Recording my own violin playing

  • Thread starter Thread starter eug_fiddler
  • Start date Start date
I believe that is a total budget, and he followed up with a post asking for a $100 mic. A pair of 414s is close to $2K, I believe.
Yes, that's right... $500 total for a pair of mics, an interface and cables/stands/etc.
 
Ethan said it well. "Ever heard a CD that blew you away? Then 44.1 and 16 bit is enough."
 
Seriously, the quality of the converters is much more important than the sample rates and bit depth.
 
So, basically, MSH, you are saying, in agreement with my points, that a 44.1KHz or 48KHz sample rate is, perhaps "adequate" in your opinion, but not ideal. And that a 96KHz rate is better or ideal. Further, as you point out, 24 bits would give better dynamic range. Which was my point exactly.

About 10 years ago, when I was 42, I would have loved to compare audiometry and hearing with your 8 year old daughter. Assuming, of course, that we could have had her around then to compare with. I had my hearing evaluated, in detail, and it wound up being better, across the spectrum, than the norm for a 19 year old. At the age of 42. This has always been a two-edged sword: I hear very well, and have trained myself on top of it. I can hear lots that others can't, and it enriches my appreciation of music and many other things. Disadvantages include the need for much better stereo equipment than most of my peers need (what do you think has driven my having been for a long time a "serious analog hobbiest handy with a soldering iron?"), awareness of the shortcomings of certain recordings or performances, and certainly a clear awareness of the shortcomings of the almighty CD. One bit of terminology that I simply hate is the mantra of "CD quality sound." UGH! That said, the fact that my hearing has always been very, very much better than average, I must sadly admit that I think I am losing some of my high end lately.

I guess I have taken it as an assumption that classical musicians would take good care of their hearing. But your point is well made. The volume level of an orchestra having at it full tilt is impressive, probably even to pop and rock fans that have never heard what over 100 acoustic instruments can do.

Thanks to eug_fiddler too for the comment about the flat-response ear plugs. I have just, literally while working on this post, have orbited around my daughter who was deep in practice, and got up close near her left ear. Had I realized that it was that loud before I would have insisted on plugs or something. Will look in to that.

PS: MSH my Revox, and the type I would recommend is a high-speed (15ips/7.5ips), half-track model. Original specs, were 30Hz to 20KHz +/- 1.5 dB, S/N ratio = better than 66 dB NO Dolby, which I can't stand the artifacts from. Now, fully tune up and tweak a deck like that for a modern +6dB tape and you gain in S/N ratio, dynamic range and using the high speed you get great high freq response (albeit at huge expense as it devours a reel of tape-about $50 with shipping-every 1/2 hour). And, yes, an A77's record pre-amps can easily manage +6 without distorting and with plenty of headroom left.

Let's turn to recommending equipment, within the guys budget, and each give our reasons for doing so. If you have any further things to discuss with me about sampling theory or whatever, please do it in a PM. As I pointed out at the beginning of this post, it seems that we are in agreement on most points at this point (44.1 is "adequate" not ideal, 96Khz is better, he wants to spend $500 max for the whole shebang, etc.). We all can learn from a reasoned discussion of the advantages of miking a violin with an omni, and the constraints it imposes vs. the advantages of using a cardioid and the constraints that would impose. Ditto large diaphram vs. small, etc.
 
Ethan said it well. "Ever heard a CD that blew you away? Then 44.1 and 16 bit is enough."

NO! Please let me rephrase that for you.

"Ever heard a CD that blew you away? Then you are deaf and your hearing untrained."
 
Ok, deaf violinist coming through...

Here's something that might blow your mind N7SC, I've heard astounding performances live on the RADIO with poor reception that blew me away. ;)
 
[edit]argh the cursed double post. might as make it worthwhile

I don't think it's such a great idea running around calling people deaf just because you don't like their opinions. I dealt with lots of audiophiles in the past who had the same symptoms... Just take it easy a bit, and don't spend $2000 on interconnects. ;) (it's a joke, i'm not saying that you do, or that it's a crime to do so)

Also good hearing has nothing to do with musicality. Beethoven was completely deaf for a good part of his career and he was one of the best musicians the world has ever known. Let's just drop the hearing thing, I think we've exhausted this topic.
 
Ok, deaf violinist coming through...

Here's something that might blow your mind N7SC, I've heard astounding performances live on the RADIO with poor reception that blew me away. ;)

Doesn't astound me a bit. Your point is well taken. Some of my favorite performances are quite old, like orchestral ones conducted by Toscanini. Still, if there is modern technology that can make clear and accurate recordings, I say we should try for the best, not just the "adequate." If we assume brilliant performances (that is your department), then it would be a disservice to mask the nuances with merely "adequate" recording technique.

