Geddes - I don't agree with the idea that we don't know anything.
If it was that audible, masses of people would complain about it, which they don't.
So even without exactly knowing the numbers, we have a decent idea of order of magnitude relative to all the other problems and hurdles.
Knowing that exact number is maybe fun as an academic exercise, but that's about it.
If it was that audible, masses of people would complain about it, which they don't.
So even without exactly knowing the numbers, we have a decent idea of order of magnitude relative to all the other problems and hurdles.
Knowing that exact number is maybe fun as an academic exercise, but that's about it.
Umm thats not a sound approach....relying on mass opinion...lol...how many people do you know with low efficiency systems vs high efficiency....or that smoke cigarettes.
Sht I'm still (was rather) trying to convince the mass of studio engineers that a 15" + horn/waveguide is a pinnacle of loudspeaker design...and they are supposed to be experts on sound reproduction....choose your "mass" wisely
Sht I'm still (was rather) trying to convince the mass of studio engineers that a 15" + horn/waveguide is a pinnacle of loudspeaker design...and they are supposed to be experts on sound reproduction....choose your "mass" wisely
Last edited:
LIR Linear Phase Crossovers
A White Paper from Linea Research
http://linea-research.co.uk/wp-cont...ssets/Tech Docs/LIR_LinearPhaseCrossovers.pdf
Linear Phase Crossovers
http://www.felusch.com/ton/Lake/004_Tech_Papers/Dolby_-_Linear_Phase_Crossovers.pdf
Phase coherent crossover networks
Pass LabsPhase coherent crossover networks - Pass Labs
A White Paper from Linea Research
http://linea-research.co.uk/wp-cont...ssets/Tech Docs/LIR_LinearPhaseCrossovers.pdf
Linear Phase Crossovers
http://www.felusch.com/ton/Lake/004_Tech_Papers/Dolby_-_Linear_Phase_Crossovers.pdf
Phase coherent crossover networks
Pass LabsPhase coherent crossover networks - Pass Labs
Last edited:
Lets nit picking arguments with extremes...., you're missing the bigger picture as well as the point here. Even in more controlled tests, the contribution isn't that obvious at all.
Obviously it's not about those extremes.
Btw, it's not about opinion here either.
Reflection, reverb, standing waves and even certain amount of distortion (above an obvious threshold) are very harmful, even to the least experienced listeners. To some smaller extend, even directivity is.
Thought experiment again; just make a list of a speaker system (incl room acoustics and what have you), of all things that can contribute to be harmful to the original signal or to a proper listening experience.
Next, prioritize them based on order of magnitude of how important/harmful they are.
I bet you will be amazed.
Not so sound approach? They have been doing this for literally centuries in other fields of science.
More than often leading to extremely accurate results and predictions.
btw, I would highly recommend reading a bit about Fermi problems as well.
Obviously it's not about those extremes.
Btw, it's not about opinion here either.
Reflection, reverb, standing waves and even certain amount of distortion (above an obvious threshold) are very harmful, even to the least experienced listeners. To some smaller extend, even directivity is.
Thought experiment again; just make a list of a speaker system (incl room acoustics and what have you), of all things that can contribute to be harmful to the original signal or to a proper listening experience.
Next, prioritize them based on order of magnitude of how important/harmful they are.
I bet you will be amazed.
Not so sound approach? They have been doing this for literally centuries in other fields of science.
More than often leading to extremely accurate results and predictions.
btw, I would highly recommend reading a bit about Fermi problems as well.
Last edited:
camplo, the frequency and time domains are linked it is what allows Fourier transforms to work, the impulse response is measured and the frequency and phase can be derived from it and vice versa. If the information is in one domain it is also in the other, that is why mabat is telling you it is the same data.
Sometimes things are easier to see in one view or another but it is there in every view somewhere.
https://acris.aalto.fi/ws/portalfiles/portal/52513428/Audibility_of_Loudspeaker_Group_Delay_Characteristics_AAM.pdf
Here is some research that could be seen as a good target to aim for. In this research the impulse responses of varying amounts of group delay were time reversed and listened to forwards and reversed to see if they sounded different. Whilst this doesn't say what the actual level of audibility is if you keep below these levels you can reverse your impulse response and likely not hear the difference. The levels where is was hard to hear and obviously different give some clues where things start to go wrong.
This image is the conclusion for those that don't want to read the whole thing
Sometimes things are easier to see in one view or another but it is there in every view somewhere.
https://acris.aalto.fi/ws/portalfiles/portal/52513428/Audibility_of_Loudspeaker_Group_Delay_Characteristics_AAM.pdf
Here is some research that could be seen as a good target to aim for. In this research the impulse responses of varying amounts of group delay were time reversed and listened to forwards and reversed to see if they sounded different. Whilst this doesn't say what the actual level of audibility is if you keep below these levels you can reverse your impulse response and likely not hear the difference. The levels where is was hard to hear and obviously different give some clues where things start to go wrong.
