Pan said:
Your post is childish.
Of course you have to show data if you claim something. My data can be found in the linked page I posted 2 times in this thread. Joining a discussion without listening to what others have to say is what I would call rude.
You are arguing like all people that look at modes with the knowledge they have from textbooks. In a real room there will be no such thing as a not excited mode. Buy software and a mic and find out for yourself.
Want to see some more real world data? Here you go:
http://www.mehlau.net/audio/dba/
Let's assume a rectangular concrete room, 5 meters long and 2 meters wide and 2 meters high.
If we place one subwoofer at the end of the longest dimension of the room and play a sweep from 20Hz to 100Hz we will see a huge standing wave at 34.3Hz as dictated by the dimension 5 meter and the speed of sound 343m/s.
Now let's take another woofer and place it at the opposite wall and run a sweep in phase with both woofers.
What will happen?
There will be NO standing wave in that room at 34.3Hz.
We have totally eliminated the resonance from the playback chain.
We will still see the 2:nd resonance at 68.6Hz in the long dimension of the room though. Now drag both woofers one 1/4 into the room and the 2:nd standing wave (68.6Hz ) is gone as well.
The next resonances we will see is the first one dictated by the distance 2 meters between the long walls and between floor and ceiling at 85.7Hz or thereabouts. Use two more woofers for a total of four, and place these up against the ceiling and the first resonance in that dimension is gone as well.
By using four woofers we have avoided to excite all axial modes in this room up to 100Hz. By moving the woofers around we can optimize the layout for surpressing the oblique and tangential odes as well.
Now perform a impulse measurement of the room and see if the time behaviouir is as bad as it was with one woofer even though the frequency response is improved greatly, or if maybe, we have improved the decay time as well.
/Peter
If we place one subwoofer at the end of the longest dimension of the room and play a sweep from 20Hz to 100Hz we will see a huge standing wave at 34.3Hz as dictated by the dimension 5 meter and the speed of sound 343m/s.
Now let's take another woofer and place it at the opposite wall and run a sweep in phase with both woofers.
What will happen?
There will be NO standing wave in that room at 34.3Hz.
We have totally eliminated the resonance from the playback chain.
We will still see the 2:nd resonance at 68.6Hz in the long dimension of the room though. Now drag both woofers one 1/4 into the room and the 2:nd standing wave (68.6Hz ) is gone as well.
The next resonances we will see is the first one dictated by the distance 2 meters between the long walls and between floor and ceiling at 85.7Hz or thereabouts. Use two more woofers for a total of four, and place these up against the ceiling and the first resonance in that dimension is gone as well.
By using four woofers we have avoided to excite all axial modes in this room up to 100Hz. By moving the woofers around we can optimize the layout for surpressing the oblique and tangential odes as well.
Now perform a impulse measurement of the room and see if the time behaviouir is as bad as it was with one woofer even though the frequency response is improved greatly, or if maybe, we have improved the decay time as well.
/Peter
markus76 said:
What!?
About your link, it would help if you gave a hint on that in the specific post that you claim something..
That said there are a lot of info missing to draw any conclusions at all from the graphs that I found.
Maybe you remember the moderator that told that is NOT the case?
Maybe you remember what i told you about motivation when people are rude?
This is a discussion board!
If you want to be rude, step out on the street and be rude to someone.
/Peter
otto88 said:
otto88,
the transient/time behaviour of a singel speaker depends on the frequency response. A narrow bandpass filter/speaker has higher GD than one with wide bandwith low Q flanks.
A closed box woofer that reach down looow on its own or with EQ will have the best time bahaviour. In a system you also have to include the crossover.
/Peter
Markus, Toole in his book discusses mode cancellation (page 221). Essentially, the second subwoofer will actively cancel out the odd-order modes in the dimension whose axis in which the two subwoofers lie. Essentially, you have changed the modal field, and these particular resonances no longer happen. Obviously, you can avoid exciting a resonance or mode by positioning a subwoofer in one of its nodes. Toole also discusses how bass frequencies essentially act as a minimum-phase phenomena (section 13.1 and 13.4), so there can in fact be a relationship between the frequency and time domains. Toole provides real world data to back this up.
