Beginner recording classical solo violin

[marco.swe]

Hello everyone,
I am totally new to audio recording and I am try to get into the subject. I will try to summarize all my struggles in a very simple way and hopefully get good advice on how to continue.

-What I need is to record classical violin solo pieces, both in recording sessions as in live performances (audio/video).
-Gears I have to use:

• RECORDER: ZOOM F8n
• MICs: RODE NT55 matched pair on a stereo bar (cardioid and omni capsules available)
• DAW: Studio One 3 Artist

-The violinist is very skilled and the violin has a beautiful sound
-We also have the possibility to record in church
-I was thinking to try both X/Y (cardioid) and A-B (omni) mic techniques in order to learn

Now the question list

1) I find the Rode NT55 a little too bright mic for the violin. Do you have any opinion and advice on that?
2) The violin has a frequency range of 196 Hz to 10 kHz (approximately). Are there advantages to use the three position High-Pass Filter (flat, 75Hz and 150Hz) directly on the mics or is better to do it in post?
3) Are there any general rules to adjust the Trim and the Fader levels before recording? How manipulating the Trim or/and the fader affect the levels and the recording?
4) Considering mic placement, how would you adjust the mic position practically? Upper and lower to change the violin tone and closer or more far away to analyze the critical distance for reverb?
5) How can I notice phase cancellation issue on A-B setup? Is it important to always keep 3-1 rule? I have seen several time 2 parallel mics 20cm apart on A-B setup more than 1m from sound source. Is this wrong?
6) Is "invert phase" something useful and when?
7) I set up the recorder so that I have: Track 1-8 + L/R (Poly WAV). This will generate a single WAV file with separate channel for each inputs (2 in my case) and a L+R mixed channel. How can I create a wide stereo image? Panning L and R? How does it works? Should I PAN the channels on pre or post? Which are the differences in working with the single channel files or with the L+R channel? Shall I also use the delay? How?
8) As I still need to learn my DAW properly, do you find is OK for me to start learning Studio One 3 Artist? Does it fits good for classical music? Are there nice plug-in for violin?
9) Is there a book or trainig material you recommend in order to start?

This is a sample of the recording I have done: Google Drive: Sign-in

I normally find my recordings with a lack of fullness and warmth. Do you have any comment?

Thanks a lot for the time you will take to read and answer my doubts and give me a feedback!

/Marco

 

[Deleted member 196982]

In Google Drive, you need to create a shareable link and post that if you have a WAV file to share. You can always attach a high quality MP3 to your post (Go Advanced button).

So, not having heard your recording, I can't make a comment on that. If it's really a solo violin, and a nice space that you want to capture, I'd probably try XY at a distance of somewhere around a meter, plus or minus, depending on the space. I feel you can get a little better sound of the instrument by being above it and aiming down, but you do need to make sure the performer is not making a lot of noise with their feet. And, if they tend to move around a lot, that will limit how close you can get, as well.

I've been in several old churches in Sweden and the acoustics are amazing, but it might be overwhelming in a recording if you get too much of the space. That distance in the picture seems a little far.

I have not used the NT55 so can only look at the spec sheet. The omni pattern has a big bump at the top, while the cardioid looks flatter, though its rolloff on the low end is not going to help there (i.e., getting a full sound), especially at a distance I'd imagine it thins out.

Never did it, because I only recorded in smaller spaces, but trying MS or just getting a good LDC (or SDC with a bit flatter low end) to augment the XY pair (or some other configuration) might allow you to play around with the amount of room you mix in.

Edits:
Unless you go to a setup where these mics are really different distances from the soloist, I can't imagine phase cancellation, though if you use a widely spaced pair, and they move around, and you mix hard LR, it can get weird... So, I would stick with XY (IME) or use a single mic for the soloist, and have a MS setup, or as was my typical, an XY on the piano, and the violin bleed would get masked by the single mic on it. A touch of reverb on a stereo bus spreads things around, though in a church it probably isn't needed.

