Crowd Size and Physics of Pi(e):Yet another sizing post
Crowd Size and Physics of Pi(e):Yet another sizing post
Good day all,
I wanted to bounce my experiences off you and see if you have guidance before I launch into my next build. I'm also searching for confirmation, since I'm reading and learning. We find our gigs, where we have to provide sound, are a mixture of indoor/outdoor from 100-500 people. My intention is to build a minimum of four (4) BFM design mains so that I have a basic rig. I may be wrong, but I find that two mains just can't do it, no matter how much spin the sales department puts on their product.
The problem is covering a 500 person audience outside. Given our catalog and style, our audience likes about 96-100dB SPL (c-weighted slow) average maximum, and 90db SPL average minimum. SPL peaks are 6-10dB above that, so I use 106dB peak as the maximum for the nearest crowd. Since SPL decreases by 6dB every time the distance doubles, it appears that I basically need an area that starts 25 feet from the mains extending to 75 feet. I know that falls off too much, approximately 10dB instead of 6dB, but even outdoors we're limited by venue "size" since they're usually blocking off streets, want people to get into the bar easily, etc. So, would you please let me know if that 25-75 foot distance makes sense to you? Or are there other ways, power shading, of setting up mains that would bring the crowd in closer and still provide the 6-10dB fall off.
I'll provide a concrete example. I'm sorry to use subs in the example, but I don't have experience with BFM mains yet. My 2 Titans hit 106dB at 40 Hz at 25 feet using about 55V (800W - 4 Ohm drivers); I need more Titans. Run that out to 75 feet, they are down to 96dB, about 10dB drop and reasonable coverage. So a 90 degree, quarter pie, wedge in-front of the stage gives me about 4000 sqft, or enough space for 500 people, assuming 6 sqft per person.
(3.14*75^2-3.14*25^2)/4/6 = 650 people
I guess what I'm saying is that sizing a PA comes down to the physics of packing people into a space with desired gradient of SPL. Is that how you size mains? I guessed at 6 sqft per person, thoughts? I'm also assuming 1/4 pie based on 45 degrees of axis, and of course that's venue specific even outdoors, but I'm also open to suggestions on the slice width. Obviously I'm simplifying the mains to a single point source with no distance between them. I can cover more people in less depth, by spreading the mains, but it's squishing a balloon of SPL that only produces so much: ie., I can only smooth out humps, not add more.
I have read many other sizing posts, and I apologize for duplicating what's in some of them, but I didn't see a direct discussion of "depth" of the crowd. It appears to me, that the depth, SPL fall off, is what actually determines the mains sizing. Pick an audience size, then some slice of a pie to fit that many people. Finally use those distances to size the mains: not too much for the close people, and enough for the far people. Or in other words, plan for the farthest experience using the amount of people that can reasonably fit into a space and plan on some complaints from the close people.
I hate to complicate it a little, but I also think some additional headroom should be built in to the system. If I expect a 106dB peak, then I plan for a mains that are capable of 112dB peak. Is that reasonable?
Best, and thank you!
I wanted to bounce my experiences off you and see if you have guidance before I launch into my next build. I'm also searching for confirmation, since I'm reading and learning. We find our gigs, where we have to provide sound, are a mixture of indoor/outdoor from 100-500 people. My intention is to build a minimum of four (4) BFM design mains so that I have a basic rig. I may be wrong, but I find that two mains just can't do it, no matter how much spin the sales department puts on their product.
The problem is covering a 500 person audience outside. Given our catalog and style, our audience likes about 96-100dB SPL (c-weighted slow) average maximum, and 90db SPL average minimum. SPL peaks are 6-10dB above that, so I use 106dB peak as the maximum for the nearest crowd. Since SPL decreases by 6dB every time the distance doubles, it appears that I basically need an area that starts 25 feet from the mains extending to 75 feet. I know that falls off too much, approximately 10dB instead of 6dB, but even outdoors we're limited by venue "size" since they're usually blocking off streets, want people to get into the bar easily, etc. So, would you please let me know if that 25-75 foot distance makes sense to you? Or are there other ways, power shading, of setting up mains that would bring the crowd in closer and still provide the 6-10dB fall off.
I'll provide a concrete example. I'm sorry to use subs in the example, but I don't have experience with BFM mains yet. My 2 Titans hit 106dB at 40 Hz at 25 feet using about 55V (800W - 4 Ohm drivers); I need more Titans. Run that out to 75 feet, they are down to 96dB, about 10dB drop and reasonable coverage. So a 90 degree, quarter pie, wedge in-front of the stage gives me about 4000 sqft, or enough space for 500 people, assuming 6 sqft per person.
