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 77

post #2281 of 9491
hello
I have been trying the EQ panel.
In my result, I get original response and predicted response in similar blu colours, which are difficult to distinguish.
Is there a way to change colours. I know how to do change colurs in the measurements panel. But how to do it in the EQ panel?

Gear mentioned in this thread:

post #2282 of 9491
Quote:
Originally Posted by LastButNotLeast View Post

How's this?

 

 

 

Start, obviously, with your graph in REW at the limits you want. Click on "Decay," deselect all but the first and 160ms boxes. If necessary, go to "Controls" and select "Fill slices" (default is red and blue). Close "Controls," capture image, save as "decay."

 

Click "Overlays." "Controls" icon on top right, "Offset" -20, click on "Add offset to data." Close "Controls" box, capture image, save as "overlay."

 

I used Photoshop Elements 8. Open decay and overlay files. Check to make sure they are the same size (sometimes they are, sometimes they aren't; go figure). On the overlay file, select the magic wand tool. I found tolerance = 12 works well; deselect antialias and contiguous. Click on the line that makes up your graph. If too much other stuff is selected, lower the tolerance. Edit -> Copy.

 

Switch to the decay graph. Select -> All. Edit -> Paste Into Selection. Move it ("nudge" in PS parlance) so the left and right endpoints of your graph are where they belong. Edit -> Stroke (Outline) Selection: 2 px is fine. Double click on the color box, change R, G and B to 255. OK.

 

Save your masterpiece. Post it here for general admiration, then get to work on the areas where the decay isn't below the 20dB line.

 

Piece of cake, right?

Is this still the "Simplified REW" thread?

 

biggrin.gif

 

Michael

 

Michael, do you think it would be easier just to import your graph in PS, trace your top curve, make it a selection on a new layer, offset by 20 dB (eyeball it with nudge) and just stroke it a few pixels in white from the middle? Tracing a path might sound harder but if you've ever "cut" objects out of an image (like a title treatment) then I've found paths is the best way to go for creating your selection.  Magic wand has improved quite a bit so for this exercise it probably would do the trick. I'll experiment some tomorrow as I plan to be taking some new measurements now that I've finally got my sub relocated.

post #2283 of 9491

That is pretty creative, Michael!

post #2284 of 9491
Quote:
Originally Posted by kbarnes701 View Post

Two subs is always much better than one. Most of us here run at least two subs. Room size isn't an issue (I run two Submersive F2s in a room of less than 1,000 cubic feet). The point is that all rooms have modes and two subs helps deal with them much better than one. You seem to have a very difficult, room-induced null centred at about 55Hz. This will be audible IMO. You cannot use EQ to combat a null because it is caused by the way the soundwave physically interacts with the room. You can do a lot by moving the single sub around as I suggested earlier and finding the place where the response is the smoothest. You already did that to some extent and that mid-wall positon helped smooth the response a lot, but did nothing for that null. You can experiment with other sub locations to see if you can do better. If you can't then a second sub is the best way to solve it. You cannot EQ a true null - think of it like pushing both your hands together - pushing harder makes no difference - it just expends more energy to no effect.

I am not familiar with the 'calculation' feature of REW that you mention. But I just tried it with my Sub 1 and Sub 2 graphs using the procedure you describe and the result is nothing like the combined response of my two subs, as you can see below (blue is actual, green is 'calculated'):

So I wouldn't rely on that!  It is highly possible that that REW feature is not intended for the purpose you are using it for. IDK.

I wanted to know more if it is possible to predict 2sub performance from single sub measurements.
So made a post on the hometheatershack website.
http://www.hometheatershack.com/forums/rew-forum/66619-can-rew-predict-2sub-performance-single-sub-measurements.html#post602295
From the answer I understand that it is not possible with a USB mic. Bad luck for this thread!
I do not understand the details and the reasons, but I hope to find out more. Is there anyone who can explain why is not possible with a USB mic, and how it works with a non-USB mic?
Edited by turboman123 - 3/30/13 at 7:53am
post #2285 of 9491
Quote:
Originally Posted by jkasanic View Post

Quote:
Originally Posted by LastButNotLeast View Post

How's this?

 

 

 

Start, obviously, with your graph in REW at the limits you want. Click on "Decay," deselect all but the first and 160ms boxes. If necessary, go to "Controls" and select "Fill slices" (default is red and blue). Close "Controls," capture image, save as "decay."

 

Click "Overlays." "Controls" icon on top right, "Offset" -20, click on "Add offset to data." Close "Controls" box, capture image, save as "overlay."

 

I used Photoshop Elements 8. Open decay and overlay files. Check to make sure they are the same size (sometimes they are, sometimes they aren't; go figure). On the overlay file, select the magic wand tool. I found tolerance = 12 works well; deselect antialias and contiguous. Click on the line that makes up your graph. If too much other stuff is selected, lower the tolerance. Edit -> Copy.

