don't want to take convo on a major detour but just to say you can do Multichannel recording in Audacity, its basic but does the job.
Ripping Vinyl - The Basics
- Thread starter GOS
- Start date
Help Support QuadraphonicQuad:
don't want to take convo on a major detour but just to say you can do Multichannel recording in Audacity, its basic but does the job.
Don't you have to set-up 4 empty mono channels first (for quad), something like that?
could be! my audio interface bus is setup in a 5.1 config by default, where 1:FL, 2:FR, 3:C, 4:LFE, 5:RL, 6:RR, so on my rig Audacity springs open with 6 channels and that seems to be the channel assignment/mapping that seems to be the most compatible with all sorts of dacs, receivers, computers, players and other bits of machinery whether its 4-channel vintage Quad LP or tape material being recorded in or not, so i've stuck with it like that tbh there may be a better way i'm far from expert on techhie stuff.
markshan
1K Club - QQ Shooting Star
You know, when I recommended the Scarlett 2i2, I forgot where I was. It only has two inputs. If you want to rip quad, you'll need to pony up a C more for the 6i6.
Scarlett 6i6
true!
i use a Motu Ultralite interface, a dinky little thing does 8 in/8 out @96/24, got a used one for a couple of hundred smackers from a local bloke on eBay.
we're technically in a non-surround part of QQ so i didn't want to detract from the original gist of Gene's thread which i think would be Stereo needledrops as he specialises in them but its all good chat i guess!
true!
i use a Motu Ultralite interface, a dinky little thing does 8 in/8 out @96/24, got a used one for a couple of hundred smackers from a local bloke on eBay.
we're technically in a non-surround part of QQ so i didn't want to detract from the original gist of Gene's thread which i think would be Stereo needledrops as he specialises in them but its all good chat i guess!
+1
markshan
1K Club - QQ Shooting Star
MOTU stuff is awesome, but I think that's about twice the budget of the Focusrite. I'm sure either is a fine choice.true!
i use a Motu Ultralite interface, a dinky little thing does 8 in/8 out @96/24, got a used one for a couple of hundred smackers from a local bloke on eBay.
we're technically in a non-surround part of QQ so i didn't want to detract from the original gist of Gene's thread which i think would be Stereo needledrops as he specialises in them but its all good chat i guess!
hobie1dog
500 Club - QQ All-Star
BGOS, thanks so much for getting this tutorial thread started (precious bodily fluids saved
Gene:
We all appreciate your time and trouble in doing this for the forum.
Marty
When I first tried to do stuff like this circa 1998, my computer just wasn't up to the task and I kept having trouble with everything seeming to go well only to find some bits and pieces were missing when I went to play it back. At the time, it made sense to go with a standalone CD recorder for the initial capture and then using the computer to finalize everything.
Of course, that limited my captures to 16bit/44.1kHz. Nowadays I use one of those little handheld digital recorders (mine is a Roland R-05, but there are a bunch of others out there that are probably fine). That allows me to capture at 24/96, which is allegedly better for downstream manipulation (ClickRepair, volume adjustments, etc.). Because it's so small, it also allows me to carry it to wherever the playback hardware happens to be at the time.
I mention that only as an option and not to even HINT that there's anything wrong with going computer all the way. Using an external recorder will not prevent you from having to learn how to use Audacity or Audition or any other computer tools, it's just an alternative method of obtaining the original capture. While I find it less trouble, I can see where others would find it more.
I also wholeheartedly endorse ClickRepair, though I strongly suggest listening closely while it's doing its work. Most of the time it works just fine, but there are certain things--in my experience, usually surprising and upredictable--that trip it up. When it goes bad, it will attempt to obliterate something that it should not.
There was some discussion above about what to do with between-track spaces. I used to quickly fade to silence and then quickly fade back up, but I've semi-recently used CllickRepair instead. I cut my initial capture into pieces containing music or what should be silence. I process the music bits through ClickRepair at conservative settings and then crank it up all the way for the inter-track bits. I have come to find the results call less attention to themselves than going to silence, but that's going to be a matter of taste (and I don't claim to have any!). I've also learned that you can often get away with those maxed-out ClickRepair settings during end-of-song fades where the music gets very quiet but the record noise is, relatively speaking, loud. It doesn't always work though and you may wind up hurting the fading music. Again, it's always smart to listen in real time while you're doing it.
The same guy who wrote ClickRepair also has a piece of free software simply called Equalizer. Its main purpose is for pre-RIAA records that have been captured using an RIAA pre-amp, but it has other uses as well. For example, I've had 45s that would not track properly when played at 45 but were just fine at 33. Equalizer has a setting that accepts a 45 played at 33 with RIAA and outputs a corrected file. It will also correct 78s played at 33 or 45. Of course, you'll need to first use SoX or Audacity or some other tool of your choice to correct the recording speed first.
I've also found that things go faster for me using a combination of GUI tools and commandline tools. My process usually leaves me with a bunch of sequentially-numbered files that need to be stitched back together. I find using repeated "Open Append" operations in Audition tedious, so I wound up automating the process at the commandline via SoX. I also use SoX to downconvert my final results to 16/44.1, but I realize some will find that blasphemous. Since I'm usually putting the results out on a server, I convert to FLAC and use a command file for that as well because I can't remember all the various switches used to get the minimum file size...I looked them up once and now they're in that file forever.