I know that seems to argue both directions at once, but, for example, as good as Toscanini's recording of, say Beethoven's Symphony in D minor (9th) is, imagine it captured with the very best, most accurate modern recording techniques instead of the early 1950's stuff that was used. Or, to go the other way, imagine, say a great performance by Shaham muddied up by being recorded at a merely "adequate" level of sampling. See my point.

However, what I think this whole thread is about is helping you get the best, most accurate, recordings that you can, so you can put your best foot (or whatever) forward to the juries you are going to submit to. I think that if we capture every nuance and subtlety, so much the better. It is up to you, the musician, the magic man, to deliver the right nuances in the right places to blow the jury away.
 
We might be talking about two different things here. By the reference to sampling rate and bit depth, APL was clearly talking about the technical aspects, as was my response to him. You, especially as a musician and thus, presumably, a music lover, are talking performance. While I get it, there is no reason to sell your self short on recording quality.
 
It is up to you, the musician, the magic man, to deliver the right nuances in the right places to blow the public away.
Fixed your post. Too much emphasis on juries and competitions these days...

Yes, like duh obviously, we're trying to find the best equipment for me... for a paltry semi-student-poor-musician's budget of $500. :)

... I'll take donations, though!
 
Fiddler, you made your point very well by directing me to my own opinions of superb performances from the dark ages. No one would, hypothetically, deny Heifetz a slot in a graduate program because of he chose to sample at 44.1KHz (if such was available to him). They might deny him because he is better than any of their professors, but that is a different story . . .

So, let's make it a game of trying to deliver the best we can for the lowest price. We'll make the usual dealing-with-a-student assumption that the budget quoted involves a bit of scrimping and scraping so we'll try to come in lower as much as possible.

BTW, I happen to be one of the few audiophiles that does not spend $2000 on interconnects. I willingly admit that I can not hear the difference. Nor do I have Shakti Stones (don't ask, you'd never believe it) sitting on my equipment. I'm strictly a rationalist, tempered by what I can hear. Actually, I have a tape splicing block, and an Anna Russell CD sitting on top of my tuner at the moment.

My crack about being deaf was not aimed at anyone in particular, just updating the bit about 44.1 and 16 bits being adequate.
 
Fiddler, you made your point very well by directing me to my own opinions of superb performances from the dark ages. No one would, hypothetically, deny Heifetz a slot in a graduate program because of he chose to sample at 44.1KHz (if such was available to him). They might deny him because he is better than any of their professors, but that is a different story . . .

So, let's make it a game of trying to deliver the best we can for the lowest price. We'll make the usual dealing-with-a-student assumption that the budget quoted involves a bit of scrimping and scraping so we'll try to come in lower as much as possible.

BTW, I happen to be one of the few audiophiles that does not spend $2000 on interconnects. I willingly admit that I can not hear the difference. Nor do I have Shakti Stones (don't ask, you'd never believe it) sitting on my equipment. I'm strictly a rationalist, tempered by what I can hear. Actually, I have a tape splicing block, and an Anna Russell CD sitting on top of my tuner at the moment.

My crack about being deaf was not aimed at anyone in particular, just updating the bit about 44.1 and 16 bits being adequate.

You're snooty and bossy, too, lol.

I'm a poseur audiophile. I had Jeff Beck's Blow By Blow in the Rotel RCD-950 the other night and was so enjoying the detail in that recording. Not quite blown away, but pleasantly into it.

So, is there a CD that sounds like a good recording to you? Do you have a CD player that's good enough to bring out the best of the CD? Do you have a room that sounds decent enough to bring out the best in your speakers?

Anyway, the sample rate argument is moot because the final product will likely be a CD. If the final product is 44.1 then there's no point in sampling faster during recording. Downsampling to 44.1 will introduce artifacts, especially at non-integer multiples. The nice part about recording at 24 bit is that you've got lots of headroom and don't have to worry so much about clipping.

And if the final product is an MP3, well...

Oh, big props to you for being a rationalist about snake oil audio products.
 
Sorry I don't have PMs active . . .

One clarification: I don't consider 96kHz ideal, I consider 64kHz ideal. But very, very few converters support it, so high rates work fine, up to 96kHz. 96kHz wastes a bit of HD space and CPU cycles, but that's not critical anymore, so it works. Above that I would consider technically worse. I tend to work at 48kHz.

As for the Revox specs, let's just agree that it is not comparable to a mastering-grade 1/2" deck. It's a consumer solution that is comparable to consumer grade digital audio. -1.5dB at 20kHz is about the same as 44.1khz digital. Digital is flat down to 20Hz; most tape decks can't manage that. That doesn't matter at all for violin. Tape may have desirable euphonic characteristics; if so, these are distortions.