 

Switch to the decay graph. Select -> All. Edit -> Paste Into Selection. Move it ("nudge" in PS parlance) so the left and right endpoints of your graph are where they belong. Edit -> Stroke (Outline) Selection: 2 px is fine. Double click on the color box, change R, G and B to 255. OK.

 

Save your masterpiece. Post it here for general admiration, then get to work on the areas where the decay isn't below the 20dB line.

 

Piece of cake, right?

Is this still the "Simplified REW" thread?

 

biggrin.gif

 

Michael

 

Michael, do you think it would be easier just to import your graph in PS, trace your top curve, make it a selection on a new layer, offset by 20 dB (eyeball it with nudge) and just stroke it a few pixels in white from the middle? Tracing a path might sound harder but if you've ever "cut" objects out of an image (like a title treatment) then I've found paths is the best way to go for creating your selection.  Magic wand has improved quite a bit so for this exercise it probably would do the trick. I'll experiment some tomorrow as I plan to be taking some new measurements now that I've finally got my sub relocated.

 

Ok, so I tried my method using my plot from a couple of posts before yours.  Probably splitting hairs as it took me about 3-4 min.  One note for anyone trying this at home, you'll need to stroke the selection first and then nudge as you can't nudge an empty selection!

 

 

Off to do some measuring...I can only hope this plot is a lot flatter with the new sub location!

post #2286 of 9491
Quote:
Originally Posted by jkasanic View Post

 

Off to do some measuring...I can only hope this plot is a lot flatter with the new sub location!

 

I hope so too, but if not, don't be discouraged.  I have been working on my bass response off and on for the last several years, and there is still room for improvement.

post #2287 of 9491
Quote:
Originally Posted by amirm View Post

I wouldn't be able to tell if IR offset is performed but assuming so, yes, it looks right. You can for example see what happens on dips. The sudden change in excess phases says they are not correctable by eq.

It's with the IR offset. The graph is actually an old measurement showing a few dips in the low region. Jerry used a mode cal chart to show me I was sitting in a null, but I wanted to verify using the excess phase graph. Great information guys.

BTW, I finally found the perfect spot for my subs, and it's not the most ideal location, because I like keeping'em out'a sight. It's right next to my mains!!!

I'm getting some badass upper bass from my two tiny subs. None of this room shaking stuff. Just some solid, don't-know-where-it's-coming-from bass that has me totally stoked. It's so good, I've decided to put in my order for two 15" Hsu subs I've had my eyes on.
post #2288 of 9491
Quote:
Originally Posted by Pres2play View Post


It's with the IR offset. The graph is actually an old measurement showing a few dips in the low region. Jerry used a mode cal chart to show me I was sitting in a null, but I wanted to verify using the excess phase graph. Great information guys.

BTW, I finally found the perfect spot for my subs, and it's not the most ideal location, because I like keeping'em out'a sight. It's right next to my mains!!!

I'm getting some badass upper bass from my two tiny subs. None of this room shaking stuff. Just some solid, don't-know-where-it's-coming-from bass that has me totally stoked. It's so good, I've decided to put in my order for two 15" Hsu subs I've had my eyes on.

 

Are you going to post some new measurements?

 

Which Hsu's?  I have ULS-15's (4).

post #2289 of 9491
I'll post later today, I'm finishing up a few more sound panels.

Frames with just batting installed... I'm using OC 705.



I'm ordering the front-firing VTF-15H. How's the ULS-15?
post #2290 of 9491

I like them.  I prefer sealed subs.  From what I hear, the VTF-15 is a nice sub as well.

 

BTW, with your subs next to the mains, are they by any chance at or near the 1/4 and 3/4 points on the front wall?  I found these spots give me the best response.

post #2291 of 9491
Quote:
Originally Posted by jkasanic View Post

 

Ok, so I tried my method using my plot from a couple of posts before yours.  Probably splitting hairs as it took me about 3-4 min.  One note for anyone trying this at home, you'll need to stroke the selection first and then nudge as you can't nudge an empty selection!

 

Off to do some measuring...I can only hope this plot is a lot flatter with the new sub location!

Nice. Whatever works. I do think it provides a lot of useful information.

You have seen the suggestion to put the sub in the main listening position and then take measurements at sub level at different places around the room? Unless you only have a couple of possible sub locations, that may be easier.

post #2292 of 9491
They're at 6 feet and 12 feet, and the front wall is 18.5 feet, so that puts them at roughly 1/3 and 2/3 points.

My last Hsu was also ported but down-firing, which doesn't work as well in my room.
post #2293 of 9491

Quote:

Originally Posted by AustinJerry View Post

Quote:
Originally Posted by jkasanic View Post

 

Off to do some measuring...I can only hope this plot is a lot flatter with the new sub location!

