or Connect
AVS › AVS Forum › Audio › Audio theory, Setup and Chat › Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How To Interpret Graphs
New Posts  All Forums:Forum Nav:

Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How... - Page 209

post #6241 of 9541
For the REW Gurus;
I have been tinkering with speaker/subwoofer placements & just went thru a before/after "Speaker placement - REW study" of my space. Many HT's seem to have some good news/bad news issues; I will start with what I hope is my good news-best result....

Are these waterfalls and waterfall settings OK ?
XT32 Off:


XT32 On:


To my understanding; it seems the results are slightly deteriorated with XT32 switched On?

Gear mentioned in this thread:

post #6242 of 9541
Quote:
Originally Posted by cuzed2 View Post

For the REW Gurus;
I have been tinkering with speaker/subwoofer placements & just went thru a before/after "Speaker placement - REW study" of my space. Many HT's seem to have some good news/bad news issues; I will start with what I hope is my good news-best result....

Are these waterfalls and waterfall settings OK ?
XT32 Off:


XT32 On:


To my understanding; it seems the results are slightly deteriorated with XT32 switched On?

Can you show the two pre/post Audyssey spectrograms (same data), but setting the range on the X axis to 45 to 105 db, instead of what seems like 45 to 110? The Y axis is reasonable at 15 to 300 Hz. And you may want to click on the frequency option on the upper right of the screen so that the settings are logarithmic (numbers oriented so that you have more emphasis on < 100 Hz than the even cuts you have on this chart) so that it's clearer what's happening in the lower bass range.
post #6243 of 9541
@ Cuz,

The measurement level is way too low to draw any conclusions from the waterfalls you provided. What level are you calibrating your mic to? IMO, it should be 90dB, or even 95dB of you can stand it that loud. You want at least a 40dB range for a waterfall, and yours looks more like 25dB

And follow Stuart's advices as well. Use the same measurement to produce a Spectrogram. It's our new go-to graph.
Edited by AustinJerry - 11/3/13 at 7:24pm
post #6244 of 9541
I was certain (or at leat thought) I was calibrating my mic (UMIK) and (including running my 4311's MV at 0) to 80db.

This was my 3rd go-round with REW, was hoping the 3rd time would be the charm ....

Oh well - lots of winter ahead to wear my REW dunce cap! That said; I do sppreciate everyone's help here
post #6245 of 9541
Quote:
Originally Posted by AustinJerry View Post

@ Cuz,

The measurement level is way too low to draw any conclusions from the waterfalls you provided. What level are you calibrating your mic to? IMO, it should be 90dB, or even 95dB of you can stand it that loud. You want at least a 40dB range for a waterfall, and yours looks more like 25dB

And follow Stuart's advices as well. Use the same measurement to produce a Ectrogram. It's our new go-to graph.

I have been thru the Guide several times - but have to ask, what is an "Ectrogram " ???

I really want to master this great tool called REW, but each time I get close the bar gets raised.
post #6246 of 9541

To refresh everyone's memory regarding Waterfall settings (excerpt from the Guide).  These are not my settings, they are Paul Spencer's.  We should all be consistent with our settings (time range is variable, of course).

 

 

 Set the control values as displayed below.  (Note:  These recommended settings are taken from a series of articles published here, titled Bass Integration Guide Parts 1-3, by Paul Spencer.  A highly recommended read.)

 

post #6247 of 9541
Quote:
Originally Posted by cuzed2 View Post

I was certain (or at leat thought) I was calibrating my mic (UMIK) and (including running my 4311's MV at 0) to 80db.

This was my 3rd go-round with REW, was hoping the 3rd time would be the charm ....

Oh well - lots of winter ahead to wear my REW dunce cap! That said; I do sppreciate everyone's help here

 

80dB is still too low.  I am in the process of updating the guide, and I'll recommend a higher setting.

 

I misspelled "Spectrogram".  I am adding a section for this graph as well.

post #6248 of 9541
Quote:
Originally Posted by AustinJerry View Post

80dB is still too low.  I am in the process of updating the guide, and I'll recommend a higher setting.

