Your Vinyl Transfer Workflow (sharing best needledrop practices)*

Thanks, I'll research this thread more thoroughly when I have the time and money to convert

 

Does anyone know what might cause these tiny interruptions during recording?
During playback rarely I'll notice a fraction of a second of music didn't get recorded,
but not because of any vinyl skip, but something else.

Can anything cause an interruption in recording?
I'm not running any other programs. Wifi and all internet connections are disabled.

Anybody ever notice this?

 

Found it. Everyone can stop researching for me now lol.
Need to change sleep settings to 1 hour on laptop.

 

I just went through a similar issue with my Focusrite Scarlett 2i2 USB interface. I 'accidentally' looked at what I was recording on my test bench using a steady 1kHz signal source. What I found was that the audio had occasional drop outs or tiny interruptions. I'm sure if I was just doing a needle drop, they would have been dismissed as a click or pop on the album. I spent many hours trying the various suggestions from Focusrite's tech support (they were very willing to help), but even after applying dozens of 'solutions' (including the sleep setting changes), I was only able to get glitch free recording using the latest beta driver from Focusrite and the Direct Sound API. The CoreAudio / WASAPI was never glitch free, no matter how radically I altered my laptop's settings.
After seeing how many things can go wrong with the audio capture using a USB interface, I finally just shelved the Focusrite 2i2 and purchased a Tascam DA-3000.
It leaves me wondering how many people are experiencing dropouts and glitches from their USB interfaces and blame it on clicks and pops.

Terry

 

Good question, No issues with the beta driver using Reaper on the 2i2, it's been solid, that Tascam is a nice unit, if my FR ever fails I'll probably grab one
Edit: I am using a desktop PC Core I5 8G memory and a slot 3.0 USB interface

 

With Reaper, you are probably using the ASIO driver. I didn't try ASIO since I'm using Goldwave which doesn't support ASIO, but the Focusrite guys suggested I'd be better off with an ASIO interface. I just didn't want to switch DAWs and have to go up the learning curve. I have been using Goldwave for quite a while and am pretty comfortable with it now. I used it for years with a desktop PC and a PCI sound card. I don't think the PCI interface is as tricky to get working with audio as the USB interfaces are (based on reading quite a bit on the web). Search the web for "DPC Latency" if you want a deep dive into PC and USB issues with real time operations.

The Focusrite drivers (non-ASIO) were a bit weird for me. When I first tried the driver that was installed, I found that even though I was recording at 24 bits / 96kHz, the bandwidth was only 22kHz. I verified it really was recording at 96kHz, but somehow the bandwidth was limited to what a 44kHz bit rate would be. Poking around the web I found others with the same experience, and that is how I found out about the beta drivers page. The beta driver cured the bandwidth problem. Also, using their CoreAudio / WASAPI driver, I found the peak level was topping out at -0.1dB. So even if I drove it hard into clipping, Goldwave didn't show it as clipping. By changing the audio properties to 'Give Exclusive Mode Application Priority', I got normal clipping operation, but the glitches / dropouts appeared. Only the Direct Sound API gave decent results, even though everything I found on the web said the CoreAudio / WASAPI is the preferred interface with Win 7 and above.
Finally I lost confidence in the thing and decided to go with the Tascam DA-3000 and be done with the whole USB interface issue.

One other thing to be aware of with the Focusrite 2i2 is that there is a large amount of noise in the DAC output above about 50kHz. I wasn't planning on using the DAC output (I only wanted the ADC for recording), and who knows what noise above 50kHz would do to the sound anyway, but it definitely affects measurements. It just seemed like poor practice not to filter it out.

The 2i2 is a pretty neat little package at a decent price, but I just got to the point that I didn't trust using it with my laptop (Dell I5 2.8GHz 16G Win 7). If everything was just right, it recorded well, but any little change in a setting seemed like it opened the possibility for getting glitches / drop outs. Probably using the ASIO interface it is fine, but I'd check it carefully before I spent a bunch of time doing needle drops with it.

Terry

 

One thing that helped the occasional drop out I saw when recording was disabling network adapters. Thought I might share that.

 

Yep. I disabled the network adapters, the WiFi, the battery charging, and pretty much every other thing I could get to including almost all of the services and processes that weren't required. I spent a couple of weeks searching the web and trying pretty much every suggestion out there. In the end I was considering replacing my laptop, but the one I have should be plenty good enough, so it would have been a crap shoot buying a new one. Would it work any better? Who knows. It is tough to find any data on laptops that relates to the DPC latency (if indeed that was my issue as Focusrite support suggested).

I even checked the DPC latency of my laptop at work which was another fairly powerful machine with a solid state drive using this tool:
http://www.resplendence.com/latencymon
(free home edition)
and even it showed that there might be issues with real time audio recording. So rater than sink money into a new laptop that may or may not have fixed the issue, I decided to go with the DA-3000 which means I don't need to rely on a computer to do the needledrop. I still use Clickrepair and Goldwave to process the files after I capture them of course. Ultimately I don't know if the issue is with my laptop or with the Focusrite drivers, but it was getting to be too much of a chore to figure it out.