P.S. If you are using the F8n (I have one and just sold my F8), you generally don't play with the fader during recording unless you are trying to get a monitor mix - at least I don't. Set your recording using the TRIM (aka gain) so everything is in that range where the peaks are well below 0dB. I would set the recorder limiter as well, though a solo violin is not likely to cause the kind of transients that will ruin a take, I just set all my tracks with some limiting and also HPF, to prevent that train rumble from getting picked up. I would just record polywave and do the mixing in a DAW. Never use the L+R option, which I always felt was more of a live-feed/quick-mix thing for video/post work.

 

[marco.swe]

Thank you very much Roger.
I have now fixed the Google Drive link so that you can see (and hear) the video recording.
The Sound has original reverb and no automation has been applied. Looking forward to your feedback.

I guess the mics were a bit too far away and it feels a bit thin with too much ambient reverberation maybe due to the big distance.
Next time I would definitely try X/Y setup with cardioid. I will also read a bit more about MS setup and maybe also try that out.

I think is a good idea as you write: Polywave without L+R. In this case, shall I pan the two channels and use some delay in the DAW in order to wider the stereo image in the mix?

 

[rob aylestone]

You've picked up lots of things, but completely missed context. The 3:1 rule, for example. This phase cancellation happens when you have things like a singer playing a guitar and the vocal mic picks up the guitar and the guitar mic picks up the vocal and the combine destructively. That ratio makes each one have more of it's wanted source and less of the unwanted source, so the cancellation effect is much less - look at the inverse square law if you like science - every inch/cm you can add to one and remove from the other has a very big impact.

With stereo ambient capture it's totally different. You need these differences. Difference in level between the mics and difference in time, as the sound travels further.

A pair of omni mics can create a HUGE sound field, yet convert directional information. A pair of cardioids creates directional information, but does it differently. In X/Y, for example, they are placed as close together (diaphragm position) as is possible, so sound arrives at both simultaneously, the stereo information being created by the polar pattern of the mics at an angle to each other. Other techniques combine the two components.

In your example, you forgot one thing. Focus. Omni in a very reverberant space capture the sound of the church really well - BUT - what they capture is a mess. You also need to evaluate what a stereo mic setup is doing. A violin is considered to be a single sound source. It has very little width, so move left to right in front of it and nothing really changes. A trumpet is a point source - even narrower. A piano however, is a BIG thing. Sit near it and the left hand note clearly comes from a different place.

If it was me, I'd have used one mic on cardioid setting much closer. I'd have also recorded the stereo church sound, and blended the two together. The church is stereo - BIG TIME!

You've also missed the point on frequency response. The frequencies you mention - they're the fundamentals - a Stradivarius will have lots of interesting overtones above that top limit, and I would not use a dynamic mic on a violin. Ever. I want all that stuff up the top. That doesn't mean that it has to be tinny sounding. It just gives you EQ scope. Every violin needs help here. Wrong or just inappropriate EQ is horrible, but gentle assistance makes them beautiful with a decent player.

For me - close perspective with the mic 18"/450mm from the f holes (maybe a tad more if it creeps into the video frame, but no further, and if possible above, looking down. They actual distance often has to be set by the player's style. Some are rock steady and don't move, others wave around crazily. The stereo mics maybe 10'/3m back, central and high.

Level wise - get the player to play the loudest part, drop the gain so you have some headroom at the top - on your recorder, that's actually easy as it has plenty of headroom, and then leave the gains alone.
The snag is you only have two mics?

The only solution here is to get the mics very close in to maximise direct sound and minimise the space contribution. Be aware though that any movement from the player will do wild things with pan.

With just two mics, I might sacrifice reluctantly stereo and record the church ambience in mono and add some stereo reverb. Not good process, but the only alternative is crossed pair, close in, on cardioid, then gradually increase the difference till it sound best in the headphones. That's a lot of faffing.

 

[marco.swe]

First of all, Thanks Rob for your very constructive critics.
I think it is very interesting to read what you write and to try thinking the setup you described: Violin in mono and ambient in stereo. Maybe I could use my NT55 pair with omni caps for the ambient and use a third mic with cardioid pattern closer to the violin. Is this reasonable?

You wrote:
"With stereo ambient capture it's totally different. You need these differences. Difference in level between the mics and difference in time, as the sound travels further."

