or Connect
AVS › AVS Forum › Audio › DIY Speakers and Subs › First build - Octo SI 18's with flat packs.
New Posts  All Forums:Forum Nav:

First build - Octo SI 18's with flat packs. - Page 14

post #391 of 790
Thread Starter 
Thanks FOH - I'm definitely open to putting in more bass traps. Luke Kamp is planning on coming over sometime with some of his acoustic panels to experiment and has been on my case for awhile to at least get bass traps in the back corner by the computer.

I just need to realize that this will be a marathon not a sprint - I can't expect to have all the upgrades NOW on my salary that's for sure. This will be a life long hobby so I need to just take it a step at a time - slow and steady.

Ok, I did a bunch of tinkering today. Thanks to Luke for giving me step by step instructions on how to set the delay times with the subs and then integrate with the speakers. I got it all done with the subs in the back and liked it, but I'm not done with experimenting with placement so I put the 4 subs up front under the screen. I quickly realized I'm going to have to keep all my subs at the front of the room.



The red line is all 8 subs up front with 4 under the screen and 4 in the corners, and the black is with subs in the front and subs in the back. Both curves the subs are optimized as far as delay goes.

It doesn't look like much but remember my numbers on the left of the graph on snapshots go by 10's not 5's, so I'm gaining 4-5 dbs from 12hz to 26hz and that's too much for me to let go especially since I was able to get a nice curve with all the subs up front anyway.




With the subs up front I experimented with delay for the subs, and the best was hardly any difference: .5 ms delay on the subs under the tv so basically nothing. The best delay on my receiver for all subs was 13.5 feet as far as integrating with the speaker(s).

Here is what I ended up with after LT boost and pulling down eq:



Here are my settings on the minidsp for the LT:

f(0) 21.5
f(p) 14 hz
Gain 7.452416969

This is less gain than I have used with any sub placement so far, so I assume that's good.

I was able to get a graph just as flat with 4 subs in the back of the room, but I had to use more boost down low and I had to have the sub out on my reciever 5 db louder to match the speakers when level matching.

I'm not going to crank it up tonight, my ears are a little tired after all the movie scene and movie cranking the last couple of days. However I am listening to music right now at around 80db and the bass sounds great, so tight and controlled just like my Submersives but like I always say, with a little extra weight behind them.
post #392 of 790
Thread Starter 
HAHAHA... biggrin.gif

I've spent all this time the past few days testing scenes and bass heavy music. Now that I'm listening to MY music again (like I do the vast majority of the time while using my system) the headroom for bass is laughable. I cranked it up very loud couple of times very briefly and the drivers are hardly moving.

Haha, this is just stoooopid. cool.gifsmile.gif

Just goes to show the amount of power/displacement it takes to reproduce the ultra low stuff in movies compared to rock/metal music. When I have the lights on during a movie clip it's distracting because you can see all the drivers going nuts - so cool though I gotta say (and a little scary eek.gif)
post #393 of 790
Quote:
Originally Posted by carp View Post

HAHAHA... biggrin.gif

I've spent all this time the past few days testing scenes and bass heavy music. Now that I'm listening to MY music again (like I do the vast majority of the time while using my system) the headroom for bass is laughable. I cranked it up very loud couple of times very briefly and the drivers are hardly moving.

Haha, this is just stoooopid. cool.gifsmile.gif

Just goes to show the amount of power/displacement it takes to reproduce the ultra low stuff in movies compared to rock/metal music. When I have the lights on during a movie clip it's distracting because you can see all the drivers going nuts - so cool though I gotta say (and a little scary eek.gif)

The structural integrity of your theater must rival that of a nuclear missile silo
post #394 of 790
I'm not sure if you saw our discussion about all this. Bosso had a theory, and I explained it was tied to all the sources being up front.
Quote:
Originally Posted by carp View Post

I quickly realized I'm going to have to keep all my subs at the front of the room.



The red line is all 8 subs up front with 4 under the screen and 4 in the corners, and the black is with subs in the front and subs in the back. Both curves the subs are optimized as far as delay goes.

It doesn't look like much but remember my numbers on the left of the graph on snapshots go by 10's not 5's, so I'm gaining 4-5 dbs from 12hz to 26hz and that's too much for me to let go especially since I was able to get a nice curve with all the subs up front anyway.

This is what Bosso saw, and I explained as greater first length mode (1, 0, 0) excitation. You're driving the length mode maximally.