Sector Boundary Errors were also mentioned above and one easy way to avoid them is to use CD Wave Editor. It's technically not freeware, but I lost my registration and, when I attempted to pay him again, I found that payment was no longer being accepted. The software is 100% functional without registration.
jimfisheye
2K Club - QQ Super Nova
- Joined
- Jan 8, 2010
- Messages
- 3,033
A note on click repair:
Izotope RX can identify clicks and has settings to determine the detection 'strength'. It removes the detected clicks by isolating the click and subtracting that from the source audio. You have the option to output just the removed clicks instead of the cleaned source. You could choose to output the isolated clicks and subtract them from the source audio yourself in a second step.
This leads to this workflow:
Generate isolated click tracks (no pun intended) form the source. Go through the audio and run isolated click tracks at different detection strengths. Find the detection strength for most of the program. Run the little bits where you have to increase the strength. Edit these rendered "click tracks" together. Manually delete the odd false triggers (that weren't a click). At the end, you'll have a track that is all the clicks and just the clicks. Now subtract THAT from the source to perfect null away all the clicks and ONLY the clicks perfectly.
Takes a little time but you can achieve literally perfect results. Still magnitudes faster than manually spectrally editing everything.
Izotope RX can identify clicks and has settings to determine the detection 'strength'. It removes the detected clicks by isolating the click and subtracting that from the source audio. You have the option to output just the removed clicks instead of the cleaned source. You could choose to output the isolated clicks and subtract them from the source audio yourself in a second step.
This leads to this workflow:
Generate isolated click tracks (no pun intended) form the source. Go through the audio and run isolated click tracks at different detection strengths. Find the detection strength for most of the program. Run the little bits where you have to increase the strength. Edit these rendered "click tracks" together. Manually delete the odd false triggers (that weren't a click). At the end, you'll have a track that is all the clicks and just the clicks. Now subtract THAT from the source to perfect null away all the clicks and ONLY the clicks perfectly.
Takes a little time but you can achieve literally perfect results. Still magnitudes faster than manually spectrally editing everything.
I brought this up and then forgot the primary reason for doing so: If you have the ability to get your turntable output into your computer but have difficulty feeding it a proper RIAA-equalized signal, the free Equalizer software will accept a non-EQed ("Flat") file as input and perform the RIAA equalization digitally. I have not personally tried this yet and assume you'll also have to apply some pretty serious amplification, but in theory it will work.
http://www.clickrepair.net/software_info/equalizer.html
The catch is that, unless I'm looking in the wrong place, Izotope RX Standard is ten times the cost of ClickRepair while Izotope RX Advanced is nearly thirty times the cost of ClickRepair. . That's not a criticism or any suggestion that it's a bad product, it's just a serious barrier for some of us. I'd love to have the chance to tinker with it some time!
jimfisheye
2K Club - QQ Super Nova
- Joined
- Jan 8, 2010
- Messages
- 3,033
Well, the catch is that either the time spent with manual repair or the compromise from tools that compromise fidelity adds up to a LOT more time spent. Time = money.
If this is only for novelty use and not really a serious digital capture of a rare album... all good.
But if this is serious and about preserving a rare artifact, the tools and technique mean a lot.
If there's a hunger for precision but budget problems, you can "demo" this software as it were.
I haven't demo'd ClickRepair. If it has an "output clicks" mode though, you're free to try the same workflow. See if you can dial it in to perfection the same way for a truly lossless click removal.
I haven't demo'd ClickRepair. If it has an "output clicks" mode though, you're free to try the same workflow. See if you can dial it in to perfection the same way for a truly lossless click removal.
It doesn't have that in the sense that it will directly output a noise file, but it does allow you to just monitor the clicks, the results of which can be captured by one's preferred capture software (Total Recorder, in my case).
The monitoring capability is an important part of the process because it's an easy way to determine if you're removing something you shouldn't.
I'm intrigued by the thought of Izotope (or anything else) managing to clean up some 78s without trying to remove the singer! One of these days...
jimfisheye
2K Club - QQ Super Nova
- Joined
- Jan 8, 2010
- Messages
- 3,033
You could do it the other way around. Run different levels of cleaned. Then edit between them.
Working with the click output tracks feels more direct. You can listen to the original raw audio and hear the clicks while looking at the click tracks and visually see the clicks as you hear them. If you ran 3 (for example) click tracks (heavy/medium/lite) and lined them up, you'd see the one with the clicks you heard vs the one missing some (if too lite) or the one with all kinds of other stuff (too heavy setting). You end up starting with the one that worked for most of the program and then pasting in bits from the heavier and the lighter run tracks as needed.
But you could run say, 3 different levels of removal strength where it outputs the cleaned audio. Then comp between them. Listen vs. the raw audio and pick as best as possible. (This is where it helps to see the clicks in front of you.)