(3.14*75^2-3.14*25^2)/4/6 = 650 people
I guess what I'm saying is that sizing a PA comes down to the physics of packing people into a space with desired gradient of SPL. Is that how you size mains? I guessed at 6 sqft per person, thoughts? I'm also assuming 1/4 pie based on 45 degrees of axis, and of course that's venue specific even outdoors, but I'm also open to suggestions on the slice width. Obviously I'm simplifying the mains to a single point source with no distance between them. I can cover more people in less depth, by spreading the mains, but it's squishing a balloon of SPL that only produces so much: ie., I can only smooth out humps, not add more.
I have read many other sizing posts, and I apologize for duplicating what's in some of them, but I didn't see a direct discussion of "depth" of the crowd. It appears to me, that the depth, SPL fall off, is what actually determines the mains sizing. Pick an audience size, then some slice of a pie to fit that many people. Finally use those distances to size the mains: not too much for the close people, and enough for the far people. Or in other words, plan for the farthest experience using the amount of people that can reasonably fit into a space and plan on some complaints from the close people.
I hate to complicate it a little, but I also think some additional headroom should be built in to the system. If I expect a 106dB peak, then I plan for a mains that are capable of 112dB peak. Is that reasonable?
Best, and thank you!
- Bill Fitzmaurice
- Site Admin
- Posts: 29072
- Joined: Tue May 02, 2006 5:59 pm
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
I consult here, as a sound level monitor, so they don't get in trouble with the town: https://www.banknhpavilion.com/
The levels there in the FOH 75 feet from the stage with the loudest acts are 110dB'C' average, with 116dB'C' peaks. Most acts come in 10dB lower than that. If you're getting 96dB at 75 feet with two T48 you'll get 102dB with four. That puts you at pro-touring levels. Four DR250 or OTop 12 should keep up with them. More is OK, for added headroom, but you shouldn't need more than six of each no matter what.
The levels there in the FOH 75 feet from the stage with the loudest acts are 110dB'C' average, with 116dB'C' peaks. Most acts come in 10dB lower than that. If you're getting 96dB at 75 feet with two T48 you'll get 102dB with four. That puts you at pro-touring levels. Four DR250 or OTop 12 should keep up with them. More is OK, for added headroom, but you shouldn't need more than six of each no matter what.
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
Thank you Bill. I did not expect a 6dB increase by going to four Titans. That would greatly reduce the pressure on the existing pair. Knowing 6 speakers max is also very helpful.
Do you have guidance on a listening curve? I find many of our listeners want a huge bump from the sub-woofer frequencies: often 12-15dB.
We don't have a problem with overall SPL due to law or ordinance since beyond the crowd is nothing but open field for miles. I use a cardioid setup, even with just my two Titans, to limit stage volume, with the added benefit that behind the stage (trailers) is town: they get next to nothing.
Do you have guidance on a listening curve? I find many of our listeners want a huge bump from the sub-woofer frequencies: often 12-15dB.
We don't have a problem with overall SPL due to law or ordinance since beyond the crowd is nothing but open field for miles. I use a cardioid setup, even with just my two Titans, to limit stage volume, with the added benefit that behind the stage (trailers) is town: they get next to nothing.
- Bill Fitzmaurice
- Site Admin
- Posts: 29072
- Joined: Tue May 02, 2006 5:59 pm
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
Bass is most prominent in the 60 to 90 Hz range, so if you want more thump that's where you boost EQ.
-
Bruce Weldy
- Posts: 8655
- Joined: Fri Nov 27, 2009 11:37 am
- Location: New Braunfels, TX
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
I do the size crowd you are talking about....have been for many years. I use two OT12s per side and 6 T-39s total.
4 OT12s with the CD horns at 1.2k crossover will serve you well. Add 2 more T48s and you should be golden. You can always V-plate those T48s and get some more oomph out of 'em.
4 OT12s with the CD horns at 1.2k crossover will serve you well. Add 2 more T48s and you should be golden. You can always V-plate those T48s and get some more oomph out of 'em.
6 - T39 3012LF
4 - OT12 2512
1 - T24
1 - SLA Pro
2 - XF210
"A system with a few knobs set up by someone who knows what they are doing is always better than one with a lot of knobs set up by someone who doesn't."
