Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How To Interpret Graphs - Page 350 - AVS Forum
Forum Jump: 
 159Likes
Reply
 
Thread Tools
post #10471 of 11856 Old 06-17-2014, 01:25 PM
AVS Special Member
 
Alan P's Avatar
 
Join Date: Jan 2003
Location: Rapid City, South Dakota
Posts: 2,380
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 474 Post(s)
Liked: 250
Quote:
Originally Posted by asere View Post
Thanks for the link on the review. The sub is really clean and offers very tight bass and has the slam for movies even in my 5400cf space. I mean when watching Tron Legacy for example I could feel the couch vibration.
Ken makes them for clean sound not really for higher spls and the 3D pulsar is for multiple subs.
Rolling off at 30hz based on the review and based on what I see with my graphs is disappointing at least on paper.
Here is another review graph for the 12012 and also the complete sub review.
As you can see the sub extends really low on the graph. I think in my case is the room size, mode and maybe better placement that I am still working on. Sounds really nice now like I said but now that I have REW I need to see it roll off much later especially if it is capable from the review.

http://www.hometheatershack.com/foru...er-review.html
OK, so after reading that review and the discussion thread I can only conclude that the numbers I posted earlier from the Sound & Vision review must be in error. The DXD-12012 definitely looks to be a solid performer...although it should be for the almost $3K price tag. For comparisons sake, that's what I paid for FOUR PSA XS15s.

To be fair though, the review you linked to was for the "Duo" (2 sub) configuration.

I wish I could find some more real-world measurements for that sub but I can't find any besides yours and the one you posted.

I'm anxious to see what you can do with proper calibration/placement/EQ....should be some impressive graphs in the end!
Alan P is online now  
Sponsored Links
Advertisement
 
post #10472 of 11856 Old 06-17-2014, 01:47 PM
Advanced Member
 
asere's Avatar
 
Join Date: Dec 2011
Location: Texas
Posts: 800
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 29 Post(s)
Liked: 23
Quote:
Originally Posted by Alan P View Post
OK, so after reading that review and the discussion thread I can only conclude that the numbers I posted earlier from the Sound & Vision review must be in error. The DXD-12012 definitely looks to be a solid performer...although it should be for the almost $3K price tag. For comparisons sake, that's what I paid for FOUR PSA XS15s.

To be fair though, the review you linked to was for the "Duo" (2 sub) configuration.

I wish I could find some more real-world measurements for that sub but I can't find any besides yours and the one you posted.

I'm anxious to see what you can do with proper calibration/placement/EQ....should be some impressive graphs in the end!
I hope so. I will get the mic boom
WOW 4 PSA'S I am sure your home needs foundation work

Last edited by asere; 06-17-2014 at 02:26 PM.
asere is online now  
post #10473 of 11856 Old 06-18-2014, 11:28 AM
Newbie
 
Rheagar's Avatar
 
Join Date: Apr 2013
Posts: 6
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 10
Quote:
Originally Posted by Selden Ball View Post
Your description is consistent with your computer using a screen resolution which is not compatible with the AVR. Most AVRs will pass only the standard home entertainment video resolutions. E.g. 1920x1080, 1280x720, 720x480 & 640x480 or the PAL equivalents for the latter two.

BIOS and startup screens are 640x480, so they're fine. However, if your computer's default screen resolution is, for example, 1366x728, that can't get through.

Most video cards support a second video output which can be set to any resolution, so that might be an option for you.
Thanks for the tips. Your suggestion lead me to eventually uninstall the graphics drivers, run a registry clean-up, and then re-install the drivers. This did the trick and my hdmi port is now functioning as it should.

