Soekris DAM1021 R2R DAC Measurements

[Jimster480]

I cannot imagine that it would be the case. But I think that measurements can be more detailed in that the same tests could be repeated with many different tones or maybe even a combination of tones to see how the results turned out. Since real music has many tones playing at the same time. Checking to see what the distortion is when there are say 10+ tones playing at the same time could also give you a better idea of what the performance is like in the real world.
Also Doing the same tests even singularly across the whole audible range and averaging them all gives you an idea aswell.

I don't believe there is anything that isn't measurable, just things we haven't yet measured.

 

[Jinjuku]

How much music out there has R2R DAC's in the mastering chain I wonder.

 

[Jimster480]

How much music out there has R2R DAC's in the mastering chain I wonder.

That is another thing that I have thought of aswell.

 

[Rene]

How much music out there has R2R DAC's in the mastering chain I wonder.

All of the HDCD releases.

 

[svart-hvitt]

How much music out there has R2R DAC's in the mastering chain I wonder.

FWIW, this guy has some Pacific Microsonics (AD only?) in his chain. Pretty impressive list of customers:

http://magicgardenmastering.com/

 

[svart-hvitt]

Agreed.

I hate it when that happens, i.e. something which is there but can’t be measured.

I am no audio engineer, but I am still skeptical to your position. Your position is (seems to be?) that engineers, after 30-40 years of digital audio, haven’t come up with intelligent ways to measure sound originating from digital boxes. How can that be? Are audio engineers the dumbest students to enter the faculty of the hard sciences?

I am not claiming that your position is wrong. I am just trying to formulate how your position could make sense.

 

[dwalme]

I don't believe there is anything that isn't measurable, just things we haven't yet measured.

Very well said.

I am no audio engineer, but I am still skeptical to your position. Your position is (seems to be?) that engineers, after 30-40 years of digital audio, haven’t come up with intelligent ways to measure sound originating from digital boxes. How can that be? Are audio engineers the dumbest students to enter the faculty of the hard sciences?

I am not claiming that your position is wrong. I am just trying to formulate how your position could make sense.

The flip side to this argument is how is it possible to know everything there is to know scientifically about this subject?

We might as well stop exploring, challenging, testing, learning as it's all been covered already and there is nothing left to learn.

It's not possible to know what we don't know before we know it.

It's only possible to believe that there is more that we can learn and then go after it.

There will be more discoveries.

 

[mindbomb]

If we are talking about noise modulation, Chord talks about measuring it. In a few presentation slides, they compare the noise floor of 1khz full scale signal to that of nothing playing.

 

[Wombat]

Very well said.



The flip side to this argument is how is it possible to know everything there is to know scientifically about this subject?

We might as well stop exploring, challenging, testing, learning as it's all been covered already and there is nothing left to learn.

It's not possible to know what we don't know before we know it.

It's only possible to believe that there is more that we can learn and then go after it.

There will be more discoveries.

If you can detect audio content from the air with your ears then it can be captured, measured and analysed with current technology.

If what you 'hear', includes your mind influencing it, it is a different situation.

 

[March Audio]

Aren’t you suggesting here that R2Rs have a sound that measurement tools can’t...measure?

If so, there is a sound aspect in R2Rs that need to be heared, not measured?

Which again raises the question: Why measure DACs if measurements are below hearing limit, and there are audible aspects that cannot be measured?

This would leave blind testing the only valid process for testing high-end audio gear.

Agree?

The problem here is proving the validity of individuals subjective statements. FWIW I like the sound of this dac however I don't think it sounds more natural. Not even sure if I know what that means

However it is clear that these dacs do measure differently. We have a spray of harmonics which are at a level which I am, well how shall I put it, uncomfortable about. I can't tell you if that's responsible for what some hear or if it's below perceptible levels, but it's a characteristic that could make them sound different.

 

[dwalme]

If you can detect audio content from the air with your ears then it can be captured, measured and analysed with current technology.

Great. Then there is no debate why ds dac vs r2r dac sound different. Measure each and we know with 100% certainty why they sound different. I've never seen anyone show measurements that clearly end the debate on why they are different.

If what you 'hear', includes your mind influencing it, it is a different situation.

Not possible for human beings to use their ears without their brains.

People are influenced by so many things which is why I prefer measurements explain the differences.

 

[March Audio]

Great. Then there is no debate why ds dac vs r2r dac sound different. Measure each and we know with 100% certainty why they sound different. I've never seen anyone show measurements that clearly end the debate on why they are different.

I'm afraid you have missed a crucial point here. We don't know they do sound different. Individual subjective anecdotes are worthless unless there has been some form of control in the comparison. Sighted?. Forget it.