Modes exist, determined by the boundary distances. How much one couples to them, or how one avoids coupling with them, determines their significane. Similarly, you can experiment with this selective mode cancelation, across the width mode of your room, not strongly affecting the 12-26hz gains you like wrt length.



Here's Bosso inquiring about them;
Quote:
Originally Posted by bossobass View Post

If the pressure is increasing and decreasing everywhere in Carp's room, explain the 6dB difference he (and all of us) measures by simply moving the subs. Does moving the subs to a new location change the longest dimension in the room?


Myself explaining what he's seeing;

"It does exactly what one expects. Of the varying attributes to utilization of multiple subs, selective mode cancellation.

The physical characteristics establish a room's modal behavior wrt frequency. This doesn't change, regardless where one places the LF sources. However, the placement does affect the magnitude of the modal excitement."


Quote:
Originally Posted by bossobass

If the pressure is increasing and decreasing everywhere in Carp's room, explain the 6dB difference he (and all of us) measures by simply moving the subs.


My response;

"I see nothing inexplicable about Carp's measurements. Moving the LF source, changes the manner in which it couples too, and drives a particular mode. Just as one can minimize that excitation, one can similarly maximize it."


Again, not sure if you saw this.

That post is here. It's not a long post.




Most importantly, I'm glad you're enjoying the results. Soon, I'm going to have precious few opportunities to post, so best of luck until I can.
post #395 of 790
Thread Starter 
Thanks FOH, very interesting reading in Austin's thread.

Thanks for all your input so far in this thread, and I hope your reasons or not being able to check the forum are not something serious or negative... (not sure how to word that, hopefully you know what I mean). Either way hope to see you soon.

By the way, I think I'm going to have 2 settings on the minidsp. One as is (pretty flat to 10hz with slight bump around 20hz like posted above a couple posts) which sounds great and will work fine for normal movie watching and music listening and another that will be when I really want to push things or have a demo.

For that setting I'm going to mimic the roll off of what dual corner stacked Submersives did in here below 30hz. It's amazing how much less boost I'll have to have to accomplish that and headroom will be off the charts.
post #396 of 790
Thread Starter 
Ha, well that will be easy. Here is a graph of one submersive compared to one SI, neither has any eq:






Now when I go to 8 SI's compared to dual corner stacked Submersives:



Gotta love that, same as the Submersives in the low end in my room with no eq. I don't think that's too shabby for how cheap these drivers are. It also points out yet again how crappy my room is for room gain down low.....
Edited by carp - 3/23/13 at 9:46pm
post #397 of 790
That says a lot for the submersives. Duals vs eight. Now I agree that the SI's are cheap, but if space is a concern that graph is a no brainer. I'm actually a little surprised it's that close.
post #398 of 790
Thread Starter 
Since I'm going graph posting crazy, one more comparing 8 subs up front before and after LT and eq:

The red is no eq or LT, black is with. I didn't have the speakers on in the red one, can't remember why not.


post #399 of 790
Thread Starter 
Quote:
Originally Posted by bass addict View Post

That says a lot for the submersives. Duals vs eight. Now I agree that the SI's are cheap, but if space is a concern that graph is a no brainer. I'm actually a little surprised it's that close.

Right but both SI measurments were level matched, so you won't see any higher spl's from the SI's than the Submersives. Actually I don't understand why the bottom end comes up for the SI's with more subs since it's level matched and the volume is brought down before measuring. You know what I'm trying to say?
post #400 of 790
Thread Starter 
That and the Submersive has built in DSP and in that graph no eq at all for the SI's.
post #401 of 790
"The best delay on my receiver for all subs was 13.5 feet as far as integrating with the speaker(s)."

that seems a little odd. 13.5ft is about one cycle at 80hz. the subs are probably one cycle behind the mains. have you tried it with zero delay on the subs and a hair of delay on the mains?
post #402 of 790
Quote:
Originally Posted by carp View Post

Right but both SI measurments were level matched, so you won't see any higher spl's from the SI's than the Submersives. Actually I don't understand why the bottom end comes up for the SI's with more subs since it's level matched and the volume is brought down before measuring. You know what I'm trying to say?

Duh, that makes sense, lol. Guess I should pay attention more and zombie read less. biggrin.gif

I was for some reason thinking that was max SPL on each one.
post #403 of 790
Thread Starter 
Quote:
Originally Posted by LTD02 View Post

"The best delay on my receiver for all subs was 13.5 feet as far as integrating with the speaker(s)."

that seems a little odd. 13.5ft is about one cycle at 80hz. the subs are probably one cycle behind the mains. have you tried it with zero delay on the subs and a hair of delay on the mains?