This image is the conclusion for those that don't want to read the whole thing
Attachments
Forgive me if I’m wrong but isn’t this all about making the best possible studio/mastering monitors possible? I would think then that the nitpicking would be relevant?
I have no idea what’s being discussed except the broad strokes, this is just an observation as to why Camplo is/might be putting so much effort towards something as ‘perfect’ as possible.
I have no idea what’s being discussed except the broad strokes, this is just an observation as to why Camplo is/might be putting so much effort towards something as ‘perfect’ as possible.
I was under the impression they were to be near field (ish) enough to be out of the rooms influence? But then again this has been like following a Netflix drama series! 😀
Oh yeah maybe you guys will like this… Linear phase can't fix group delay/Time to peak energy… Unless it addresses the amplitude envelope....(I don't think it does but does it?)
Excess time to Peak energy is effectively the same as dynamic compression with an instant attack but release time similar to the m/s read as GD at the corresponding frequency
Dynamic compression a.k.a. dynamic attenuation in the terms of terms excess time to peak energy, caused by pressure, is a result of the diaphragm not being able to over come resistance by the end of the first cycle...as the frequency continues...each oscillation is another chance, aided by all sources of rebound (air spring, suspension, voice coil, translates to momentum gained by the time the diaphragm reaches center on its way to opposing amplitude) Until max velocity is reached.
.......yeeaaeeee =)
Excess time to Peak energy is effectively the same as dynamic compression with an instant attack but release time similar to the m/s read as GD at the corresponding frequency
Dynamic compression a.k.a. dynamic attenuation in the terms of terms excess time to peak energy, caused by pressure, is a result of the diaphragm not being able to over come resistance by the end of the first cycle...as the frequency continues...each oscillation is another chance, aided by all sources of rebound (air spring, suspension, voice coil, translates to momentum gained by the time the diaphragm reaches center on its way to opposing amplitude) Until max velocity is reached.
.......yeeaaeeee =)
Is anyone taking notice to how fluids posted graphs supports moving the xo 1.5-2x above fb....this is fun =)
Remember my 350hz horn example with the 3ms gd peak at 350hz?
Whats that? 2.7ms at 350hz....group C? Mostly Judged different? For real? Whats that? Lets put a filter on top of it to exacerbate GD even further...oooookkkkkkkk =(
Bad sq within crossband = bad crossover performance
Gd is also not good in areas around fb....move xo 1.5-2x above fb = better gd performance before adding filter and afterrrrr
Remember my 350hz horn example with the 3ms gd peak at 350hz?
Whats that? 2.7ms at 350hz....group C? Mostly Judged different? For real? Whats that? Lets put a filter on top of it to exacerbate GD even further...oooookkkkkkkk =(
Bad sq within crossband = bad crossover performance
Gd is also not good in areas around fb....move xo 1.5-2x above fb = better gd performance before adding filter and afterrrrr
Last edited:
Yeah I also have that thought for 800 pages as well.
Thought it was just a small side track.
But I can't keep up with the ping ponging
Well, seen from a "perfect" point of view, I would start with the recording 😀
I guess we shouldn't stop you from wanting to move the crossover higher. Even if moving it 1.5 to 2x would get you at what frequency... somewhere between 1050 to 1400? Really? You wanted 300 not too long ago. 🙄 You are talking about octaves, right? One octave above 350 being 700...
Last edited:
So you're looking for half to a whole octave higher. Not 1.5 to 2 octaves at all.
If that's your answer, be sure to EQ the output flat below fb before applying the crossover. Or you won't get what you're after if using the 1.5 number.
(I'd still say 2 x 350 is 700, yeah I noticed the "Last edited by camplo; Today at 11:26 PM." 😉)
If that's your answer, be sure to EQ the output flat below fb before applying the crossover. Or you won't get what you're after if using the 1.5 number.
(I'd still say 2 x 350 is 700, yeah I noticed the "Last edited by camplo; Today at 11:26 PM." 😉)
Last edited:
I didn't say that we "know nothing", I said that we know some things but not others. Big difference.
You didnt see that
The idea is to have flat initial fr through the cross band. With a sharp enough xo I would think that whatever is behind fb is useless so you must be talking about low order xo like 1 and 2? Im aiming for the strict side, hence the 150hz horn....I think this will allow 1st order at 300hz? Im not certain