Working with the click output tracks feels more direct. You can listen to the original raw audio and hear the clicks while looking at the click tracks and visually see the clicks as you hear them. If you ran 3 (for example) click tracks (heavy/medium/lite) and lined them up, you'd see the one with the clicks you heard vs the one missing some (if too lite) or the one with all kinds of other stuff (too heavy setting). You end up starting with the one that worked for most of the program and then pasting in bits from the heavier and the lighter run tracks as needed.
But you could run say, 3 different levels of removal strength where it outputs the cleaned audio. Then comp between them. Listen vs. the raw audio and pick as best as possible. (This is where it helps to see the clicks in front of you.)
I hope all this technical talk doesn't scare off folks who simply want to rip some vinyl. It doesn't have to be so complicated. It depends on what you want.
For those of us who have a stash of vinyl, that may not be perfect, yet want to have them in our digital library with little fuss...well. That was my original intent.
My way is far from perfect, but it brings me a heck of a lot of pleasure. The vinyl has flaws, with scratches, scrapes, etc. The way I described, can go a long ways toward having a handy copy of your vinyl, that (for the most part) is missing the glaring clicks and yet retaining the warm fidelity that many of us love.
There were comments about USB players being inferior. I don't know. The final product I get is far from inferior. I have no intention of comparing to a much higher dollar setup. That's not where I want to spend my money. It's like anything else. We can all rip apart the components, methods, call them inferior. But it all comes down to what we have, what we can afford and what we are willing to sacrifice.
For those of us who have a stash of vinyl, that may not be perfect, yet want to have them in our digital library with little fuss...well. That was my original intent.
My way is far from perfect, but it brings me a heck of a lot of pleasure. The vinyl has flaws, with scratches, scrapes, etc. The way I described, can go a long ways toward having a handy copy of your vinyl, that (for the most part) is missing the glaring clicks and yet retaining the warm fidelity that many of us love.
There were comments about USB players being inferior. I don't know. The final product I get is far from inferior. I have no intention of comparing to a much higher dollar setup. That's not where I want to spend my money. It's like anything else. We can all rip apart the components, methods, call them inferior. But it all comes down to what we have, what we can afford and what we are willing to sacrifice.
These QQ posts might be helpful:
Battery powered Needledrop amplifier design (I haven't built mine yet):
https://www.quadraphonicquad.com/fo...software-demodulator.24900/page-2#post-385180
Using Audacity for Needledrops (I haven't downloaded Audacity yet):
https://www.quadraphonicquad.com/fo...software-demodulator.24900/page-2#post-383882
Kirk Bayne
Battery powered Needledrop amplifier design (I haven't built mine yet):
https://www.quadraphonicquad.com/fo...software-demodulator.24900/page-2#post-385180
Using Audacity for Needledrops (I haven't downloaded Audacity yet):
https://www.quadraphonicquad.com/fo...software-demodulator.24900/page-2#post-383882
Kirk Bayne
markshan
1K Club - QQ Shooting Star
It does and this is very true. When I did a drop of "It's a Beautiful Day", I had to manually declick "Time" because it grabbed the cowbell even at it's lowest setting.I haven't demo'd ClickRepair. If it has an "output clicks" mode though, you're free to try the same workflow. See if you can dial it in to perfection the same way for a truly lossless click removal.
Having skimmed the thread I wanted to toss in a couple points that I hope wonât be controversial and that werenât already mentioned. These both apply only in cases where your final product is going to be a RBCD, so youâre going for 44.1/16 files.
1. Record at 24 bit depth and set your levels conservatively, i.e. on the low side. Then before you split up an album side normalize it with a db or so of headroom. This will preserve the relative dynamics of the side but make the most of the 16bits/dynamic range you end up with. Basically youâre choosing the most significant 16-bits out of the available 24 and buying yourself some extra insurance against âoversâ.
If you have a way to âditherâ during the 24->16 bit reduction process, it can sometimes make an audible difference, not one Iâve experienced myself but Iâve read others testimony that seem very convinced. Probably depends a lot on the nature of your program material.
2. If you have the option to record at a sample rate thatâs a multiple of 44.1 (which means 44.1, 88.2 or 176.4) you can avoid subjecting your file to the more involved math required to go from 192 or 96 or 48->44.1 CD sample rate- which might or might not make an audible difference depending on how well your software handles the process. But why not avoid the process?
1. Record at 24 bit depth and set your levels conservatively, i.e. on the low side. Then before you split up an album side normalize it with a db or so of headroom. This will preserve the relative dynamics of the side but make the most of the 16bits/dynamic range you end up with. Basically youâre choosing the most significant 16-bits out of the available 24 and buying yourself some extra insurance against âoversâ.
If you have a way to âditherâ during the 24->16 bit reduction process, it can sometimes make an audible difference, not one Iâve experienced myself but Iâve read others testimony that seem very convinced. Probably depends a lot on the nature of your program material.
2. If you have the option to record at a sample rate thatâs a multiple of 44.1 (which means 44.1, 88.2 or 176.4) you can avoid subjecting your file to the more involved math required to go from 192 or 96 or 48->44.1 CD sample rate- which might or might not make an audible difference depending on how well your software handles the process. But why not avoid the process?
Similar threads
