[upstateguy]

Quote:

Has anyone established if DiffMaker actually works? I want to know if it ever finds any differences.

Of course it does Shah.  Check out the demos and see for yourself. 
Better yet, make some yourself with various EQs and see if it picks it up.  You will find that it will. 
It's just an automated nulling program. 

 
I know you have a lot invested in it's not working, but maybe you might try it with your EE dac.  Tubes Vs Solid State.  You can hear the difference, see if the diffMaker can  find it.

 

[upstateguy]

Quote:

The former could be the case - I am using ESS9018 which I believe decodes the SPDIF data on-chip (I cant recall what oversampling it does etc).  We are also talking the difference between adaptive USB and asynchronous.  Or it could be that the DI is just really that inferior.  I can't personally say whether this is better or not to a stock hiface as I have not compared.  But the statement I made WRT magnitude of difference is definitely not an overstatement.  If one had adequate equipment and methodologies I am sure one should be able to measure a difference - but I don't have access to studio grade equipment.  What I can do, and have done is set up the mk3 and DI to run off separate computers into my DAC allowing easy A/B listening comparison, and from there the differences are easily discernable, not SPDIF cable swapping sort of guesswork.  A thick layer of sludge has been lifted.

You also can make some DiffMaker files.  Especially if you have two computers.  If there is a difference it will be there loud and clear.
 
Check out the Marching Band demo on the Audio DiffMaker site.

 

[Shahrose]

Quote:

Of course it does Shah.  Check out the demos and see for yourself. 
Better yet, make some yourself with various EQs and see if it picks it up.  You will find that it will. 
It's just an automated nulling program. 

 
I know you have a lot invested in it's not working, but maybe you might try it with your EE dac.  Tubes Vs Solid State.  You can hear the difference, see if the diffMaker can  find it.

Actually for me, positive or negative results would be of no value. I only care about what I'm hearing, remember?
 
As for the positive comments about the MK3, I wouldn't go as far as to say it's like changing headphones, but it did noticeably improve my system (and I was using the MK1/MK2 before that, which are pretty good already).
 
If I compare USB (of all my current/previous DACs) and the MK3, there's just no contest.
 

 

[upstateguy]

 
Quote:

Actually for me, positive or negative results would be of no value. I only care about what I'm hearing, remember?
 
As for the positive comments about the MK3, I wouldn't go as far as to say it's like changing headphones, but it did noticeably improve my system (and I was using the MK1/MK2 before that, which are pretty good already).
 
If I compare USB (of all my current/previous DACs) and the MK3, there's just no contest.
 

Humm, I think it was you who said you'd like to know if the diffMaker ever finds any differences. 
 
That would be a way to find out that it does.  Then, of course, you could experiment and see which differences really existed and which were imaginary.  Finding out isn't fatal.... just humiliating.

 
You know, your resistance to science is almost approaching phobia level. 

 
Anyway, in a few weeks we'll have around 1000 O2s floating around and we'll see if anything else gets debunked. <shrug>
 

 

[drez]

Don't forget though - just because an experiment fails doesn't conclusively prove something.  There could be problems with the tools you are using, their level of accuracy, or whether the test is able to pick up the kind of information you are looking for.  For example with diffmaker - the results of any difference are meaningless noise, no matter what the difference in equipment being measured - it won't tell you which sample is clearer, has better soundstage, better imaging, smoother treble, or tighter bass response.  It will only tell you if there is a measurable difference within the tolerance of your equipment and give you some idea of the magnitude.  Another thing to consider is that the absolute magnitude of the differences might not be what is important in the perception of these aspects of sound reproduction, there may be more complex factors in the signal which have not been studied.
 
Personally Sound engineering is not my area of study, so it is not a good application of my knowledge and talents to try and set out to measure perceivable aspects of sound reproduction.  If I were doing a PDH and had a research grant, then maybe it would be worth my effort because I would 1)have the requisite knowledge 2)have proper tools and 3)have funding.  Numerous sound engineers may try to suggest that certain aspects of audio equipment performance are not important, however this does not prove that it is so - bodies of research (If there are any that are relevant to the performance of DD converters) are overturned all the time.  And also don't forget there is a thriving pro audio sector that invests a lot of engineering into jitter management.
 