-
himhimself
- Posts: 644
- Joined: Fri Nov 13, 2015 9:34 pm
- Location: Vancouver BC & Bloomington IN
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
+1 to Bruce and Bill's recommendation of a couple 0tops per side. I've remained stunned at the output my Otops give. They'll easily support a whole herd of subs, which is where you need to put the focus for overall effortless massive sound that isn't harsh or exhausting to listen toBruce Weldy wrote: ↑Sat Feb 07, 2026 10:18 am I do the size crowd you are talking about....have been for many years. I use two OT12s per side and 6 T-39s total.
4 OT12s with the CD horns at 1.2k crossover will serve you well. Add 2 more T48s and you should be golden. You can always V-plate those T48s and get some more oomph out of 'em.
2xT30 (20", 3012LF)
2xT30 (21", 3012LF)
4xOtop J-array (Beta12, melded/straight)
Truck Tuba (MCM)
Next up: 2xJack12L or family of table tubas
2xT30 (21", 3012LF)
4xOtop J-array (Beta12, melded/straight)
Truck Tuba (MCM)
Next up: 2xJack12L or family of table tubas
-
Bruce Weldy
- Posts: 8655
- Joined: Fri Nov 27, 2009 11:37 am
- Location: New Braunfels, TX
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
And even in an inside venue where one per side would be enough, run two anyway. Once the OT12s couple, the mids get sweet and warm....don't know why, but they do.himhimself wrote: ↑Sat Feb 07, 2026 10:25 am+1 to Bruce and Bill's recommendation of a couple 0tops per side. I've remained stunned at the output my Otops give. They'll easily support a whole herd of subs, which is where you need to put the focus for overall effortless massive sound that isn't harsh or exhausting to listen toBruce Weldy wrote: ↑Sat Feb 07, 2026 10:18 am I do the size crowd you are talking about....have been for many years. I use two OT12s per side and 6 T-39s total.
4 OT12s with the CD horns at 1.2k crossover will serve you well. Add 2 more T48s and you should be golden. You can always V-plate those T48s and get some more oomph out of 'em.
6 - T39 3012LF
4 - OT12 2512
1 - T24
1 - SLA Pro
2 - XF210
"A system with a few knobs set up by someone who knows what they are doing is always better than one with a lot of knobs set up by someone who doesn't."
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
This is a good place to start
-
DaveMacKay
- Posts: 100
- Joined: Tue Aug 03, 2021 3:10 pm
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
That’s quite similar to what I measure at the main (MLP) and secondary (2LP) listening positions in my listening room (sorry for drifting a bit off-topic). I created 3 different listening profiles, they differ only in how much the bass is boosted (blue = +20 dB SPL, green = +15 dB SPL, red = +10 dB SPL). I differ from the graph that Seth posted in that I also boosted the HF a little bit (for no good reason since I can't hear above 14 kHz anyway).
I have tri-amped La Scalas that cross over to my two THTLPs at 80Hz using a 4th order L-R. Each THTLP is driven by a 250W plate amp. The plate amps have a -24 dB/octave LPF at 125 Hz and a -12 dB/octave HPF at 25 Hz.
Creating the different configurations (listening profiles) took quite a lot of work.
- I used REW to help me configure the MiniDSP Flex Eight DSP output channels to get the 1m measurements from each speaker as flat as I could.
- I used REW to take measurements for use with MSO (Multi-Sub Optimizer) to minimize the subwoofer variation between the two listening positions. MSO generated configurations for the DSP output channels for each THTLP.
- After that, I worked with Dirac by adjusting the target curves to tailor the sound to my taste and to deal with some of the challenges my room presents. Dirac generated configurations for the input channels on the DSP.
Last edited by DaveMacKay on Mon Feb 09, 2026 9:57 am, edited 5 times in total.
- Bill Fitzmaurice
- Site Admin
- Posts: 29072
- Joined: Tue May 02, 2006 5:59 pm
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
That amount of bass makes perfect sense when you look at an equal loudness chart.
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
As a musician, I don't like a huge hump in EQ if it's too steep or starts/ends in weird places, because it's a struggle to keep dynamics nice and smooth through a scale. Most often that's due to an unexpected hump when a newer sound guy doesn't know how to EQ a room: hat-tip @DaveMacKay. However equal loudness charts certainly have a huge fall off which the 12-15dB hump would help with. I've never heard it called equal loudness, thank you @Bill Fitzmaurice, I always approached it through the A-weighted human perception of loudness, thank you!
