Your Vinyl Transfer Workflow (sharing best needledrop practices)*

Has anyone had success using RX Standard to fix sibilance?

I know the immediate answer is "No, dummy. Get a better pressing or tweak your TT/cart". However, those don't seem to be options in this situation.

I have a record with sibilance on a few songs, so I played it on a different table/cart. Still there! So then I got a totally different pressing (from Japan!) with different matrices. Same sibilance on that one, on both 'tables.

I normally hate any such "revisionism", and even stopped myself from repairing a tape drop just yesterday. But sibilance is just sooooo annoying

 

@Grant
Do you swap the channels in the Voxengo VST or outside? I chose the swap channels in the VST and it does weird things with the signal, like attenuate the file. I was able to gain normalize after combining the M/S back to L/R. Auditioning after all the monkeying around doesn't seem to show anything weird.

 

I've had better luck fixing sibilance with other programs like Sound Forge. I can do a good job with RX but it really takes too much trial and error than I have the patience for.

 

I don't use Voxengo.

 

I've had great success fixing sibilance in needledrops by switching to Mid/Side then using the Spectral Repair Attenuate at a strength of 2 on patches of sibilance in the side channel only. In my experience, most sibilance on records occurs in the Side channel. This is not always the case but most of the time, it is. RX also has a decent De-Ess module, but what I like about fixing siblance in the Side channel only is that the original sibilant is left intact and not muffled.

As for revisionism, don't sweat it. I fix tape drops when I can, if it can be done without sounding fake. You're doing this for fun and personal use, not to sell to anyone as being the real master or anything.

 

Do I have mid/side on RX 7 Standard? I don't see it. Seems like it should be an option under "Mixing".

 

Should be in the drop down "MS Encoder" and "MS Decoder"

 

See my explanation in post #3452

 

I don't see that, unless I'm looking in the wrong spot?

Thanks!

 

I went ahead and bought Ozone 9 Elements (only $20).

I guess I still don't understand Mid/Side. If I highlight a selection, pick "Mid", then render, now what should I be doing. Run de-click (for example) on the Mid, then re-render the section as Side?

 

Sam, under Utility select MIXING. In the MIXING dropdown select MS DECODER or MS ENCODER

 

OK, I can make a preset for MS Encoder and MS Decoder. You're showing 50/50/50/-50 for Encoder, what about for Decoder? I thought some people were using 71/71/71/-71?

 

Change the settings to 71 71 71 -71. They worked better for me than the program presets of 51

 

If you use the 50/50/50/-50 ones, you need separate presets for going to and from MS. The 71/71/71/-71 preset I use acts like a toggle in that it goes both ways. As I wrote previously, I tend to be keyboard oriented, so I have this preset assigned to Ctrl+Shift+M and use that combo to toggle back and forth. The one thing I would caution about this is that you need to do the entire file or at least a selection starting and ending at gaps between tracks because the borders of the selections can result in artifacts if you select a portion of a song. Once you do it a few times, it becomes really second nature.

 

Which tools work better in LS mode than in regular LR mode? I'm talking about RX.

 

Thanks. Do you use the Ozone plugin? I'm on a Mac, and am having trouble with Presets, so I'm wondering if I can use M/S, exclusively using the Ozone plugin.

Slightly different topic: a few things I've learned... For 'drops of electronic music (New Order, Radiohead), even the lightest de-click results in transient loss when I listen on my main system. So I run a "output clicks only", and find sections that don't look like any music beats, let it process those sections. Then I do some manual de-clicking. I'm getting better at. I'm not trying to zap 100% of clicks. If they're not obvious listening my Audio Technica ATH-A900X headphones while monitoring via my Apogee Duet, then I leave them be.

For regular rock music, I do OK with a de-click of 1.5.

 

In extreme cases of clickiness, I will go through the entire file second by second and deal with each click in different ways, even going as far as literally editing it out if need be. So, going through a four minute file can take hours.

