Hmmm - scratches chin...
-
Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Macro Photos of Record Grooves
- Thread starter ray_parkhurst
- Start date
Thinking about it some more, a video feed and turntable wouldn't even be needed. A single hi-res image of each side of the full record could be used. The AI routines could still read the grooves from the image and do the same processing.
How about a slot loading record scanner capable of scanning both sides at the same time?
That would get us close to eliminating almost all the drawbacks of playing vinyl records. No need to get up to change sides. Could have pause, next and previous options.
somebodyelse
Master Contributor
- Joined
- Dec 5, 2018
- Messages
- 5,870
- Likes
- 5,573
How many DPI will it need for mechanical-equivalent performance? How much does a scanner with that resolution cost?
Good question. Some document imaging software I wrote over 15 years ago worked with 4800 dpi scanners. On the other hand the OCR scanners we used were only 300 dpi and could handle very small text.
My phone can take 200 megapixel images. I'm sure the technology already exists to capture the grooves with more than enough resolution.
somebodyelse
Master Contributor
- Joined
- Dec 5, 2018
- Messages
- 5,870
- Likes
- 5,573
http://pspatialaudio.com/stylus_grooves.htm gives maximum groove modulation of 76µm pk-pk, 16 bit = 65536 levels. If I didn't cock up the calculation that's equivalent to ~2M dpi, 4 terapixels per square inch. If your 200 megapixel sensor is square that's 143 photos per linear inch which could be automated. It seems plausible but nontrivial unless there's some optics problem.
You don't need 65536 pixels per cycle. If with the 2d image we can measure the depth of the groove (we can't) then sampling theorem can be used and only 2 samples per cycle is needed.
But here is the problem - 2d imaging is insufficient. We need to detect the 3d shape of the groove.
I was thinking about that. A slightly angled image should be usable since we can know the measurements of the image we would be using and can then calculate the depth of the grooves.
Or just use laser scanning, but not by trying to emulate a stylus over a spinning turntable. Do a complete 3d scan of each side of the record. Then process that data and extract the audio.
Looks like it is already being done:
arrival3d.com
Looks like it is already being done:
Flipping the LP Record Wax With DJ 3D Laser Scanning
Those old LP records have made a comeback and now they can be played without even touching the wax on a 3D laser scanning device.
arrival3d.com
Phonograph records made from vinyl have been mass-produced since ~1950. Surely some type of rigorous study on the wear and life span of a record has been done by now. I'm sure Columbia and RCA did this when the format and materials were developed.
Previous research on this topic may be buried in manufacturer archives and be somewhat inaccessible. I wouldn't be surprised if the first marketing of 'perfect sound forever' was actually done with vinyl records long ago...
Please do and post somewhere (here at ASR would be nice) a literature-search on the topic. A literature review of previous research is a prerequisite to real scientific research. I'm sure this research you want to do has been done long ago.
Previous research on this topic may be buried in manufacturer archives and be somewhat inaccessible. I wouldn't be surprised if the first marketing of 'perfect sound forever' was actually done with vinyl records long ago...
Please do and post somewhere (here at ASR would be nice) a literature-search on the topic. A literature review of previous research is a prerequisite to real scientific research. I'm sure this research you want to do has been done long ago.
It doesn't matter if previous studies took photographs of record wear or measured in some other manner. You want to research the life-cycle of vinyl records. This has been done. Please post the previous literature you want to ignore just because it did not use your photographic methodology.
The photographs you take won't get around that fact that your photos are a 'sample size of 1.' It is statistically invalid to base a result on a sample size of 1. Your pictures may be interesting, but unless you apply true scientific rigor, they are not really even a useful data point.
The photographs you take won't get around that fact that your photos are a 'sample size of 1.' It is statistically invalid to base a result on a sample size of 1. Your pictures may be interesting, but unless you apply true scientific rigor, they are not really even a useful data point.
Yup, science is hard. And you ain't doing any.
It'd help if you'd "reply" to the specific posts you're commenting on so people know what you're talking about.
OP
ray_parkhurst
Active Member
- Joined
- May 18, 2023
- Messages
- 109
- Likes
- 102
- Thread Starter
- #95
Recent posts deleted as all I get from this forum is grief from people who have no idea what "science" is.
For folks interested in following my experiments, go to the various SWS threads on VE and SHF.
For folks interested in following my experiments, go to the various SWS threads on VE and SHF.
Wow, I feel like I've ascended to another level at ASR. I made the groove-photographer mad and he deleted his posts and ran back to the SHF. I did this all by myself, without the usual ASR method of ganging-up on subjectivists.
His response to my 1st post was that in his literature searches no one had done anything remotely to what he plans to do.
His response to my 2nd post was "I remember why I don't post here."
Did I destroy the entire thread? I gotta admit, I'm more proud than sorry if I did.
His response to my 1st post was that in his literature searches no one had done anything remotely to what he plans to do.
His response to my 2nd post was "I remember why I don't post here."
Did I destroy the entire thread? I gotta admit, I'm more proud than sorry if I did.
BTW, @ray_parkhurst , you can always put me on 'ignore' instead of destroying the thread that some folks find interesting.
OP
ray_parkhurst
Active Member
- Joined
- May 18, 2023
- Messages
- 109
- Likes
- 102
- Thread Starter
- #98
Yes, you did destroy the thread since the OP won't be posting here any more, though you could continue posting if you like.
I am always mystified by "scientists" who have little understanding of the scientific method yet are quick to judge others who are applying it. You and others on this forum seem to believe that "science" demands literature searches and studies, large statistical samples, and covering all known and suspected variables, but none of those things are actually required by the scientific method.
I don't ignore people since it gives them license to post without my knowledge. If you were worried about destroying the thread, perhaps you should have been more careful, though you do say you are "proud" to have destroyed it. Good job.
Happy holidays. Sheesh... 
If people do want to follow this 'science' further, here's the link for @ray_parkhurst's thread at the Steve Hoffman Forums. He's even getting grief about his methodology over there...
Last edited:
Similar threads
- Locked
