Phono Cartridge Response Measurement Script

[DrCWO]

What is the best Shure V15 version

The measurement above is from V15MR with original NOS stylus. Reading in the forums V15MR is better as V15xMR made in Mexico. But I never tested as I'm happy with my V15MR.

I moved to a new place last year and now say the the Azimuth got misadjusted during transport. With correct azimuth the crosstalk is same for both channels.

 

[JP]

Also checked with Chrome and Firefox. WAV file output is always perfect, FLAC differs between browsers and between WAV.
Seems the Audio API implementation of the browsers is buggy (and different) in all browsers for 24/96

Then nothing can be supported other than WAV. There are no python native decoders for FLAC. EDIT I think there's a JS module. Can look at that later.

 

[JP]

Runtime errors along with the traceback are being passed back to the UI now instead of it just processing forever. In the traceback dropdown ("Show Technical Details") is a copy icon to make life easier.

 

[DrCWO]

Then nothing can be supported other than WAV. There are no python native decoders for FLAC.

Client Challenge

pip3 install pyflac

 

[JP]

Client Challenge

pip3 install pyflac

It's a wrapper for libsndfile, so not native Python.

I'm reverting back to WAV only support. JS implementation adds more complexity and dependencies. Maybe another look later.

 

[DrCWO]

I'm reverting back to WAV only support.

I understand that you like to get a stable release first. But please keep it in Mind
It is so cool that you made it run in the browser.

 

[DrCWO]

B.T.W. I have contact to all the German HiFi papers.
If the whole thing is up and running reliably I can offer to ask them for a promotion in their publications.
Would you like me to do that?

 

[DrCWO]

Found this FLAC decoder without any dependencies.

https://www.nayuki.io/page/simple-flac-implementation

And here the source code. Quite quite small but should be sufficient for the task.

https://www.nayuki.io/res/simple-flac-implementation/simple-decode-flac-to-wav.py

 

[JP]

It is so cool that you made it run in the browser.

It was on my winter projects list to package it in a self-contained executable. Then I briefly thought about porting it to JS, but that's a lot of work and I didn't want to maintain two code bases. Also, python is very easy for people to hack around with and modify to do specific tasks. JS not so much.

It's also nice that this is hosted on GitHub Pages right out of the repo.

Would you like me to do that?

That's great if you want to help make people aware of the tool, and hopefully increase library contributions. I've no interest in any interviews or anything like that.

 

[JP]

Found this FLAC decoder without any dependencies.

https://www.nayuki.io/page/simple-flac-implementation

And here the source code. Quite quite small but should be sufficient for the task.

https://www.nayuki.io/res/simple-flac-implementation/simple-decode-flac-to-wav.py

"around 100× slower than the Java code" Ouch.

 

[DrCWO]

That's great if you want to help make people aware of the tool, and hopefully increase library contributions

Please get me a "GO" to start that.

 

[DrCWO]

"around 100× slower than the Java code" Ouch.

But the file is only some few minuts long...

 

[JP]

I'm trying to optimize - seconds add up quickly. I've an early proof of concept version where AI re-wrote the plotdata functions. That one could spit out a two-sweep plot in about 2.7 seconds. Figuring that out is the next step after this web version is stable enough.

 

[JP]

Okay it's now at its permanent home: https://sjplot.com/online/

I think just some UI clean-up and it's ready for its debut.

 

[JP]

@USER I added a CTA for you that pops after analysis successfully completes.

 

[USER]

@USER I added a CTA for you that pops after analysis successfully completes.

Oh, that's nice! Thanks.

 

[somebodyelse]

"around 100× slower than the Java code" Ouch.

Entirely expected - there's a reason the heavy lifting is usually delegated to native code rather than pure python. If there's a way to call out to something in WASM you might get near-native speed.

 

[dlaloum]

The measurement above is from V15MR with original NOS stylus. Reading in the forums V15MR is better as V15xMR made in Mexico. But I never tested as I'm happy with my V15MR.

I moved to a new place last year and now say the the Azimuth got misadjusted during transport. With correct azimuth the crosstalk is same for both channels.

Yes the V15MR is superior to the later V15xMR - the xMR is basically a M97 family body with a V15MR stylus fitted - it uses the wider stylus fitting of the mass market models, and the generator is a lower spec generator too.

As the primary determinant of performance is the stylus the xMR is still excellent (and superior to pretty much anything currently made...) - but the straight MR was the peak of the Shure range, when serious R&D was going on. (debatably, the Ultra400 might have matched it, and was originally designed to replace it! - the Ultra500 was a rebranded V15V)

 

[JP]

UI is usable and we're feature complete for this simplified version, so over to y'all to find wonderful ways to break it. https://sjplot.com/online is its permanent home, so no issue for anyone to share that.

@DrCWO no trade push just yet - let's let it gel for a bit.

I also want to add GA4 to it and expose a page of stats so anyone can see how/where/what. Thought that'd be fun. Or depressing. We'll see.

 

[USER]

Updated the main library page. Do let me know if I can word things better. This is a game changer!