Thanks to all for your help. Now the real work begins!
Rheagar is offline  
post #10474 of 11856 Old 06-18-2014, 03:42 PM
AVS Special Member
 
djbluemax1's Avatar
 
Join Date: Oct 2004
Location: MI
Posts: 2,261
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 275
Quote:
Originally Posted by 3ll3d00d View Post
All that review shows is room gain, it is a bit of a daft comment from the author to say it extends to 4.5Hz. You may well need 4 of them to get the extension you appear to be after.
Yep, ESPECIALLY when you consider that the measurement was taken at 75db. As I mentioned my 400 watt Bag End subs played flat to 7hz at 75db, but it wasn't flat when you turned it up any higher than that. You need a lot of displacement and/or a lot of power to produce high SPLs in the lowest octaves which is why MANY subs can't do much of anything below 30Hz.


Max
djbluemax1 is offline  
post #10475 of 11856 Old 06-18-2014, 05:48 PM
Senior Member
 
KevinG's Avatar
 
Join Date: Apr 2002
Location: S. Jersey
Posts: 462
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 30 Post(s)
Liked: 14
Quote:
Originally Posted by Audionut11 View Post
It's all about the ETC.

What makes sense for the measurements, depends on how willing you are to understand the effects the panels have. If you run a measurement now, you can at least confirm your placement of the panels.

The general consensus is to have all unwanted reflections at -20dB as soon as possible in the time domain.
Here's my mdat file.

https://drive.google.com/file/d/0B-G...it?usp=sharing

I think it is showing me that my panels are doing very very little. I'm studying my left speaker.

There seems to be an "echo" at 2.97ms (-4db), and another at 6.57ms (-16db).

The panel seems to be doing an excellent job BETWEEN those two spikes. It brings the broad spike at 3.95ms down from -13.14db to -21.24db, and the broad spike at 5.35ms from -19.45db to -22.9db. Is this what I should expect from a panel so close to the left front speaker?

What distance should I be looking at to figure out what is causing the spike at 29.67ms (-6.4db)? And, maybe more importantly, the one at 2.88ms (-5.24db)?

I'm not below -20 until 67.77ms, so I've got plenty of work to do. :-)

Thanks for any and all assistance!
-Kevin
KevinG is offline  
post #10476 of 11856 Old 06-19-2014, 12:07 AM
Senior Member
 
Audionut11's Avatar
 
Join Date: May 2008
Posts: 354
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 52
http://www.sengpielaudio.com/calculator-soundpath.htm

So if we enter 2.88ms, the result is 98.78cm (3.241ft). This is the distance over and above the source signal. So if your speaker to listening position is 384cm, you need to add 98.78cm to find the total distance path of the reflection. In this example, 482.78cm.

@jim19611961 mentions,

Quote:
Originally Posted by jim19611961 View Post
The size of the room (psychoacoustically) is determined mostly by the first significant reflection
So I think it's wise to work on these early reflections first, and be happy with any effects to the later reflections as a result.

Here is what I would do. Move that large round (sub?) object in front of the left speaker, and measure again. This can probably be moved pretty easily, all things considered, so lets check what effect it is having on your results.

I would then try moving the speakers (L+R) a little more towards the listening position. Depending on your speaker-speaker-listening position triangle, this may mean moving the speakers a little closer together, or simply moving the speakers directly towards the listening position, maintaining the existing triangle. Air is an insulator, so as you decrease the distance from speaker to listening position, you increase its amplitude, while decreasing the amplitude of the reflections, and, moving the reflections further away in the time domain.

Next, I would remove the panels, and just leave 1 leaning up against the wall near the left speaker. Measure, move the panel a little, measure, move the panel, measure. Rinse and repeat. Concentrate on the early reflections (1-5ms). You have a fairly live room, so the overall levels of the reflections, won't be reduced until you get all of the intended panels installed.

The panels as installed now, hit that spot at 3.95ms really well. Pay around with placement to try and hit that one at 2.97ms also.
Audionut11 is offline  
post #10477 of 11856 Old 06-19-2014, 04:06 AM
Senior Member
 
KevinG's Avatar
 
Join Date: Apr 2002
Location: S. Jersey
Posts: 462
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 30 Post(s)
Liked: 14
Quote:
Originally Posted by Audionut11 View Post
The panels as installed now, hit that spot at 3.95ms really well. Pay around with placement to try and hit that one at 2.97ms also.
Excellent. I will be playing around some more tonight. Thanks for the assistance!