I should have given more info. My mains have a delay of 12 and 11' 10" so it's real close to what you are saying right?
post #404 of 790
Thread Starter 
Quote:
Originally Posted by bass addict View Post

Duh, that makes sense, lol. Guess I should pay attention more and zombie read less. biggrin.gif

I was for some reason thinking that was max SPL on each one.

Haha, oh man I'd be pissed if that were the case!!!! biggrin.gif I would picture all that work in the garage just to end up with the same output as I had with the Submersives - god that would have sucked! biggrin.gif
post #405 of 790
Quote:
Originally Posted by carp View Post

It also points out yet again how crappy my room is for room gain down low.....

Interesting, but it also is illustrative of how dominating a room is to the response. It's all about the room,..at least that's the mantra .. cool.gif

A couple years ago, the two brothers pulled off a comparo of subs ~$1k.

The room is dominating, as shown here.



Reading your post, I remembered this comparo. Fwiw, it was in this thread. Fun, and they really put a great deal of effort into the presentaion component. The images, etc, were sweet.
post #406 of 790
Quote:
Originally Posted by carp View Post

Actually I don't understand why the bottom end comes up for the SI's with more subs since it's level matched and the volume is brought down before measuring. You know what I'm trying to say?

What are you saying?
post #407 of 790
Thread Starter 
What I mean is, with the 8 SI drivers compared to 1 I changed the volume on the amp and sub out from the receiver so that the output from 1 SI and 8 SI's matched on my spl meter.

So, since that's the case I thought the low end would look the same. Each driver is working much less, but wouldn't the end result be the same? I understand why the curve is smoothed out since there are multiple subs in different places now so yeah things get smoother, but why the low end boost? Obviously it would happen if I left the volume settings on the amps alone but that's not what I did.
post #408 of 790
Which graph?
post #409 of 790
One potential reason, you have less an opportunity for destructive modal interaction, within the room axis the subs are spread.
post #410 of 790
Thread Starter 
Oh yes, I remember that thread well it's one of the best I've ever seen.

It also was posted right around the time of my re-awakening into the hobby. It got me thinking and prompted me to post this thread.

http://www.avsforum.com/t/1319822/svs-vs-1000-dollar-shoot-out-subs-lots-of-pictures-of-my-system-room#post_20106224


Back when I posted that I bet I had like 70 posts tops. My thread led me to where I am today.

Very interesting reading through my thread. Later in it I meet Luke and Jonathan for the first time. Luke invites me over to hear his VTF-15 and it's all over for me, I gotta have some bass!!!!!

So, I owe a lot to the thread you linked. If I hadn't seen it I wouldn't have posted mine wouldn't be in audio heaven and wouldn't have met the KC guys.
post #411 of 790
Thread Starter 
Quote:
Originally Posted by FOH View Post

Which graph?

First graph on post 396.
post #412 of 790
Thread Starter 
Oops, I meant both graphs on post 396. The 2nd is where the SI's close the gap down low.
post #413 of 790
Call me confused.

The upper graph, the single SubM, single SI

CREATOR: gd-jpeg v1.0 (using IJG JPEG v62), default quality

The delta could easily just be the inherent lift EQ built in the dsp program. The onset is about right. Don't know really.




Then, the 8 SI, dual SubM

CREATOR: gd-jpeg v1.0 (using IJG JPEG v62), default quality

Any delta around 6hz, and 65dB can't be trusted.

The mid-modal range smoothing is perhaps the axis smoothing I mentioned above in #409.
post #414 of 790
Thread Starter 
Quote:
Originally Posted by FOH View Post

Call me confused.

The upper graph, the single SubM, single SI

CREATOR: gd-jpeg v1.0 (using IJG JPEG v62), default quality

The delta could easily just be the inherent lift EQ built in the dsp program. The onset is about right. Don't know really.




Then, the 8 SI, dual SubM

CREATOR: gd-jpeg v1.0 (using IJG JPEG v62), default quality

Any delta around 6hz, and 65dB can't be trusted.

The mid-modal range smoothing is perhaps the axis smoothing I mentioned above in #409.

I know you had to leave, but I'll post if anyone is interested. My point was that between 10hz and close to 30hz the SI driver closed the gap on the Submersive. The volume output of the SI was the same for 1 SI as it was for 8 SI's since I matched them to the same db level on my meter.


