not to look at the results of meaningless data so how are these tests helpful to a prospective buyer?
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,386
Likes
24,752
Location
Alfred, NY
...how are these tests helpful to a prospective buyer?

They show the flaws in the engineering for less good equipment and show the performance merits of better engineered equipment. If I can get X performance for $1000 or $100, personally, I'd rather have the latter. And often it's X performance for $1000 and 2X performance at $100. That's nice to know!

Benchmarking equipment performance and accuracy seems pretty useful to me, but if you're buying a story rather than a piece of gear meant to accurately reproduce the musical data, perhaps it's not as useful.
 

Bds3151

Member
Joined
Sep 20, 2018
Messages
57
Likes
23
Where did I mention a "story" anywhere? As Amirm says, you won't be able to tell the difference in a DB test. How is that extra measured performance useful to audio enjoyment if you can't hear it? Just buy the cheapest and you don't need to worry about measured performance. Unless you break out the spec sheet to show off to your fellow audiophiles for bragging rights. If you like doing that, I see the benefit.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,386
Likes
24,752
Location
Alfred, NY
Or if one has an esthetic preference for higher performance. I doubt I could hear the difference between my $9 Apple dongle or Behringer 404HD and an RME ADI-2 Pro or a Benchmark, but if I had the money, I'd be happy to buy either of the latter two because they perform better and have more features and better build quality. That's my particular irrationality. :cool:

If you like the Soekris story ("multibit is better") and are willing to pay extra and accept lower performance (albeit unlikely to be audibly inferior to higher performance, cheaper units), go for it. That's the beauty of the free market.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,506
Location
Seattle Area
After reading this and several other test results, some that have listening reviews attached, no matter what the test says it's not audible. Buy on price alone with no other criteria. Is this the overall result being communicated? I buy audio equipment to listen to, not to look at the results of meaningless data so how are these tests helpful to a prospective buyer?
We have not done any kind of universal blind test across many people, content, levels, etc. So you can't draw that conclusion unless distortion products get beyond the dynamic range of our hearing. That translates into a SINAD of 116 dB. So many products can't be proven to be transparent.

But yes, odds are against vast majority of audiophiles hearing higher distortions than above limit. So you can shop based on features, price, looks, etc. While you are at it though, might as well get one that measures better than another for the same cost.
 

Bds3151

Member
Joined
Sep 20, 2018
Messages
57
Likes
23
SIY, I get that (irrationality, features). And I'm very happy to be irrational as often as I want for whatever reason. And features I can get from 100 other sites. But we shouldn't be deluded into thinking these measurements have any real value beyond numbers on a web page. A kind of entertainment for audio nerds (like me)!
Maybe there is a couple of products Amirm has heard an audible difference (DB confirmed of course) that he could test and compare? That might be helpful - if it exists?
 

Bds3151

Member
Joined
Sep 20, 2018
Messages
57
Likes
23
We have not done any kind of universal blind test across many people, content, levels, etc. So you can't draw that conclusion unless distortion products get beyond the dynamic range of our hearing. That translates into a SINAD of 116 dB. So many products can't be proven to be transparent.

But yes, odds are against vast majority of audiophiles hearing higher distortions than above limit. So you can shop based on features, price, looks, etc. While you are at it though, might as well get one that measures better than another for the same cost.

Amirm, So if the limit is 116db, are you saying you can DB test those and tell a difference reliably? Assuming that number is much lower, could you do a test that measured that? That seems like it could be useful if it was verified by a few others? It would seem to make all your hard work more meaningful.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,386
Likes
24,752
Location
Alfred, NY
But we shouldn't be deluded into thinking these measurements have any real value beyond numbers on a web page.

Again, it separates good performers from bad- look at the Schitt stuff for example, where you wouldn't know how poorly it performs without someone independent measuring it. Likewise, the headphone sections of DACs vary quite a bit, and trying to match one to your needs is best done with actual data rather than marketing claims.

Moving outside of DACs, performance data can be even more important- if Amir and I hadn't measured the Lounge Audio phono preamp, for example, no-one would know how piss-poor it is. I'm doing measurements at the moment on another phono preamp which has some strong and weak points, and my data will allow prospective buyers to decide whether the advantages outweigh the disadvantages for their needs and priorities.
 

Bds3151

Member
Joined
Sep 20, 2018
Messages
57
Likes
23
I had a thought about how to tie the measurements to some user experience. I believe Amirm has stated previously that he has a particular dislike for tube equipment. Assuming this was a listening test, maybe comparing the a tube dac vs the best dac and determining the appropriate measurement differences would develop an initial benchmark. It wouldn't be perfect due to the small sample but it's a start.
 

Jimster480

Major Contributor
Joined
Jan 26, 2018
Messages
2,880
Likes
2,032
Location
Tampa Bay
Amirm, So if the limit is 116db, are you saying you can DB test those and tell a difference reliably? Assuming that number is much lower, could you do a test that measured that? That seems like it could be useful if it was verified by a few others? It would seem to make all your hard work more meaningful.
its not that simple. There are other factors to take into account.
For example different amps behave differently with different loads from different headphones. This creates a lot of variables as to what you are actually listening to.
Amir measures what is the best performance available from a specific device and tries to get a baseline of what the "reasonable expectation" of performance could be with his various tests.
Therefore gear which performs better simply performs better.
"How it sounds" isn't a measurement since its totally subjective and pointless. Some people like garbage sound with tons of distortion... some people like overblown and muddy bass.
We are a forum of objectivity; benchmarking what is objectively performing better.

From my own experience; the data on this site does translate into real world listening results. Although at the high end of scores; it would be very difficult if not impossible to tell most devices apart (DACs) through the same Amp.
Often telling Amps apart is easier as there are different topologies and using different headphones with different amps can result in wildly different performance on a per-case basis.

In conclusion the data on this site allows you to make an informed purchasing decision about the reasonably expected performance of a given device at a given pricepoint.
 
Top Bottom