Any chance the one at 2.97ms is coming from the floor?
KevinG is offline  
post #10478 of 11856 Old 06-19-2014, 04:20 AM
Senior Member
 
Audionut11's Avatar
 
Join Date: May 2008
Posts: 354
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 52
I very much doubt it. Have a look at sliced ETC, that reflection is 4000-8000khz, and your floor is carpeted.
Judging from your screenshots, at a rough guess, I would expect your ceiling to be one (or more) of those spikes at 13-16ms, so that won't be it either.

So I'd say it's either the wall, or the microphone boom, if you are using one.


edit: Feel free to place one of your panels on the floor, and measure it.

Last edited by Audionut11; 06-19-2014 at 04:39 AM.
Audionut11 is offline  
post #10479 of 11856 Old 06-19-2014, 06:38 AM
Senior Member
 
KevinG's Avatar
 
Join Date: Apr 2002
Location: S. Jersey
Posts: 462
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 30 Post(s)
Liked: 14
Quote:
Originally Posted by Audionut11 View Post
I very much doubt it. Have a look at sliced ETC, that reflection is 4000-8000khz, and your floor is carpeted.
Judging from your screenshots, at a rough guess, I would expect your ceiling to be one (or more) of those spikes at 13-16ms, so that won't be it either.

So I'd say it's either the wall, or the microphone boom, if you are using one.


edit: Feel free to place one of your panels on the floor, and measure it.
I looked around a bit, and couldn't see how to look at "sliced ETC", so I'll go off and google that in a bit. Knowing the frequency should help. The floor is carpeted, but it is rubber padding (for moisture resistance) over concrete, so I wasn't sure how effective it would be at absorption.

I was planning to start building my next three wall treatments, but I think I'm going to put that idea aside, and build a ceiling panel next. That should allow me to focus all of my attention on one speaker and get an idea of what is needed for the rest. I plan to make the ceiling trap twice as thick (two sheets of 3" roxul afb) and hang it two inches down from the ceiling (though, I still haven't figured out the best way to mount them like that). Also, unlike the wall traps, I plan to leave the sides (edges) of the ceiling traps open.

I am using a microphone boom, but I can't think of how this would introduce a peek like this. I'll get a pic of my mic position/stand and post it this evening.

I'll also lie one of the panels on the floor and get some measurements. It will be a great way to see if my floor is contributing much as a FRP.

Re: your post a couple back and moving my front speakers. I can't move them tighter together without first moving them a considerable distance out into the room (the screen fits just between them when in the down position). What is it that I am trying to achieve by moving, aiming them? Anything I can check with REW? Also, yes, that big black cylinder is my SVS PC13 Ultra. That was previously determined to be the best position in the room (of the limited options) by using REW. I can and will move it to see how it is effecting the ECT of the front left speaker, but it will probably have to be moved back to that position anyway.

More pics of the room to get an idea of the screen/speaker interaction:
http://imgur.com/a/YOLoo
KevinG is offline  
post #10480 of 11856 Old 06-19-2014, 07:51 AM
Senior Member
 
Audionut11's Avatar
 
Join Date: May 2008
Posts: 354
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 52
Sliced ETC is in the bottom left hand corner of the "filtered IR" graph.

http://i.imgur.com/cQj0V40.png

I'll answer your other questions shortly.
Audionut11 is offline  
post #10481 of 11856 Old 06-19-2014, 08:30 AM
Senior Member
 
Audionut11's Avatar
 
Join Date: May 2008
Posts: 354
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 52
Quote:
Originally Posted by KevinG View Post
I was planning to start building my next three wall treatments, but I think I'm going to put that idea aside, and build a ceiling panel next. That should allow me to focus all of my attention on one speaker and get an idea of what is needed for the rest. I plan to make the ceiling trap twice as thick (two sheets of 3" roxul afb) and hang it two inches down from the ceiling (though, I still haven't figured out the best way to mount them like that). Also, unlike the wall traps, I plan to leave the sides (edges) of the ceiling traps open.
Good idea. If you have the time, inclination, and funds, consider making it rather large. This will ensure you capture all reflections.
Have you purchased the roxul yet? Roxul is great where space is limited, however, when you have space for 6" or more, pink fluffy will perform better.