But anyway, my point is, accept the limitations of you methodology and research background - that is unless you have a PhD in sound engineering and are working actively in academic research.  There are always two sides to every academic research field, just it turns out researchers tend to argue from one side only to avoid schizophrenia.

 

[upstateguy]

Quote:

Don't forget though - just because an experiment fails doesn't conclusively prove something.  There could be problems with the tools you are using, their level of accuracy, or whether the test is able to pick up the kind of information you are looking for.  For example with diffmaker - the results of any difference are meaningless noise, no matter what the difference in equipment being measured - it won't tell you which sample is clearer, has better soundstage, better imaging, smoother treble, or tighter bass response.  It will only tell you if there is a measurable difference within the tolerance of your equipment and give you some idea of the magnitude.  Another thing to consider is that the absolute magnitude of the differences might not be what is important in the perception of these aspects of sound reproduction, there may be more complex factors in the signal which have not been studied.
 
<snip>

 You have no idea what you're talking about.

 

[drez]

Quote:

 You have no idea what you're talking about.

 
exactly.  Now you try and admit the same.

 

[Sapientiam]

Quote:

Anyway, in a few weeks we'll have around 1000 O2s floating around and we'll see if anything else gets debunked. <shrug>
 

You think debunking is doing science? Really?

 Next (and here's a testable hypothesis for ya) you'll be telling us Randi's a first-rate scientist.

 

 

[Willakan]

Quote:

 
exactly.  Now you try and admit the same.

Upstateguy has a point. You are horribly distorting the concept of the burden of proof to justify a logical absurdity.


 

 

[drez]

Nope I think that's over extending my argument.  I am merely pointing out that nobody is an expert here, and that even among experts there is no lasting consensus on this matter or any other.  If there is a consensus it is merely and agreement of opinions based on interpretation of experimental data.
 
Additionally, seeing as there are numerous highly qualified professional engineers who dedicate much of their careers to jitter management in digital audio systems, the experiments and opinion of an enthusiast, or even that of another academic and his researchers is not Fact, just opinion based on interpretation of [lesser] data.
 
The designers of any decent DAC go to great lengths to manage jitter before it reaches the DAC chip.  If low levels of jitter were not important, then why not tell these engineers they are wasting their time?
 
JK hiface is doing the same thing any decent USB receiver is doing, just IMHO better.

 

[Willakan]

But there is a consensus about jitter. There are lots of papers on it and audibility (Benjamin and Gannon springs to mind). The audibility threshold with test tones, let alone real music, is really very high.
 
Giving the impression that this is some undecided, cutting-edge field full of debates is simply misleading. Jitter is largely a solved problem - indeed there is no real reason for separating distortion products from jitter from more conventional distortion products. There is a very good reason you don't find DACs with 0.0005% THD into 1khz that have 10ns of jitter on their analog outputs.
Engineers do have to make provisions to reduce jitter, but these are hardly particularly onerous. My relatively cheap DACMagic reduces jitter from the digital inputs far (excessively far) into the realm of irrelevancy.
 
 
 

 

[drez]

Quote:

But there is a consensus about jitter. There are lots of papers on it and audibility (Benjamin and Gannon springs to mind). The audibility threshold with test tones, let alone real music, is really very high.
 
Giving the impression that this is some undecided, cutting-edge field full of debates is simply misleading. Jitter is largely a solved problem - indeed there is no real reason for separating distortion products from jitter from more conventional distortion products. There is a very good reason you don't find DACs with 0.0005% THD into 1khz that have 10ns of jitter on their analog outputs.
Engineers do have to make provisions to reduce jitter, but these are hardly particularly onerous. My relatively cheap DACMagic reduces jitter from the digital inputs far (excessively far) into the realm of irrelevancy.
 