I have seen @Seth's chart before, and I freely admit Seth, that I stole and use your "crossover to create the bottom end EQ" math in addition to your cardioid math: thank you! And not to leave @Bruce Weldy out, I have followed your threads and am generally using them as a guideline for sizing / validation. In fact when I started I was going to do exactly what you did, but then I let my helper talk me into T48s because if we're building, might as well build.
I'm sorry @himhimself, I don't have experience with BFM mains, but soon. I do expect, as you and Bruce noted, that I'll always use at least two per side: it just makes sense to even out the frequency response (to me at least.) Although even now, I sometimes only use "one side" based on room / crowd size. I never mix stereo, just mono.
Can any of you speak to DR280/DR250/OTop 12? I'm a bit nervous about the complexity of the DR line, but then it's just wood so it might take a few tries. I also think that since I'm retired, there's no reason not to build the "top of the line" even if it takes extra time. However, if I can get what I need out of OTop 12's, why would I invite extra complexity if it's unnecessary? I would like to crossover to T48s pretty low 80Hz - 100Hz. That's because I'm try to avoid the "weird" sound cardioid arrays make outside of their tuning zone. It looks like any of those three mains would work fine. A DR250 has a significant lead on an OTop 12 at 100Hz, but then I'll get a boost from the subs at crossover so it doesn't really matter. (That's all just how I see it and a question, please and thank you!) In the end I lean towards build DR250s as it looks like the best bang for the buck, even though it'll take me more time.
EDIT: I forgot to mention, that I think I'd build OTop 12 JArrays, so the difficulty between a OTop12 JArray / DR 250 doesn't seem that big?
I have seen @Seth's chart before, and I freely admit Seth, that I stole and use your "crossover to create the bottom end EQ" math in addition to your cardioid math: thank you! And not to leave @Bruce Weldy out, I have followed your threads and am generally using them as a guideline for sizing / validation. In fact when I started I was going to do exactly what you did, but then I let my helper talk me into T48s because if we're building, might as well build.
I'm sorry @himhimself, I don't have experience with BFM mains, but soon. I do expect, as you and Bruce noted, that I'll always use at least two per side: it just makes sense to even out the frequency response (to me at least.) Although even now, I sometimes only use "one side" based on room / crowd size. I never mix stereo, just mono.
Can any of you speak to DR280/DR250/OTop 12? I'm a bit nervous about the complexity of the DR line, but then it's just wood so it might take a few tries. I also think that since I'm retired, there's no reason not to build the "top of the line" even if it takes extra time. However, if I can get what I need out of OTop 12's, why would I invite extra complexity if it's unnecessary? I would like to crossover to T48s pretty low 80Hz - 100Hz. That's because I'm try to avoid the "weird" sound cardioid arrays make outside of their tuning zone. It looks like any of those three mains would work fine. A DR250 has a significant lead on an OTop 12 at 100Hz, but then I'll get a boost from the subs at crossover so it doesn't really matter. (That's all just how I see it and a question, please and thank you!) In the end I lean towards build DR250s as it looks like the best bang for the buck, even though it'll take me more time.
EDIT: I forgot to mention, that I think I'd build OTop 12 JArrays, so the difficulty between a OTop12 JArray / DR 250 doesn't seem that big?
- Bill Fitzmaurice
- Site Admin
- Posts: 29072
- Joined: Tue May 02, 2006 5:59 pm
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
Building a DR250 is not more difficult than an OTop 12 JArray, it just takes longer.
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
I built two DR250s on the hood of my retired Volvo wagon in my back yard, years ago when I was just starting. That means that it is attainable for you mtglass.
By the way, have you looked at the OTop8?
TomS
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
Good day,
Thanks Bill, I'm pretty sure I'm going to start with 4 x DR250s and two more T48s. @Tom Smit, thank you for the suggestion, I had not given them serious consideration. Their response graph is quite impressive, but I can't build them any cheaper than DR250s (I think) so I don't see the value proposition; I worry about the parallel planes in the cabinets; and, I could splay them in an array but it looks like the OTop 12 J-Array and DR series are built for splay. I may be all wet, I'm all ears.
I don't want to start a flame war, but there are people who seem very impressed with Compression Horns, and people who are very impressed with Piezos. I have no way of making an informed decision. It looks like Bill says it's really dependent on throw: Piezos for dispersion and less throw. But what is "less"? I'm going to have audiences from about 6' to 100': most of my planning is for the 25-75' zone. Center fills don't really come into play at that 25-75' distance, in my opinion.