That is the point. For the same instrument being played back, due to room modes, the focus shifts as the same instrument creates notes ad different frequencies.janneman said:
The purpose of high fidelity equipment is to reproduce the original recording to provide an illusion as real to the original equipment as possible. No matter how technically you describe it, if the goal is not to perform this function, then it is just another musical instrument playing in your room. There are actually many recordings that use to mics only. So when playing those recordings, one would expect the experience to be the same. When we want you evaluate high fidelity equipment, we should select recordings that had the intent to reproduce the live experience because this is what we can compare. We cannot go to a studio and ask them to play back thier original recording for us to listen to, but we can go to a music hall and select the right seat as close to where they normally place the mics as possible. This way, we have a better reference rather than guessing what the original recording should sound like.markus76 said:
I do accept that not all people are looking for this kind of fidelity and have different preferences. But for me, I do not like one instrument floating around when it should be stationary when I listen to music. This does not even have anything to do with sound field but the basic reproduction of one single musical instrument.
Yes but likely you will be able to find a single driver with T/S parameters that do the same job as those two drivers used in isobarik.. but then again you could use two of those drivers and end up with an even smaller box. 
Problem is the smaller the box, the higher the distortion and the more severe thermal compression.
/Peter
Problem is the smaller the box, the higher the distortion and the more severe thermal compression.
/Peter
markus76 said:
What is the significance of the top to bottom plots? How many things are you changing at each treatment/test/setup point? Are the top to bottom plots at different locations for each sub treatment/setup?
I can easily select 4 locations in my room with 3 subs that give pretty smooth plots; likewise I can find locations with huge modal resonance points as well... not sure of the real world significance of your data...
Hopefully the smoother respnse is where you want to listen...
Might wanna have a look at some multivariate experimental setups... could save yourself alot of trials and might pick up some confounded effects as well...
Looks like left to right you've traded a prominent resonance at ~35 hz for an equally prominent one at ~70 Hz or so. Does look a bit smoother from upper right to lower left, ignoring the 70 hz resonance...
Soongsc,
So you're switching to fullrange drivers? If not I don't understand the point you're trying to make. Yes the initial transient of the stick hitting the drum will come out of the mid and tweeter not the sub. When a cello plays sound will come out of every speaker in the system. That's the nature of the beast, unless you switch to fullrange drivers.
Not a very pleasing discussion.
I will say that I do not consider "mode cancellation" to be the goal, nor do I consider this approach to be effective. For every mode that is cancelled another is acentuated - i.e. doubled. This does not lead to a smooth overall response. Talking about cancelling "axial" modes in only one direction also seems like a dead end to me as there are lots of other types of modes and there are three directions. What about all those modes? Quite simply, the problem is too complex and multi-dimensional to treat in anything but a statistical manner.
Viewed statistically cancelling a mode is not going to yield the best statistics. You have to look at the total picture of all the modes and all the sources combined into a single system. The discussion gets lost when you try and narrow it down to the point of talking about a single mode in a single direction.
I will say that I do not consider "mode cancellation" to be the goal, nor do I consider this approach to be effective. For every mode that is cancelled another is acentuated - i.e. doubled. This does not lead to a smooth overall response. Talking about cancelling "axial" modes in only one direction also seems like a dead end to me as there are lots of other types of modes and there are three directions. What about all those modes? Quite simply, the problem is too complex and multi-dimensional to treat in anything but a statistical manner.
Viewed statistically cancelling a mode is not going to yield the best statistics. You have to look at the total picture of all the modes and all the sources combined into a single system. The discussion gets lost when you try and narrow it down to the point of talking about a single mode in a single direction.
poptart said:
I have switched to full range for small rooms and at lower listening levels, but we also must know that full range has other drawbacks. It's important what one places priority in when setting up a system. If more bass is desired and takes priority, the other areas may be compormised. It's best to know what you gain and what you lose by different design and setup methods.
Let's take the cello for example: Most of the time change in direction of the bow creates a small transient that allows you to pinpoint where the cello seems to be; however, if the note is low that excites room modes, then you will not get a feeling that the cello is at that original location. Whereas, if you listen to cello played in a hall, you will almost hear the note coming from the cello where it is located.