Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How To Interpret Graphs - Page 196 - AVS Forum
Forum Jump: 
 118Likes
Reply
 
Thread Tools
post #5851 of 11702 Old 10-29-2013, 11:30 AM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 7,331
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 611 Post(s)
Liked: 745
Quote:
Originally Posted by kbarnes701 View Post
 

 

 

EDIT: here's the spectrogram of the same thing:

 

 

It isn't all that bad, is it?? It has a lot more red/orange in it than yours though, Jerry. I don't understand, looking at our two waterfalls, why that should be the case - anyone?

 

According to the scale on the right, the difference in colors simply indicates your base measurement level was somewhat higher than mine.  No need for concern, IMO.


__________________________________________________

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
AustinJerry is online now  
Sponsored Links
Advertisement
 
post #5852 of 11702 Old 10-29-2013, 12:01 PM
AVS Special Member
 
jkasanic's Avatar
 
Join Date: Dec 2006
Posts: 1,169
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 41 Post(s)
Liked: 62
Quote:
Originally Posted by AustinJerry View Post
 
Quote:
Originally Posted by jkasanic View Post
 

 

Ok, I'll be the guinea pig and ask for some help interpreting my Spectogram plots!  I know my measurements are on the low side but I thought a real world example might help others visualize the text of your post:

 

 

 

IMO, the Waterfall with Audyssey shows a textbook issue, the mountain at 65-70Hz.  Since this is a pretty important frequency for a lot of things, it should be a primary focus.

 

Nice analysis.

 

Thanks (as usual) for the response Jerry.  Could you clarify a little bit on the textbook portion of your statement?  I guess I'm a little confused as to what makes what is going on between 65-70Hz and what's happening between 40 and 50Hz as well 85-90Hz different (besides the former being more difficult to treat)?  The peak at 85Hz is in the crossover region and I did apply the sub distance tweak to smooth out the null that Audyssey created.  During my next round of measurements, I was planning to experiment with a 90Hz crossover as I'm currently using 80Hz for all speakers.

jkasanic is offline  
post #5853 of 11702 Old 10-29-2013, 12:06 PM
Advanced Member
 
artur9's Avatar
 
Join Date: Mar 2013
Location: near philly
Posts: 725
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 47 Post(s)
Liked: 40
Quote:
Originally Posted by kbarnes701 View Post

It isn't all that bad, is it?? It has a lot more red/orange in it than yours though, Jerry. I don't understand, looking at our two waterfalls, why that should be the case - anyone?

More red just means yours was louder at the same frequency. What's really incredible about your spectrogram is that all the frequencies appear to have the same volume. Talk about flat responses!

(I used to spend hours looking at spectrograms back when I worked on natural language speech recognition systems).
artur9 is offline  
post #5854 of 11702 Old 10-29-2013, 12:14 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 7,331
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 611 Post(s)
Liked: 745
Quote:
Originally Posted by jkasanic View Post

Thanks (as usual) for the response Jerry.  Could you clarify a little bit on the textbook portion of your statement?  I guess I'm a little confused as to what makes what is going on between 65-70Hz and what's happening between 40 and 50Hz as well 85-90Hz different (besides the former being more difficult to treat)?  The peak at 85Hz is in the crossover region and I did apply the sub distance tweak to smooth out the null that Audyssey created.  During my next round of measurements, I was planning to experiment with a 90Hz crossover as I'm currently using 80Hz for all speakers.

I think it is textbook because the mountain starts immediately at the top and has closely spaced slices all the way to the bottom, meaning early and persistent ringing. Go back and read J's description again.

__________________________________________________

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
AustinJerry is online now  
post #5855 of 11702 Old 10-29-2013, 12:17 PM
Advanced Member
 
artur9's Avatar
 
Join Date: Mar 2013
Location: near philly
Posts: 725
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 47 Post(s)
Liked: 40
Quote:
Originally Posted by LastButNotLeast View Post

Just when you think things are getting too esoteric, the real world steps in with something different.
We recently got a new refrigerator (stay with me, here).
After a while, I noticed a high-pitched noise that was driving my tinnitus crazy.
I used the Real Time Analyzer of REW (see, I told you) to localize the sound to about 2kHz.




The Generator confirmed that that was the sound I was hearing.
The service tech suggested putting the fridge on carpet. I'm pretty sure you guys can do better.
So put down your base traps and help me find something to absorb sound at around 2kHz.
Thanks.
Michael

I came across this site listing a bunch of materials recently. Owens Corning 701 has a good rating at that frequency. But can you tell us more about where the fridge is located, etc?

I have a fridge near my MLP in a kind of closet. I've been stuffing the closet with discarded eggshell foam from work to control the sound. 'Course, now people at work think I'm strange. rolleyes.gif
artur9 is offline  
post #5856 of 11702 Old 10-29-2013, 12:38 PM
AVS Special Member
 
jkasanic's Avatar
 
Join Date: Dec 2006
Posts: 1,169
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 41 Post(s)
Liked: 62
Quote:
Originally Posted by AustinJerry View Post
 
Quote:
Originally Posted by jkasanic View Post

Thanks (as usual) for the response Jerry.  Could you clarify a little bit on the textbook portion of your statement?  I guess I'm a little confused as to what makes what is going on between 65-70Hz and what's happening between 40 and 50Hz as well 85-90Hz different (besides the former being more difficult to treat)?  The peak at 85Hz is in the crossover region and I did apply the sub distance tweak to smooth out the null that Audyssey created.  During my next round of measurements, I was planning to experiment with a 90Hz crossover as I'm currently using 80Hz for all speakers.

I think it is textbook because the mountain starts immediately at the top and has closely spaced slices all the way to the bottom, meaning early and persistent ringing. Go back and read J's description again.