Yes these test and others attempt to measure the audibility of jitter, and show that the threshold is well below what most decent transports provide.  I can understand that well enough.  
 
What I don't understand is that all these SPDIF transports are meant to be bit perfect, and many of them sound different.  Now we are met with a condition of two seemingly mutually exclusive conditions.
 
We can either accept that:
 
A) That these two conditions are not mutually exclusive - ie that the tests were correct, but limited in the scope of their conclusions.
 
OR
 
B) that bit perfect audio sounds different even though jitter has no effect (ie the results are not universally applicable)
 
OR
 
C) everyone is hallucinating
 
In my opinion the most likely of these is A), but you may dissagree and support  B) or C).  Personally I don't find B) or C) very likely, and would suggest the burden of proof there is to prove that everyone is hallucinating, or that there is some other factor at play.
A) to me represents the most logical and supportable of these possible scenarios.  I cannot prove it, but neither can I prove B) or C).
 
Either way any decent Thesis abstract or conclusion should at least hint at A)
 

 

[slim.a]

Quote:

mk3 has arrived and charged!  OOTB this thing is miles ahead of the AGD DI - were not talking changing SPDIF cable maybe-yes-maybe-not different, this is literally on par with changing headphones, DBT proof changes.
I can only imagine if burn in improves things.  Very happy with this little box, it's worth every cent.

 
Quote:

I have a new Redwine HPA with a  minwatt tube replacement. With  Denon ah d7000 headphones the HPA is very clear, articulated, holographic and  natural sounding. I have been really enjoy this combo. Previously I was using a Apogee ensemble. Which was nice but it has been used primarily for recording music. I was using a stock usb cable with the HPA. Which sounded good as it is galvanic  isolated. But I came down with the audiophile bug and wanted and even better sound. I was looking at a locus axis usb cable, but decided a spdif converter would improve the sound more. It was also cheaper. When I first got my Hiface Mk 3, I plugged in my old  Sennheiser HD 280 pro's to burn in the unit. But to my utter disbelief this combo sounded better in a lot of ways, than the HPA with the Denon 7000's. This was without proper burn in! Don't misunderstand me. I greatly prefer in every way, the Denons over the Sennheiser HD 280 pro's. I kept this combo in place. Because I figured by waiting to put my good headphones in would be like getting another new toy. Boy was it! The mk3 is NOT subtle at all, it gave the HPA a huge improvement in sonic performance. Similar to drez's post, it's like changing the headphones or DAC, not a cable. The sound stage became even more 3D. Allowing a more accurate placement of instruments which were deeper and wider. It felt like a huge veil was lifted with a much blacker background. That I did not know that was there. As the HPA is quit black to begin with. Micro details within songs not only became more readily apparent but were clearly defined in space. The bass was tighter and much more articulated. The high end sounded extended and clearer, without grain. Cymbal began to sound as if you hearing them live, with proper attack and decay. Everything had more body and was richer . Bringing everything closer to that elusive  "REAL" sound. Anybody who does not believe a hiface mk3 makes a difference in sound, only needs to hear one! Well truthfully, in my experience I can only say with a Redwine HPA it will make a HUGE difference. Making a GREAT sounding unit even better!
 
 

First of all, thanks for all those who shared their comments about how the different versions of the modified Hiface sound in their system (Shahrose, drez, tuberock...). This thread usefulness (in my opinion) is indeed about sharing people experiences with different converters and not about discussing theory of jitter.
 
Let me also share an anecdote about the Jkeny MK1 usb converter. Around the time I got the prototype of the Jkeny MK1, I went to my brother's place to try a few things in his desktop system. My brother is an engineer and he is far from being an audiophile. However, when it was time to to replace some "broken" desktop speakers he had been using mainly for gaming, I encouraged him to get a cheap amp and some bookshelf Klipsch speakers in order, to have more "flexibility" if he ever needed upgrades later on.
 
