I thought Wayne's recommendations for RIAA plugins (and da bad ones) were on his supa dupa preamp threads (which Bill has Shanghaied by claiming he found them)
I cited you first time I mentioned them
But futher links were to things you HADN'T mentioned. Just in case you missed ithttp://www.diyaudio.com/forums/analogue-source/298896-digitizing-vinyl-9.html#post5053286
Sorry if I missed protocol somewhere but I never claimed I had found them.
Note from the forum janitor. Keep the forum clean! Please do NOT quote the entire post just above yours. See the rules section. Extraneous quotes have been removed.
There was an interesting interview on Radio 3 this morning with Andrew Rose of Prisine classical, discussing various things around music restoration and playing some samples. What was particularly interesting was how he said that he works with the sonograms on the PC and its the visual input that really helps home in on what needs to be done to the recording. Still not totally sold on the ambience addition that they do, but apparently people like toscanini insisted on really dry recordings.
I've noticed that a lot of old symphony recordings are rather dry. Did Arturo and his contemporaries want to her the sound they knew at the podium, rather than the sound of the hall?
I have an old Denon CD with anechoic recordings of a symphony orchestra. It's odd sounding, and a number of people I've asked to listen to it thought it was a pre-war recording.
I have an old Denon CD with anechoic recordings of a symphony orchestra. It's odd sounding, and a number of people I've asked to listen to it thought it was a pre-war recording.
Some more information on the process they use here https://www.pristineclassical.com/pages/xr-remastering
Also have a listen here https://www.pristineclassical.com/collections/artist-louis-armstrong/products/pajz006 which is a restored acoustic recording. I have to say mightily impressed with what they can do these days with the right software and someone who knows how to use it.
Also have a listen here https://www.pristineclassical.com/collections/artist-louis-armstrong/products/pajz006 which is a restored acoustic recording. I have to say mightily impressed with what they can do these days with the right software and someone who knows how to use it.
Not to mention, someone who is willing to do a whole lot of work. More on how its done here: Andrew Rose: Restoring Old Recordings |
Keith Snook has posted a book on sound restoration techniques. I couldn't get a link to the book to work.
~ Stuff you really should Read ~
~ Stuff you really should Read ~
That's quite interesting if you compare the sound on sound article from 6 years ago with the link I posted. He is still evolving the techniques. I hadn't thought about leaving the denoising until after the eq, but makes sense. He works closely with Mark Obert-Thorn so they are probably pushing the limits of what can be done. Certainly not something we could manage (sw costs alone for some of the specialist stuff).
Richard (kgrlee) seemed to be having difficulty finding the links I have to RIAA vst plugins. It seemed like a good idea to put them here in one place.
The first link may be the ones Scott wrote(?) for jiiteepee. They use SynthEdit Pro Audio Design Forum • View topic - RIAA VST Plug-Ins I got these from a client using the flat preamp to vinyl stream.
Olaf Matthes has also written some vst plugins that I've also installed in Audacity: Pro Audio Design Forum • View topic - More RIAA VST Plugins
I did want to correct the "record" (pun intended) that Diamond Cut 8's RIAA is not vst. It does sound accurate in casual listening. DC8 is available for a limited 15 day trial period for those that want to try it. I got busy and my trial expired before I could really dig into it. http://www.diamondcut.com
As an experiment I want to use AudioDiffMaker to compare the various 96 kHz vst plugins to analog RIAA.
The "Flat Phono Preamp Based on John's P10 and 2SK389" thread being cited here is a development thread and like most of my work meanders as I get from "A" to "B." In it I briefly become distraught at the thought of bias current in the cart. I got over it. http://www.proaudiodesignforum.com/forum/php/viewtopic.php?f=6&t=423
The final "production" balanced input flat preamp and RIAA monitor switcher designs are in this thread: Pro Audio Design Forum • View topic - Phono Transfer System
There is also an experimental ZTX851-based flat MC front-end design here that I really should finish: Pro Audio Design Forum • View topic - A Low Noise Balanced In Moving Coil Preamp Using the ZTX851 I throw a lot more current at the problem just to see how well the ZTX851's perform.
I saw a lot of referrals from this thread and thought I should pop in, say hello and link to the RIAA plugins.
Wayne
The first link may be the ones Scott wrote(?) for jiiteepee. They use SynthEdit Pro Audio Design Forum • View topic - RIAA VST Plug-Ins I got these from a client using the flat preamp to vinyl stream.
Olaf Matthes has also written some vst plugins that I've also installed in Audacity: Pro Audio Design Forum • View topic - More RIAA VST Plugins
I did want to correct the "record" (pun intended) that Diamond Cut 8's RIAA is not vst. It does sound accurate in casual listening. DC8 is available for a limited 15 day trial period for those that want to try it. I got busy and my trial expired before I could really dig into it. http://www.diamondcut.com
As an experiment I want to use AudioDiffMaker to compare the various 96 kHz vst plugins to analog RIAA.
