Seeking education about those ultra-expensive interconnects - Page 87

I asked a Dolby employee. I've offered it to others for a very long time. Also, when I'm interested in a paper whose access to is problematical or might entail dropping $20 on, I'll find the email of the author and request a copy. Scientists are generally flattered that the public takes an interest in their work. Also I know scientists, some of which you know, who have access to some periodicals at the university they do research and teach at.

Impugn my integrity? Think of the worst thing you've been called and I'll one up it.

Say what? You are not having this much difficulty reading my English are you? The difference was in the equipment under test, not the measurement equipment. I have worked and run hardware projects that shipped on both sides of the ocean. Except in rare cases, they were different boxes. And power supply change is the primary change made. The output of the Miller test shows power supply spurrs that are not in John's. No wonder you are not convinced. After explaining half a dozen times, you still not following the answer .


That is code for you didn't pay for it? If so, no I can't take your copy. I am fascinated that you did not quote anything form it though .


See, you really can't follow the answers given. I explained at length his statement and here you are, confusing the *jitter frequency* with source music frequency. Jitter frequency can be anything it wants. It is created by equipment design and activity. It has nothing to do with what is in your music.


But you are Chu. You constantly ask the other side to go and run exercises for you. It is classic debating tactic to shut down the conversation unless and until someone produces what you are demanding. Why don't you go and find a tool that does what you want, just the way acquired that Dolby paper?


Yes, but I did it like a scientist. I showed how measurements can be inaccurate due to varying samples, test day condition, the mileage on the car, the wind direction. It took a marketing person to translate a jitter variation to a precise *2* second change in 0 to 60 time. You lecture how you are about detail and accuracy and then pull a stunt like that.


For all but a select few which are hand tested unit by unit and tweaked, yes. The Berkeley DAC for example gets burned in for a week and then hand tuned. Ditto for my Mark Levinson DAC although they offered a non-tweaked version.

The standard line in every spec sheet is what Chu? "Specifications subject to change." So if you think anyone guarantees specifications of any audio equipment, you are in for a rude awakening.


Clearly when we explain the technology at hand, there is trouble following the answer. So I thought I use an analogy that is easier to understand. What was the result? Misdirection. Instead of say "OK, for those I see that there can be variation." But instead, you turn that into a statement about me. Are you still standing by your generalization that no matter what I measure, I am going to find identical results even if the subject under test varies?


Nothing.


I didn't bail. I had left the thread before he showed up. You quoted him here and I answered it. And twice. If you have trouble following my answers, then you have no business claiming to believe what he has said.


Oh I don't know.... Maybe managing development of technologies that have shipped in a few billion products? Three of them won this kind of award:


Indeed, you probably have trouble putting together a system without it using a technology coming out of my team. To answer the next question of personal involvement, there are many but you can look the list of patents for a bit of that: http://www.boliven.com/patents/searc...Majidimehr%22). And this little book: http://www.amazon.com/Optimizing-UNI...2647103&sr=8-1

As a marketing guy, you really don't want to ask open ended questions like this from the other guy, do you? But no, I have been fortunate enough to have a job that mirrors what I talk about here so in that sense, I am not wasting time here.

Well obviously he connected one of them backwards! Audioquest RCA cables are clearly marked which end is the input and output side. Hook them up backawrds, well it will just sound like crap

I'm troubled by the thought that we can't trust JA's jitter measurements to guide us in purchasing decisions because performance varies so much from unit to unit. If there is such a great variation from unit to unit, perhaps such measurements cannot be used to support arguments on the forum either.

By the way, did the DUT in fact have a different power supply? Or was it a SMPS that automatically adjusted to voltage and line frequency, eliminating the need for different design and components for models sold in different countries?

further to that AS, just out of sheer curiosity...has anyone ever done a test on (say) ten units 'purchased' from ten different places??

What IS any normal unit to unit variation? How important might that variation be?

Now that would be interesting, and with more measurements than just jitter. Be an expensive proposition though.

I inverted the terminal's polarity in one of he left speaker conectors

Yes, it is sad but it is not a rule. Many objectivists keep a decent debating level.

Because I own a lot of SACDs that I used to play in an old Denon 3910 multiformat player that died a while ago. A friend of mine offered me that Marantz and I only paid 1200 dollars

This is the same case of my Pass Labs amp. It cost me only 3700 bucks, but I refuse to say how I got that incredibly low price

No. They are Monitor Audio Silver RS8 (that I don't own anymore), B&W Nautilus 804 and Revel Performa F32.

I put them together just to make noticeable the difference in size.

You want to see pictures of my cables? Acoustic research and Phoenix cables. The most expensive is less than 80 dollars and the average cost could be some 40 dollars or so

yeah, but I put purchased in inverted commas..are they always purchased or does the manufacturer lend them for review?

If you asked for 'ten', you'd like them to be random, hence the 'different locations'.

more a thought experiment I guess. It might help put things into perspective (the random variation could be greater than any specific figure as an example) but COULD also explain why one can hear things and not another?

I'd love Johns thoughts on the possibility.

There is a version of above test already. Namely, left and right channels in stereo systems. Often you see channel variations. Here is an example from airport express:



I have two identical Onkyo AVRs. Alas, one is in one house and the other, at our primary residence. I am too darn lazy to rip the one out and bring it home to compare . Maybe I will try it one day.

So is it audible at normal listening levels? If not, your troubles deciding what to purchase are over.

If amirm is to be believed, it can be when the jitter is above 500ps. With that litmus test in mind, if JA measures a DAC at significantly below 500ps and Paul Miller measures that same DAC at significantly above 500ps, and the discrepancy is attributed to normal unit-to-unit variation, does the DAC go on the short list or not? What if only one of them does the measurement?

Better yet, what if more do jitter measurements.

Seems the spec could be turned into a marketing opportunity.

In the case of the Nagra CDP. this wasn't "normal unit-to-unit variation." One was a UK model designed to be powered from 50Hz 240VAC; the other was a US model for 60Hz 115V operation. The difference in the jitter measurements was the appearance of high-level sidebands at +/-50Hz in the UK sample.

This is obviously related to the way the power supply is engineered. Perhaps the transformer was incorrectly specified and its core saturated with 50Hz but not 60Hz? (With a full-wave rectified power supply, interference at the powerline frequency tends to be magnetic in nature.) Perhaps the HiFi News sample was from pre-production and Stereophile's from actual production with beta errors fixed? I don't know, but this is not "normal unit-to-unit variation."

John Atkinson
Editor, Stereophile

Would someone be so kind as to outline the method by which 'jitter' is being measured by each source of information in this thread?

Try this post: http://www.avsforum.com/avs-vb/showp...postcount=2332

I'm more interested in the test setup. Was the device fed with real random jitter? Was the device sent frequency correlated jitter? was the device fed with LSB toggling method to trigger jitter? etc.

My understanding is it was the last, Xianthax. Amir is of the opinion that's realistic due to the complexity of music.

time to re-boot this thread, it has relationship to the "Are audio companies all involved in a huge conspiracy?" recent thread