I misspelled "Spectrogram".  I am adding a section for this graph as well.

Something that the group should discuss is when a waterfall will be more useful for diagnostics than a spectrogram, above and beyond the FR plot and ETC that we already have in our tool kits as well. Based on the past few days, for looking at bass decay it's a slam-dunk that spectrograms communicate more useful information on one chart than a waterfall. You might want to make sure that the settings that are the defaults in the spectrogram are the ones we want to use, or add the ones that any experts are recommending.

BTW, I did approximately 85 db on my plots on Saturday and I had to plug my ears. Picking up noise-limiting protective devices for your ears is HIGHLY recommended for serious measurement above 85 db IMO (and depending on where you're sitting when measuring, IMO maybe 80 db). Hopefully there's an 85 db exception for special cases (urban living in our case or WAF), with 90 db as the lowest GAAP otherwise. That's an abbreviation for generally accepted acoustic practice on the thread, with apologies to accountants LOL. Yes, we've gotten that formal...rolleyes.gif

DK about you, Jerry, but I notice a fairly drastic difference between 75 and 85 db. Maybe I'm not deaf enough from listening to rock close up in my misspent "youth" (the 80's)....
Edited by sdrucker - 11/3/13 at 7:55pm
post #6249 of 9541

But if you're listening at 80dB, what's the point of measuring at 90dB?

post #6250 of 9541
I'm going on a road trip, but will revisit these efforts when I return with hearing protection at the ready, I really do appreciate everyone's help here ...

Tomorrow Nite (from a hotel), I might put up some more 80db sweeps for troubleshooting input, and my continued tutoring ...
post #6251 of 9541
Quote:
Originally Posted by LastButNotLeast View Post

But if you're listening at 80dB, what's the point of measuring at 90dB?

I raised that some point about a week ago on one of Keith's plots, because I didn't see the point of ear-splitting volume when we don't listen to a 95-100 db peak in practice. I don't think I ever listen at more than 80 db on average, and I wonder how many people here crank up the volume at more than that. However, I can see 85 db as a safeguard for measuring, though, because it's above what we normally listen to and fall into the +/- 5 db range from target on a FR curve that's deemed "acceptable compared to 80 db". But I think the depth of the measurable noise floor has more to do with the level of volume needed than what we listen to. Hence 85-90 db as a minimum (preferably 90)., with a USB mic noise floor of 50-55 db.
post #6252 of 9541
Quote:
Originally Posted by Alan P View Post

....does anyone have a link for a good room design program?

I've used Sweet Home 3d very happily.
post #6253 of 9541
Quote:
Originally Posted by AustinJerry View Post

Here are several questions that would help interpreting the graphs:

2.  Looking at 500Hz, 1000Hz, and 4000Hz, I seem to be seeing different reflections.  I recall from previous discussions that reflections that are present at higher frequencies, but missing at lower frequencies, are caused by smaller surfaces.  Is that correct, or is it the other way around?

IIRC, different materials can reflect different frequencies at different rates. Something like, if you have a leather chair vs a coffee table.
Quote:
Understanding the impulse response measurement has become very important, IMO.

This is one of my main objectives. Reflections wreak havoc on dialogue intelligibility.
post #6254 of 9541

Version 3.1 of the REW Guide has been posted.  Changes in this version are:

 

 

3.1

Nov 3, 2013

-          Added note 3 (p. 3) describing using custom mic calibration files.

-          Minor revisions to Waterfall Graph and Decay Graph sections, based on best practices since the original write-up.

-          Added new section on generating Spectrograms (p. 63)

 

Click on the link in my signature to access a link to download the revised Guide.  Please review the updated sections and provide feedback.

post #6255 of 9541
Quote:
Originally Posted by LastButNotLeast View Post
 

But if you're listening at 80dB, what's the point of measuring at 90dB?