See here.


Quote:
Originally Posted by KevinG View Post
I am using a microphone boom, but I can't think of how this would introduce a peek like this.
It is a reflective surface.


Quote:
Originally Posted by KevinG View Post
I'll also lie one of the panels on the floor and get some measurements. It will be a great way to see if my floor is contributing much as a FRP.
Indeed. Of course, having absorption on the floor, full time, is not an easy task. But it will show you which reflections are related to the floor, so you're not chasing those reflections later on.


Quote:
Originally Posted by KevinG View Post
Re: your post a couple back and moving my front speakers. I can't move them tighter together without first moving them a considerable distance out into the room (the screen fits just between them when in the down position). What is it that I am trying to achieve by moving, aiming them?
Reflections are a relationship between the direct source signal, and the reflections themselves. As you move the source signal closer to the listening position (LP), you reduce the level of all reflections (all other things being equal). You also push the reflections further away in the time domain (increase spaciousness).

So you can use panels to target specific reflections, whereas, you can also place an effect on all reflection simultaneously with speaker/LP changes. My next suggestion would be to move the couch closer to the screen. Get the couch in front of that pylon, speaking of which, I would wrap that pylon in absorption material also. You will need to determine your speaker-speaker-LP triangle, whether you can move those side speakers appropriately, the effect of placement with regards to the other couch etc.

Try running a measurement with the microphone a foot or so in front of that pylon (towards the screen), and see if the differences are worthwhile.


Bear in mind, I am the type of person who places sound quality above everything else. Ergonomics/WAF are words that hold no meaning to me, they should be scrubbed from dictionary. So I make suggestions with regards to maximum sound quality (as I understand sound quality). I leave it to the reader to determine suitability of my suggestions, for their own personal circumstances.



Quote:
Originally Posted by KevinG View Post
Also, yes, that big black cylinder is my SVS PC13 Ultra. That was previously determined to be the best position in the room (of the limited options) by using REW. I can and will move it to see how it is effecting the ECT of the front left speaker, but it will probably have to be moved back to that position anyway.
Yeah. As per the floor treatments above, even if that position is fixed, at least you will understand the effect it is having on your measurements, so you're not left scratching your head at a later date, wondering about some measurement anomaly.

Last edited by Audionut11; 06-19-2014 at 08:37 AM.
Audionut11 is offline  
post #10482 of 11856 Old 06-19-2014, 09:10 AM
AVS Special Member
 
jkasanic's Avatar
 
Join Date: Dec 2006
Posts: 1,172
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 45 Post(s)
Liked: 62
Could someone please remind me the trick for doing "sliced" ETC overlays in REW? I seem to remember something about saving individual measurements multiple times but can't for the life of me recall how to do the overlays at the various slices since you can only overlay the ETC and not the filtered IR directly.

@AustinJerry: Perhaps the benefit of sliced ETC's and how to generate the overlays in REW would be a useful addition to the ETC section of the guide?
jkasanic is offline  
post #10483 of 11856 Old 06-19-2014, 09:17 AM
Advanced Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 512
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 195 Post(s)
Liked: 56
You can use trace arithmetic to copy a measurement. Go to the all SPL view, pick the same curve into the A and B slots then calculate (A+B)/2
3ll3d00d is online now  
post #10484 of 11856 Old 06-19-2014, 09:25 AM
AVS Special Member
 
jim19611961's Avatar
 
Join Date: Mar 2011
Location: Texas
Posts: 1,396
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 84
Quote:
Originally Posted by jkasanic View Post
Could someone please remind me the trick for doing "sliced" ETC overlays in REW? I seem to remember something about saving individual measurements multiple times but can't for the life of me recall how to do the overlays at the various slices since you can only overlay the ETC and not the filtered IR directly.