 

I hope so too, but if not, don't be discouraged.  I have been working on my bass response off and on for the last several years, and there is still room for improvement.

 

Well Jerry, today was my day of reckoning! biggrin.gif Holy crap...let me start off by saying I thought I was going to have a nervous breakdown!  I had taken some measurements yesterday with my AT screen down just to make sure I was still directionally correct with moving the sub.  Graphs looked better but I didn't bother to run Audyssey.  Today, with the screen in place, I was getting some CRAZY weird measurements.  Higher dB output, massive dropoff above 10 kHz etc.  Then I noticed my noise floor was closer to 30 dB with the UMM-6?! confused.gif  I must've tried calibrating SPL at least 5 times until finally, I shutdown REW and restarted everything from scratch.  I went through the guide like I was measuring for the very first time!  Wow, all I can say is what a relief!  Anybody that starts noticing unusual measurements, do yourself a favor and start from scratch!

 

Here's the full range FR L+R+Sub with (blue) and without (green) Audyssey as well as with a sub distance tweak (red):

 

 

As you may recall, I'm only running one sub which I relocated from the right corner to the mid front wall.  Audyssey initially set my XO's to 80 for C, 90 for L/R and 40 Hz for surrounds.  I bumped the surrounds to 90Hz.  There was a big dip in the XO region so I gave the sub distance tweak a try (not having much luck with it in the right front corner, expectations were low).  Anyway, here's a closeup of the XO region with the initial C and Sub only with Audyssey measurements plus their combined response as well as a few tweaks I tried.  Audyssey initially set the sub at 30' so I tried 20', 15' and 17.6' in that order:

 

 

Much to my surprise, the tweak actually improved the nasty dip at the XO to keep my overall FR within +/- 5dB with no room treatments.  Having come from the green plot to the red plot below I was very encouraged:

 

 

Here's the L+R+Sub with Audyssey Waterfall and Decay plots:

 

 

 

 

Both show considerable improvement over where I started:

 

 

 

A HUGE thanks to this thread but especially to you, Keith and Sanjay for all of your help getting me to this point!  I certainly wouldn't have gotten here on my own that's for sure!  I think I'm going to start enjoying my new home theater for a while before I start considering making any further modifications...at least for a week or so anyway! cool.gif

post #2294 of 9491

I agree, the results now are much better.  Congratulations!  It's nice that the distance tweak resulting in a nice improvement.  I would say you have earned some time off to enjoy...

post #2295 of 9491

^^^^Thanks Jerry!  I was hoping it wasn't just me being biased!  I've been holding off watching Skyfall since I've heard such great things about it (although not sure about the audio?!).  Supposed to rain tomorrow so I may have a date with the theater!  I'll be interested to hear your thoughts on where my focus should be once I've had some time to actually listen to a relatively flatter base response?

post #2296 of 9491
Quote:
Originally Posted by jkasanic View Post

Much to my surprise, the tweak actually improved the nasty dip at the XO to keep my overall FR within +/- 5dB with no room treatments.
Your subwoofer to speaker blend is all but invisible. What a smooth transition! Remarkable improvement.
post #2297 of 9491
Quote:
Originally Posted by sdurani View Post

Quote:
Originally Posted by jkasanic View Post

Much to my surprise, the tweak actually improved the nasty dip at the XO to keep my overall FR within +/- 5dB with no room treatments.
Your subwoofer to speaker blend is all but invisible. What a smooth transition! Remarkable improvement.

I know...couldn't believe the sub distance tweak had such a profound influence! Thanks again Sanjay...after all, it was your interpretation of my standing waves plot that lead me to try the sub in the mid front wall location!
Edited by jkasanic - 3/30/13 at 6:30pm
post #2298 of 9491
Just battled with REW and my new UMIK USB mic for two hours. Kept getting an REW error message saying my soundcard is not sending "input data". ARGH!!!

Will try a different laptop next week..
post #2299 of 9491
Quote:
Originally Posted by AustinJerry View Post

Are you going to post some new measurements?

Which Hsu's?  I have ULS-15's (4).


Here's a graph showing the improved FR. Man, the more and more you measure the clearer it gets just how hard it is to deal with a null. Raising the volume just doesn't work; it only raises the peaks all around the dips. After a few dozen measurements you really learn to appreciate REW.