How would you mix these 3 tracks together then? I guess I should aim to "wide up" the ambient stereo recording and add the mono violin track on top. Do you have any suggestion in how I could try to do that? Was thinking more regarding levels, panning and delays between the tracks.

Thanks for helping out!

 

[rob aylestone]

No - it's huge enough already. The ambient nature of churches means that the stereo recording of it is awash with all kinds of timing and level information - in fact, exactly the kind of thing the reverb plug-in try to generate. The mixing of this is actually pretty trouble free. If you can grab a half-dec ent condenser for the violin, you should get a cracking recording. The first task is getting the recording, making sure the distance to the violin is helping keep it as clean as possible, then simply bring up the stereo space source and blend to taste. You'll be surprised how little you need. You can also experiment with gradually delaying the violin track. This sometimes makes the violin sound stronger and more focussed. The only issue is sometimes what reverb it has collected fighting with the stereo reverb track, but it's just a set of adjustments to make. Treat the violin as the lead - bring it up panned central and tweak the EQ. Then with the stereo track panned hard left and right, creep it in, as you would creep in a reverb plugin. In your original post you did mention the invert phase facility (which of course is the term we have all used for 50 years wrongly, as it's simply polarity not phase!) - ALWAYS worth prodding this button, in this case on the mono violin track. It will make a positive or a negative impact. Pick the best sounding position.

Note that this technique is a godsend in lovely sounding spaces, and falls flat on it's face in rotten ones. In those places where RT60 is high - over a second or two, it's always worth it. Back when I started out, we did lots of TV work in UK churches, and a condenser on a tube extension would be first choice - usually back then, an AKG 451cardioid capsule and the long thin tube making it look acceptable to the vision guys. We'd use these on orchestral soloists, opera singers and any singer who could project. We struggled for delays back then, so often the reverb was just added with the fader to 'taste'. Our only option. We'd often roll off the bass, if the ambience was a bit boomy.

Give it a go - I think you'll like it!

 

[Deleted member 196982]

...
I have now fixed the Google Drive link so that you can see (and hear) the video recording.
The Sound has original reverb and no automation has been applied. Looking forward to your feedback.

I guess the mics were a bit too far away and it feels a bit thin with too much ambient reverberation maybe due to the big distance.
Next time I would definitely try X/Y setup with cardioid. I will also read a bit more about MS setup and maybe also try that out.

I think is a good idea as you write: Polywave without L+R. In this case, shall I pan the two channels and use some delay in the DAW in order to wider the stereo image in the mix?

Way too much room in the audio in that recording, at least for my ear.

I think rob & I are kind of saying the same thing. You have to be closer to the violin to control the amount of room, and it's easier to do if you record those aspects separately. You just have to experiment, but a 3rd mic of good quality for just the violin will give you more flexibility. (With that church ambience, you do not need delay or reverb.)

I have found it hard to get in that .5m range with many players when they are standing if they move much, and still keep the mic out of frame. Of course, if you are zoomed like the video, you can get very close, but then the mechanical noises of playing, like fingers and bow (and breaths and other vocalizations from some players!) can start to be intrusive.

Just a point on that zoom/crop view in the video - I often had kind of the opposite problem, where I had close micing and direct recording of live, amplified performances, but the camera placement would give a more distant view. I personally found it can be a little off-putting (for lack of better vocabulary) when the audio distance does not match the video. Eg., in my case, the sound could be almost studio like, but it was clearly a live performance "viewed" at a distance. So I would mix in more ambience (from the camera mic in this case). You have kind of the opposite problem, for my ear/eye, at least. The very close view is contrasted with an audio track that leaves the impression of listening from 5+m. I think you'll be more satisfied if you can make the video and audio have a more consistent feel. (And, a 2nd camera can be used for occasional close ups, or you can use some pan/zoom FX in the video editing to provide the close view when it's interesting. Just my $.02.)

On the F8n, I assume you have already imported the polywave into your DAW and it breaks it into the separate tracks Ok? Otherwise you can just set the mode for separate tracks, of course, but I found the polywave so much easier to use.

 

[marco.swe]

Hello again Keith.
Another very good critic. I agree! It is almost like adding a huge reverb on a video recorded outside (I have done tha too eheheh). I will definitely consider that aspect for next produtions and yes, Polywaves on f8n will be perfect!