 

Ok, what I understood from J's explanation seems like it would apply more to the 80-90Hz region (or both hence the reason for my question)?  The 65-70Hz region has a shelf (shown circled in blue and implying a different rate of ringing) while the 83'ish Hz ringing appears to be consistent the full length of the waterfall (as designationed by the straight blue line) or have I misunderstood?! :confused:

 

jkasanic is offline  
post #5857 of 11702 Old 10-29-2013, 12:48 PM
AVS Special Member
 
jkasanic's Avatar
 
Join Date: Dec 2006
Posts: 1,169
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 41 Post(s)
Liked: 62

WRT to the Spectrogram plots, what does the area below t=0 actually represent (seen circled in blue below)?

 

jkasanic is offline  
post #5858 of 11702 Old 10-29-2013, 01:12 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 7,331
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 611 Post(s)
Liked: 745
Quote:
Originally Posted by jkasanic View Post
 

 

Ok, what I understood from J's explanation seems like it would apply more to the 80-90Hz region (or both hence the reason for my question)?  The 65-70Hz region has a shelf (shown circled in blue and implying a different rate of ringing) while the 83'ish Hz ringing appears to be consistent the full length of the waterfall (as designationed by the straight blue line) or have I misunderstood?! :confused:

 

 

Yea, we are talking about the same mountain.  I was extending a straight line down to the horizontal scale, forgetting that the scale is actually offset to the left.  Not a feature I like about the way the graph is presented.  It makes it too easy to misinterpret where the problem is.

 

All good now.


__________________________________________________

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
AustinJerry is online now  
post #5859 of 11702 Old 10-29-2013, 01:14 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 7,331
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 611 Post(s)
Liked: 745
Quote:
Originally Posted by jkasanic View Post
 

WRT to the Spectrogram plots, what does the area below t=0 actually represent (seen circled in blue below)?

 

 

Wow, that blue plume at 30Hz!  Is your system on fire?  :p


__________________________________________________

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
AustinJerry is online now  
post #5860 of 11702 Old 10-29-2013, 01:38 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 17,940
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 1437 Post(s)
Liked: 1712
Quote:
Originally Posted by AustinJerry View Post
 
Quote:
Originally Posted by kbarnes701 View Post
 

 

How about that ceiling/wall corner over the fireplace?  Can you not get a couple more 244s in there?

 

I have been looking into Skopus traps too but I don't think I have any room for any more.

 

The Height speaker gets in the way, unfortunately.  And on the left side of the fireplace, there is also no room.  I talked to Bryan Pape at GIK, and he says that the Scopus trap should be placed flat on the wall, so the space on the front wall seems to be a good candidate.

 

Can you put the Skopus anywhere in the room?  If so, then I might have a spot....



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
kbarnes701 is online now  
post #5861 of 11702 Old 10-29-2013, 01:40 PM
Advanced Member
 
artur9's Avatar
 
Join Date: Mar 2013
Location: near philly
Posts: 725
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 47 Post(s)
Liked: 40
Quote:
Originally Posted by jkasanic View Post

WRT to the Spectrogram plots, what does the area below t=0 actually represent (seen circled in blue below)?



From http://www.hometheatershack.com/forums/rew-forum/53408-spectrogram.html
Quote:
The spectrogram shows us both the rise and the fall of the SPL by looking down from above the waterfall graph. It can now show the rise of SPL (negative time) and the fall of SPL (positive time).
Quote:
Originally Posted by AustinJerry View Post

Wow, that blue plume at 30Hz!  Is your system on fire?  tongue.gif

It does look like smoke! But since the rise time precedes the other frequencies I think it must be something in the room. A fish tank?
artur9 is offline  
post #5862 of 11702 Old 10-29-2013, 01:41 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 17,940
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 1437 Post(s)
Liked: 1712
Quote:
Originally Posted by AustinJerry View Post
 
Quote:
Originally Posted by jim19611961 View Post


Open "E" on a bass guitar is about 42hz, and all rock music will hit this note fairly often. A 5 or 6 string bass reaches 31.5hz, again, not uncommon. Piano goes down to 27.5hz. Pipe Organ has a fundamental at 16hz.

Just saying smile.gif

 

I stand corrected, Jim, thanks.  I have ordered four Scopus tuned membrane traps, the T-40's tuned to 40Hz.  The actual range is 30-55Hz, which seems to be the area on my Waterfalls that deserves the most attention.  I'll look forward to measuring and posting the results.

 

 

Way to go Jerry!  I will await your report. If they have something that can handle 50-80Hz-ish then I might be able to accommodate them... ;)



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
kbarnes701 is online now  
post #5863 of 11702 Old 10-29-2013, 01:42 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 17,940
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 1437 Post(s)
Liked: 1712
Quote:
Originally Posted by AustinJerry View Post
 
Quote:
Originally Posted by kbarnes701 View Post
 

 

 

EDIT: here's the spectrogram of the same thing:

 

 

It isn't all that bad, is it?? It has a lot more red/orange in it than yours though, Jerry. I don't understand, looking at our two waterfalls, why that should be the case - anyone?

 

According to the scale on the right, the difference in colors simply indicates your base measurement level was somewhat higher than mine.  No need for concern, IMO.

 

Thanks. Understood now (thanks to Joe too).



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
kbarnes701 is online now  
post #5864 of 11702 Old 10-29-2013, 01:45 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 17,940
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 1437 Post(s)
Liked: 1712
Quote:
Originally Posted by artur9 View Post
 
Quote:
Originally Posted by kbarnes701 View Post

It isn't all that bad, is it?? It has a lot more red/orange in it than yours though, Jerry. I don't understand, looking at our two waterfalls, why that should be the case - anyone?

More red just means yours was louder at the same frequency. What's really incredible about your spectrogram is that all the frequencies appear to have the same volume. Talk about flat responses!

(I used to spend hours looking at spectrograms back when I worked on natural language speech recognition systems).

 

Oh, that is great news artur - thanks. Yes I do have a fairly flat bass response now - after about 12 solid months (or more) of trying, using treatments, Audyssey, speaker and sub placement optimisation, REW and a LOT of help from the guys in this thread. There is just that one niggle left that I would like to address if possible. I can’t bring the MLP forward any more as the room is too small - but the Skopus traps look promising, depending on Jerry's findings.



