Freesound Forums

I recently discovered this terrific site, and I'm looking forward to uploading many field recordings.

As a start, I just uploaded a quick comparison of stereo microphone techniques that some may find useful. I made three simultaneous recordings (from about the same position) with three different standard microphone arrangements: Mid/Side, X/Y, and Jecklin disk. Since the three recordings were made from exactly the same source, at the same time and from (close to) the same position you can directly compare the resulting stereo images and other characteristics of these standard mic arrangements. I'll upload additional comparision sets for various other kinds of field recording sources (nature, auto traffic, etc) if there's interest.

This first set records a moving train. Each 5'34" recording captures a long freight train passing approximately 10 feet in front of the microphones (from left to right). Note that the volume range is VERY wide here, with very faint sounds audible at the begining and end and with the input levels set to accomodate the loudest sounds. A passing train really pushes the dynamic range and noise floor of a microphone, especially in 24 bit recordings.

The Mid/Side recording was made with a pair of Sennheiser MKH-800 microphones inside a Rycote blimp. The "mid" microphone was set to a "wide cardioid" pattern and the "side" was set to "figure-8" . The recording was then mixed down to standard A-B stereo.

The X/Y cardioid recording was made with a Rode NT4 X-Y stereo microphone (with a Rycote "Windjammer".

The Jecklin recording was made with a pair of Audio-Technica AT-3032 omni-directional microphones (with Rycote "ball gag" windscreens) mounted on a Jecklin disk.

Everything was captured at 96KHz/24 bit with a Sound Devices 744T and a Marantz PMD-671, with the final mix (to slightly normalize levels and do the M/S processing) done in Logic Pro.

Compressed (256Kbps) MP3 versions of the set can be found in the microphone-comparison-freight-train-mp3 sample pack. Uncompressed 96K/24 WAV versions are available in the microphone-comparison-freight-train-wav sample pack. (Warning: the uncompressed recordings are seriously BIG -- about 190MB each!)

Thanx for all the info. I'll link here the sample pack for convenience of interested readers: http://freesound.iua.upf.edu/packsViewSingle.php?id=1416

saludos

And thanks for your very kind comment on the M/S sample, Dobroide.

mab

Thanks, I'll listen to it when I have a chance... very useful

Hi,

Havent had time to dig into your recordings properly, but im curious why you didnt choose to do an x/y with the mkh-800's (im jealous). It seems a bit unfair to compare a 400 dollar setup with a 6000 dollar setup (mike wise) not even considering different recorders. On the other hand, i see the point of having simultaneous recordings. Hmm anyways, this is still a very interesting comparision of three totally different setups, i guess all im saying is that the subject of the thread is a little misleading with all the other variables.
Thanks!

Anton,

Yes, absolutely, and if anyone wants to send me a few more pairs of MKH-800's, I'd be glad to redo the recordings. As it is, I made these mostly for my own use, to compare the tools I have at my disposal.

Seriously, though, I think these samples are mainly useful to compare the stereo images produced by the various mic arrangements (which depends more on the microphone patterns than on the quality of the capsules) without regard to the quality of the audio reproduction itself (which of course depends quite heavily on the quality of the capsules). Actually, in this particular case, while I think the MKH-800 M/S does indeed beat the NT4 on both sound quality and imaging, the sample I personally like the best overall is probably the one from the Jecklin disk with the AT-3032s -- which is the least expensive system of the three I used.

Best,

mab

Really, I was surprise at how *good* was the NT4 was at getting a stereo image of distant sounds (i.e., beginning and end of the recording). My personal opinion is that XY outperforms M-S in this respect. But, boy, as the train got closer... WHOA, the MKH worked damn good!
I expect getting an affordable figure-8 AKG in the near future to combine with a Senn ME66 in M-S. This combination will probably yield a more 'fair' comparison (as regards price) with the NT4. I'll let you know then.

saludos

As being the creator of this site I can only say two things:

1. it pleases me tremendously to see people so dedicated to high quality field recordings join the site!!
2. I hope to see many more of your recordings show up on freesound, as these are already quite exceptional

Did you ask the machinist to blow the horn continuously??

- bram

I've added a new set of stereo microphone technique comparisons, with the same three arrangements but a (slightly) different kind of source: traffic passing on a country road. Again, three recordings were made simultaneously, with mid-side, Jecklin, and X-Y mics at about the same position. This is perhaps a more interesting stereo image comparision than the passing train, because a prolifieration of crickets provides a mix of "static" content along with the "moving" content across the soundstage. (And again, for my money, the Jecklin disc provides the most persausive stereo image here -- even though captured with the least expensive microphones used in the three recordings -- but people of good will might disagree with me).

These recordings can be found in the mab-microphone-comparison-traffic-wav sample pack at http://freesound.iua.upf.edu/packsViewSingle.php?id=1431

mab

I do prefer the Jecklin recording. The other two leave me with an uncomfortable lack of sensation in one ear after a car has passed.

I can't hear the crickets, so can't comment on them!

Mike

I've added another set of stereo microphone technique comparisions, this time of the sound of choppy water breaking against a rocky shore. Mics used were an MKH-800 mid-side pair, a pair of AT-3032s in a Jecklin disk, and, to be a little different, a Crown PZM SASS-P MKII quasi-binaural "head". The recordings were made at the edge of of the San Francisco Bay in Berkeley, CA, with the microphones positioned approximately 6 feet (2 meters) from the Bay's edge, oriented toward the water. A small airplane flying overhead becomes audible at about one minute into the recordings. Again, the recordings were made simultaneously and from approximately the same position.

The stereo images and overall characters of the recordings are far more different from one another here than they are with the train or traffic comparison samples. (Personally I prefer the SASS recording, but again, others may disagree).

For some reason these recordings seem to suffer a quite a bit more than usual from the "preview" compression, so you may need to actually download them rather than preview them to make meaningful comparisions.

http://freesound.iua.upf.edu/packsViewSingle.php?id=1451

mab

mab
For some reason these recordings seem to suffer a quite a bit more than usual from the "preview" compression, so you may need to actually download them rather than preview them to get any meaningful comparision.

mp3 doesn't like noisy source material or source material with an inherent noisy character (think wind, waterfalls, water, etc etc)...

- bram