 

It has been said quite a few times in this thread, the validity of a graph showing bass decay comes into question if there isn't at least 40dB headroom between the measurement level and the noise floor.  So, if you measure a noise floor of 50dB, you need to measure at 90dB.  If your noise floor is 40dB, then perhaps an 80dB measurement is OK, as long as you set the lower limit on your graphs to 40dB as well.  If you don't believe this recommendation, take two measurements without moving the mic, one at 80dB and one at 95dB, generate the waterfalls, and observe the differences.  The graph generated with the 80dB measurement will mask bass decay information important to decision-making.

 

And yes, Stuart, I purchased a box of inexpensive foam earplugs that I am now in the habit of wearing during REW sessions.

post #6256 of 9541
Quote:
Originally Posted by jkasanic View Post

First things first...get yourself some blackout shades for those windows! eek.gif If you find anything with good absorption properties then let me know as I've been looking for something to treat my sliding glass door on the rear wall that is dual purpose! wink.gif

As for the soffits, corners and ceilings are one thing but I think floor would be a tough sell for anyone especially considering WAF let alone the "shrinkage" factor (I guess it's not just cold water that causes this!). If you can treat the floor then I agree soffits all around would be very intrusive especially if you treat the ceilings this way. My only point was that if you're going to straddle panels at the ceiling boundaries, a square soffit is the best bang for your buck as far as I can tell.

Hey guys. Just discussed it with my Fiancé. 15x15 soffits will go in front RL corners on tv wall (the right will have a chunk missing due to the in wall AC). Here is the good part! She agreed 15x15 soffits all along every ceiling to wall junction!!! If...I don't do the floors. So there I will build some custom panels. 24 inch wide at a 45 degree angle will protrude a little too far into the room. So perhaps about a 16 inch. I can come out about 8 inches from the wall. Anyway good news on the big soffits up high! Can't wait to get it done and get some measurements!

On a side note I was reminded of clean bass when I put on my nice Denon headphones. It's been awhile since I put them on. I don't know all the technicalities but I don't think there is much decay in a high quality headphones.

So in this pic imagine big black micro suede pink fluffy filled 15x15 soffits all along the ceiling to wall. (Will cover a strip of the top of the two windows, about 7 inches of it). And the vertical corners minus a gap for the AC. I really hope this makes enormous gains in my waterfall and spectrogram wink.gif

Edited by jlpowell84 - 11/4/13 at 12:27am
post #6257 of 9541
Quote:
Originally Posted by jlpowell84 View Post

I don't know all the technicalities but I don't think there is much decay in a high quality headphones.
There's none. You're hearing the sound minus the room.
post #6258 of 9541
Quote:
Originally Posted by jlpowell84 View Post

24 inch wide at a 45 degree angle will protrude a little too far into the room. So perhaps about a 16 inch.

I wouldn't imagine it would need to straddle the corner with the center of the bass trap in center with where the wall meets the ceiling.
I guess this would be ideal, but I'm also sure if you adjusted the angle so that it was closer to the wall at the top of the bass trap where it's near the ceiling, having more of the bass trap down the wall would be better then a narrower bass trap that straddled the corner evenly.
post #6259 of 9541
Quote:
Originally Posted by sdurani View Post

There's none. You're hearing the sound minus the room.

Sanjay, what are your thoughts to my treatment approach? Obviously this is bass first, then reflections...
post #6260 of 9541
Quote:
Originally Posted by Audionut11 View Post

I wouldn't imagine it would need to straddle the corner with the center of the bass trap in center with where the wall meets the ceiling.
I guess this would be ideal, but I'm also sure if you adjusted the angle so that it was closer to the wall at the top of the bass trap where it's near the ceiling, having more of the bass trap down the wall would be better then a narrower bass trap that straddled the corner evenly.

No, I am putting those at the floor to wall junction as I specified above in my planned approach. 15x15 soffits along the floor is much to intrusive and I am quite happy with my wife to be (I gave up saying the engaged term due to constant corrections by Keith on my mis-spelling) being okay with those Big long squares all around up high smile.gif
post #6261 of 9541
Quote:
Originally Posted by sdurani View Post

There's none. You're hearing the sound minus the room.