Dynamic range is a non-issue for any digital format in recording solo violin. I think my Soundblaster, used with care, would be fine in that respect--and it's a noisy piece of crap. Any reasonable tape solution would also be fine, provided that gain staging is managed properly (which is also a requirement of digital, but less critical at 24 bit).

OK, more interesting topics: polar response and its effects on microphone selection. I don't have much to say that was not said by the great Harvey Gerst in this thread:

https://homerecording.com/bbs/showthread.php?t=27030

But I know that is long. Instead, look at this page, while we are talking about the C414 series:

http://www.akg.com/site/products/powerslave,id,782,pid,782,nodeid,2,_language,EN,view,diagram.html

One of Harvey's basic posts describes the process of creating a directional microphone--it involves sound incident to the rear of the microphone arriving at the rear of the capsule slightly delayed so as to cancel with the arrival at the front. That is a frequency-dependent effect, and it creates a resonant peak. That large resonant peak is the simple case; it is further canceled by making a more complicated arrangement which reduces that large resonant peak into smaller resonant peaks. Once you are done, you generally have a fairly flat response.

Added to that will be resonances from the case, grille, and anything else the manufacturer throws in there.

The C414 (B-XLII in this case) has a series of varying resonant peaks, according to the polar pattern. Sometimes you select a polar pattern based upon the need for off-axis rejection. That varies in intensity and orientation--figure eight has near-perfect off-axis rejection at 90 degrees. It is useful in close-micing ensembles where isolation is desired, for example. Moving along to hyper/supercardioid, that has very good rejection at 120 degrees, very good for stage use where wedges need to be rejected. Then there is the usual cardioid, good rejection at 180 degrees, finally subcardioid and omni.

OK, so which do we use for a solo instrument? As you can see, the resonances vary according to the polar pattern. With a single mic in a good room (or truly dead room), you would select the pattern that complements the instrument the best. I have no way of knowing which that is in advance of a session (although I will allow myself to note the omni pattern's response lacks a presence peak :) )

The complicating factor is stereo recording. Stereo recording involves first selecting a stereo micing technique, which will indicate a polar pattern. Further complicating is the off-axis mic orientation in many stereo recording patterns. That will yield a different response, in a lesser mic than the AKG, that could be a good thing.

Some mic techniques used for solo violin (and other instruments) are not true stereo arrangements, but rather spaced pairs, for example above and below the instrument, or near and farfield mics. In that case, it is not necessary to use identical microphones.

Microphones in the low-budget category that could be considered for solo violin:

- Small diaphragm condenser mics. More accurate transient response; more consistent off-axis response. Really cheap cardioid ones tend to have significant presence peaks, which I really think is a bad idea for violin. The cheapest reasonably flat cardioid mic I can think of is the Shure SM81, which is very commonly available used for about $200. Also popular, although I haven't used them, is the Oktava MC012 (darkish), and the MXL 603 or 604 (bright, but tameable). I believe the 604 is the model where you can get omni caps. That's true of the Oktava too; it used to be true of the Shure, but the omni cap for the SM81 is long discontinued and impossible to find on the used market. Audio-Technica has a range of inexpensive small-diaphragm mics that are worthy of consideration; we use I think 4041s as choir overheads (and off-axis pipe organ!) at church and they work well. They also have entries in their 20 and 30 series. I would avoid the AKG C1000, universally derided as perhaps their worst microphone. There are probably better choices in their range, but I don't know them.

I believe choices in polar pattern are good . . . but a reasonable cardioid mic used in a typical stereo configuration will be between 45 and 55 degrees off-axis, where they are flatter and darker.

- Large diaphragm condenser mics. Greater variance in off-axis response; slower transient response. Often viewed as euphonic, especially on vocals. Quieter than SDCs, due to higher capsule output. Not used as frequently in off-axis stereo patterns, but popular for techniques such as spaced pair (on drum overheads, for example), or on-axis stereo such as mid-side or Blumlein. However, slim pickin's for solo violin in the $100 range. How much is the CAD M179? People seem to like that one, and it's switchable pattern. Beware of the SDC hiding in an LDC body--not a bad thing per se, it could sound great. Just don't be fooled into thinking you will get the characteristics of an LDC. You might get something clever, or you might get an SDC with added case resonance.

Before we leave the topic of condenser microphones, you will hear recommendations for tube mics. Everybody loves tubes! I love tubes! I love tubes on violin! Some designs are better than others, but I think $100 is probably pushing it.

lastly:

- Ribbon mics. Smooth high-frequency response, but often not much of it. Low sensitivity. Most are figure-eight pattern. Ribbon mics are hugely trendy. I'm not a ribbon fanboi, and that's not because I make condensers. I've actually done a lot of work with ribbon mics, but at $70 (watch for sales) the Chinese have cornered the cheap end of the market. I think they are good at what they do, I just don't think they do everything. People that use them on drum overheads, I think they should buy better cymbals :p but that is my bias.