 

That's one of the reasons I waiting so long before jumping in to needle dropping. Editing on song/side for 30 minutes takes the fun out of the end result. Different strokes, of course.

 

To me, it's the process that I like. It's the satisfaction of knowing that I created something that sounds so pristine. Diff'rent strokes, and all that.

 

Our processes are undoubtedly different, but you've put your finger on some truth. I get great satisfaction out of the process I follow. When I'm in the midst of it, the feeling for me is very similar to the feeling when I've been in the midst of writing good code.

 

I must restate my thoughts a bit better. I am not great at creating things, but have a talent for improving on what others have created. In this case, I can improve on something that exists, and that is a noisy recording. My personality is that i've never really been completely satisfied with whatever has been presented to me. I always look for ways something can be improved or fixed. Everyone's different, though. Some people are happy with authority, or decisions that are made for them.

 

Good points. However I stand by my tape specs of 1960's tape machines. They really had a
lot of trouble recording the last half octave. Not talking about a brick wall filter. They really couldn't record anything past 15 maybe 16 khz. The heads and transport designs of that era just couldn't pull it off. Not to mention that there was no good tape until 1970. The first good modern tape was the Scotch 306. Ironically it would turn out to be a pain in the butt suffering from SSS along with it's sister tape the 307. (307 echoes for 4 seconds and then we hear wind and a wolf howl. Fade to black) It wasn't impossible just very impractical. The unsteady tape transports made the last half octave almost impossible to record if at all.

So the Studer J-37 was rated at 30 - 15 000 hz +-2db. But the specs didn't contine as if 16 khz was at -3 db and 17 khz was at -4 db, etc.
If you tired to record a 18 khz tone on a
Studer J-37
Ampex MM-1000 16 / 24 ***
Ampex 300 - 4
Ampex 300 - 3
Ampex 300 - 8
Scully 284 - 8 **
Scully 284 - 12 (only 60 made!) **
3M M23

you wouldn't get any tone back at all. That simple. You want proof? The Ampex test tapes of the 1960's were only 30 - 15 000 hz. If the machines could record and playback higher tones then why limit the test tapes? And also take a listen to the 15 khz tone at the end of Pepper. Do you hear all that noise and buzzing around the tone? Even at the rated spec it was having trouble.

Excellent post. I agree. But the cutting lathe just had a big problem with anything above 18 khz.
Or at least 20 khz. It wasn't a cut off filter. NO. if you to cut a 22 or 23 khz tone onto vinyl you would just get buzz noises back. 19 khz dip - 10 db at 19 khz and - 14 at 20 khz. This is not a brick wall filter just the character roll off of the cutting lathe. You will not hear anything at those levels.

Most 24 tracks were 40 - 20 000 hz +-2db @ 15 ips between 1971 - 1984. ¥
30 - 21 000 hz +2db / - 3db
25 - 22 000 hz +2db / - 5 db
After this things really dip. The machine just won't play back super high tones. It is not aligned and calibrated for that. 25 khz might be there but you ain't gonna heat it at - 12 db. Try and record a 20 khz tone on a 2 head $160 deck back in 1983. You will get back hiss. Same here.

One of the most popular mics used to record guitar amps, snare and toms are the Shure Beta 57/58. They are designed on purpose to roll off around 80 hz and 15 khz. I have owned several.
The famous original Nueman vocal mic rolled off sharply at 14 khz. Why? When it was designed no tape machine could go above 15 khz anyway. And the company found that rolling off the top octave just made the mic sound better. For singing anyway.





**
Damn you! O.k. at 30 ips the Scully machines in question could do 50 - 18 000 hz +-2db.
But these speeds back then were only used for copying. Mostly...At night mostly. And that would murder your bottom end which no one could hear back then anyway. Mmmmm??

*** God, what a piece of garbage!
The Ampex MM-1000 - 24 track was released in 1969. The same year as the superior and better behaved MM-1000 16 track. Unfortunately the MM-1000- 24 had a problem with it's head stack cracking. So Ampex solve the problem by replacing it with a staggered head design.

