Optimizing subwoofers and integration with mains: multi sub optimizer - Page 3 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 166Likes
Reply
 
Thread Tools
post #61 of 797 Old 11-14-2015, 01:50 AM
AVS Forum Special Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 9,947
Mentioned: 115 Post(s)
Tagged: 0 Thread(s)
Quoted: 4254 Post(s)
Liked: 1654
Hi Andy

Is there a simple way to duplicate filters (PEQ in this case) to other channels?

The Graph Properties panel can't be resized to see all of the left column:
Attached Thumbnails
Click image for larger version

Name:	Screen Shot 2015-11-14 at 10.48.29.png
Views:	163
Size:	53.4 KB
ID:	1056313  

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
markus767 is offline  
Sponsored Links
Advertisement
 
post #62 of 797 Old 11-14-2015, 01:56 AM
AVS Forum Special Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 2,919
Mentioned: 103 Post(s)
Tagged: 0 Thread(s)
Quoted: 1666 Post(s)
Liked: 646
It would be nice to able to define a template set of filters and then just choose to assign that template to a channel.
3ll3d00d is online now  
post #63 of 797 Old 11-14-2015, 06:49 AM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
Quote:
Originally Posted by markus767 View Post
Is there a simple way to duplicate filters (PEQ in this case) to other channels?
At present, there is no way to do this. This, or something like what 3ll3d00d refers to below seems like it would be a good addition.

Quote:
Originally Posted by markus767 View Post
The Graph Properties panel can't be resized to see all of the left column:
Yes, the general situation with dialog resizing is something I need to work on. Microsoft treats C++, and especially MFC, as a second-class citizen. Unbelievably, they only added a dialog and child control resizing framework to MFC in April of this year! I think I remember you saying you are a developer? If so, you might find this article on the new MFC dialog resizing entertaining, especially the comments section in which MS takes a pounding. They've been dragging their feet on this for almost 20 years. In the meantime, I'll make the tree view portion of the property sheet wider in the next release as a temporary workaround. I'm an EE (retired) by trade and a self-taught C++/Windows development guy, so I tend to depend more on what the vendors provide for UI features than a hardcore Computer Science UI guru might.

Quote:
Originally Posted by 3ll3d00d View Post
It would be nice to able to define a template set of filters and then just choose to assign that template to a channel.
I like this idea. It may be a while before I incorporate something like this, or what Markus suggested above, as after I finish up the changes I'm working on now, I have a DIY sub project to work on that's been on hold for some time. I haven't used the software in my own system yet!
andyc56 is offline  
 
post #64 of 797 Old 11-14-2015, 06:50 AM
Senior Member
 
AV_mike's Avatar
 
Join Date: May 2009
Posts: 431
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 120 Post(s)
Liked: 106
Hi Andy,

Great application - I ran 25 measurements using a timing reference (HDMI input, ADC monitoring Left Surround pre-amp output - with actual speaker disconnected) to record multiple positions for my Left, Right, Centre, Sub 1 and Sub 2. Presently I am just optimizing the two subs with the centre (but I hope to compare the results using left & right as well, hence the extra measurements).

I did have a few 'exceptions' when generating graphs - I'll have to see if I can repeat my steps and make notes - might help in catching bugs, etc.

With regards to crossovers - have you thought about having some form of 'link' between crossovers - so that they could be optimized, but would be forced to track each other with regards to frequency? Also, maybe a setting that would limit available choices to the usual 10Hz or 20Hz steps?

One last thought - would the ability to store the best five 'results' be helpful in seeing if the best settings were all 'tending towards common values' or if they were more random/chaotic in nature?

So far this software has demonstrated to me that raw subwoofer responses are no guide to possible integration when multiple units are available - my starting points are awful - but the end results (if they translate in the real world.......................) could be the best yet!