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
kbarnes701 is online now  
post #5865 of 11702 Old 10-29-2013, 01:46 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 7,331
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 611 Post(s)
Liked: 745
Quote:
Originally Posted by kbarnes701 View Post

Can you put the Skopus anywhere in the room?  If so, then I might have a spot....

Yes, anywhere is fine. However, Brian indicated that it takes a number of Scopus traps to have the desired effect. I was going to get three, and he talked me into four (good salesman). You might want to ask Bryan directly if a single trap would be worthwhile.

__________________________________________________

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
AustinJerry is online now  
post #5866 of 11702 Old 10-29-2013, 01:57 PM
AVS Special Member
 
jkasanic's Avatar
 
Join Date: Dec 2006
Posts: 1,169
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 41 Post(s)
Liked: 62
Quote:
Originally Posted by AustinJerry View Post
 
Quote:
Originally Posted by jkasanic View Post
 

WRT to the Spectrogram plots, what does the area below t=0 actually represent (seen circled in blue below)?

 

 

Wow, that blue plume at 30Hz!  Is your system on fire?  :p

Yeah, switching to the Spectrogram just makes those hard to deal with areas (aka below 40Hz) look even more problematic!

jkasanic is offline  
post #5867 of 11702 Old 10-29-2013, 04:42 PM
AVS Special Member
 
fotto's Avatar
 
Join Date: Nov 2003
Location: N.E. OH
Posts: 1,868
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 19
I guess my earlier post on spectograms got overlooked....
The bottom or negative side of the "mountain" represents ramp up from the initial signal. You don't get to a max spl level from initial signal instantaneaously right? Has to be some ramp of time to reach pressure.

So you're looking at an aerial view of a mountain range. That range has peaks and valleys along it's range corresponding to you particular SPL at that point. You walk up one side of the mountain from ground zero (negative side) and then down the opposite side to ground zero again. Some points on the range have a drop off to sea level which are fast decay, while other areas you have a gradual walk/slope down the mountain, or long decay times. Higher SPL is this color coding represent higher spl (hot) while blue represent cooler temp of lower SPL, with other shades in between.

Consistent color over the range correspond to flatter spl, with overall decay times in Y axis.
jkasanic likes this.

Floyd

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
fotto is offline  
post #5868 of 11702 Old 10-29-2013, 04:53 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 7,331
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 611 Post(s)
Liked: 745
Sorry for not responding to your original post a week ago, Fotto. I appreciate the feedback. Between your feedback, and the input from J, the Spectrogram is coming into focus now.

__________________________________________________

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
AustinJerry is online now  
post #5869 of 11702 Old 10-29-2013, 05:20 PM
AVS Special Member
 
LastButNotLeast's Avatar
 
Join Date: Feb 2007
Location: 08077
Posts: 4,570
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 178 Post(s)
Liked: 245
Quote:
Originally Posted by artur9 View Post


I came across this site listing a bunch of materials recently. Owens Corning 701 has a good rating at that frequency. But can you tell us more about where the fridge is located, etc?

I have a fridge near my MLP in a kind of closet. I've been stuffing the closet with discarded eggshell foam from work to control the sound. 'Course, now people at work think I'm strange. rolleyes.gif

Thanks for that. That should help. The refrigerator is, well, in the kitchen, pretty much where you'd expect a refrigerator to be, on a vinyl floor next to the stove. A pile of egg cartons in the kitchen would not have a high WAF, but I'm thinking something BEHIND the fridge should help. Customer service suggested carpet underneath; I may try something on the wall. 


Downloadable FREE demo discs:

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
 

Did you really need to quote that entire post in your reply?
LastButNotLeast is offline  
post #5870 of 11702 Old 10-29-2013, 11:06 PM - Thread Starter
Senior Member
 
jevansoh's Avatar
 
Join Date: Oct 2010
Location: Southern, Ohio
Posts: 471
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 47
Quote:
Originally Posted by LastButNotLeast View Post

Quote:
Originally Posted by jevansoh View Post

If you cut 80hz I'm sure it's working. It's nulls that you cannot boost, but cutting should never be a problem.


Here's what I "think" is happening.


You are using a Mini-DSP, right? That's only hooked up to your subs. So you are cutting 80hz on the subs but the subs aren't what is boosting the 80hz. The 80hz peak is probably coming from your mains.


You need a way to cut 80hz in the mains.


To verify, set the Mini-DSP back to 0dB cut/boost at 80hz then unplug your mains. Do a sweep. Then cut 6dB, 12dB, etc, and see if it works. It should. wink.gif


This will verify that the peak is coming from your mains and you either need a new speaker location, seating location, combination of both, acoustic treatments, and/or EQ for the mains.


Hope this helps.... err.. At least hope it helps explain what's going on. smile.gif


--J
How about a higher crossover? Too easy? biggrin.gif

tongue.gif I didn't think about that lol. However, that "could" cause other problems too, the least of which would be possible localization of the subs depending on where they're located.

For diagnostics though and to verify I'm right about the speakers being the cause of the peak, temporarily setting a higher crossover may just be the answer! wink.gif

Thanks to EVERYONE that Helps Make These Threads so Awesome!

jevansoh is offline  
post #5871 of 11702 Old 10-29-2013, 11:24 PM - Thread Starter
Senior Member
 
jevansoh's Avatar
 
Join Date: Oct 2010
Location: Southern, Ohio
Posts: 471
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 47
Quote:
Originally Posted by AustinJerry View Post

Quote:
Originally Posted by jevansoh View Post

Waterfall Interpretation:


Since there has been some confusion lately, I'm going to try and clear a few things up.


 

Ah, J, you are a man of such few words!  Seriously, this has been a very useful post, especially regarding interpreting the Waterfall graph, and for the introduction to the Spectrogram, which we have not been focusing on.  I look forward to interesting future discussions regarding phase, as I am sure Markus has some research in that area as well.