That's nice because I can have a reference point to come back too. At least 30hz up that is. But I do often hear detail in them I don't catch in my room. But then again I have yet to get some high end speakers in there. JTR is going to have price increases Jan 1st. So I may make a push sooner than intended for my speaker upgrade. Then world class detail will become a reality!
post #6262 of 9541
Quote:
Originally Posted by LastButNotLeast View Post

But if you're listening at 80dB, what's the point of measuring at 90dB?

It doesn't matter as long as your far enough from the noise floor. I'd recommend about 20dB.
post #6263 of 9541
Quote:
Originally Posted by cuzed2 View Post

For the REW Gurus;
I have been tinkering with speaker/subwoofer placements & just went thru a before/after "Speaker placement - REW study" of my space. Many HT's seem to have some good news/bad news issues; I will start with what I hope is my good news-best result....

Are these waterfalls and waterfall settings OK ?
XT32 Off:


XT32 On:


To my understanding; it seems the results are slightly deteriorated with XT32 switched On?

 

Audyssey has pushed up the average SPL when attempting to flatten the frequency response. The natural corollary of that is that the bottom of the graph is also 'pushed up'.

 

I did think at first glance that you need to raise the SPL for the test measurement as it seems a little low on the initial graph.  The spectrogram will reveal more - can you post one of those of the same basic measurements?  Also, the dB scale of the graph needs to be in 5dB increments for consistency with everyone else.

post #6264 of 9541
Quote:
Originally Posted by AustinJerry View Post
 

To refresh everyone's memory regarding Waterfall settings (excerpt from the Guide).  These are not my settings, they are Paul Spencer's.  We should all be consistent with our settings (time range is variable, of course).

 

 

 Set the control values as displayed below.  (Note:  These recommended settings are taken from a series of articles published here, titled Bass Integration Guide Parts 1-3, by Paul Spencer.  A highly recommended read.)

 

 

Jerry - these are substantially different settings to the ones we have been using to date. For comparison purposes, the graphs below are the 'new way' and the 'old way'.

 

This is it with the settings you post above (other than this uses 450ms which I always use):

 

 

And this is the waterfall with the REW default settings:

 

 

Here are the before/after with the time range set to 300 - the biggest impact is the X, Y, Z changes.

 

 

 

Is it recommended now we change the graph presentation to the 'new way'?  Could someone who understands these things please explain the differences?  I agree it is important we all use the same presentation methods for ease of comparison and viewing.  Will these new settings be in an updated Guide?

post #6265 of 9541
Quote:
Originally Posted by LastButNotLeast View Post
 

But if you're listening at 80dB, what's the point of measuring at 90dB?

 

Measuring and listening aren't really connected in this way. The idea of measuring as loud as your ears/equipment will permit, and in any case at 85dB, is to create a bigger 'gap' between the noise floor and the peak measurement so that meaningful conclusions can be drawn.

 

85dB is movie Reference Level average, so it doesn't really seem all that loud to me TBH.

post #6266 of 9541
Quote:
Originally Posted by jlpowell84 View Post
 
 
On a side note I was reminded of clean bass when I put on my nice Denon headphones. It's been awhile since I put them on. I don't know all the technicalities but I don't think there is much decay in a high quality headphones.

 

 

Headphones remove the room 100% from what you are hearing.

 

EDIT: I really must learn to engage readahead mode... ;)

post #6267 of 9541
Quote:
Originally Posted by jlpowell84 View Post
 
Quote:
Originally Posted by Audionut11 View Post

I wouldn't imagine it would need to straddle the corner with the center of the bass trap in center with where the wall meets the ceiling.
I guess this would be ideal, but I'm also sure if you adjusted the angle so that it was closer to the wall at the top of the bass trap where it's near the ceiling, having more of the bass trap down the wall would be better then a narrower bass trap that straddled the corner evenly.