Whatever it is I like it because it means that I'm getting the same low end from my SI subs as I had with the Submersive BUT with no eq/boost/dsp LT/ etc. and with tons more headroom!! Haha, no wonder I was absolutely loving it the other night when I turned off the minidsp and just cranked all 8 subs "naked". That was when I gave in before I saw a red clip light even flicker.

I won't run the subs that way often, since I don't need those levels very often. But - it's nice to know it's there when I need it! smile.gif
post #415 of 790
might just be an optical illusion.

smooth off the peaks with multiple subs, then raise the whole level to level match.

post #416 of 790
Thread Starter 
Quote:
Originally Posted by LTD02 View Post

might just be an optical illusion.

smooth off the peaks with multiple subs, then raise the whole level to level match.


That makes sense.
post #417 of 790
Thread Starter 
Still, I think crazy demo mode nuttiness has to be no eq. I think I might get a little crazier with the volume than most (not often, but when I do...), way past reference level bass. Last GTG here we tripped a breaker when it was at +12 master volume and god knows how hot above that we were running the dual opposed dayton's and dual submersives. NUTS!!
post #418 of 790
Is this using track 3 on the Omnimic to level match? What did you use in the first graph to level match the SI and the SubMersive? I ask because it doesn't look like they would be level matched, since they are mostly equal but the SubMersive clearly has more going on down low. Actually it looks a lot like the graph you'd see if you matched them with the SubM in pgm1 and then, without re-matching, swapped it to pgm2. So, for them to be truly matched in your Omnimic level matching procedure you'd have to shift the red curve up some.

Then, when you add go from 1->8 SI's you're getting the constructive-only gain down low (below the Super Magic Infinitely Rigid pipe Frequency, which I'm now dubbing the SMIRF) and a variety of constructive/destructive gains up high so, when you level match, it looks like more of the contribution to that level is coming from the lower frequencies then before. At least that's my best interpretation, I'm sure one of the smarter guys will have a better explanation.

Also, earlier I was trying to look up something about the Omnimic and it looks like there's a v4 of the software that has some integration with miniDSP. It might be worth a read, it starts on page 25 of this manual.
post #419 of 790
Thread Starter 
Quote:
Originally Posted by MrSmithers View Post

Is this using track 3 on the Omnimic to level match? What did you use in the first graph to level match the SI and the SubMersive? I ask because it doesn't look like they would be level matched, since they are mostly equal but the SubMersive clearly has more going on down low. Actually it looks a lot like the graph you'd see if you matched them with the SubM in pgm1 and then, without re-matching, swapped it to pgm2. So, for them to be truly matched in your Omnimic level matching procedure you'd have to shift the red curve up some.

Then, when you add go from 1->8 SI's you're getting the constructive-only gain down low (below the Super Magic Infinitely Rigid pipe Frequency, which I'm now dubbing the SMIRF) and a variety of constructive/destructive gains up high so, when you level match, it looks like more of the contribution to that level is coming from the lower frequencies then before. At least that's my best interpretation, I'm sure one of the smarter guys will have a better explanation.

Also, earlier I was trying to look up something about the Omnimic and it looks like there's a v4 of the software that has some integration with miniDSP. It might be worth a read, it starts on page 25 of this manual.

It was using the tones from my receiver, and either the radio shack spl meter or the omnimic meter not sure since I use both. Too bad I shipped the Submersive this week or I could do some more comparing. I know the Submersive was in program 1 though.

Cool thanks for the link.

Ahhh yes the SMIRF. Haha biggrin.gif
post #420 of 790
I wonder what the spectral content of the pink noise test signals on your receiver is, a simple google search isn't turning up anything definitive. Actually I find that whole process a little frustrating anyways. Say for the sake of argument your receiver was putting out 20-80Hz pink noise for calibration. Then, if you had flat (or even hotter) 20->10 Hz your calibrated level would be the exact same as someone who had nothing below 20Hz but was flat above. Similarly, if the pink noise instead was 10-80Hz and you both level matched, stuff without much going on below 20 (say, Avengers) would be much louder and more impressive to that other guy (assuming the same performance in the frequency range where you're both flat). To the guy whose subwoofer system rolled off at 25-30Hz he might think and post about how the Avengers track is the greatest bass track ever created, because to level match he'd be running those frequencies hot compared to you, even though there's really nothing either of you are doing "wrong".
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: DIY Speakers and Subs
AVS › AVS Forum › Audio › DIY Speakers and Subs › First build - Octo SI 18's with flat packs.