So, using your explanation as a guideline, Here is my current waterfall once again, with the lower limit set to 45dB:




Looking at the mountains, it would seem that I have issues at 48, 62, and 78Hz (at least these are the worst issues).

Now the Spectrogram:




With the cursor marking 450ms, I don't see any flames that reach this level.  I interpret this as a good thing (yes?).

Second Spectrogram:




This time, the cursor shows the level of the highest blue flame, and it registers at 385ms.  Based on your explanation, it would seem that this is still a reasonably good result.  The most serious issue below 100Hz is at ~60Hz, which is consistent with the worst mountain in the Waterfall.

So, am I getting closer to understanding Waterfalls and Spectrograms?  BTW, I would like to place a link to this posting in the Guide, acknowledging your contribution.

Now you've got it!! biggrin.gif

With the exception of between 15-20hz (good luck ever doing anything about that) see how ONLY the frequencies you mentioned appear to start pushing forward near the top of the waterfall? Those are the problems.

You have other "mountains" that appear at the bottom of the graph that don't end by 450ms but they don't begin to protrude until several 10's of db's down from the top so they aren't really much to be concerned about. In fact, they could be measurement artifacts or even noise (electronic or acoustic) and if you measure again without changing anything it may even look different.

But those peaks/mountains pushing forward right about in the middle of your graph are actually issues.

When I say they are issues, I only mean you don't have an even decay. It's true you have nothing of concern that pushes past 450ms but 450ms isn't your average either.

This is why I like using the Spectogram because it's easier (for me) to get a quick glimpse at the average and the "real" problem peaks that stick out (up in this case). It shows your average is actually around 300ms. Look at the Green to Light Blue transition and see how it's mostly a horizontal line which is what we want. There are flames at the problem frequencies that rise above the 300ms average but 300ms IS the average.

So, even though you don't have anything peaking past 450ms, in my opinion, there is still room for improvement, because the ultimate goal is consistency and you do have (referring back to the waterfall) mountains that push forward.

I feel a little silly referring to mountains and flames and am having to hold myself back from getting a bit more technical but I think at this point it's probably best to take it slower and make sure everyone is on the same page.

I would work on the exact frequencies you mentioned with the ultimate goal to have those peaks that are pushing forward from the middle and starting nearer the top of the waterfall to decay by 300ms or so (or at least closer than they are now) to even things out.

This will tighten up your bass and will possibly even reveal more subtleties and nuance in other parts of the track you're listening to that could be covered up or muddied up by the response you're hearing now.

It's important to note that your response is absolutely not terrible and in fact is right at the point I'd call "good."

It could be better but could also be a heckuva lot worse! smile.gif

Hope this helps and by all means feel free to reproduce in part or in full and feel free to modify and/or rewrite for better clarification for others.

Both of my parents were teachers then principals all their lives until just recently retiring yet I most definitely did not inherit the ability to teach! I am not even good at explaining the plot of a movie and hate it when people ask me what such and such movie was about, lol.

Thanks,

--J

Thanks to EVERYONE that Helps Make These Threads so Awesome!

jevansoh is offline  
post #5872 of 11702 Old 10-29-2013, 11:43 PM - Thread Starter
Senior Member
 
jevansoh's Avatar
 
Join Date: Oct 2010
Location: Southern, Ohio
Posts: 471
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 47
Quote:
Originally Posted by jkasanic View Post

Quote:
Originally Posted by jevansoh View Post

PS After proofreading this, I see it is blatantly obvious that this topic isn't blatantly obvious. biggrin.gifThis hopefully will start a discussion that through questions and clarifications will get everyone on the same page, understanding what the actual goal is and why, how to interpret these charts and graphs, and eventually, what to "do" about it to actually fix these issues.

EDIT:  I should've known Jerry would beat me to it! biggrin.gif   I also added waterfalls to my post.

Ok, I'll be the guinea pig and ask for some help interpreting my Spectrogram plots!  I know my measurements are on the low side but I thought a real world example might help others visualize the text of your post:

FR with and without Audyssey for L+R+Sub no smoothing:




L+R+Sub Waterfall without Audyssey:




L+R+Sub Waterfall with Audyssey:





L+R+Sub Spectrogram without Audyssey:




L+R+Sub Spectrogram with Audyssey XT32:




Just a reminder, room is currently completely untreated but I'm working on adding absorption behind the false wall of my projection screen (AT GOM fabric plus AT screen) as well as treat the L/R contra lateral reflection points on my side walls.  I'm subscribing to the approach of not treating reflections that are coming from where they are supposed to but rather treating the ones coming from where they are not supposed to (e.g. surrounds reflecting off the front wall, contra laterals as previously mentioned and possibly some back wall treatments for reflections of mains).

You need help. biggrin.gif

Ok, look at your first waterfall before Audyssey.

See how you basically only have a long decay issue on the left hand side of the graph where the mountain hasn't fully decayed by the graph limit? Other than that problem, PRE-EQ, what this tells me first is that it's possible (highly probable IMO) you could do better with subwoofer placement and/or integration with main speakers and/or adding more subs.

Now look at what Audyssey has done to the waterfall. It took down one peak and made that problem on the left hand side a little better but by having to raise basically everything else it made all the other problems in the time domain much much worse.

This is why EQ is bad IMO. Yeah, it is sometimes necessary and even has its place for implementing house curves and when there is no other option.

Those of us in this thread though are here because we want other options.

I'm not going to get into phase right this second, but do remember the comparison with your waterfall before and after Audyssey and remember the goal is to flatten out that frequency response without introducing time domain issues. Audyssey did the former but failed miserably at the latter.

By adding treatments, if you size them and place them properly, and assuming this is AFTER you have found the best placement for speakers, listener, and subs you'll not need as much correction and Audyssey can then be used without doing so much harm.

Refer to the post above in response to Jerry's waterfall/spectogram for some more detail in what to look for and how to interpret them and if you have questions, be sure to ask. I figure it's best to give you a chance to read that post first instead of repeating a bunch of stuff since my posts already tend to be so long I'm afraid a bunch of people probably skip over them. wink.gif

Also, make sure when considering treatments you don't just go buy a bunch of stuff and spend a bunch of money.