 

[amirm]

Not possible for human beings to use their ears without their brains.

They can use their brains. We just need to make sure nothing other than the ears inform them. When we do that, all the things people say they readily hear unfortunately disappears. This leaves nothing to investigate.

 

[March Audio]

Not possible for human beings to use their ears without their brains.

People are influenced by so many things which is why I prefer measurements explain the differences.

But it is possible to control the test to remove or minimise the bias influences

 

[Wombat]

The industries that supports ethereal audio qualities are the ones that should be proving their statements. Have they been objectively active in this regard?

Extracting signals which are below noise levels is commonly practised in other spheres of interest.

 

[Rene]

FWIW, this guy has some Pacific Microsonics (AD only?) in his chain. Pretty impressive list of customers:

http://magicgardenmastering.com/

I didn't see a mention of the PMI in his studio list, but he only mentioned the Bricasti dac (another one that should be tested).

The Pacific Microsonics A-D uses a two stage flash converter. A flash converter has N-1 comparitors, each of which has a different reference level equal to each bit weight. Thus, a 10 bit flash has 1023 comparitors strung on a voltage divider spanning the reference voltage and divided into 1024 parts. A two stage converts the analog input signal to N bits, feeds this to a very accurate dac, subtracts the dac output from the input analog signal, then converts the remainder and adds the two conversion results.
In order for this trick to work, the input signal must be sampled and held until the conversion is finished. It may sound cumbersome, but flash converters can be made to operate at nanosecond speeds. The slowest part of the operation is the settling time for the sample and hold circuit.

 

[Rene]

I hate it when that happens, i.e. something which is there but can’t be measured.

I am no audio engineer, but I am still skeptical to your position. Your position is (seems to be?) that engineers, after 30-40 years of digital audio, haven’t come up with intelligent ways to measure sound originating from digital boxes. How can that be? Are audio engineers the dumbest students to enter the faculty of the hard sciences?

I am not claiming that your position is wrong. I am just trying to formulate how your position could make sense.

Dumb and getting dumber with age and experience. I still don't know what we aren't measuring that matters.

We rarely measure phase difference between channels, a very important parameter for some sources. It explains much of the perceived difference in phono preamps, but no one mentions it; just amplitude.
Our best dynamic measurement method involves subtracting out the original signal from the converted one and looking at the difference. While this works for linear amplifiers, it is hard to see how to apply it to a dac.

 

[RayDunzl]

We rarely measure phase difference between channels, a very important parameter for some sources.

Ooh, I did...

---

I agree on Channel Phase Difference having importance, I would think that plays games with the imaging.

---

Phase difference, measured at preamp output:

Measured last October, in my amateur capacity.

Setup: REW digits 48kHz optical to switch to Behringer DEQ2496 to miniDSP OpenDRC-DI to Benchmark DAC2 HGC, analog output hobbled by Krell KCT preamp, signal capture hobbled by a 25 foot cheap cable, and capture finally hobbled by the PC onboard ADC:

Left and right channel matching, with a little induced electrical noise (60Hz) and its harmonics picked up along the way:

Since the channels matched so well (to my eye), despite a bit of curve (likely from the cable or preamp as much as from the DAC), I didn't worry about it further.

---

Add a little room correction DSP, and the prior flatness becomes moot, and even the close matching has been tossed in preference of what gets sent onward toward the listening position.

A rather large change in the graph scale has occurred to accommodate the modified signal:

Change the DSP to accomodate an alternate set of speakers, and get this phase measurement:

 

[March Audio]

Dumb and getting dumber with age and experience. I still don't know what we aren't measuring that matters.

We rarely measure phase difference between channels, a very important parameter for some sources. It explains much of the perceived difference in phono preamps, but no one mentions it; just amplitude.
Our best dynamic measurement method involves subtracting out the original signal from the converted one and looking at the difference. While this works for linear amplifiers, it is hard to see how to apply it to a dac.

You can easily compare phase between channels on a dac. One is the reference. The reason you dont see this measurement on DACs is because it is rarely anything less than perfect. If there is any filtering of the signal then of course absolute phase may start to change but once you see what speakers and crossovers do to it what the dac is doing is somewhat irrelevant.

 

[FrantzM]

Hi

I do understand the difficulty of blind tests. It would have been instructive however to compare a few samples of R2R DAC to other kind of DAC in a blind test. I understand their unpracticality and their difficult logistics but ...
I am struggling with the idea and cannot admit that all DAC (or electronics in general) when competently designed sound the same ... It could be the remnants of my long audiophile conditioning ...