Your Vinyl Transfer Workflow (sharing best needledrop practices)*

I don't think with my tinnitus I can hear it either way but if I can denoise with no harm I'm up for trying.

 

Ok, I forgot about that. I have RX 8 and they changed the behavior in that. In previous versions, if you move the quality slider to B, C, or D, certain options gray out. Try the denoise on A. The quality isn't bad. I know we all lik to go for maximum quality, but sometimes it's overkill. Using A, the denoise is actually a bit stronger, so you can back off a bit in the strength.

 

So here's my 2 pass run using pink noise at -85db I must be misunderstanding you about the 2 passes. But it sounds pretty good and the wave gets much cleaner.
Dropbox - Herbie Hancock sound system2 passes.flac - Simplify your life

Also are you sure going to A quality is worth getting use of the release time? It processes very fast.

 

S

So adding noise is a Jedi mind trick. Because this is blowing my mind. Is this what they say with tube gear adding noise and the result in a more musical sound?

 

Stefan, you are correct about professor Brian Davies, the mathematician and statistician, who developed ClickRepair. More on that in a moment. I purchased ClickRepair from him back in 2005/06 time frame. About that time, I sent him a mono recording of some American rockabilly, a 45 I wanted to add to a compilation dub CD I was making. The recording was noisy and needed to be cleaned up somehow, but the recording had this magnificent soundscape or aural sound stage that added immense sparkle to the recording. Once I processed it through ClickRepair, all of that wonderful space was eliminated and the recording sounded dull. So Dr. Davies asked me to make specific runs and to send him those results. All I know is that over the next few years, he found a way to eliminate those processing effects. Further, he worked with some other audio gurus along the way to really ramp up the capabilities of CR as a tool. If you don't know the back story, he used his considerable skills to write code for multiple math/stat approaches to identify issues to be corrected in his conversion of his classical record collection to digital. The one area that I have elected to avoid using CR is on electronic music, with at least back then gave CR fits with false reading to repair. But I think the program is simply brilliant.

I communicated with Brian Davies last night actually (and before I saw you mentioned him). Asked him about his distribution platform and updates.

This he wrote me: You need to keep good backups of the software and the license. I am too ill to continue and I can’t speak on behalf of where things might go. Nor can I run the old site or sell new licenses.

When I responded, I said this: ...you did a stellar, magnificent job on and with ClickRepair. I will forever be grateful.

He replied because he remembered that I use a MacBook Pro: Thanks for that. The Java “jar” file ClickRepair.jar will run on any platform using Java 1.8, license is also cross platform.

So yeah. We are may soon lose a legend in the vinyl-to-digital community. Dr. Davies has always been generous with his time, at least with me.

 

We love distortion, of the right kind, in the right places. This is entirely true.

 

So sad to hear about Davies. Actually, I was one of those folks sending him ideas along with pbthal. I recall once sending him the suggestion of adding a reverse mode to Clickrepair and he mentioned that pbthal had just sent him the same idea, so I guess he and I were on the same wavelength! His other programs for denoise and the free EQ program were quite good too. I always enjoyed reading his manuals as they weren't just step-by-step instructions. I always remember his description of denoising as being similar to trying to remove cream from coffee!

 

Well tubes actually add harmonic distortion and a bit of analog compression as well as smoothing transients. I sem to recall our host saying once that a given release that had been recorded with tube equipment would have sounded like mush if it were cut through all-tube cutting equipment.

 

I tried this on another drop and it did not work so well noticeably dulling the sound. My be a result of the A quality. Trying it again in C

 

Did I understand you correctly you ran the same setting 2 times?

 

I experimented with De-click and I agree that it must be used on the low, low end of the settings. Last night, I took a section of music that had obvious clicks. I ran a simulation, listened to the removed clicks, did it again, listened to the music. Went back and listened to the original track. As I ramped it up, I did get to a point where I could hear that De-click was removing music.

 

Thank you then for that suggestion! He did a great job with it, and that is how I process using ClickRepair.

 

I've never had A settings dull the sound too much, it just seems a bit stronger. In fact, I find I have to increase the strength when I go to C or D.

 

The settings in the module didn't change, but I processed then relearned the noise sample before applying the second time. After learning, I then adjusted the strength until the yellow line again lined up with the higher frequencies of the orange line.

 

It always depends on the material. On old records in bad shape, I've sometimes run Clickrepair at 36. It doesn't dull transients as much as you'd think if the material is an older analog recording of acoustic instruments As you mentioned before electronic music doesn't do so well There's the famous sawtooh intro to Van Halen's Jump that Clickrepair absolutely demolishes! I also ran into problems with one of the early Cars albums. I've noticed similar problems with declicking in RX.

 

Same here. Surprisingly, it doesn't seem to make any difference in RX. I've tried reversing a file )Shift+R) then applying declick, but it doesn't seem to make any difference. I really think the two programs approach declicking quite differently.

 

Just like a good tube guitar amp.

 

I'm glad you brought this up. I almost never use a noise sample taken from the beginning of an LP side to denoise a whole side as the resonance is usally stronger at the beginning. I'll typically use the gap between the first and second track, and sometimes I'll even relearn for each gap.

That reminds me of sdvice the Clickrepair author Brian Davies wrote in one of his manuals, that denoising before declicking results in better declicking.

 

Sorry about deleting my earlier post. I've just been at this too long today, I think this may be a huge step forward. I just cringe using the A quality twice. I will probably upgrade I'm going to do it sometime.

 

Don't sweat it. Is it that A quality is actually giving you bad results or is it just because you know it's the lowest quality setting? I often use A and never find it to be bad.

 

[QUOTE="Stefan, post: 26258663, member: 8892”] Brian Davies wrote...that denoising before declicking results in better declicking.[/QUOTE]

Absolutely. De-noise removes the grunge, opening up the music in the track. Makes it so much easier for ClickRepair to detect, isolate, and scale the actual click.

 

I just sometime jump to conclusions only to have to change my mind the next day. Here is a sample I actually eq-ed a little bass in it thinking the record needed a little. The 1st is the raw sample I bumped the gain up trying to match the levels. I very interested what people think.
Dropbox - Frank Zappa Hot Rats denoised.flac - Simplify your life

 

That sounds really nice.

 

The following is an interesting video I just read about that features RX, Ozone and some other plugins. Although it's not all relevant for needledropping, some of the new rebalancing technology in Ozone and RX could be useful for restoration. It's definitely amazing how far this technology has come.

 

Here's one for the Motown experts. I have this 80's Motown compilation double LP called 25 #1 Hits From 25 Years.


It's not bad soundwise: all stereo except for a mono version of The Temptations My Girl and the subject of this post, a strange sounding version of Papa Was A Rollin' Stone. My first instinct was to consider it fake stereo but if the guy who mastered it (John Matousek) did fake stereo for it, why not also for My Girl. The thing is, it doesn't sound like normal fake stereo in that there seems to be some stereo reverb and slight separation. However, it's also not the usual stereo mix of the song. I'm tempted to just fold it down for finishing my needledrop but I thought I'd throw it out there. It also might be the mono mix in the left channel with some of it mixed into the right with EQ to cut the highs along with some reverb in the right channel. I know Grant knows his Motown quite well but I open this up to anyone. Thoughts?

Here's a sample: Dropbox - Strange Temptations.wav - Simplify your life

And here's a mono version using the left channel only: Dropbox - Strange Temptations Left.wav - Simplify your life