On violin, they can be useful on a bright instrument where close-micing is required. They're good at instant fiddle. I would not select one of the cheaper ones for distant micing a violin, there's just not enough there. Beyerdynamic makes a couple of midrange models worthy of consideration; I satisfy myself with an old Shure 315. Perhaps once you've got a setup and are looking for another tone, but before then I'd look in the SDC range for a primary pair.


OK? Interfaces I already said I don't really know. I think in the same price range they are probably comparable. On the lower end, you probably won't get a brilliant analog front-end, but for your use I don't think you need that. USB vs. Firewire is not gonna be an issue for stereo recording.

Ordinarily I would never recommend mixing on headphones, but if you're not doing much processing on a stereo recording, you're probably fine. Just don't do any crazy EQ changes without rechecking in your car or something. Actually, cans might be good, since reverb stands out more, you will resist the temptation to drown your recordings with the cathedral setting :D

Building can amps sounds fun :) . . . I really am not quite sure how I stumbled into microphones. I'd rather do a small metal box with tubes sticking out of it in strange places . . . but it's a living, I guess.
 
Fiddler, you made your point very well by directing me to my own opinions of superb performances from the dark ages. No one would, hypothetically, deny Heifetz a slot in a graduate program because of he chose to sample at 44.1KHz (if such was available to him). They might deny him because he is better than any of their professors, but that is a different story . . .

My recording of Heifetz performing the complete solo violin works of Bach is 44.1kHz . . . of course originally to tape. It served me well in college, when somebody was in my room that I wanted to leave, on went the Chaconne. Very few uninitiated people could take that . . . those that could, I married them.

Darn laws against polygamy . . . :mad:
 
Downsampling to 44.1 will introduce artifacts, especially at non-integer multiples.

Oh Mr. p-l, you know better than that. SRCs don't work by division and interpolation.
 
My recording of Heifetz performing the complete solo violin works of Bach is 44.1kHz . . . of course originally to tape. It served me well in college, when somebody was in my room that I wanted to leave, on went the Chaconne. Very few uninitiated people could take that . . . those that could, I married them.

Darn laws against polygamy . . . :mad:

Unfortunately IMO Heifetz playing Bach = sacrilege anyway :P
 
Unfortunately IMO Heifetz playing Bach = sacrilege anyway :P

Well, you know what I like? It's like reading the score, except I don't read too well, so Heifetz is like a machine spitting it out, and I can just think about the counterpoint, not what the performer wanted to interpret. And when I do think about his performance, he's like "you can't play this", and speaking for myself, he's right :o

Does anybody get Bach's solo violin stuff right? In the post-Romantic period, it's probably tough. Bach liked to play viola . . . which I have noted, for some unseen reason, involves a lot less swaying than violin :D

Pipe organ's the good stuff anyway :cool:

You know, I just popped this in, and it's recorded rather badly. 1952 . . . gear now consider classic to be sure, but it's boxy, full of unpleasant resonances, and noisy to boot. I should remaster this thing ;)
 
Well, you know what I like? It's like reading the score, except I don't read too well, so Heifetz is like a machine spitting it out, and I can just think about the counterpoint, not what the performer wanted to interpret. And when I do think about his performance, he's like "you can't play this", and speaking for myself, he's right :o

Does anybody get Bach's solo violin stuff right? In the post-Romantic period, it's probably tough. Bach liked to play viola . . . which I have noted, for some unseen reason, involves a lot less swaying than violin :D
Wow that sounds the opposite of what I think his Bach sounds like... but then again it's been a while since I've heard it, maybe I'll give it another listen!

IMO the way to play solo Bach on violin is to try your damndest not to be anything like a violinist - i don't mean like a machine. I mean, Bach had like 18 kids, he was a MAN you know. But, you have to get out of the linear frame of mind that violinists are trained in, and start thinking vertically in CHORDS. At least, that's what I'm striving for...
 
Oh Mr. p-l, you know better than that. SRCs don't work by division and interpolation.

There was a couple of articles on all that at the old prorec.com by a guy who, iirc, wrote src algorithm code or something for a living, and that's what he said.
 
And when I do think about his performance, he's like "you can't play this", and speaking for myself, he's right
Although, this is EXACTLY why I never feel "happy" after hearing him. It's like hearing a space-alien do something amazing in front of you and all you can do is be awed by it, but have 0 hope of actually achieving that yourself... Someone like David Oistrakh on the other hand makes music that sounds "human". And somehow that's comforting...
 
Back
Top