 

[ecc83]

Opera, two off and now classical violin! HR is getting better and better for me!

I am not qualified to comment in this company but I would really like to hear the results of crossed cardioids. I am also wondering if a modest amount of absorbent material could be used? That window could be making the sound brighter? Any chance of borrowing/hiring a pair of ribbons?

As ever, DO ask at soundonsound.com There will certainly be at least one in depth article about this. My dim brain thinks it was covered not many months ago?

Just a thought. There is a dodge in small, dead rooms of standing a guitarist say, on a hard surface, we use hardboard. Maybe the inverse works here? Put the gal on a rug. If nothing else it stops 'foot sound'.

Q. How should I mike-up a violin?



Dave.

 

[marco.swe]

Thanks a lot guys for all your very nice and informative comments!
Dave, ribbons would be amazing to try but I guess I have a lot to experiment and learn before jumping at that level. I could probably hire those once I will feel a little bit more confident with the subject

 

[marco.swe]

I have now re-rendered the audio after hard panning the channels L and R (and raising a bit the volume as well)

Looking forward to record next time! I have plenty to read and to learn in meanwhile.
Thanks a lot for amazing top support

 

[marco.swe]

Hello again guys.
Today I have managed to do a test recording but I have some confusion in my head (even more, yes )

I had the following setup (with ZOOM F8N):

-Matched pair RODE NT55 on X/Y setup on channel1 and channel2
-Matched pair NEUMANN K184 on X/Y setup on channel3 and channel4

Recording setup: Tr1-8+L/R (Poly)

The ideas was to perform 2 different stereo recordings with 2 different types of mics.
As results I have got a single Wave file with 6 audio waves in it (see attached pic). In order to work with these data in post, it looks I need to split the tracks in separate mono files first and import them again.

The question is: which is the benefit in using a polywave file when you anyway need to split the track in multiple files in order to work with them?

What I expected as output was a single wave file that, once imported in my DAW (Studio One Artist), It would have shown each channel on a separate track.

What am I doing/thinking wrong? How do I know which file is which channel once I split the polywave file in separate files?

Thanks for help!

 

[TalismanRich]

The only reason I could see for having a Polywave would be that its an easy way to contain all the tracks in one file. No chance of losing anything if you store it for archiving. Consider it to be like a ZIP file.

I'm curious how the comparison of the NT55 and the K184 came out. Was there a big difference? Which did you like better? I've done that on some recordings where I used multiple mics so that I could compare and pick the one I prefer.

 

[marco.swe]

I liked more the Neumann but now I am more struggling in getting things to work

How do I know which channel is which now?

If I split the polywave in single mono files I doesn't say anything to me on which was on channel 1,2,3, or 4

 

[TalismanRich]

I know that in Reaper, if you split a multi track file into singles, they come out in the order of the original file top to bottom. A L-R stereo track goes to 1-2.

I guess you could set it up the way it was when the original recording was made and then tap each mic to identify the track that was assigned. I don't know how Zoom set up the polywave, but I would assume it was just channels first, then the L/R....?

 

[marco.swe]

When I split the file into mono files, I get separate files with a number that, probably, is the same as the channel. Anyway the L+R track was numbered 5 and 6 while those channel were not used ....

 

[Deleted member 196982]

They should be 1,2,3,4,L,R but let me set up my F8n and confirm how they break up in Logic. I never record LR, but have always used polywave, starting with the F8. Might as well try it...

(The guy next door has been trying to power-wash his entire house, it seems, with a tiny gas power-washer and it has been brrbapbapbrrbapbap--- ALL DAY LONG, so there will be no actual audio recorded today...)

 

[marco.swe]

They should be 1,2,3,4,L,R but let me set up my F8n and confirm how they break up in Logic. I never record LR, but have always used polywave, starting with the F8. Might as well try it...

(The guy next door has been trying to power-wash his entire house, it seems, with a tiny gas power-washer and it has been brrbapbapbrrbapbap--- ALL DAY LONG, so there will be no actual audio recorded today...)