It's ok and in fact encouraged to explain your limits in budget, space, time to build (you can get so much more for your money by DIY'ing and can do so much better plus it's actually very easy!) or preference to buy, along with offering some pictures of your room so I can recommend treatments.

I've seen much worse graphs before treatment, but based on what I do see with yours I feel with enough dedication you may not even need EQ when you're done, other than to set a house curve or for a bit of touch up.

You mentioned having a false wall. If you have at least 10" or so on the front wall I totally recommend covering it in R-30 (or even R-38 which is the same thing only thicker) insulation. You can hang this vertically so it doesn't compress and want to fall over after you've stacked a few pieces.

If, however, you only have 4" to spare or so, then you'll need more expensive stuff like Safe N Sound which you can get at your local Lowes or HD.

In certain areas you can use the smaller, but more rigid/dense/expensive stuff where you don't have as much room, but for corners (remember, there are 12 corners in a room) and in hidden areas or areas with a lot of blank space (anywhere where you can use at least 8" or more of material) the absolute best stuff to use is the cheapest and easiest to obtain standard insulation.

Let me know if this clears some things up and feel free to share some more info so we can get you on a treatment plan to clear up your issues.

Hope this helps,

--J

PS For answers to your other questions, refer to the reply to Jerry's post above and if I've forgotten/missed anything be sure to let me know.

Thanks to EVERYONE that Helps Make These Threads so Awesome!

jevansoh is offline  
post #5873 of 11702 Old 10-30-2013, 12:02 AM
AVS Special Member
 
jlpowell84's Avatar
 
Join Date: Dec 2012
Location: Eugene, OR
Posts: 4,079
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 177 Post(s)
Liked: 225
Quote:
Originally Posted by jevansoh View Post

You need help. biggrin.gif

Ok, look at your first waterfall before Audyssey.

See how you basically only have a long decay issue on the left hand side of the graph where the mountain hasn't fully decayed by the graph limit? Other than that problem, PRE-EQ, what this tells me first is that it's possible (highly probable IMO) you could do better with subwoofer placement and/or integration with main speakers and/or adding more subs.

Now look at what Audyssey has done to the waterfall. It took down one peak and made that problem on the left hand side a little better but by having to raise basically everything else it made all the other problems in the time domain much much worse.

This is why EQ is bad IMO. Yeah, it is sometimes necessary and even has its place for implementing house curves and when there is no other option.

Those of us in this thread though are here because we want other options.

I'm not going to get into phase right this second, but do remember the comparison with your waterfall before and after Audyssey and remember the goal is to flatten out that frequency response without introducing time domain issues. Audyssey did the former but failed miserably at the latter.

By adding treatments, if you size them and place them properly, and assuming this is AFTER you have found the best placement for speakers, listener, and subs you'll not need as much correction and Audyssey can then be used without doing so much harm.

Refer to the post above in response to Jerry's waterfall/spectogram for some more detail in what to look for and how to interpret them and if you have questions, be sure to ask. I figure it's best to give you a chance to read that post first instead of repeating a bunch of stuff since my posts already tend to be so long I'm afraid a bunch of people probably skip over them. wink.gif

Also, make sure when considering treatments you don't just go buy a bunch of stuff and spend a bunch of money.

It's ok and in fact encouraged to explain your limits in budget, space, time to build (you can get so much more for your money by DIY'ing and can do so much better plus it's actually very easy!) or preference to buy, along with offering some pictures of your room so I can recommend treatments.

I've seen much worse graphs before treatment, but based on what I do see with yours I feel with enough dedication you may not even need EQ when you're done, other than to set a house curve or for a bit of touch up.

You mentioned having a false wall. If you have at least 10" or so on the front wall I totally recommend covering it in R-30 (or even R-38 which is the same thing only thicker) insulation. You can hang this vertically so it doesn't compress and want to fall over after you've stacked a few pieces.

If, however, you only have 4" to spare or so, then you'll need more expensive stuff like Safe N Sound which you can get at your local Lowes or HD.

In certain areas you can use the smaller, but more rigid/dense/expensive stuff where you don't have as much room, but for corners (remember, there are 12 corners in a room) and in hidden areas or areas with a lot of blank space (anywhere where you can use at least 8" or more of material) the absolute best stuff to use is the cheapest and easiest to obtain standard insulation.

Let me know if this clears some things up and feel free to share some more info so we can get you on a treatment plan to clear up your issues.

Hope this helps,

--J

PS For answers to your other questions, refer to the reply to Jerry's post above and if I've forgotten/missed anything be sure to let me know.

Wow this really helps me understand as well. I want to get best naturally before any EQ. What are your thoughts on Roxul AFB insulation? Hopefully not too bad lol as I just bought a bunch smile.gif

DIY Sound Group Volt 10 surround speaker build

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

My Setup Thread

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
jlpowell84 is offline  
post #5874 of 11702 Old 10-30-2013, 12:10 AM - Thread Starter
Senior Member
 
jevansoh's Avatar
 
Join Date: Oct 2010
Location: Southern, Ohio
Posts: 471
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 47
Quote:
Originally Posted by jlpowell84 View Post

Quote:
Originally Posted by AustinJerry View Post

This would be unusual. What mic do you have? Those of us with USB mics have observed that the mic's sensitivity is such that even in a dead quiet environment, the mic measures ~50dB. If you are getting measurements that low, I would question the mic's calibration. Did you calibrate it as per the instructions in the guide?

I don't think any of us are measuring at 105dB. That is pretty loud. I generally measure at either 85 or 90dB, and my noise floor measures 50dB.

Yea I got 30-34 at first then re calibrated and now get 44-45 with my UMM-6. Perhaps an important note is that I live in the country and it's very quiet out there at night.
Quote:
Originally Posted by jevansoh View Post

If you cut 80hz I'm sure it's working. It's nulls that you cannot boost, but cutting should never be a problem.