No, I am putting those at the floor to wall junction as I specified above in my planned approach. 15x15 soffits along the floor is much to intrusive and I am quite happy with my wife to be (I gave up saying the engaged term due to constant corrections by Keith on my mis-spelling) being okay with those Big long squares all around up high smile.gif

 

And now your fiancée is happy :)

post #6268 of 9541
Quote:
Originally Posted by AustinJerry View Post

It has been said quite a few times in this thread, the validity of a graph showing bass decay comes into question if there isn't at least 40dB headroom between the measurement level and the noise floor.  So, if you measure a noise floor of 50dB, you need to measure at 90dB.  If your noise floor is 40dB, then perhaps an 80dB measurement is OK, as long as you set the lower limit on your graphs to 40dB as well.  If you don't believe this recommendation, take two measurements without moving the mic, one at 80dB and one at 95dB, generate the waterfalls, and observe the differences.  The graph generated with the 80dB measurement will mask bass decay information important to decision-making.

And yes, Stuart, I purchased a box of inexpensive foam earplugs that I am now in the habit of wearing during REW sessions.

This might prove useful smile.gif :
http://www.gcaudio.com/resources/howtos/loudness.html

In context, even 95 db as part of the environment doesn't sound terribly loud, or problematic unless you're measuring for four hours a day LOL (there are probably a few people here who may do that on occasion).

However, note that the 85-95 db we're measuring at is measured at the MLP. Depending on where you're sitting or standing relative to the MLP, the actual db you hear when testing will differ . When in doubt it's smart to whip out the Radio Shack SPL and measure where you may be in the room at the time, consider a chart like the one I posted, and get those ear plugs when in doubt.
post #6269 of 9541
Quote:
Originally Posted by kbarnes701 View Post

Jerry - these are substantially different settings to the ones we have been using to date. For comparison purposes, the graphs below are the 'new way' and the 'old way'.


Is it recommended now we change the graph presentation to the 'new way'?  Could someone who understands these things please explain the differences?  I agree it is important we all use the same presentation methods for ease of comparison and viewing.  Will these new settings be in an updated Guide?

The settings are not different. These are the settings that were published in the guide from the beginning, and have always been different from the REW default settings that many people use, probably because they didn't realize the difference. Back when I was developing that section of the guide, I did some research into what settings to use. Following my research, I still didn't have a good understanding of what effect different settings had WRT the presentation of the data, or it's readability. I thought the referenced article was a good one, so I decided to use those settings. I don't think the actual settings are that important, but consistency of presentation here in the thread is important. How many times have we stared at a Waterfall graph, trying to figure out why it looks peculiar, only to realize it is using different settings?

So, no guide update necessary...
post #6270 of 9541
Quote:
Originally Posted by AustinJerry View Post
 
Quote:
Originally Posted by kbarnes701 View Post

Jerry - these are substantially different settings to the ones we have been using to date. For comparison purposes, the graphs below are the 'new way' and the 'old way'.


Is it recommended now we change the graph presentation to the 'new way'?  Could someone who understands these things please explain the differences?  I agree it is important we all use the same presentation methods for ease of comparison and viewing.  Will these new settings be in an updated Guide?

The settings are not different. These are the settings that were published in the guide from the beginning, and have always been different from the REW default settings that many people use, probably because they didn't realize the difference. Back when I was developing that section of the guide, I did some research into what settings to use. Following my research, I still didn't have a good understanding of what effect different settings had WRT the presentation of the data, or it's readability. I thought the referenced article was a good one, so I decided to use those settings. I don't think the actual settings are that important, but consistency of presentation here in the thread is important. How many times have we stared at a Waterfall graph, trying to figure out why it looks peculiar, only to realize it is using different settings?

So, no guide update necessary...

 

Thanks Jerry. I guess there isn't all that much difference then, because I seem to have always been using the REW defaults for some reason and I have never had a single comment that my waterfalls were 'non-standard'. I have gone into REW and set the parameters in accordance with your suggestions now and will use these for future posts. In fact, they make my waterfalls looks slightly better, so that makes me feel good too LOL.

 

Here is my current waterfall using the recommended settings, but with my customary 450ms time window.

 


Edited by kbarnes701 - 11/4/13 at 6:23am
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Audio theory, Setup and Chat

Gear mentioned in this thread:

AVS › AVS Forum › Audio › Audio theory, Setup and Chat › Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How To Interpret Graphs