Great, thanks
(you need a brrbapbapbrrbapbap-cut filter )

 

[Deleted member 196982]

They should be 1,2,3,4,L,R but let me set up my F8n and confirm how they break up in Logic. I never record LR, but have always used polywave, starting with the F8. Might as well try it...

(The guy next door has been trying to power-wash his entire house, it seems, with a tiny gas power-washer and it has been brrbapbapbrrbapbap--- ALL DAY LONG, so there will be no actual audio recorded today...)

Great, thanks
(you need a brrbapbapbrrbapbap-cut filter )

Haha.

I just tried it, and I was wrong. The first two tracks were the LR, and then it was 1, 2, 3, 4. Here's a pic that isn't real clear, but you might be able to discern where I spoke into the mics I had laid out left to right, plugged into inputs 1, 2, 3, 4. (Plus the bump where I laid each down after talking into it.) The first 2 tracks are the L+R recording, followed by the individual mics. (You would then pan those hard L and R to start. Not sure what the L+R recording gives you, but I'm probably repeating myself.)



P.S. (edit) I noticed after the fact that I had some panning set in the tracks - most noticeable is 3/4 which were L40 and R40, though 1/2 were L20 and R20. That shows up in the LR tracks, but the mono tracks are just the straight mic, with no panning. It *looks* like the LR tracks are actually R/L???

P.P.S. (edit 2) Just to clarify the F8n setup, the PAN setting (access via PFL is how I do it) really only affects what you hear in the headphones, and also what is recorded *if* you opt to save the LR tracks (i.e., in addition to the input tracks, which are *not* affected by the PAN setting - they are all/each just MONO tracks). I used PAN to help monitor live recordings, but used different panning in the DAW mix, as appropriate. I am not familiar with your DAW, so it might be less confusing to not use polywave, and if you really want the LR track, just use that mode on the second SD card. Then, when you go to copy your files, you'd only use the SD card with just the actual recorded tracks, and save the SD card with the LR recording if you needed it. Polywave does have the advantage, assuming your DAW supports it, of just popping everything in to the project exactly aligned as recorded, vs. dragging up to 8 files into the DAW.

 

[marco.swe]

Haha.

I just tried it, and I was wrong. The first two tracks were the LR, and then it was 1, 2, 3, 4. Here's a pic that isn't real clear, but you might be able to discern where I spoke into the mics I had laid out left to right, plugged into inputs 1, 2, 3, 4. (Plus the bump where I laid each down after talking into it.) The first 2 tracks are the L+R recording, followed by the individual mics. (You would then pan those hard L and R to start. Not sure what the L+R recording gives you, but I'm probably repeating myself.)

View attachment 106533

P.S. (edit) I noticed after the fact that I had some panning set in the tracks - most noticeable is 3/4 which were L40 and R40, though 1/2 were L20 and R20. That shows up in the LR tracks, but the mono tracks are just the straight mic, with no panning. It *looks* like the LR tracks are actually R/L???

P.P.S. (edit 2) Just to clarify the F8n setup, the PAN setting (access via PFL is how I do it) really only affects what you hear in the headphones, and also what is recorded *if* you opt to save the LR tracks (i.e., in addition to the input tracks, which are *not* affected by the PAN setting - they are all/each just MONO tracks). I used PAN to help monitor live recordings, but used different panning in the DAW mix, as appropriate. I am not familiar with your DAW, so it might be less confusing to not use polywave, and if you really want the LR track, just use that mode on the second SD card. Then, when you go to copy your files, you'd only use the SD card with just the actual recorded tracks, and save the SD card with the LR recording if you needed it. Polywave does have the advantage, assuming your DAW supports it, of just popping everything in to the project exactly aligned as recorded, vs. dragging up to 8 files into the DAW.

Hello Keith, exactly the answer I was looking for. Thank you!
I have recorded the L+R track just for test/learning purposes; I actually do not need it and yes, also for me, it was track 1 and 2.
I will also try to make some extra test after work so that I will have it more clear. Maybe it helps to configure some metadata for each channel directly in the f8n to figure out which is left and which is right afterwards (talking about mono tracks)?

Probably using the "mono/stereo wav" instead of the Polywave recording setup will make life simple just for that.
I will try more.

PS: thanks for the panning clarification