Here's what I "think" is happening.

You are using a Mini-DSP, right? That's only hooked up to your subs. So you are cutting 80hz on the subs but the subs aren't what is boosting the 80hz. The 80hz peak is probably coming from your mains.

You need a way to cut 80hz in the mains.

To verify, set the Mini-DSP back to 0dB cut/boost at 80hz then unplug your mains. Do a sweep. Then cut 6dB, 12dB, etc, and see if it works. It should. wink.gif

This will verify that the peak is coming from your mains and you either need a new speaker location, seating location, combination of both, acoustic treatments, and/or EQ for the mains.

Hope this helps.... err.. At least hope it helps explain what's going on. smile.gif

--J

Yes I thought about that and the fact that is where I have the crossover set. I can always bump it up to 90 or down to 70 if I wanted. I have not played with main speaker placement at all yet either. I have a whole bunch of materials coming to build my own acoustical treatments. I will do a lot of setting raw insulation all around and running measurements to see what I need and where. I am not dead set on my house curve that I had developed either I might end up flattening it out a little. Time listening with music and movies will tell. Although last night I ran through a bunch of Bass demos off one of the demo discs. It was freaking nuts.

Hold on a minute..... It's great you are willing to treat your room and even better (much smarter use of money and CAN be more beneficial) you are DIY'ing but....

You MUST use the proper materials in the proper places.

Exactly WHAT do you have coming? Where CAN you place it? How thick can it be?

I cannot stress enough how important it is to use the proper density (more accurately Rayl/s) and thickness of insulation if you want to successfully treat your room.

Some pictures of your room notated with where you're willing to put treatment (if it ends up necessary to place in that spot) with notations of how much it can protrude into the room in each spot (how thick it can be) will tell me exactly what type of insulation you can go with.

Compare your problems on your FR, Waterfall/Spectogram to models you create for treatment.

I've mentioned this several times and don't know if people are using it or not but everyone thinking about adding treatment NEEDS to use this site/modeling tool.

http://www.acousticmodelling.com/porous.php

It makes you enter dimensions in MM instead of inches. All you need to know is 25mm = 1inch.

In the field where it says "flow resistivity" you will enter the following numbers depending on the type of insulation to get accurate results.

Any type of standard fluffy insulation you'll find at Lowes/HD - Fiberglass insulation from R13 to R38 for instance enter "5000"

Roxul Safe N Sound or other similar types of Mineral Wool with medium densities enter "12500"

Our European friends will find it easier (and cheapest) to source different densities of Rockwool/Mineral wool (the stuff that is more rigid and not as floppy as Safe N Sound although I'm sure there is an equivalent in Europe??):

From 24kg/m^&3 to 32kg/m^3 (converted for our European friends) enter "18000" and for the even more dense 48kg/m^3 stuff enter "31500"

Owens Corning 703 or equivalent enter "27000" (This is from a NASA study so even though there have been several different numbers thrown around for this, including from the manufacturer themselves, I'm going with NASA)

Owens Corning 705 or equivalent enter "38000"

If the material you are using is not on this list there is a very comprehensive list at http://www.gearslutz.com/board/studio-building-acoustics/625978-common-gas-flow-resistivity-numbers.html which lists the GFR (Gas Flow Resistivity) of just about every known material in use for these purposes.

With this information it is very easy to model your traps and "planned" treatment plan before spending money and/or time and trouble so you don't have to redo everything.

I HIGHLY RECOMMEND using this tool!!! smile.gif (BTW - For our purposes, ALWAYS select Random Incidence)

The first person to tell me whether it is better to use 12" (300mm) of OC703 or 12" of Pink Fluffy to treat a problem at 80hz gets a cookie!

After you do this exercise try modeling 17" and then ask yourself which insulation is best (not even taking into consideration ease of procurement or cost) to use for those Superchunk corner absorbers that people make at 17" x 24" x 24" (that's the SMALLEST version!)

To sum up, it's great you have a bunch of gubbins coming, to steal a phrase from Keith, but you need to make sure they're the right gubbins for the job. wink.gif

Hope this helps,

--J

Thanks to EVERYONE that Helps Make These Threads so Awesome!

jevansoh is offline  
post #5875 of 11702 Old 10-30-2013, 12:31 AM
AVS Special Member
 
jlpowell84's Avatar
 
Join Date: Dec 2012
Location: Eugene, OR
Posts: 4,079
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 177 Post(s)
Liked: 225
I will take an in depth look into that link. I know this will cause a palm to face moment perhaps for you been around the mountain guys but I figured if ATS acoustics used Roxul AFB in their bass traps and broadband panels then it was good enough for me. I found a ton for cheap so I'm not out anything if I made the wrong choice. But it seemed right. Here is what I have







I can provide pics or a link back a couple weeks to my room if needed. I am willing to place a bunch but I can go 100% all out since it is a rental and a living room rather than a dedicated room. I was thinking corner bass traps in front right and left corners, panels all over the walls, and panels angled at the floor to wall junction about 8ft back from the front right and left corner bass traps which is the tv wall. I would also place all over the tv wall. I do have a window to deal with on the tv wall

DIY Sound Group Volt 10 surround speaker build

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

My Setup Thread

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
jlpowell84 is offline  
post #5876 of 11702 Old 10-30-2013, 12:31 AM
Senior Member
 
garygarrison's Avatar
 
Join Date: Aug 2013
Posts: 217
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 27 Post(s)
Liked: 55
Quote:
Originally Posted by LastButNotLeast View Post

Just when you think things are getting too esoteric, the real world steps in with something different.
We recently got a new refrigerator (stay with me, here).
After a while, I noticed a high-pitched noise that was driving my tinnitus crazy.
I used the Real Time Analyzer of REW (see, I told you) to localize the sound to about 2kHz.




The Generator confirmed that that was the sound I was hearing.
The service tech suggested putting the fridge on carpet. I'm pretty sure you guys can do better.
So put down your base traps and help me find something to absorb sound at around 2kHz.
Thanks.
Michael

Put the fridge in the back yard.