@AustinJerry: Perhaps the benefit of sliced ETC's and how to generate the overlays in REW would be a useful addition to the ETC section of the guide?
Open the same measurement multiple times and assign each a different filter.


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.

Rega - Apollo, Rega - DAC, Goldpoint Passive, (2) Classe CA-100 bridged power amps (350w)

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
jim19611961 is offline  
post #10485 of 11856 Old 06-19-2014, 09:35 AM
AVS Special Member
 
jkasanic's Avatar
 
Join Date: Dec 2006
Posts: 1,172
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 45 Post(s)
Liked: 62
Quote:
Originally Posted by jim19611961 View Post
Open the same measurement multiple times and assign each a different filter.
Thanks Jim! I had forgotten that the ETC overlay would still work for the filtered IR as long as you change the filter.
jkasanic is offline  
post #10486 of 11856 Old 06-19-2014, 09:42 AM
Senior Member
 
KevinG's Avatar
 
Join Date: Apr 2002
Location: S. Jersey
Posts: 462
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 30 Post(s)
Liked: 14
Quote:
Originally Posted by Audionut11 View Post
Good idea. If you have the time, inclination, and funds, consider making it rather large. This will ensure you capture all reflections.
Have you purchased the roxul yet? Roxul is great where space is limited, however, when you have space for 6" or more, pink fluffy will perform better.

See here.
I already have 2 bundles (16 sheets) of the roxul. But that doesn't matter much, since pink fluffy is so inexpensive. What would you consider "rather large"? I was planning on 3 or 4 2x4x8" deep panels on the ceiling, depending on what the mirror test shows me. I've already built a 14 foot by 7.5 foot by 15inch deep entire front wall, so I would say I'm pretty committed. :-) It seems that front wall might have been better stuffed full of pink fluffy instead of 7.5" of SnS + 7.5" of air, but that's behind me now...

I'll do some measurements forward of the current couch position, but distance to the screen (which is 133" diagonal) is also a consideration here.

Thanks for all the help! Any pointers on hanging 2x4x6" panels two inches from the ceiling? This must have come up a million times before?
KevinG is offline  
post #10487 of 11856 Old 06-19-2014, 10:00 AM
AVS Special Member
 
jim19611961's Avatar
 
Join Date: Mar 2011
Location: Texas
Posts: 1,396
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 84
Time Deviation Correction for Sliced ETC

Since we are on the topic of ETC slicing again, let me share a few things I have learned recently.

Click image for larger version

Name:	06 19 14rew etc sliced overlay.jpg
Views:	20
Size:	372.2 KB
ID:	126242

Here is my overlayed ETC slices. What I want everyone to notice is how the lower frequency slices move away from the zero timing reference.

Click image for larger version

Name:	06 19 14 rew 500hz slice.jpg
Views:	21
Size:	198.8 KB
ID:	126250

Looking at just the 500hz slice, one can see this slice has moved 2.1ms.

Click image for larger version

Name:	06 19 14 rew 500hz slices time corrected.jpg
Views:	21
Size:	191.4 KB
ID:	126258

Using this screen, one can correct the slice so its time reference is zero again using the controls panel and the t=0 offset.

Click image for larger version

Name:	06 19 14 rew sliced overlay time corrected.jpg
Views:	22
Size:	372.8 KB
ID:	126266

Going through each slice, and correcting each for its time reference deviation, the ETC sliced overlay now looks like this. Notice how every slice now has the same time reference.


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.

Rega - Apollo, Rega - DAC, Goldpoint Passive, (2) Classe CA-100 bridged power amps (350w)

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

Last edited by jim19611961; 06-19-2014 at 10:30 AM.
jim19611961 is offline  
post #10488 of 11856 Old 06-19-2014, 10:01 AM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 7,399
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 645 Post(s)
Liked: 762
Quote:
Originally Posted by KevinG View Post
Here's my mdat file.

https://drive.google.com/file/d/0B-G...it?usp=sharing

I think it is showing me that my panels are doing very very little. I'm studying my left speaker.

