That's why I mentioned other issues. Again, you need to consider the power of software to analyse, dissect information. If the program is comprehensive in its scope, sufficiently adept, then provided all the data is captured cleanly, then there's nothing to prevent the algorithm coming to the table from all sorts of angles, digesting and filtering the raw data in all sorts of ways. Two recordings with no-one in the room, nothing altered inbetween, will give a baseline on the intrinsic repeatability: as it is, the electronics will be in different states of warm up, stability, memory effects; as will the speakers.
After all, this is exactly why playback does vary with time, us humans are obviously sensitive to it, so what hope do we have of making sense of these audible artifacts unless we go to a very precise methodology to analyse the situation ...
Frank
Using highly precision instrumentation in a climate controlled acoustic lab environment, using calibrated devices would be the only way to accurately and precisely "measure" differences. Very costly. It also needs to be repeatable to be of any absolute true value.
Using instrumentation computers to run the tests and synchronize sweeps/test passes would be needed. Baselines would need to be developed/determined as references.
It would cost a lot of money to do this with very little learned or useful...IMHO.
Good luck!
Using instrumentation computers to run the tests and synchronize sweeps/test passes would be needed. Baselines would need to be developed/determined as references.
It would cost a lot of money to do this with very little learned or useful...IMHO.
Good luck!
And there is always Audio Diffmaker for synchronizing things and extracting the diffence. Bill Waslo, the author, is a member here.
My point was with exactly the same stimulous same gear, etc. there is a fundamental limit to repeatability. Averaging is fine to collect data and reduce the noise floor but in listening we have the same experience only within the one to next one difference.
You made a comment about subtle issues, I wish folks would even begin to say these things are subtle rather than so obvious we are deaf idiots.
Yes, I'm familiar with DiffMaker, though haven't used it. To my mind ADDC would need to be some variation on that theme, perhaps with other "smarts" to improve, or provide a different slant on interpreting the variations. From the POV of meaningful answers on audibility, it would be handy to have a good chunk of psychoacoustic understanding of how sound registers in the brain incorporated in such a program, so the finger can be quickly pointed to the offending differences. Perhaps down the track ...
Frank
Yes, I can extract differences with the software, but I'm not sure what difference they make! It can be hard to tell what it might sound like, tho the software does allow you to fold it back in to hear it.
LOL, yes. But when clutching at straws, catching one often feels like a big bamboo pole. Changes get mysteriously more subtle as time goes by. Maybe the emotion of finally catching that straw weighs more than the straw itself.
Yes I know this wonderful piece of software and I cited it in my research as a very good attempt of working out a solution to the alignment issue.
I came across to it roughly three years ago and I also kept in touch with Mr Waslo, who was very, very kind. To tell you the truth I felt a bit sad at first as I felt he had won the "measure race" but, fortunately enough, his path was completely different from mine, so the two measures are basically different in their approach to the problem.
Thank you but, as I said, before I add anything I would prefer to wait from a response from Mr.Pedersen (who is taking a quite long time, to be honest), as I'm sure he is the most qualified scientist to assert if I am wrong or not.
By the way, to get a first response from him any help from any of you would be really appreciated!
What I can say (and I proven this by comparing the results coming from my measure with the listening impressions of a selected audience) is that there are differences which don't have any effects on the listening pleasure, until they go beyond a threshold.
One of them is distortion (and this is literature) and another one is the acoustic environment: you can't imagine how the captured signal changes from one session to another just moving stuff inside an anechoic room (I had the chance of using one of them) or even not moving anything inside it. But the listening impressions don't change at all, and this is exactly what happens while listening to our domestic equipments.
If my wife or my children come in while I am listening to my stereo, my (and yours) ears are smart enough to not ear any changes. So we have to use good sense and match our impressions with the results we have.
The matter is to get facts and figure as precise as possible, to figure out what is important and what is not by a listening prespective, no being deceived by any approximations or misconceptions, and this is exactly the aim of the ADDC in the audio field, which can reach the precision levels shown in my web site thanks to it bans any approximation at every step (and this is proven).
What is needed is then an as precise as possible statistical base only focused on the measured phenomenon, not mixing any measure limits in an uncontrollable way, and you can guess I strongly hope the ADDC is able to do this!
At the same time, if applied in other fields, it could immediately measure what is crucial of a given equipment.
Sorry to read this: if it had been a deep dark secret, I wouldn't have sent my full report to Mr Pedersen. As I said, I'm waiting for a trusted response from AES, and in my opinion this is far more serious than let unknown people on a forum claim whatever they want about the measure details: I hope you understand this would result in a big mess.But it's a deep dark secret.
It took five years to me to work out the measure. I dont' think a few more time to check if it is trustworthy (or not) will be too much.
Lastly, have I to be blamed if I hope to get some sort of acknowledgment from it?
From your last few comments some might quibble with your exact meaning of "unmeasurable" in this context. Certainly two people coming in and sitting down near you is a rather easily measured effect. A metric that "knows" to ignore it is a lot different than measuring it or not.
I don't think the same logic applies to laboratory/medical instrumentation where the desired results do not leave the same wiggle room.
I don't think the same logic applies to laboratory/medical instrumentation where the desired results do not leave the same wiggle room.
Then why did you even start this thread?
I don't understand why you posted on a forum and announce your "measure" method at all then. People who develop real tangible engineering don't do that.
How exactly did you expect people in a public forum to react?
All your doing is posting a claim with nothing to back it up and nothing to show the community. It's like standing up at a scientific conference and telling your colleagues I have made a discovery, but I can't tell you anything about it!!! Ridiculous! So what do we do?.....it's a non-event...it really means nothing to us!!
That's why when people post threads like this (and there's another poster on here doing a similar thing right now), it either has no substance and it's BS trolling or something else is going on....talk is cheap. It has no value without substance!
Then perhaps it might have been prudent to avoid telling the world that you have something great and wonderful until you knew you had something great and wonderful.
Sorry I didn't realize you were not the OP.
Just callin em as I see em...I don't know you and you don't know anyone on here so all too often there is some vetting process that needs to take place before anybody "believes" new claims or theories. It's not like your a famous Nobel scientist who just discovered the existence of the Higgs boson particle.
Still....you came on here and took credit and the topic is discussing the validity of your test process, so you can't expect not to have to defend it with some clarity and substance we can sink our teeth into.
You still haven't!
I was not being impolite....that's not a trait you find much on open forums.
I was merely being honest in my observation and opinion about the topic.
That gives you a little insight as to where were coming from.
Need some meat with those potatoes!
from: http://www.clear.rice.edu/elec301/Projects02/empiricalMode/
...could it be that dabudabu have applied the "Empirical Mode Decomposition" (EMD) to create his "ADDC" measure?
Among the other things, the EMD have been proven quite effective in building special "filters" to extract weak signals from strong background noise...
Last edited:
Dear all,
sorry for my late reply.
I only would advise whoever could be interested that the full report of the ADDC measure is now available under a common creative non commercial share alike license.
If you wish you can have a look at its site:
Measure hardly detectable signals with a really high precision
I strongly hope the measure will improve thanks to our common support.
sorry for my late reply.
I only would advise whoever could be interested that the full report of the ADDC measure is now available under a common creative non commercial share alike license.
If you wish you can have a look at its site:
Measure hardly detectable signals with a really high precision
I strongly hope the measure will improve thanks to our common support.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.