Seriously, though, our modern refrigerator is noisier than previous ones ... and the noise is at a higher frequency, probably at about 2K and above. Manufacturers now seem to not care a damn about noise pollution. We should have called Sears and had them come take it away, but wimped out. A solid core door between the kitchen and the music room, with an air-tight seal helped ... all but eliminated the problem. Go ahead and put the fridge on a carpet as well.
garygarrison is offline  
post #5877 of 11702 Old 10-30-2013, 12:37 AM - Thread Starter
Senior Member
 
jevansoh's Avatar
 
Join Date: Oct 2010
Location: Southern, Ohio
Posts: 471
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 47
Quote:
Originally Posted by AustinJerry View Post

All this discussion on Waterfalls and bass response has me itching to try something new.  And it irritates me to no end that my friend Keith has managed to find room for additional treatments!  So, I have one remaining space in my listening room that I can use to add additional traps.




Note the space immediately below the flat panel, behind the center speaker.  I currently have a couple of old 2" ATS panels there to hide some cabling.  The ATS panels are cosmetic only and are not contributing to controlling bass ringing at all.

I have been investigating the Scopus Tuned Membrane bass trap .  These come in three flavors, tuned for 40Hz, 70Hz, or 100Hz, and are 24"x24"x4.25" (10.25" for the 40Hz model).  I am considering four panels, placed vertically on their edges, along the space below the flat panel.

I need opinions on which model to choose.  My Waterfall and Spectrogram graphs were posted earlier.  GIK is recommending the T-40 model.  However, if my focus is on music, which doesn't really have content down to 40Hz, wouldn't I be better off with the T-70?  Has anyone had experience with the Scopus trap, or tuned membrane traps in general?  And finally, based on my measurements, do I really need more traps (other than to keep up with my friends)?

We all know Keith's motto, "You can never have too much woofage".  Should there be another motto, "You can never have too much bass trapage"?

Jerry,

First of all I'm so proud of you!!!! I thought we'd never get past talking ONLY about broadband bass traps! biggrin.gif

Tuned traps are more complicated, more expensive to buy, harder to build, and harder to place, but absolutely do the job where most of the time it's not feasible to use pink fluffy.

Here's some information for you.

First, if you want a simple solution and have more money than time or interest I can absolutely recommend GIK.

However, if you like to tinker and want to spend maybe 5-10% of the cost and have the absolute minimal amount of tools and skill, you're in luck.

The same calculator I linked to above also models perforated panels/Helmholtz absorbers.

I took the liberty of creating two models for you to compare just so you can see what is possible.

Here's the list of things you need to make these.

2' x 4' pre-cut handy panels from Lowes/HD
2-by standard lumber from same place
Some 1x2 furring strips (scraps will even work) to keep the insulation in the spot you want it)
Glue/clamps
Drill/drill bit


That's it.

In your case you don't even need fabric or paint to cover these because you can just put your 2" panels in front of them for a perfect fit, great looks, and to still absorb the higher frequencies so you don't lose anything (re: absorption) you have now.

Look here at what you can accomplish. PERFECT absorption at 50hz (this is just an example as I figured since I was taking the time to make the model I might as well make it for my own one remaining problem frequency biggrin.gif )




Now for a few caveats...

First of all, Tuned traps such as these (whether you buy or build) are pressure based absorbers (the traps made from insulation are velocity based and do not have the following requirements) which means they MUST be at the pressure maximum to work.

What this means is simply if the mode you are trying to get rid of/tame is a length mode, then placing the traps on your front or back wall will work just fine, but if the mode you are trying to get rid of is a width mode, placing them on your front or back wall won't do much if anything.

So you first need to determine if you're stuck on placement on the front wall there. If so, then you can automatically discount building or buying a trap for a frequency that isn't length mode related.

These traps work great in corners, too, BTW.

Secondly, look at the two models. You'll note they are two different sizes (the width and length of the traps don't matter and I recommend 2' x 4' - NOTE: GIK's are only 2' x 2'!!!) regarding the depth and they each use a different thickness for the front panel. (The rear of these MUST be sealed - no air gaps - these don't work that way) and they both target the same frequency.

However, the larger one has a smaller Q which means it covers MORE frequencies and also means you don't have to be quite as precise about building it to get great results but if you do want to target just one frequency and either have pretty good DIY skills and/or don't mind possibly having to buy an extra 2'x4' handy panel for a few bucks if you mess up then you can absolutely build the smaller one with great results.

I'd recommend building 2 of the 2'x4' models or buying 4 of the GIK (BIG difference in cost) to see some great results. Less than that in your size room with your problems may not give you the results you want.

There is no reason you cannot build (or even buy, but I obviously recommend building or at least trying it out) different panels for different frequencies and take care of all your modes!

So remember, these are "a bit" more complicated (or very much more expensive) depending on which way you go but the results are better than you'd ever be able to do with any reasonably sized velocity based trap.

You just have to decide which modes are length modes if you must place them at the specified location so you get the results you want instead of building (or buying) some that target width modes then placing them in the wrong spot and being very disappointed. wink.gif

Hope this helps and really hope you'll be a guinea pig and show some folks here how you can actually FIX your problems with proper treatment!!!

It'll be so great when we can stop spending so much time and effort on talking about EQ!! (I don't mean in this thread - just in general. People spend so much effort, time, and expense on EQ and most of the time it only does more damage than good. If the same amount of expense, time, and effort were spent on "treating the problems" instead, there wouldn't be a need for it and options for other equipment and listening modes would suddenly open up expanding the possibilities of this hobby much more for many more people but I digress.

Hope this helps,

--J

Thanks to EVERYONE that Helps Make These Threads so Awesome!

jevansoh is offline  
post #5878 of 11702 Old 10-30-2013, 12:48 AM
AVS Special Member
 
jlpowell84's Avatar
 
Join Date: Dec 2012
Location: Eugene, OR
Posts: 4,079
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 177 Post(s)
Liked: 225
I just read that REW can upload its own suggested cuts to my mini dsp that way I don't do it manually and it does a spot on job. Anyone else do this or know about how to do it?