I do recommend that anyone using a USB audio interface to capture needledrops be aware that dropouts can happen. I don't know how common the issue is. Maybe I just got 'lucky'. I suspect that if I hadn't looked at the recorded signal from a stable test oscillator, I probably would have assumed that the glitches I was getting were just clicks or pops on the vinyl and processed them out. Ignorance is bliss is guess.

Terry

 

Hi All, Hope you can help?

I have a few mono records that I want to needle drop. I intend to go the Y cable route. So I have ordered an RCA Stereo to RCA Mono cable - 2 x female RCA to 1 x male RCA from Ebay. I already have Y cables I use for needledrops that are 2 x stereo RCA to 1 x stereo RCA (or vice versa). The cable I have ordered will take the L - R stereo channels and sum them to one mono channel. Can I use the cable I already have to receive the mono and split it into a L - R mono signal (or will it revert back to stereo)? Or is there a different 1 x mono to 2 x mono cable I need?

Secondly, is it best to run the Y cable between TT & phono preamp or between phono preamp & sound card/receiver?

Finally, when I capture the mono needledrop using Izotope RX4. Should I set the RX4 to capture as mono or as stereo (L - R mono)?

Thanks in advance.

 

If you use a Y cable (best is between phono preamp and sound card), then capturing as mono in RX4 will suffice.

But for a cleaner signal I suggest bypassing the cable altogether, record a stereo signal in RX4 and then using the Channel Operation fold it 50/50/50/50 to mono.

 

Thanks, but I don't understand what you mean by folding it 50/50/50/50?

 

Just look inside RX4. You'll see what I mean. Channel Ops module, Mixing. Or just choose preset "Mix to mono".

 

Got ya, thanks.

 

Just thought I'd mention to any of you here who use iZotope RX or are thinking of getting it, the newest version RX5 Audio Editor was released yesterday.

I've had the pleasure of beta testing this new version and I'm very happy with the results and the new features such as the Instant Process Tool, Module Chaining, De-Plosives, RX Connect for using RX with other DAW software packages, a very much improved built-in EQ that's got some really powerful options, and so on.

For vinyl restoration work, I find the De-Plosive module is particularly effective because does a great job dealing with those recurring, low-frequency thump-thump-thumps that occur when there are vinyl defects. I used to have to select each of those and try various tools to remove them without mangling the bass, but with De-Plosibe, you just select the entire section and with one click they're gone 99% of the time and without harming the bass.

Anyway, just thought I'd pass the news on. iZotope has some special upgrading pricing on right now.

Check out the new features here: https://www.izotope.com/en/products/audio-repair/rx/

 

Oooh! That sounds really awesome. Thanks for the heads up. If that De-Plosive module is what you describe it to be, then that alone makes it well worth the upgrade for me.

 

Question, folks: after having a passel of people at my house over the past two weeks, I'm spending a nice Turkeyday in an empty, quiet house reading, listening to records and doing some needledropping, and I've got two questions from a needledropping newbie.

Currently, I've got the Focusrite Scarlett 2i2 hooked up to the line-out on my preamp, so I can still listen to the music over my speakers while I'm recording. Is that going to have a negative effect on the needledrop? Should I be going straight from my phono pre to the ADC and restrict myself to listening using the headphone-monitoring capability the Scarlett has while I'm recording?

Also, the 2i2 can only do 24/96: should I set Audacity to record 32-bit float/96 anyway and "dither it down" (in quotes because these are words I've read in needledropping discussions and I don't really know how to do that, but I can figure it out, if it would be best!), or should I record straight to 24-bit PCM? I've seen some folks say recording in 32 and dithering to 24 is better, but I wasn't sure if that was the case only when you can record in 32, or if the "extra room" is good even when you can only record in 24.

 

Happy t-day! On the first question, not sure, but don't think there would be any issues, quite a few people use a tascam portable via line out and seem to do ok, the second part, 32 bit should be used if you are doing any processing, de-click etc, if not don't worry about it

 

Pretty well all audio interfaces that any of us can afford input 24-bit audio streams. There are some 32-bit devices, but they're very pricey last time I checked. You can easily record in 24-bit audio without problem. Once you have the file and do anything with it, you want to work in 32-bit or higher floating point audio because as soon as you do anything, even a simple volume change, there's math involved and floating point format preserves all the information to the right of a decimal point whereas integer-based audio streams do not. Then once you're ready to save your finished audio to a format for playback, you can reduce the word length to 24- or 16-bit audio as the format dictates. Be sure to use dither for 16-bit audio, and it won't hurt to use it for 24-it audio as well.

 

If you want to know how much acoustic feedback makes it into your needledrop, you can do this simple test. Put the stylus on a stationary record, play a CD, turn the volume up to your normal listening level, and record the output of the turntable to your PC. If you're comfortable with how much "acoustic information" makes it through to your turntable with your speakers playing, you're good to go.

 

For recording, it's always better to monitor through headphones. This is a useful test to see just how much feedback enters the system, just listening to a record on the turntable.

 

Why would you downsample from 192 to 48 and then upsample back to 192? Wouldn't this adversely affect the sound?

 

I didn't..I adjusted the sample rate, effectively slowing it down, stopped doing that years ago because clickrepair got better​

 

So the adjustment down to 48 doesn't equal downsample to 48? I don't get the difference.

 

It is making it 4 times slower..go try it yourself..Edit Adjust Sample Rate...which is different than Convert Sample Type...one is reversible..one is not