There seems to be an "echo" at 2.97ms (-4db), and another at 6.57ms (-16db).

The panel seems to be doing an excellent job BETWEEN those two spikes. It brings the broad spike at 3.95ms down from -13.14db to -21.24db, and the broad spike at 5.35ms from -19.45db to -22.9db. Is this what I should expect from a panel so close to the left front speaker?

What distance should I be looking at to figure out what is causing the spike at 29.67ms (-6.4db)? And, maybe more importantly, the one at 2.88ms (-5.24db)?

I'm not below -20 until 67.77ms, so I've got plenty of work to do. :-)

Thanks for any and all assistance!
-Kevin
Looking at the Waterfall graphs, I see a significant improvement 35-40Hz for the left speaker, and a modest improvement 60-70Hz. I don't see any improvement for the right speaker. there is a modest improvement for the center speaker at 60Hz.

Looking at the ETC graphs, I see one reflection for the left speaker at 4m that has improved, and one improvement for the right speaker at 8.1m, but no improvement for the center speaker.

So, the existing treatments have had some positive impact, but I agree, there is room for more improvement.

__________________________________________________

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 #10489 of 11856 Old 06-19-2014, 10:04 AM
Senior Member
 
Audionut11's Avatar
 
Join Date: May 2008
Posts: 354
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 52
Quote:
Originally Posted by KevinG View Post
What would you consider "rather large"?
Anything bigger then a suggested minimum. Ideally, one that covers all 3 speaker reflection points. A dirty great big rectangle thing, with lighting, if you like lighting.
For me personally, as a minimum, I would use double width of your current panels over each speaker reflection point.


Quote:
Originally Posted by KevinG View Post
I'll do some measurements forward of the current couch position, but distance to the screen (which is 133" diagonal) is also a consideration here.
Looks like you are already at around 40 degree viewing angle. So yes, that rules out moving the couch closer. Do your best to acoustically deaden the pylon.


Quote:
Originally Posted by KevinG View Post
Thanks for all the help! Any pointers on hanging 2x4x6" panels two inches from the ceiling? This must have come up a million times before?
Happy too. No help from me on the hanging, and I can't recall any previous discussion in this thread.
Audionut11 is offline  
post #10490 of 11856 Old 06-19-2014, 10:06 AM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 7,399
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 645 Post(s)
Liked: 762
Quote:
Originally Posted by jim19611961 View Post
Going through each slice, and correcting each for its time reference deviation, the ETC sliced overlay now looks like this. Notice how every slice now has the same time reference.
What causes the time reference deviation?

__________________________________________________

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 #10491 of 11856 Old 06-19-2014, 10:10 AM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 7,399
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 645 Post(s)
Liked: 762
Quote:
Originally Posted by KevinG View Post
Thanks for all the help! Any pointers on hanging 2x4x6" panels two inches from the ceiling? This must have come up a million times before?
If you don't mind a reasonably-priced commercial solution, take a look at the GIK Ceiling Mounting Brackets: http://www.gikacoustics.com/product/...ting-brackets/.

Or, you can look at the pics of my ceiling-mounted panels in the MySetup link in my sig (which were a DIY solution and not that easy).

__________________________________________________

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 #10492 of 11856 Old 06-19-2014, 10:16 AM
AVS Special Member
 
jim19611961's Avatar
 
Join Date: Mar 2011
Location: Texas
Posts: 1,396
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 84
Frequency Bias for Sliced ETC

Here is something else interesting. I have always wondered why the full range ETC is so biased towards the higher frequencies. I think I have discovered why and by how much.

First bring up the ETC IR filtered overlay again. Now turn off the normalization. Move the pointer to time zero.

Click image for larger version

Name:	0619 14 rew etc slices overlay un-normalized.jpg
Views:	18
Size:	255.8 KB
ID:	126298

Notice that REW sees each slice at a different magnitude. This is represented in db at the bottom of the screen for wherever you put the pointer. In this case, we want t=0. When we normalize a fullrange ETC, it normalizes to the highest value it sees. So the higher frequencies predominate the result.