DIY Sound Group Volt 10 surround speaker build

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.

My Setup Thread

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
jlpowell84 is offline  
post #5879 of 11702 Old 10-30-2013, 12:54 AM - Thread Starter
Senior Member
 
jevansoh's Avatar
 
Join Date: Oct 2010
Location: Southern, Ohio
Posts: 471
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 47
Quote:
Originally Posted by jim19611961 View Post

Quote:
Originally Posted by jevansoh View Post


This hopefully will start a discussion that through questions and clarifications will get everyone on the same page, understanding what the actual goal is and why, how to interpret these charts and graphs, and eventually, what to "do" about it to actually fix these issues.

This cuts to the heart of the matter.

Here is my take:

Frequency response has subjective elements (absolutely flat isn't necessarily what sounds best), but I think we can agree it needs to be smooth and without sudden large deviations in terms of peaks and nulls.

For the first 20ms, the best sound seems to be where you hear the direct sound only (in the ideal). That means you hear no room modes & no reflections. I think that means you dont hear residual decay either. This is all in ideal terms in my perception. What one does with room energies AFTER 20ms I think deserves its own discussion. But right now I want to focus on the first 20ms.

Now, lets examine the tools at our disposal:

1) Waterfalls
2) Decay
3) Spectrograms
4) ETC
5) FR

Let me focus on 1-4 for the time being. Each of these show in various forms what is present in the room that is NOT the direct energy in addition to a representation of the direct. Each shows the data in a different view or perspective. Yet, none of them is the complete picture or the best tool depending on what your trying to focus on. And each has variable parameters.

So if better sound is the goal, lets look at what we are hearing.



This gives a general view of common instruments and their frequency ranges.

I would contend that a instrument is best heard when the room doesnt modify the direct sound (first 20ms). And what this chart brings to light to me is that most of what we hear in terms of fundamentals falls in the 100hz - 1000hz range. The harmonics are key as well, but frequencies above 1000hz are rather easy to control so the hard work lies below that.

So the question becomes: What tool or tools best illustrates or conveys whats happening in our room in the 100-1000hz range. Again, the frequencies above and below this are important. But I wanted to focus on those frequency ranges hardest to deal with and therefore least dealt with properly even in treated rooms.

I think Ill stop here and see what direction the rest of you are going before continuing.

I agree with almost all of what you're saying. In fact, in principal, I agree with all of it.

In practice, it isn't quite so easy to answer your question though.

Here's why. Let's assume you are right and only 100hz to 1000hz matters when it comes to fixing problems.

The problem is 100hz is in a different region than 1000hz.

If you were to say you wanted to concentrate on fixing 300hz to 1000hz I'd say use the ETC.

The problem is from 100hz to about 300hz you are in the modal or transition range and in that case you need to use the waterfall or spectogram (I use both).

The ETC is frequency blind and also skews its results to the higher end of the frequency response.

So if you want to concentrate on fixing problems in the 100hz to 1000hz range, here's what I would do.

I'd limit the measurement itself from whatever starting frequency (I believe it's VERY important to look at problems with slow decay below 100hz but in your case and for the sake of this conversation we'll say the starting frequency is 100hz) to 300hz. Then use the waterfall/spectogram to see problems. You'll only have 100hz to 300hz to work with but that's all that is necessary to use the waterfall/spectogram.

Then I'd take a different measurement and limit it from 300hz to 1000hz. Note, I'm not talking about the graph limits to zoom into this range, rather actually ONLY measuring 300hz - 1000hz and then use the ETC to isolate and treat your issues.

This way the results won't be skewed for either measurement and you will know you're treating the problems you wish to treat (and only looking at the problems you feel are in deed problems.)

Now, again, I agree with everything you wrote except only focusing on fixing problems in that limited frequency range but that's a different topic.

Hope this answers your question. If I totally missed the mark, let me know, as that's been known to happen before at 4:00 A.M. wink.gif

--J

Thanks to EVERYONE that Helps Make These Threads so Awesome!

jevansoh is offline  
post #5880 of 11702 Old 10-30-2013, 12:58 AM - Thread Starter
Senior Member
 
jevansoh's Avatar
 
Join Date: Oct 2010
Location: Southern, Ohio
Posts: 471
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 47
Quote:
Originally Posted by jkasanic View Post

Quote:
Originally Posted by kbarnes701 View Post

 
EDIT: here's the spectrogram of the same thing:




It isn't all that bad, is it?? It has a lot more red/orange in it than yours though, Jerry. I don't understand, looking at our two waterfalls, why that should be the case - anyone?

Looks pretty good to me based on what I've read about Spectrograms.  I think the red is just an indication of the dB output so measurements at higher SPL levels will show more red.

Exactly re the red. That's all it means and nothing to be concerned about. Just a hotter measurement. Only pay attention to the light blue to dark blue transition area and above and other than that the overall smoothness, horizontally, which is your average decay rate and the ultimate goal of the surrounding/problem frequencies.

Keith, looking good except that ONE spot. I honestly need pictures at this point to be able to make more recommendations although I will say briefly here (have a much more detailed post planned) that the improvements you did get, while a bit unexpected in some ways (although totally understandable now after the fact and after I thought about it for a few minutes) are more than just a minor improvement IMO and not too shabby at all!

Although the target problem wasn't resolved, we're on the right track.

I'll have more to say on this. Planned on it tonight but then other stuff in this forum came up and I've been posting for a few hours! smile.gif

At least I figured out how to put pictures in the post and now feel stupid since it couldn't be easier! BTW, is there a limit as to how many pictures I can put in one post or how many total pictures I can put in all my posts over time??

--J

Thanks to EVERYONE that Helps Make These Threads so Awesome!

jevansoh is offline  
Reply Audio theory, Setup and Chat

Tags
Dayton , Dayton Audio , Room Equilizer Wizard Rew

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off