You can never have too much cowbell....
RTA is from the eighties i would say
Why wouldn't a transfer function suffice?
There are multiple possible methods, they boil down to:
- just run all subs at unity gain and only apply global EQ
- EQ each sub individually, and then apply global EQ
- use mso (i don't know of an alternative) to measure each sub's individual response and model the sum (and EQ/delay/gain/other filters applied to each sub as you see fit). This last approach labor intensive but is especially useful for multiple positions optimization.
Yep, Boiling it down to the three methods you mention makes intuitive sense to me too.
FWIW, I was using transfer functions only, until a recent 3 day training class for Smaartlive.
The instructors did a good job of demonstrating that indoors, where the room is so much a part of the measurement (both above and below Schroeder), that a good ole RTA often gives a better picture and more useful information than transfer.
They took a speaker and adjusted it to have the same transfer at two different locations in the room, and then let us listen to how different the two locations sounded.....and measured with a RTA.
Perhaps even better at giving good indoor info was Smaart's recording spectrograph capability...which I'm just beginning to learn to use...
I'm certainly still learning here, but my layman's takeaway is a transfer function is kinda forced by its math to make a mag and phase traces that tie together throughout the spectrum, no matter how many time arrivals from various locations are hitting the mic, no matter how disjointed the spectrum really is. So sometimes we get big averaging traces that don't really match 'what the heck?' is really going on..
And that transfers are best for speaker building in quasi-anechoic conditions, and usually for outdoor tuning unless too many reflections exist.
Whereas indoors it's often good to turn back to RTAs / spectrographs.
Mark100
It would go far afield to discuss the theory of room measurements here, but yes, sometimes a transfer function is more revealing and sometimes an RTA is going to work better. As I stated before, the consensus in the Pro field is that a sliding window on the transfer function works best (I don't think that Smaartlive can do this, so they won't promote it.) But again, these are all large room (non-modal NM) approaches and not relevant here.
It would go far afield to discuss the theory of room measurements here, but yes, sometimes a transfer function is more revealing and sometimes an RTA is going to work better. As I stated before, the consensus in the Pro field is that a sliding window on the transfer function works best (I don't think that Smaartlive can do this, so they won't promote it.) But again, these are all large room (non-modal NM) approaches and not relevant here.
Yep, good acoustic measurements are surely a science/art of their own...for me one of the "the more I learn the less I know fields"...
Smaartlive does allow sliding window measurement, but it's very clear they don't advocate using it.
They prefer their normal method of splitting the spectrum into chunks that have increased sampling decimation as freq decreases, for enhanced low end capability.
And specifically caution against using the windowing for low frequency work, putting a red line at the frequency where below which measurements are suspect.
I guess it depends on what you call the Pro field, but I think of it as the commercial install market, where afaict most installers use Smaartlive, with a smaller set using AFMG's Systune, of Meyer's Sim.
And I guess I also wonder about when a large room goes NM, given how RT60 inevitably goes up as room size increases.
The few numbers I've kicked around make it appear it's dang hard to get Schroeder below 40-45Hz.
I dunno.. indoor bass is one tough nut
But when clean and powerful.... for me, it's the beautiful foundation that makes the rest come alive
If only indoors were as easy as out....
I saw this post and began researching MSO. I started reading the tutorial, and found some comments about this thread. What are your thoughts?
Earl did describe his method, somewhere near the beginning of the thread, expanding on it a hundred or so posts later. One post in particular.
I followed these instructions and found working solutions. There was no mention of formulas but I knew what to vary from the instructions. Do these people want it handed to them on a platter?
I followed these instructions and found working solutions. There was no mention of formulas but I knew what to vary from the instructions. Do these people want it handed to them on a platter?
Earl,
Could you please share your thought on this:
Once I measured all the possible locations for my stereo speakers. The optimal position (according to the FR) was 35 cm from the front wall. That leaves no room for a sub in a corner. If I move the speakers further from the wall, some dips/peaks would get bigger.
My question is if this would be solvable with more subs or is it better to start with optimal stereo position and place the subs elsewhere?
By the way, I did some really quick measurements, not by Geddes method but used an auto EQ option for two pairs of subs at equidistant locations.
No matter what I do there is this huge dip at 103 Hz. It just came from nowhere! I'm pretty sure it wasnt there last time I was doing measurements!
