Markus, the only RTA I have on my mac is Signal scope http://www.faberacoustical.com/products/signalscope_pro/
It has a 30 day trial that should be enough to do this job. You can't set the bandwidth lower than 2k but the averaging is flexible. I'd be interested in any open source alternatives too.
It has a 30 day trial that should be enough to do this job. You can't set the bandwidth lower than 2k but the averaging is flexible. I'd be interested in any open source alternatives too.
Iv'e locked my DEQ2496 to 44.1KHz from a Poppulse USB soundcard using the AES/EBU connector, so don't have the upsampling issues. Is working very well.
After the DAC dividing up the signal I'm using Rod Elliot ESP, P09b Linkwitz Riley 3 way 24db crossover board. I have 2 x 10" 6th order bandpass subs, they have so much output and my room is so small it doesn't make much difference where I put them.
There is a way that you can setup the P09b as a 12db per octave bessel for a less steep slope but I don't really need it for my setup. Might be good for you though. http://sound.westhost.com/project81.htm
check out Rod's website there is loads of good stuff on there.
col.
After the DAC dividing up the signal I'm using Rod Elliot ESP, P09b Linkwitz Riley 3 way 24db crossover board. I have 2 x 10" 6th order bandpass subs, they have so much output and my room is so small it doesn't make much difference where I put them.
There is a way that you can setup the P09b as a 12db per octave bessel for a less steep slope but I don't really need it for my setup. Might be good for you though. http://sound.westhost.com/project81.htm
check out Rod's website there is loads of good stuff on there.
col.
Originally posted by poptart Markus, the only RTA I have on my mac is Signal scope http://www.faberacoustical.com/products/signalscope_pro/
poptart, thanks a lot! The realtime FFT Analyzer is awesome. Just what I needed. 768000 spectral lines from 0 to 2756.2 Hz should be detailed enough 🙂
Best, Markus
Originally posted by col Iv'e locked my DEQ2496 to 44.1KHz from a Poppulse USB soundcard using the AES/EBU connector, so don't have the upsampling issues.
Didn't we talk about the DCX instead of the DEQ?
The DCX has the bug due to the use of the Cirrus CS8420 chip. Don't know if the DEQ makes use of it too.
Best, Markus
Best, Markus
Marcus,
I beg you to just temporarily try a different room orientation. While an absorber behind the head will "help" improve things in the current setup, there will always be reduced soundstage depth and spaciousness, and (even with a huge, thick absorber) you'll always be getting some coloration from being too near a boundary.
If the couch were swung 90 deg clockwise (resulting in the short leg at the sidewall), you would have sidewall symmetry (the Nathan reportedly thrives in a narrow but symmetrical room), and lots of diffuse, distant space behind you (the kitchen).
Looks like you got lucky with that substantial ceiling height! I'm jealous there.
No speaker in the world will remedy the wall behind your head.
No (practical) absorber or diffusor will work as well as simply having adequate distance.
I had made a similar change a couple years ago when I switched from traditional towers to more directional dipoles. Having nothing near the head makes a huge difference.
Just try it!
-- Mark
I beg you to just temporarily try a different room orientation. While an absorber behind the head will "help" improve things in the current setup, there will always be reduced soundstage depth and spaciousness, and (even with a huge, thick absorber) you'll always be getting some coloration from being too near a boundary.
If the couch were swung 90 deg clockwise (resulting in the short leg at the sidewall), you would have sidewall symmetry (the Nathan reportedly thrives in a narrow but symmetrical room), and lots of diffuse, distant space behind you (the kitchen).
Looks like you got lucky with that substantial ceiling height! I'm jealous there.
No speaker in the world will remedy the wall behind your head.
No (practical) absorber or diffusor will work as well as simply having adequate distance.
I had made a similar change a couple years ago when I switched from traditional towers to more directional dipoles. Having nothing near the head makes a huge difference.
Just try it!
-- Mark
Tubamark, tried the following setup yesterday:
Much better but not practical due to a couple of none-acoustic reasons 😉 So everything's back to the initial setup but with the couch moved 25 cm away from the back. This alone helps tremendously. Have to fetch some foam from the basement storage later today...
Best, Markus
An externally hosted image should be here but it was not working when we last tested it.
Much better but not practical due to a couple of none-acoustic reasons 😉 So everything's back to the initial setup but with the couch moved 25 cm away from the back. This alone helps tremendously. Have to fetch some foam from the basement storage later today...