It goes like this. Ampex thought, If those guys at Scully could build a good sounding 1 inch 12 track then surely we here at Ampex (who know way more) can build a great 2 inch 24 track. Right? 2 inch transports and tape already exist for computers and video recorders. Hell, all the 2 inch video recorders are built by us anyway. (Any worth having)

The Scull 284 -12 had the same specs as it's smaller 8 track old brother. The 284 -12 was only one db noiser than the 284 - 8. You double the track count for 24 tracks but you double the tape width as well. What could possibly go wrong?

One small issue. The geniuses at Ampex had never built a 1 inch 12 track before. Apparently the weight or design of those early 2 inch 24 track head stacks caused it to CRACK. Eventually Ampex and other tape designers figured out a way to make a non-staggred head stack that didn't crack. 1969 was probably too early for 24 track, but those early years of development, and design (1968 -1970) led to the modern reliable 24 track of today. Elvis Presley's - Aloha from Hawaii was recorded with an MM -1000 - 24. Not bad....Not too bad. The film sync pulse is on track 24. Leakage is a bugger with that machine.

Although it might be another model. Thing is in 1969 no one else was making a 24 track. And the MM-1100 I believe was the upgrade to the earlier sucky model. Not until 1970.

And it was noisy as sheeet. The new and better Ampex MM-1100- 24 had a better design. Or so the manual says! Studios didn't really start using 24 track until 1971. This is when tape machine technology really took off. DO NOT buy a 24 track recorder before 1971. DON'T. Or I will cry, wet myself (52. Doing that a lot these days.), hold my breath and pound the floor until I turn purple. So go ahead and try it....I dare you!



¥ exceptions:
The Ampex ATR124 (1979)
25 - 22 000 hz +1 / -2db @ 15 ips
40 - 24 000 hz +1 / -2db @ 30 ips

The Studer A800-24 Mark 1 (1973)
30 - 20 000 hz +1 / -2db @ 15 ips
50 - 22 000 hz +1 / -2db @ 30 ips
Which by the way was the frequency response spec of the leading 16 track machines back then. Yep Yea!! (15 ips spec)

 

As a follow-up to what I was explaining about fixing distorted sibilants in RX while working in Mid/Side, I thought I'd post some samples from a rather nasty pressing of the debut Dire Straits LP I own.

Here's a section from the song Wild West End. Lots of nastiness here!
Dropbox - sibLR.wav - Simplify your life

After switching to MS, you can really hear that most of the distortion is in the Side channel (plays in the right channel on playback)
Dropbox - sibMS.wav - Simplify your life

Here's the MS version after I went in and fixed them with Spectral Recovery's Attenuate.
Dropbox - fixedMS.wav - Simplify your life

And here's what it sounds like once you convert it back to LR. The sibilants sound much smoother but haven't lost their natural sound the way they do when treated with too much de-essing.
Dropbox - fixedLR.wav - Simplify your life

 

I'm a newbie to the thread, 10 years later, and I've just read a shedload of the pages to try and get up to speed.
Thanks for having me, and hoping I didn't miss this answer already in the previous pages.

My current setup:
Pioneer PL-50L table
Denon DL103R cart (modified alum body and paratrace stylus)
Lehmann Black Cube SE phono
and currently ....the Berringer UCA222 into a laptop via USB.

OK - I think you see the potential weak link here.

Purpose:
I want to needledrop single album tracks only, with zero or minimal change to the raw sound output from the above chain, straight into wav's for use in videos on YouTube.
So I'm not looking to color the recorded files at all - the less the ADC adds, and the more it shows off the quality of the original chain above it, the better.

So given I'm agnostic about the ADC after the Black Cube, be it either external or soundcard based, what would you recommend?

 

Welcome to the thread and the hobby. I highly recommend the Steinberg line of interfaces, which are made by Yamaha. I've been using the UR44 for the past few years for my needledropping PC and I recently picked up a UR22C for a second computer. These have great specs, low latency, good ASIO drivers, are built really solid and are reasonably priced.