When I visited him, I brought along an Audio-gd dac100 (I wasn't using anymore) as well as the JK MK1. I wanted to try the audio-gd dac100 in his system and see if we could notice a difference with the X-Fi soundcard in an entry level system.
Switching between the Dac100 and the X-Fi, I could barely detect any differences (and I was doing the switching) and he couldn't.
It was until I placed the JK MK1 in the chain as a transport that he started to hear differences. He wasn't even aware that I inserted something new in system as I was making the switching and he was listening.
Weirdly he couldn't say which one he preferred (he is not an audiophile) but he could detect differences between the MK1+DAC100 and the X-Fi while he couldn't detect any differences between the Dac100 (run from its usb input) and the X-FI.
 
In that particular experience, it seemed that the quality of transport mattered more than what was actually in the DACs (dac chips, output stages,...). Obviously, no general conclusion can be drawn from such an entry level system, but I thought it could provide an interesting insight.
 
 
Edit -- The only "comment" I am going to make regarding the discussion above is rather to point to a quote from Daniel von Recklinghausen : "If it measures good and sounds bad, -- it is bad. If it sounds good and measures bad, -- you've measured the wrong thing."  

 

[Sapientiam]

Quote:

What I don't understand is that all these SPDIF transports are meant to be bit perfect, and many of them sound different.  Now we are met with a condition of two seemingly mutually exclusive conditions.
 
We can either accept that:
 
A) That these two conditions are not mutually exclusive - ie that the tests were correct, but limited in the scope of their conclusions.
 
OR
 
B) that bit perfect audio sounds different even though jitter has no effect (ie the results are not universally applicable)
 
OR
 
C) everyone is hallucinating
 

 
Your proposed options aren't exhaustive. I favour another hypothesis D) which is that while jitter does differ between transports, its a secondary issue and not the main cause of audible differences. Transports are in general bit perfect but have different EMC signatures. The EMI couples into the DAC via the cabling between the two and its this EMI coupling which is the primary cause of audible differences.
 
 

 

[Willakan]

Quote:

 
We can either accept that:
 
A) That these two conditions are not mutually exclusive - ie that the tests were correct, but limited in the scope of their conclusions.
 
OR
 
B) that bit perfect audio sounds different even though jitter has no effect (ie the results are not universally applicable)
 
OR
 
C) everyone is hallucinating
 
In my opinion the most likely of these is A), but you may dissagree and support  B) or C).  Personally I don't find B) or C) very likely, and would suggest the burden of proof there is to prove that everyone is hallucinating, or that there is some other factor at play.
A) to me represents the most logical and supportable of these possible scenarios.  I cannot prove it, but neither can I prove B) or C).
 
Either way any decent Thesis abstract or conclusion should at least hint at A)
 

Calling involuntary bias "hallucination" is also misleading. Current science suggests that people hear differences that are not defined by the performance of the device they are listening to. This is not "hallucination", or stupidity, or incredibly unlikely - it's being human.
Your claim to hear things is essentially unfalsifiable - I cannot prove with absolute logical certainty that you do not hear differences. However, under such circumstances the burden of proof falls upon the person with the unfalsifiable claim if your claim is to be justified as rational (Thank you Bertrand Russell). Science has reasons why you hear differences, which audiophiles have yet to prove incorrect.
 
Any conclusion made using the same rational principles with which we almost unconsciously approach much of our everyday world should suggest that existing scientific mechanisms are to blame for the differences (unavoidable human bias) rather than assuming otherwise without any evidence.
 
Alternatively, your DACs could be truly terrible at jitter attenuation and some of your transports could actually be adding considerable amounts of jitter, which is then audible. It seems unlikely, but in the absence of measurements for many pieces of gear remains entirely possible. Indeed, I believe Bel Canto actually released a USB to SPDIF converter which actually added huge amounts of jitter to the signal!
 
 
Anyway, I'll be going now! Discussions such as these I have discovered to have considerable ban-generating potential via thread derailment.