Hi John,
Good to see a noted DAC designer joined our forum and hope you enjoy your stay, even though the set-up of this forum is a bit leaning towards the 'objective' side of things.
You are welcome to discuss your products here and vent your ideas as long as the 'heat' remains switched off
.
Discussions held in a pleasant and informative manner are MORE than welcome !
I would like to take the opportunity to point everyone to
our forum rules in this matter.
In developing these latest Ciúnas devices (DAC & converter), I used an Amanero USB receiver board & tried many different isolation chips & configurations with it. I experimented with the full galvanic isolation which involved the USB board being powered from USB or battery, isolators, one half of which was powered from the same supply as Amanero & the other, clean side, powered by battery - clocks powered by a separate battery, reclocker & or DAC or USB transmitter. powered from separate battery. Everything was isolated - The I2S signals & the control signals from the Amanero board went through the isolator to the clean supply - the clock & control signals back to the Amanero board went through the isolator boards, . So the configuration completely isolates the PC/USB side from the clocks/reclocker/DAC/SPDIF converter side (within the capabilities of the isolator - all isolators let through some frequencies). Yes, I tried low jitter isolators (Potato chips), GMR, etc.
Result: no difference to the sound - it was still excellent with/without the isolator in place.
Why? There are probably a number of reasons
Well there is at least one very obvious reason for me. Your test system at home is of good quality and the influence of common mode signals is either not present or below an audible threshold and thus will remain inaudible in your system and very likely also in those of many other people.
That, of course, doesn't mean isolation is completely pointless and does NOT affect SQ at all for OTHER people on OTHER systems and working in the field of EMC and common mode related problems I can assure you that a configuration of equipment in situation A can differ CONSIDERABLY from situation B... even with the same equipment used but connected differently or even at a different physical location.
I am afraid a subjective assessment that there is no (substantial ?) difference in your experiments with or without an isolator in place is no definitive proof that it may not have (sonic or safety related) benefits in other cases.
What IS good to hear is that with and without isolator the sound is still excellent (no differences by your own admission) and could deduce from that statement galvanic separation (isolation) is not detrimental to the sound even though it (provable) adds certain types of jitter.
One could even go as far as stating that everything upstream (connections wise) isn't of any importance at all but won't go that far
.
For this reason alone I would suggest that adding an isolator (in the sense of galvanic separation not re-clocking) may be beneficial under different circumstances and is not detrimental to the sound. That makes an isolator a major asset for it to work under MORE circumstances IMO but adds to the costs and complexity.
Purists may well say... every component in the path is detrimental.. so loose it.
I am fine with that by the way, it even has a slight grain of truth in it.
Finally, all the stories & technical detail will not tell how devices actually sound. I'm happy to put my DAC or converter in the company of or up against much more expensive devices.
So will every other manufacturer that has a pride towards their product(s) / deign(s) gladly put up his/her device(s) against those of others.
I am
quite convinced your DAC sounds good by the way. There is
NO doubt in my mind it is.
Just like there is no doubt in my mind many other products sound good as well, regardless of their pricetag.
You are free to feel/think/be convinced this is because your ears, and those of others, tell you so and it is, amongst other aspects, because of the usage of a lithium battery.
I have a feeling you do not believe in specifications and measurements.
I can't find much of it except output voltage, a meaningless THD number and S/N ratio for the older DAC's, but no info on how they were obtained/measured/determined on your website. Perhaps there are some third party measurements you can link to for those that like to know ?
Obviously you have researched it judging from the first part of your post.
I assume you did use some 'standard' test methods and may deliberately leaving test results out because you do not believe they matter or tell anything about SQ and make a statement this way.
Do you care to share/post some of those measurements anyway or do you consider them trade secrets or not revealing of sound quality and leave it in the 'subjective listening only' area ?