I feel like having a bunch of tools in the tool box would be best: melded array in the bottom speaker(s), flat array above that, and then Compression drivers above those for however many cabinets I end up with. Does that follow best practice? Or maybe the flat array doesn't make any sense, jump straight to compression driver?
@Tom Smit, I drove a 1978 245 for a long time. I have very basic rules for my automobile: go, stop, turn left, turn right. It quit turning right, so I sent it to the crusher. Great car, ran like a top. I wish I had the energy to take the drive train out of it before I crushed it.
Happy hump-day!
Thanks Bill, I'm pretty sure I'm going to start with 4 x DR250s and two more T48s. @Tom Smit, thank you for the suggestion, I had not given them serious consideration. Their response graph is quite impressive, but I can't build them any cheaper than DR250s (I think) so I don't see the value proposition; I worry about the parallel planes in the cabinets; and, I could splay them in an array but it looks like the OTop 12 J-Array and DR series are built for splay. I may be all wet, I'm all ears.
I don't want to start a flame war, but there are people who seem very impressed with Compression Horns, and people who are very impressed with Piezos. I have no way of making an informed decision. It looks like Bill says it's really dependent on throw: Piezos for dispersion and less throw. But what is "less"? I'm going to have audiences from about 6' to 100': most of my planning is for the 25-75' zone. Center fills don't really come into play at that 25-75' distance, in my opinion.
I feel like having a bunch of tools in the tool box would be best: melded array in the bottom speaker(s), flat array above that, and then Compression drivers above those for however many cabinets I end up with. Does that follow best practice? Or maybe the flat array doesn't make any sense, jump straight to compression driver?
@Tom Smit, I drove a 1978 245 for a long time. I have very basic rules for my automobile: go, stop, turn left, turn right. It quit turning right, so I sent it to the crusher. Great car, ran like a top. I wish I had the energy to take the drive train out of it before I crushed it.
Happy hump-day!
-
Bruce Weldy
- Posts: 8655
- Joined: Fri Nov 27, 2009 11:37 am
- Location: New Braunfels, TX
Re: Crowd Size and Physics of Pi(e):Yet another sizing post
The melded array do have better dispersion. The CDs sound better. I don't remember, are you doing DJ work or live sound?mtglass wrote: ↑Wed Feb 11, 2026 9:25 am
I don't want to start a flame war, but there are people who seem very impressed with Compression Horns, and people who are very impressed with Piezos. I have no way of making an informed decision. It looks like Bill says it's really dependent on throw: Piezos for dispersion and less throw. But what is "less"? I'm going to have audiences from about 6' to 100': most of my planning is for the 25-75' zone. Center fills don't really come into play at that 25-75' distance, in my opinion.
I started (OT12s) with CDs, then built another pair with the melded array. I run the CDs on top. After having done this for years, I've found that in my situation, the additional dispersion of the melded isn't worth the quality I give up by not having CDs in them. But, I'm doing only live sound. Honestly, I don't think there would be as much difference with recorded music. CDs with the lower crossover option are really great for live vocals. Using CDs with the 2k crossover is going to be pretty much the same as the melded array - same crossover point.
I don't really use my system in situations where I need the additional dispersion of the melded array, so I would rather have CDs at 1.2k in all of them. Have I changed 'em out? No, I'm old, lazy and don't think I'd get enough out of it to warrant the time and energy. If I built 'em all over again, I'd definitely use the CDs.
So, take all of that with a grain of salt - you have to look at your situation and make a call. One thing to consider though, is that while we think of having equal coverage all over the crowd - don't forget that a lot of people who sit in the wings and way to the back are there precisely because they don't want it to be as loud.
Bottom line - they both sound great. I you are doing live sound, go with the CDs at the lower crossover.
As to the issue of arraying the speakers. Honestly, with only two per side - I'd build standard OT12. Way easier to stack when transporting and storing. Way easier to stack on a speaker pole. And unless you are flying them way over the crowd, there's really no need to tilt the bottom one down. They'll hear it just fine.
There ya' go......two cents from the grumpy, old sound man.
6 - T39 3012LF
4 - OT12 2512
1 - T24
1 - SLA Pro
2 - XF210
"A system with a few knobs set up by someone who knows what they are doing is always better than one with a lot of knobs set up by someone who doesn't."