Edit: I'll try a new measurement tomorrow with my Anti-Mode. Maybe it will settle the peak at 108. I'm not expecting much change. Maybe if I reset the XO above 100.
Edited by Pres2play - 3/30/13 at 9:58pm
post #2300 of 9491
Anyone in the SF bay area have REW and wouldn't mind showing me how it works thanks.
post #2301 of 9491
Hi, jsil. That's what this thread is all about. To get started, all you need is a mic, like the UMM-6, and Jerry's Step-by-Step Guide. You can order the UMM-6 at PartsExpress.com or CrossSpectrumLabs.com. If you order from CrossSpectrumLabs, they will calibrate the mic, so that it reads more accurately.
post #2302 of 9491
Sorry to barge in, but does this guide work with the Dayton UMM-6 mic as well? I heard the UMIK mics are unreliable for under 20 Hz over at the Shack so I decided to order the Dayton mic instead. Just want to know if there are any big changes to the set up procedure or if the current guide is good enough. Thanks!
post #2303 of 9491

^^^^Welcome Heinrich!  Only change is you'll need to do an spl calibration of the mic.  If you have an spl meter, it takes all of a minute and is detailed in the setup guide linked in AustinJerry's sig.  If you don't have an spl, download a free app for your phone or tablet to get you started.

post #2304 of 9491
Hi jkasanic! I have a Galaxy CM-140 SPL meter calibrated by Cross Spectrum. So you say the SPL calibration is simple? I'm all for simple. smile.gif So I only have to do it once? If I decide to go to a friends house to measure his room, I don't have to do it again?

I haven't gone through this entire thread but I'll definitely look at the guide. When I have more time I'll definitely read through the pages, but 77 pages is a lot. eek.gif What is your opinion on the UMIK vs the Dayton? Is it really as bad as what people are saying? They say the readings are inconsistent with the UMIK.
post #2305 of 9491

I have the UMM-6 so I can't speak for the UMIK-1.  If you read a couple posts earlier then you'll see I had some issues with REW and my configuration that was giving me incorrect measurements.  Save yourself a lot of time and trouble and read Jerry's guide.  There's a lot of pictures so it reads pretty quick.  You should only have to calibrate the mic once but like I said, REW was acting up for me so I went through the whole setup again.  Takes about 5 min including the SPL calibration.  Just take your SPL meter with you when you go to your friend's house.  Post any issues you have here or show us your plots.  Be sure to read the guidelines for posting measurements (in the guide).  Good luck!

post #2306 of 9491
What problems were you experiencing and how did you resolve them? Why did they occur in the first place?

Edit : sorry, I'll read above.
post #2307 of 9491

They weren't related to the UMM-6 per se.  Full disclosure - my laptop battery died with REW running so even though I powered up clean, I think something weird happened with Windows and REW playing nice.  Chalk it up to OE...

post #2308 of 9491
Quote:
Originally Posted by Pres2play View Post

Quote:
Originally Posted by AustinJerry View Post

Are you going to post some new measurements?

Which Hsu's?  I have ULS-15's (4).


Here's a graph showing the improved FR. Man, the more and more you measure the clearer it gets just how hard it is to deal with a null. Raising the volume just doesn't work; it only raises the peaks all around the dips. After a few dozen measurements you really learn to appreciate REW.



Edit: I'll try a new measurement tomorrow with my Anti-Mode. Maybe it will settle the peak at 108. I'm not expecting much change. Maybe if I reset the XO above 100.

Sorry if I missed it but are you using Audyssey? Looks like you may have some issues at your XO.  Instead of changing the XO, maybe try the sub distance tweak? I just had some very good results in my room with a similar null.  It's a little more tricky with 2 subs but still worth a try IMHO.

post #2309 of 9491
Quote:
Originally Posted by Heinrich S View Post

Hi jkasanic! I have a Galaxy CM-140 SPL meter calibrated by Cross Spectrum. So you say the SPL calibration is simple? I'm all for simple. smile.gif So I only have to do it once? If I decide to go to a friends house to measure his room, I don't have to do it again?

 

The microphone calibration step is designed to determine what setting on your AVR master volume results in a specific sound level (e.g. 80dB), as measured by your mic at the primary listening spot.  If you take your mic to your friend's house, there is no guarantee that the same MV setting on his AVR will produce the same sound level, so you will meet to run the calibration again.  

 

However, for subsequent uses of REW for your own setup, you don't need to calibrate the mic again, as long as you remember the MV setting from the first calibration.  For example, I know that a MV setting of -14 on my AVR results in an 80dB sound level, so I don't need to re-calibrate every time.

post #2310 of 9491
Quote:
Originally Posted by AustinJerry 
If you take your mic to your friend's house, there is no guarantee that the same MV setting on his AVR will produce the same sound level, so you will meet to run the calibration again.

So it's normal? So if you go to your friend and want to measure his room, you'll have to calibrate the SPL levels. The next person, the same thing...etc etc? If it's just for measuring the room, how important is the SPL calibration?

I think I'll take a look at your guide! Everyone I know has given you a big thumbs up.
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