Topic: Zoom F8 for Classical recording

[voltronic]

A short sample from the final recital:

https://s3-eu-west-1.amazonaws.com/robs-music-files/sample10.wav

As per Voltronic's suggestion, I persuaded them to have the piano lid fully open.  I also modified the microphone set-up again:

Main array, set about head height:

- OM1s at 67cm
- DPA 2011Cs at 41cm

Spot X/Y mics on violin: my vintage Calrec CM652Ds.  These are cardioids and excellent mics (recommended by Tony Faulkner in a 1980s review he did of microphones in a UK Hi-Fi magazine [a copy of which I still have!]), but they don't have the fine clinically-accurate detail of the DPAs, which I figured would work in my favour to help get the warm, woody sound I wanted from the violin.

See what you think, but this is definitely my favourite sound of the 4 sessions.  To my ears the piano has much more delicacy - opening the lid definitely helped.

Now we're getting somewhere!  This is a big improvement in tonality, and it now sounds like a real piano in a real space.  Nice natural hall decay, and the instrument itself just sounds so much better.  I'm also guessing it was tuned between your last sample posting and this one.  What was the distance of your main array from the piano, and how was it aligned?  (As in, was it lined up in front of the hammers or farther over towards the bend of the piano?)

My only criticism is that the violin is very strong left in the image and the piano very strong right, which makes me think your main array was set more to the left (in line with the hammers).  I'm not getting much of the piano in the left channel.

I think your spot sounds nice - I wouldn't want any more "detail" than that.  Would you be able to please post the same excerpt with the spot muted for comparison?  I'm curious to hear how much of the violin is captured by your main array.

[Gutbucket]

Good timbral and depth balance on that one to my ear.  Violin is no longer over-present, over-close and over-bright, breaths are less prominent, piano is far less woolly.  Personally I don't mind the strong left/right channel separation so much for this duet, the ambience doesn't seem to have any appreciable hole in the middle - I'm listening on headphones, which makes me think it should be fine on speakers as well.  The HVAC low frequency rumble defining the noise-floor is pretty noticeable and my only criticism, mostly noticeable at the start and a few other places.

[voltronic]

I'm listening again on my monitor speakers, and need to revise / amend some of my earlier comments as they are more revealing in the treble than my headphones (and obviously tell you more about imaging):

1. The piano was not tuned in between your last two sessions (but it's not far out - this isn't an annoyance; I'm just realizing I was wrong earlier).

2. The violin is not as far left as it had sounded in headphones - it's just left of center, which is a fine-sounding place for the soloist.

3. The piano is definitely placed hard right, even more so listening on speakers.  I get where Gutbucket is coming from where it's nice to have the separation, but for an instrument as large as a piano there should be a lot of information in both channels (IMO).

4. The HVAC rumble is easily taken care of with iZotope RX.  I can send you a lightly denoised sample if you want.

[robtweed]

Thanks for the comments, guys.  It's all very helpful for future reference.

I agree about the tuning of the piano - in fact it's not a great instrument and really needs replacing, and they could certainly do with tuning it more regularly.  The tuning of a piano is something I always acutely notice.

Unfortunately in these situations, you get what you get and have to make the most of them.  In fact, I reckon I've made the piano sound a lot better in the recordings than it sounds in real life! :-)

The positioning of the main array was such that the 90 degree angled mics pointed a bit to the left of the violinist (LHS) and towards the end of the piano (RHS).  The mid-way point of the array was focused directly between the pianist and violinist.  ie I aimed to get the piano's full width from about the mid-point all the way right.  It's interesting that Voltronic feels the piano is hard-right. That wasn't the intention!  I mixed for speakers (using my Naim/PMC set-up which is very revealing), and checked on headphones (Sennheiser HD650) - no panning was applied to the main array mic signals in the mix.

I'll post the 3 separate mic pair recordings from that sample a bit later and see what you think.

