Maybe some different use of terms operating here.
Pano has a mic stand taped to the floor and does before-and-after measurements of the XO. An XO is a pretty simple gizmo and easy to imagine you can anticipate the resulting FR fairly accurately with slopes, etc behaving as they should. You may not be able to produce exactly the FR you'd like.
But the question of sound at your ear is more complicated and involves directivity, reflections, and much else.
B.
Pano has a mic stand taped to the floor and does before-and-after measurements of the XO. An XO is a pretty simple gizmo and easy to imagine you can anticipate the resulting FR fairly accurately with slopes, etc behaving as they should. You may not be able to produce exactly the FR you'd like.
But the question of sound at your ear is more complicated and involves directivity, reflections, and much else.
B.
Speakers are normally measured in an anechoic like manner. So does that mean the room doesn't matter?
The room matters a whole lot, however the room is not random. The design work done before building and measuring a speaker is about creating the speaker to room relationship. This entails not only how smoothly the room participates but where it comes in and when.
The room matters a whole lot, however the room is not random. The design work done before building and measuring a speaker is about creating the speaker to room relationship. This entails not only how smoothly the room participates but where it comes in and when.
Not at all. You going to have a filter for each individual like you do for a calibrated measurement microphone?? Wait I have a cold change filter, opps I am moody change filter! You need repeatable measurements not some facsimile of the ear brain interface.
Rob 🙂
B,
thanks for commenting but I am not sure if we are discussing the same thing here or my initial comment may not have been exactly on topic. I was answering the the question in a more global way, this was the "prompt" that I "heard" in my head, "Do measurement systems combined with circuit simulation software give accurate results" and my answer without any exaggeration is yes, they are about 100% accurate, "crazy accurate" and after using this software for almost 30 years, I am still blown away with how accurate and reliable they still are. If your measurements were done correctly, what you see in the simulation is what you get in real life. Of course there are many variables that the operator can bring into the system and some non-linearities / tolerances in components as well as non linearities in electro mechanical drivers that can change the outcome, these differences are usually very small or non-existent. Again, I can only comment on the systems that I have worked with that include LEAP, LMS, LISTEN, MLSSA, Kippel and Audio Precision.
Did this clear up my viewpoint?
G.
True, but that wasn't the question - at least it didn't include those terms. It's important to realize that measurements might be taken at one point in space. For a given set of measurements simulations can be very good indeed, @gdillon points out. Getting the sound you want at the listening position is much more complex, but that's not the fault of the software, it's just procedure and technique. The software is accurate with the data that it is given. How you obtain and use that data is another matter. If it were easy, then it would be easy. 😛
Pano,
You brought up a good point and a complicated one, I believe that the essence of this thread is Loudspeaker design and in room measurements so I will give my 2 cents. When I design a system, I usually measure at 1 or 2 meters out and at tweeter level but raised up about 1,5 meters above the ground. My main goal is to have a smooth on axis frequency response, secondary goal is to have a smooth off axis response (power response). When drivers are put into a box and when a box is put into a room, all hell breaks loose. On axis response is typically ruled by creating a crossover that sums flat. Off axis response is typically ruled by a drivers off axis response, crossover slop, crossover frequency and inter driver spacing, Then you have this nasty thing called baffle step which is caused when wavelengths are relatively long when compared to baffle size, you are transitioning from hemispherical space to spherical space (ideally), this causes a -6dB loss at wavelength below the width of the baffle (ideally) . (cant remember the formula) So with all that said, a room is a messy thing. (if anyone is interested, I can go into how I deal with each issue) Generally speaking keep the microphone at tweeter level, stay at least 1 meter out (possible closer if you have to measure in a small room) and take your measurements. If you are using gated sine waves or have an impulse based system, in room measurement's are not ideal but will totally work. Measuring low frequencies accurately is much more difficult . If you can, measure outside, makes things much easier. I now step down from my soapbox!
You brought up a good point and a complicated one, I believe that the essence of this thread is Loudspeaker design and in room measurements so I will give my 2 cents. When I design a system, I usually measure at 1 or 2 meters out and at tweeter level but raised up about 1,5 meters above the ground. My main goal is to have a smooth on axis frequency response, secondary goal is to have a smooth off axis response (power response). When drivers are put into a box and when a box is put into a room, all hell breaks loose. On axis response is typically ruled by creating a crossover that sums flat. Off axis response is typically ruled by a drivers off axis response, crossover slop, crossover frequency and inter driver spacing, Then you have this nasty thing called baffle step which is caused when wavelengths are relatively long when compared to baffle size, you are transitioning from hemispherical space to spherical space (ideally), this causes a -6dB loss at wavelength below the width of the baffle (ideally) . (cant remember the formula) So with all that said, a room is a messy thing. (if anyone is interested, I can go into how I deal with each issue) Generally speaking keep the microphone at tweeter level, stay at least 1 meter out (possible closer if you have to measure in a small room) and take your measurements. If you are using gated sine waves or have an impulse based system, in room measurement's are not ideal but will totally work. Measuring low frequencies accurately is much more difficult . If you can, measure outside, makes things much easier. I now step down from my soapbox!
Ramista,
I just took a look at REW and it looks like a possible replacement to LMS. I already own a good quality sound card. I will let you know if I can get it integrate with LEAP, looks promising, thanks for the info!
GD