Just a couple of things to say about ringing:
The ringing issue with most OS DACs is due to the linear phase filters which introduce pre-ringing at ALL frequencies. The audibility of which can be argued, like so many things in audio & probably depends on the rest of your system - particularly your speakers & how phase coherent they are. I use single drive fullrange Jordan speakers on which I can hear the effects of time smearing. I agree, btw, that it may not be the pre-ringing per-se that is the problem but other factors in digital filter implementations
Can I ask you to provide some 'proof' in the form of plots or graphs or oscilloscope screenshots that pre- (and post-) ringing is present in all frequencies and not just those in the vicinity of the Nyquist frequency ?
I am dying to finally see those as I cannot produce them myself for some obscure reason, nor have others succeeded AFAIK (and granted I do not know much about this subject).
Of course you may not 'cheat' and use any audible band filters (such as digital EQ) in line and can only use a DAC and bit-perfect source to prove that point.
99% of all headphones is phase coherent (more so than a speaker) and is not inflicted by room echoes. Because of this (the absence of echoes) can show the effects of time smearing better than a single speaker. at least in the frequency range where the hearing is dependent on it. Needles to say that just like using a good speaker you will also need a
good headphone that is driven adequately.
I also have a bit of a pause when I hear statements like "Fortunately our hearing is pretty imperfect" Yes, our hearing introduces all sorts of distortions & is not linear - it's what gives it it's special properties & allows it to be so efficient.It's also backed up by a very powerful processing engine which through feedback, can change it's operational parameters.
A matter of 'religion' IMO. Every one is free to disagree with the opinions of others and just like your opinion is 'the truth' for you, you could consider 'your truth' will not be the ONLY correct one in the universe.
It's just the way you see it based on your own observations and those you feel have the same 'truth' on their side.
For exactly the same reason I, and everyone else on this planet, also is not absolutely 'correct' in a universal sense of it being the truth.
So you have your opinions and others have theirs.
All I ask is to treat other opinions with respect and while debating provide some proof other than anecdotal (he and I heard this or that and that makes it true as it is 'confirmed' subjectively by him or her)
Subjectivism is almost obligatory in audioland and am not against it.
The only circumstance I can have a pause with it if it is used as 'definitive proof' based on perceiving alone.
Our ears just aren't analysers and subjective 'proof' is based on the fact we think our ears are far more capable than scientific methods.
I happen to question that theory but do feel our ears
can be used to enjoy music though.
it's primary biological function is to quickly analyse & highlight what is important in the audio realm for survival & has evolved over millennia to fulfill this role. It's not a measuring instrument.
Agreed the brain is wired for survival and able to differentiate between something 'dangerous' and something that can safely be ignored. That last part is where digital compression is based upon.
I also agree the hearing is not a measurement instrument.. it's even my motto.
The problem is as soon as someone states I hear this and that and because I hear it it IS there implies hearing is used as a measurement/analysis tool and not 'questioned' but measurements seem to be highly questionable because there seems little to no relation between what is heard and what is measured. One may ask why..
The survival bit (which I agree with) doesn't mean music is interpreted in the same manner as there is no danger factor (unless the sound is too loud ?) involved, other processes such as recognition of individual instruments and placement is based on earlier experiences that is. The brain is excellent with single sounds but less 'perfect' when drawing conclusions based on what has entered our ear canal(s).
Perceiving (hearing) is very complex and not nearly fully understood as it involves MANY processes in peoples brains which also differs in each individual.
This doesn't mean electronics and how they work is not understood either by those working in that field. Of course, I am aware that this is just my opinion and could be wrong but so can anyone else be barking up the wrong tree.
In the area of music, a better understanding of it's operation (psychoacoustics) would seem to me to be the best approach to creating better audio reproduction devices. Accuracy of waveform reproduction has brought us a fair distance in our reproduction systems but has also led to an over-emphasis on this particular linear approach (THD, etc) - maybe we need to find out some more about the full workings of hearing & direct our attention to these aspects to speed up our progress?
I am ALL for that.
Actually, I am still hoping someday a fruitful cooperation might be there but it would involve blind testing and some science I am afraid.
Nothing will be taken at face value and needs to be confirmed in a scientific way IF a real head-start is to be made that will 'change' the way we think about audio.
That generally puts the cooperation in the icebox again.