Best, Markus
Markus
You need to move then😀
Seriously I agree with Tubamark. The Summa line are ideal for long narrow rooms and no speaker can correct for the back wall problem.
You need to move then😀
Seriously I agree with Tubamark. The Summa line are ideal for long narrow rooms and no speaker can correct for the back wall problem.
Here are some frequency response diagrams. Distance of Mic was 1m, 1.66 m from floor. Measured with a swept sine signal, Window 5 ms.
Smoothing 1/3 octave, 0° - 90° in 5° steps:
Smoothing 1/24 (!) octave, 0° - 90° in 10° steps:
Best, Markus
Smoothing 1/3 octave, 0° - 90° in 5° steps:
An externally hosted image should be here but it was not working when we last tested it.
Smoothing 1/24 (!) octave, 0° - 90° in 10° steps:
An externally hosted image should be here but it was not working when we last tested it.
Best, Markus
it kind of looks like the -3db point is 200hz, is that right? Is this just the tweeter, cant be. I know the Nathan doesn't have a lot of bass on its own, but that just seems shocking. I guess the -10 is around 50hz, that is more sensible. Maybe its just the scale that looks funny. Anyway, the rest looks great, the listening axis response is smooth with the expected smooth increasing directivity.
Since I am buying the Abbey's, I'm excited to finally get them and get them assembled. It's too bad I can't start on some of the work now so I don't have as long to wait for getting them up and running. I should also have measurements of that when all is said and done. My measurement system is different, but the results should be interesting none the less.
Oh I'm an idiot, I just realized why your low end looks like it does. To give a more pseudo anechoic response, try splicing in a close mic response of the woofer, that's typically how responses are given in publications and such.
Since I am buying the Abbey's, I'm excited to finally get them and get them assembled. It's too bad I can't start on some of the work now so I don't have as long to wait for getting them up and running. I should also have measurements of that when all is said and done. My measurement system is different, but the results should be interesting none the less.
Oh I'm an idiot, I just realized why your low end looks like it does. To give a more pseudo anechoic response, try splicing in a close mic response of the woofer, that's typically how responses are given in publications and such.
pjpoes, window is 5 ms so the diagram is valid from 200 Hz on or even more likely from 400 Hz on.
This is a nearfield measurement of the woofer:
Crossover point (-6 db) is around 84 Hz.
Best, Markus
This is a nearfield measurement of the woofer:
An externally hosted image should be here but it was not working when we last tested it.
Crossover point (-6 db) is around 84 Hz.
Best, Markus
I suspect that Markus has a back wall in his data - right? This is exactly what I would expect in this case. Fill in the low end with some subs and you will have exactly what I seek in an in room response which is a slight rise at the low end. This depends on typical playback levelsof course as a rise at LF will sound bass heavy at high volumes, but natural at low volumes.
No there's no room influence in the data I've shown from around 200 - 400 Hz on. The woofer data is a nearfield measurement so there's no room influence as well.
Is that with FuzzMeasure Markus? Did you use a marked rotating table or just copiously did the 5deg intervals by using a protractor and turning the speaker by hand? Looks like a very steady measurement lot of curves.
markus76 said:No there's no room influence in the data I've shown from around 200 - 400 Hz on. The woofer data is a nearfield measurement so there's no room influence as well.
The wide response peak in the 200-500 Hz region is not in my data, but mine is completly free field. How did you get away from the room boundaries? Not in the apartment that you showed!
Otherwise our data is pretty much identical, except that after about 45 degrees my data falls much more than your does. Again, I attributed this to room boundary effects. No so?
the in-room measured responses (near&far field) are actually very good (similar curves to my listening taste)
which software did you use Markus?
which software did you use Markus?
Salas, moved the speaker by hand...
Earl, speaker and mic were 1.66 from the floor in the middle of the room. So there's a reflection free time window of 7.6 ms. I applied an even shorter window of 5 ms that makes the data valid from 200 Hz on.
All measurements were done with FuzzMeasure.
Best, Markus
Earl, speaker and mic were 1.66 from the floor in the middle of the room. So there's a reflection free time window of 7.6 ms. I applied an even shorter window of 5 ms that makes the data valid from 200 Hz on.
All measurements were done with FuzzMeasure.
Best, Markus
- Status
- Not open for further replies.
- Home
- Loudspeakers
- Multi-Way
- Setting up the Nathan 10