Could you please share your thought on this:
Once I measured all the possible locations for my stereo speakers. The optimal position (according to the FR) was 35 cm from the front wall. That leaves no room for a sub in a corner. If I move the speakers further from the wall, some dips/peaks would get bigger.
My question is if this would be solvable with more subs or is it better to start with optimal stereo position and place the subs elsewhere?
By the way, I did some really quick measurements, not by Geddes method but used an auto EQ option for two pairs of subs at equidistant locations.
No matter what I do there is this huge dip at 103 Hz. It just came from nowhere! I'm pretty sure it wasnt there last time I was doing measurements!
Attachments
First let's discuss the timeline from my perspective. I realized while doing my PhD on "LF sound fields in small rooms" that having more than one source would reduce the spatial variability of the modes. That was way back in 1981. Sometime in the early 90's I began experimenting with multiple subs and implemented a version in my own system and had refined my approach by about 2003. This can be confirmed by Duke Lejune at Audio Kinesis who uses this in his sub configuration called "The Swarm."
When Welti's paper came out I was already well versed in the technique and I sent a letter to the AES editor (which was published, but somehow people don't seem to want to mention this) with a critique Todd's paper. I basically agreed but had some small suggestions for improvement (like random placement rather than Todd's symmetrical one.)
Then the DIY thread started.
Much later I had tried MSO and found it unworkable, so I contacted the author. I asked why if he were trying to implement my ideas, did he not make any attempt to discuss it with me? I suggested that we work together to improve his code to work better along the lines of what I was trying to do. He was not interested.
I find his comments to be pretty rude, but I am not surprised since I found most of my interactions with him to be unpleasant. He was annoyed that I would not recommend his software, but how could I do that when it did not work for me.
People want you to hold their hand and lead them by the nose. I don;t do that.
One of my main points with my technique is that it allows one to put the mains where the best performance for HFs (imaging etc.) can be achieved. Then you simply don't worry about what the mains do at LFs, because you fix this with the subs. Hence, it is certainly best to place the mains in their optimal position first.
I've contacted the author of the software. He says he's been banned from this forum, but he tells me he's going to put the entirety of this email exchange on his website so people can judge for themselves.
Thank you, that makes sense.
One more thing. What is this suppose to mean?
Why not place it exactly on the main listening position?
And one of the points in using multi subs, is to get an even FR in more than one place. And when doing it right in one place - the multi sub principle, actually "automaticly" straighten out the FR in almost all other places too.... in the same room of course..... no miracles here
A few things I wonder about:
How slow should a sweep be, to get a good measurement?
I heard and tried the DRC way, with delaying the signal almost 2 seconds, to try and FIR correct the base. But it seems overly complicated and I found little difference. Is this approach silly?
Why ever use anything else but closed subs for home, when amps and drivers are easy to get?
How slow should a sweep be, to get a good measurement?
I heard and tried the DRC way, with delaying the signal almost 2 seconds, to try and FIR correct the base. But it seems overly complicated and I found little difference. Is this approach silly?
Why ever use anything else but closed subs for home, when amps and drivers are easy to get?
I'm aware this is one of the main advatanges. However, I'm focusing primarily on one seat so it is still necessary to move the mic around in this case?And one of the points in using multi subs, is to get an even FR in more than one place. And when doing it right in one place - the multi sub principle, actually "automaticly" straighten out the FR in almost all other places too.... in the same room of course..... no miracles here
As TNT writes... +/-50cm is not a bad idear, cause it takes very little before the sound changes.
So your Xover between subs and mains, are at a 100hz?
If so - Geddes wrote that you need to let your mains run lower, so that you get and overlap.
Let's say your subs runs to a 100hz. Then your mains should run to at least to 60hz with a low order HP(if they dont like the deep notes). This could help with the dip. But it could also be a measuring issue, cause the speakers distance to the rear wall playes tricks on your measurement.
The crossover on these plots was set to 80 Hz. But like I said, this was not set by "Geddes approach", I just measured my current home cinema setup with auto calibration system.
After that I did some experimenting with RTA. I tried to move the mic a little bit forward and the 100Hz dip got smaller. That could explain this anomaly: during extensive watching in the last couple of years my seats (which are not screwed to the floor) moved backwards and therefore the 100 Hz dip was getting bigger.
I will do some more testing, I took some extra days off!