Further proof of this can be seen when you do a fullrange ETC and compare it to each slice in a filtered ETC. You will find the fullrange ETC corresponds most closely to the 8K 1/1 slice than any of the others.


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.

Rega - Apollo, Rega - DAC, Goldpoint Passive, (2) Classe CA-100 bridged power amps (350w)

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

Last edited by jim19611961; 06-19-2014 at 10:31 AM.
jim19611961 is offline  
post #10493 of 11856 Old 06-19-2014, 10:20 AM
AVS Special Member
 
jim19611961's Avatar
 
Join Date: Mar 2011
Location: Texas
Posts: 1,396
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 84
Quote:
Originally Posted by AustinJerry View Post
What causes the time reference deviation?
This is a question for Johnpm. I am afraid I do not know. But I have noticed that this is the case in all the sliced ETC's ive seen posted. Yours and others show the same phenomena.

Not only do the slices deviate in time, they deviate in magnitude (although the later is corrected by normalization when looking at individual slices).


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.

Rega - Apollo, Rega - DAC, Goldpoint Passive, (2) Classe CA-100 bridged power amps (350w)

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

Last edited by jim19611961; 06-19-2014 at 10:35 AM.
jim19611961 is offline  
post #10494 of 11856 Old 06-19-2014, 10:36 AM
Senior Member
 
Audionut11's Avatar
 
Join Date: May 2008
Posts: 354
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 52
On my measurements with loopback timing, the lower frequencies move in the opposite direction.
Attached Thumbnails
Click image for larger version

Name:	etc.png
Views:	13
Size:	286.8 KB
ID:	126370  
Audionut11 is offline  
post #10495 of 11856 Old 06-19-2014, 10:39 AM
AVS Special Member
 
jim19611961's Avatar
 
Join Date: Mar 2011
Location: Texas
Posts: 1,396
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 84
Quote:
Originally Posted by Audionut11 View Post
On my measurements with loopback timing, the lower frequencies move in the opposite direction.
Oh dear! Hmmm. That is odd. Still, the deviation can be corrected though.

Good thing for us to be aware of.

Worthy to note my measurements do NOT use loopback timing (i.e. USB mic).


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.

Rega - Apollo, Rega - DAC, Goldpoint Passive, (2) Classe CA-100 bridged power amps (350w)

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

Last edited by jim19611961; 06-19-2014 at 10:45 AM.
jim19611961 is offline  
post #10496 of 11856 Old 06-19-2014, 10:55 AM
Senior Member
 
Audionut11's Avatar
 
Join Date: May 2008
Posts: 354
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 77 Post(s)
Liked: 52
Quote:
Originally Posted by jim19611961 View Post
Worthy to note my measurements do NOT use loopback timing (i.e. USB mic).
When using ASIO, plug any old output, into any old input, and select the appropriate settings in REW. "Timing reference output" + "Timing reference input". When you have an device connected to the input, you gain the use of that input in REW for timing, rather then just the USB mics.

@3ll3d00d has a handy reference here.


It's well worth the while also. I was surprised how far my timings were out when not using a reference signal.

edit: I've only just started using timing references in the last day or 2. Great for time aligning multiple speakers.

Last edited by Audionut11; 06-19-2014 at 11:04 AM.
Audionut11 is offline  
post #10497 of 11856 Old 06-19-2014, 10:56 AM
Senior Member
 
JohnPM's Avatar
 
Join Date: Sep 2004
Location: UK
Posts: 450
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 19
Quote:
Originally Posted by AustinJerry View Post
What causes the time reference deviation?
The bandpass filters specified in ISO 3382 have phase shift and hence time delay. The Time Reversed Filtering option reduces this, from the help:

The Time Reversed Filtering control applies the octave band filters backwards in time, this reduces the filter's own contribution to the measured decay. When using 1/3 octave filters at low frequencies the filter decay time can be significant, over 200ms for a 100Hz 1/3 filter, for example. Applying the filter in reverse reduces this decay to less than 50ms, but it does affect the response somewhat, such that Early Decay Time (EDT) figures using Time-Reversed filters may not be valid.