The "Flat Phono Preamp Based on John's P10 and 2SK389" thread being cited here is a development thread and like most of my work meanders as I get from "A" to "B." In it I briefly become distraught at the thought of bias current in the cart. I got over it. http://www.proaudiodesignforum.com/forum/php/viewtopic.php?f=6&t=423
The final "production" balanced input flat preamp and RIAA monitor switcher designs are in this thread: Pro Audio Design Forum • View topic - Phono Transfer System
There is also an experimental ZTX851-based flat MC front-end design here that I really should finish: Pro Audio Design Forum • View topic - A Low Noise Balanced In Moving Coil Preamp Using the ZTX851 I throw a lot more current at the problem just to see how well the ZTX851's perform.
I saw a lot of referrals from this thread and thought I should pop in, say hello and link to the RIAA plugins.
Wayne
Last edited:
Hi Wayne,
Thanks for popping on. I have to say seriously considering building up the ZTX851 as vero board. There is something very satisfying about two 80p transistors and one 50p opamp getting to sniffing distance of fundamental limits.
And lots of other good stuff I want but currently have zero use for
Thanks for popping on. I have to say seriously considering building up the ZTX851 as vero board. There is something very satisfying about two 80p transistors and one 50p opamp getting to sniffing distance of fundamental limits.
And lots of other good stuff I want but currently have zero use for
Thanks JP for the kind words. Nice clean build!
I should mention that the jiitepee vst plugins based I link to work fine in Audacity 2.1.2 in Win XP. On my Win 7 x64 machine running Audacity 2.1.2 (32 bit) they do not.
Olaf's RIAA vst runs fine on both, is compact (about 136K) and also offers Inverse RIAA. There is only one plugin for multiple sample rates.
I think I recall seeing a discussion in this thread about flat headroom. I use about -23 dBFS @ 1 kHz @ 5 cm/s RMS lateral for my flat recordings. With LPs I typically see 4-6 dB of headroom (excluding clicks). -23 dBFS always seemed low to me but in practice its not.
The International Association of Sound and Audiovisual Archives provides a good reality check for flat recordings.
More take-aways from the IASA here: Pro Audio Design Forum • View topic - International Association of Sound and Audiovisual Archives
One thing I've wanted to experiment with is declicking in sum/difference. Most clicks are vertical. Clicks should be easier to detect in difference (particularly when recorded flat) and it seems like the artifacts from the repair would tend to be masked by the more level-dominate sum when decoded back into L and R. Sum/difference processing works well for warp removal - it might work well for click removal.
I should mention that the jiitepee vst plugins based I link to work fine in Audacity 2.1.2 in Win XP. On my Win 7 x64 machine running Audacity 2.1.2 (32 bit) they do not.
Olaf's RIAA vst runs fine on both, is compact (about 136K) and also offers Inverse RIAA. There is only one plugin for multiple sample rates.
I think I recall seeing a discussion in this thread about flat headroom. I use about -23 dBFS @ 1 kHz @ 5 cm/s RMS lateral for my flat recordings. With LPs I typically see 4-6 dB of headroom (excluding clicks). -23 dBFS always seemed low to me but in practice its not.
The International Association of Sound and Audiovisual Archives provides a good reality check for flat recordings.
More take-aways from the IASA here: Pro Audio Design Forum • View topic - International Association of Sound and Audiovisual Archives
One thing I've wanted to experiment with is declicking in sum/difference. Most clicks are vertical. Clicks should be easier to detect in difference (particularly when recorded flat) and it seems like the artifacts from the repair would tend to be masked by the more level-dominate sum when decoded back into L and R. Sum/difference processing works well for warp removal - it might work well for click removal.
I only gave him the coefficients based mainly on an ancient Bob Orban post from some DSP forum. At 44.1k or 48k one can get closer with two bi-quads. You can use the Nyquist prompt which should always work in Audacity on any platform. I've posted the error curve for 96k here several times +-.006dB with one bi-quad.
There is a great reference on audio restoration that I seem to have misplaced (heavy on the math). It was especially good on the recreation of the gaps, IIRC there was even some code examples based on it in a now abandoned Sourceforge project.
EDIT - It's least squares autoregressive interpolation and the pop finding was a statistical technique not spectral and seemed particularly good for pre-RIAA use. Still can't find the AES paper, Bill you said it was one of your profs do you remember?
EDIT - Never give up http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.50.6120&rep=rep1&type=pdf
Last edited:
Thanks Scott. Other vst plugins, including Olaf's do work on that machine. Only the jiitepee vst seem to break on that particular machine. Not sure why yet. It writes an empty file when rendered.
Has anyone posted a ready-to-use RIAA.ny file?
This looks to be an interesting resource for alternative curves: Playback equalization for 78 rpm shellacs and early 33? LPs - Audacity Wiki
Has anyone posted a ready-to-use RIAA.ny file?
This looks to be an interesting resource for alternative curves: Playback equalization for 78 rpm shellacs and early 33? LPs - Audacity Wiki
I have the pdf here. <google>
https://www-sigproc.eng.cam.ac.uk/Main/SJGSpringer
Looks like this is the book he wrote, plus some interesting examples.
Edit: the book, at 346 pages has a lot more in it than I had before. That's a few evenings reading sorted
https://www-sigproc.eng.cam.ac.uk/Main/SJGSpringer
Looks like this is the book he wrote, plus some interesting examples.
Edit: the book, at 346 pages has a lot more in it than I had before. That's a few evenings reading sorted
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.