Wednesday, December 20, 2006

Padrick Dunns: localization Test

For localization I selected six different test rigs. They are tested in groups of
two. Starting with:

13'' spread side-facing capsules w/baffle Em 158 and 21'' spread side facing capsule w/baffle

T01-01 and T03-01

I listened to each one about six time while I drew a circle and placed each mark so I could localize each mark. This goes for all of the tests. At first I did it looking at the computer but it took my attention away from where the sound was really coming from. When I closed my eyes I came up with this conclusion. Starting with the T01-01 13'' spread the separation of the sound was much more clear than the 21'' spread this could be because ot the piece of cardboard in the middle. It seemed that the variation is much different from the T010-01 to the T03-01. The sound from T01-01 has separation and you can tell where each sound is coming from where the sound from T03-01 10-12 degrees is way off they sound like their are on top of each other. Where 12-1 their is a large seperation. In conclusion the sound is much more clear when the mics are closer together in this case.

The next two are:

Front-facing w/Baffle 13'' spread em158 and Front-facing w/baffle 21'' spread em158 mics

T01-02 and T03-02

Starting with the T01-02 13'' spread 12-1 sounded right next to each other but 11-12 sounded like there was a large amount of separation. Where in the T03-02 10-11 sounded right next to each other. 12-2 also sounded on top of each other. Each group had its separations but it seems that the closer the mics are together the better the localization of each sound.

The next two are:

13''spread ORTF Em 158 mics baffle and 21'' spread ORTF Em 158 mics baffle

T01-03 and T03-03

The most interesting group was the last only becaule it was different from the first two. As I started with the T01-03 the degree of 10-12 sounded right next to each other as well as the 12-2. It seems that this was the worst test for localization. It was hard to distinguish between the degrees of 10 and 12, also 12 and 2. Then I shifted gears and listened to the T 03-03. The further the spread in teh ORTF baffle the better the separation of the sound. The localizaton was much better and much more clear. In conclusion the further the mic the better the localization in the ORTF.

Conclulion:

If I could pic one mic out of the six for localization it would be the 21'' spread ORTF Em 158 mics baffle. If I could pic a second runner up it would be the 13'' spread side facing capsules w/baffle Em 158. I found it weird that the closer the side facing baffle the better the sound quality and the further apart the ORTF the better the sound quality.

1 comment:

Rob D. said...

Hi Padrick--

(1) T01-01 vs T03-01: there is considerable support for the conclusion that a baffle helps.

(2) T01-02 vs T03-02: better localization closer at 13", there is considerable support for the conclusion that separations on the order of the spacing of the ears help=-- especially when the results are being tests with headphones

(3) T01-03 vs T03-03: So 13" vs 21" spread both at ORTF and the wider one is better? These factors get down to a case by case, location specific level of discernment, really. I'd be much more confident with the decision if the rigs were both on the center line with same sized baffles and recorded the exact same impulses, You video is in a format I can't play so I get see how close the stations are to each other. We had so many baffles "up" for that particular test that I fear we blocked the direct sound from some impulses for some rigs more than once! In playing your comparison I found the 2nd rig better at hinting at the 10 and 2 positions where is 8-10 is pretty much stationary at with the first rig. In any case, the localization with these two rigs is much poorer than other rigs like the parallel boundaries or the styrohead, so if localization for headphone playback is the goal,..we could leave the long dowels at home, at least for this type of space and subject to mic distances (30').

Front facing mics tend to create a brighter, louder center which is more symmetrical tonally and more frontal even if too bright in the middle.

Side facing mics should make steps 8-10 clearer and ORTF should make 9-11 clearer but the problem is the "holes" created the Side-Facing and ORTF rigs are warmer in frequency response and the strikes there seem more diffuse and "further away." So the stereo field with side facing and ORTF, though sometimes more dramatic, often has more bumps in it than front facing. This portrayal of the spatial imaging relations between the rigs would probably be argued among professionals though. You have gotten off to a good start on forming your own impressions over time. Rob D.