John

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
JohnPM is offline  
post #10498 of 11856 Old 06-19-2014, 11:00 AM
Advanced Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 512
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 195 Post(s)
Liked: 56
Quote:
Originally Posted by jim19611961 View Post
This is a question for Johnpm. I am afraid I do not know. But I have noticed that this is the case in all the sliced ETC's ive seen posted. Yours and others show the same phenomena.

Not only do the slices deviate in time, they deviate in magnitude (although the later is corrected by normalization when looking at individual slices).
I'm not sure it is correct to shift the impulse back to align to the peak. When you time align using an IR then you either align impulses with the same spectral content via the peaks (because the shape of the impulse should be comparable) or you align to the initial rise of the impulse as this is when the impulse started at the speaker. This makes aligning a sub to a mains using an IR from a full bandwidth sweep rather tricky (and hence various narrow bandwidth methods were born).

Basically the IR gets lower, in absolute magnitude, and slower as frequency decreases & wavelength/period increases hence if you are shifting by the peak alone then you are really shifting the impulse back in time so that the impulses no longer originated at the same point in time.

This will be quite clear if you take a series of sweeps at different frequency ranges then compare the IRs.

EDIT: given John's reply, I'm even less sure whether this is strictly relevant to a sliced ETC but anyway....

Last edited by 3ll3d00d; 06-19-2014 at 11:06 AM.
3ll3d00d is online now  
post #10499 of 11856 Old 06-19-2014, 11:04 AM
AVS Special Member
 
jim19611961's Avatar
 
Join Date: Mar 2011
Location: Texas
Posts: 1,396
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 84
Quote:
Originally Posted by 3ll3d00d View Post
I'm not sure it is correct to shift the impulse back to align to the peak. When you time align using an IR then you either align impulses with the same spectral content via the peaks (because the shape of the impulse should be comparable) or you align to the initial rise of the impulse as this is when the impulse started at the speaker. This makes aligning a sub to a mains using an IR from a full bandwidth sweep rather tricky (and hence various narrow bandwidth methods were born).

Basically the IR gets lower, in absolute magnitude, and slower as frequency decreases & wavelength/period increases hence if you are shifting by the peak alone then you are really shifting the impulse back in time so that the impulses no longer originated at the same point in time.

This will be quite clear if you take a series of sweeps at different frequency ranges then compare the IRs.
JohnPM's response suggests the filters are the reason for the delay.


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.

Rega - Apollo, Rega - DAC, Goldpoint Passive, (2) Classe CA-100 bridged power amps (350w)

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
jim19611961 is offline  
post #10500 of 11856 Old 06-19-2014, 11:11 AM
AVS Special Member
 
jim19611961's Avatar
 
Join Date: Mar 2011
Location: Texas
Posts: 1,396
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 126 Post(s)
Liked: 84
Quote:
Originally Posted by JohnPM View Post
The bandpass filters specified in ISO 3382 have phase shift and hence time delay. The Time Reversed Filtering option reduces this, from the help:

The Time Reversed Filtering control applies the octave band filters backwards in time, this reduces the filter's own contribution to the measured decay. When using 1/3 octave filters at low frequencies the filter decay time can be significant, over 200ms for a 100Hz 1/3 filter, for example. Applying the filter in reverse reduces this decay to less than 50ms, but it does affect the response somewhat, such that Early Decay Time (EDT) figures using Time-Reversed filters may not be valid.
John,

Would it be possible to normalize the timing as well as the amplitudes via software in future REW versions via extra control features?

Additionally, would it be possible to make fullrange ETC's reflect even amplitudes for all frequencies rather than favoring the upper registers?


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.

Rega - Apollo, Rega - DAC, Goldpoint Passive, (2) Classe CA-100 bridged power amps (350w)

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

Last edited by jim19611961; 06-19-2014 at 11:16 AM.
jim19611961 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