Thanks again, guys! (I'm hoping others are finding this voyage of discovery helpful and interesting too)

[robtweed]

OK so here's the 3 separate mic pairs, raw, straight off the F8:

First the DPA 2011s in the main array.  Actually I think this almost works on its own - see what you think:
https://s3-eu-west-1.amazonaws.com/robs-music-files/sample10-dpa.wav

Then the OM1s:
https://s3-eu-west-1.amazonaws.com/robs-music-files/sample10-om1.wav


and finally the Calrecs (X/Y spots on the violin):
https://s3-eu-west-1.amazonaws.com/robs-music-files/sample10-calrec.wav

[robtweed]

The HVAC rumble is easily taken care of with iZotope RX.  I can send you a lightly denoised sample if you want.

I'd not heard of iZotope RX before you mentioned it, Voltronic - looked it up and although it's pretty expensive, it looks extremely powerful.  Not sure I can really justify that kind of expense but it's definitely on my "if only I had the money" list :-)

[voltronic]

Thanks for the comments, guys.  It's all very helpful for future reference.

I agree about the tuning of the piano - in fact it's not a great instrument and really needs replacing, and they could certainly do with tuning it more regularly.  The tuning of a piano is something I always acutely notice.

Unfortunately in these situations, you get what you get and have to make the most of them.  In fact, I reckon I've made the piano sound a lot better in the recordings than it sounds in real life! :-)

The positioning of the main array was such that the 90 degree angled mics pointed a bit to the left of the violinist (LHS) and towards the end of the piano (RHS).  The mid-way point of the array was focused directly between the pianist and violinist.  ie I aimed to get the piano's full width from about the mid-point all the way right.  It's interesting that Voltronic feels the piano is hard-right. That wasn't the intention!  I mixed for speakers (using my Naim/PMC set-up which is very revealing), and checked on headphones (Sennheiser HD650) - no panning was applied to the main array mic signals in the mix.

I'll post the 3 separate mic pair recordings from that sample a bit later and see what you think.

Thanks again, guys! (I'm hoping others are finding this voyage of discovery helpful and interesting too)

Actually, I think it's a very nice sounding instrument, or at least your recording makes it sound so.   It just wasn't tuned before this session, but it's not that far out that it's a big deal.

I also wasn't suggesting that you applied panning at all, but that I believed your mics were positioned a bit far to the left, making the piano appear predominantly in the right channel.  I just listened to your isolated tracks, and that's what I hear in all 3 stereo pairs.  Your explanation of the mic position confirms this. 

Maybe this is just my preference as a pianist, but I would prefer that the main array be centered on the piano, not the center of this particular ensemble.  If this were a string quartet or woodwind quintet then that's a different story.  But the harmonic foundation of the music here is the piano, and that should be much more centered (again, IMHO).  Otherwise all of the mid / low frequency info is coming from the right.  Think of it from a concert hall perspective.  A violin sonata, opera aria, etc. is always going to have the piano centered (or close to it) with the soloist directly in front of the piano or off near the tail.

Side question - how did the pianist and violinist maintain eye contact if the violinist was left of the piano?  Was the piano not facing the audience, and angled somewhat with the tail forward?  That would also help explain the imaging I'm hearing.

[voltronic]

The HVAC rumble is easily taken care of with iZotope RX.  I can send you a lightly denoised sample if you want.

I'd not heard of iZotope RX before you mentioned it, Voltronic - looked it up and although it's pretty expensive, it looks extremely powerful.  Not sure I can really justify that kind of expense but it's definitely on my "if only I had the money" list :-)

Yeah, but I get it for half price because of the educator discount. Otherwise I probably wouldn't have shelled out for it.  But since I almost always am recording acoustic music in schools or churches with loud HVAC, it's pretty much mandatory for my recordings not to be a rumbling mess.  It does require a gentle touch and a lot of experimentation to get it to really work with music in a transparent way, and it's taken me a long time to arrive at a good starting point, and I adjust from there as needed.  You also need a good clean passage of just the hall noise without coughs, claps, stand squeaks, etc. for your noise profile.  I'm probably making it sound like it's a very difficult program to use, but it's not.  Nothing else I've ever seen comes close to it for noise reduction, and then there's all of the other great things that it does as well.

Here's your latest mix after about 5 minutes of work in RX to give you an idea what it can do.  I could have removed a bit more (or a lot more) of the noise, but then it starts to deaden the natural ambiance, and beyond that can introduce artifacts.
http://www.filedropper.com/sample10denoised