[musicreo]

i would prefer recording a good graph and i noticed that you don't actually have a negative preamp, you only have a 8.5db on low frequency which is great, how you achieved that?

Impulcifer will usually give you a recording that has no clipping but there is also a gain setting in the post processing.

 

[Xam198]

I guess there is some noise that is detected as signal before the sweep starts and that causes the warning.


I had a few capsules with lot of noise and some that were very good but it is possible that this was to my soldering.

Humm , that could be it, because one mic/vxlr is "humming" and when i switch the plug mics (without changing them in my ears) , the delay decrease.
I wonder if i would have better results/less noise whith sound professional mics...
The "humm" is typically a bad ground : when i put the hand on the vxlr+ the hum drecrease.

 

[reter]

I wonder if i would have better results/less noise whith sound professional mics...

That was my initial goal but sadly i didn't find any european website that sells these professional mics, no one, so i prefered risking some tax from england with pretty cheap mics

Jaakkopasanen have the noise of every binaural mics in his guide, seems that the primo are almost literally equal to sound professional, i don't know if this is reliable tho

would be great if someone actually tried both

 

[morgin]

@jaakkopasanen im currently trying the mesh2hrtf and finding it a lot more difficult than impulcifer. But I had a thought.

I have my scans of my basic head and detailed ears. Can’t I place a virtual speaker (or several) in blender and use impulcifer for the measurements. That way we get the best of both worlds. We get a room with no background noise and we also get things like headphone compensation. I believe the way mesh2hrtf works is in reverse instead of the sound coming into the ears they have the sound come out of the ears bounce around till they find the speakers.

 

[jaakkopasanen]

@jaakkopasanen im currently trying the mesh2hrtf and finding it a lot more difficult than impulcifer. But I had a thought.

I have my scans of my basic head and detailed ears. Can’t I place a virtual speaker (or several) in blender and use impulcifer for the measurements. That way we get the best of both worlds. We get a room with no background noise and we also get things like headphone compensation. I believe the way mesh2hrtf works is in reverse instead of the sound coming into the ears they have the sound come out of the ears bounce around till they find the speakers.

I have no experience with mesh2hftf. You'd need a some kind of acoustic space simulation software for your idea but I have no experience with those either. I doubt Blender will do anything in this case

 

[reter]

I wanted to share that in the meantime I have changed the mic placement. I removed the yellow foam and place the capsule now really into the ear canal opening as shown in the image. Overall it improved the quality of the measuremt. The second image shows how I did my last measurements. I'm really happy with the results.

@musicreo sorry for the old reply but is this still the latest best results you achieved? i mean, plugging the mics this much inside the ear canal

 

[musicreo]

@musicreo sorry for the old reply but is this still the latest best results you achieved? i mean, plugging the mics this much inside the ear canal

Yes I still use the measurements from that time. I put the mics in my ear as deep as it still feels comfortable, mark the positions on the cable and fix it on the wire construction. That way the deep mic positioning is repeatable. However, I prefer my partial open ear canal measurements over my blocked ear canal measurements but that does not mean other methods can't get similar good results.

 

[reter]

@morgin do you have a picture of your mics?
I'm trying to get how can i glue these * with a pair of foam but it's so rigid that i don't want to risk breaking the cable integrity, how did you do?

Also, does the mics actual position in the ear affect the measurements? can i just glue it sided or i have to glue it so the mic is turned outwards the ear?

 

[morgin]

Sorry about the quality of the pic but you should be able to see that I stripped the housing. When inserting into the ears I looped the stripped part so it went into the ear with the foam keeping the mics facing outwards. Others may have a better solution.

Yeh keep the mics unblocked and facing more or less outward.

 

[reter]

ok this is my first experiment, and i don't actually like it... i don't even know how i can plug this mess inside my ears

maybe i should strip a lot more so i can do a serpentine inside the foam so the cable can do a U curve and exit

 

[morgin]

Just be careful where the wires are soldered on, too much wiggling around breaks the connection and soldering later can cause the mics to burn out.

 

[reter]

Just be careful where the wires are soldered on, too much wiggling around breaks the connection and soldering later can cause the mics to burn out.

i noticed copper around the white cable... how did you manage to strip a lot of the black cable without cutting the copper? it's very risky

 

[morgin]

With a sharp blade and patience. If it went wrong I could always resolder. Check the other posts tho their method might work better for you

 

[Xam198]

Humm very strange, i've tested with another - not mine - computer and little speakers, but same mics, same zoom H5, same headphones : no warning on delay left/right ear. So i've tested again with my computer and the integrated sound card rather than the - very good - EMU 1820 : again the warning ! Either it is the computer (!) or the amplifier (mix of digital and analog). I'm lost

 

[reter]

With a sharp blade and patience. If it went wrong I could always resolder. Check the other posts tho their method might work better for you

i think i will go for something like this https://micbooster.com/microphone-holders/89-primo-6-mm-rubber-holder-158a-fc034.html

i ordered a pair, but i should be able to make them from anti vibration rubber... i will not even try to make measurements with this mess i made, i'm sure would be a waste of time