Please keep up the good work, regards, Mike (latest optimization still running........................................... ...............)
AV_mike is offline  
post #65 of 797 Old 11-14-2015, 08:28 AM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
Hi Mike,

Quote:
Originally Posted by AV_mike View Post
I did have a few 'exceptions' when generating graphs - I'll have to see if I can repeat my steps and make notes - might help in catching bugs, etc.
Aargh. If you have a repeatable way to generate these errors, then there's a good chance I can fix them quickly. A brief "steps to repeat" list, along with a description or screen shot of the error message, together with a zipped copy of the .msop and .msow files would allow me to debug the problem. The .msop is of course the project, and the .msow contains the configuration of the tabbed windows such as graphs and filter reports.

Quote:
Originally Posted by AV_mike View Post
With regards to crossovers - have you thought about having some form of 'link' between crossovers - so that they could be optimized, but would be forced to track each other with regards to frequency? Also, maybe a setting that would limit available choices to the usual 10Hz or 20Hz steps?
One feature I considered when I was starting out was "variables". The idea was that you'd be able to define one or more variables per configuration, and these could be shared across multiple filters. So you might define, say, a crossover frequency variable and then the cutoff frequencies of the high-pass and low-pass filters could be associated with this shared variable rather than being independently stored in the filters as they are now. But at the time I was somewhat overwhelmed by the complexity that the project had already taken on, and decided not to push it too far to start out, to make it easier to see the "light at the end of the tunnel" as it were. It's still a possibility for the future.

The quantization of crossover frequencies is a more thorny problem. There are many categories of optimization problems and optimizers to solve them, and one such distinction is discrete optimization (also called "combinatorial optimization") vs. continuous optimization. The optimization code I use is of the continuous optimization type, in which the assumption is made that all parameters are continuously variable between their lower and upper limits. This can be tweaked for small parameter steps by internally quantizing the values according to their "legal" values, but if the steps are large, as they are with AVR crossover frequencies, the convergence of the optimization can suffer badly. As you've probably noticed, the optimizer is already pretty slow.

Quote:
Originally Posted by AV_mike View Post
One last thought - would the ability to store the best five 'results' be helpful in seeing if the best settings were all 'tending towards common values' or if they were more random/chaotic in nature?
That's another possibility for the future as well. After the next release in about a week or so, I'll be setting aside working on it for a while until I catch up on my own sub project and other home-related items as well. I anticipate more active development during the summer months when I am mostly stuck inside in the Texas heat.
andyc56 is offline  
post #66 of 797 Old 11-14-2015, 10:28 AM
Senior Member
 
AV_mike's Avatar
 
Join Date: May 2009
Posts: 431
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 120 Post(s)
Liked: 106
Hi Andy,

I was just trying out a few things, getting familiar with graph manipulation when I got an exception. It appears to be related to setting the Project Options to compensate for sub measurements that have an AVR LPF applied. I had my AVR set to 250Hz (as high as it will go), so I set this value in the option. If the setting is made after sub measurements have been loaded then up comes the exception. Sorry for the lack of more info, but I was just using some spare time to experiment - I'll try to get more info later.

Regards, Mike.
AV_mike is offline  
post #67 of 797 Old 11-14-2015, 10:45 AM
Senior Member
 
AV_mike's Avatar
 
Join Date: May 2009
Posts: 431
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 120 Post(s)
Liked: 106
Hi again Andy,

Hopefully the attached files will assist in debug. The 'msow' file does not appear to have been created at this stage.
This time I set the sub LPF correction option first, then loaded Mains & Sub measurements - all went fine.
Then I tried to create a graph of the three 'mlp' measurements, C - S1 - S2, I was able to select the check boxes, but when clicking OK - up comes the exception.

Please let me know if I can provide any more info, regards, Mike.
Attached Thumbnails
Click image for larger version

Name:	MSO exception.jpg
Views:	128
Size:	154.2 KB
ID:	1057153  
Attached Files
File Type: zip MLP 4-position start.zip (306.6 KB, 41 views)
AV_mike is offline  
post #68 of 797 Old 11-14-2015, 11:21 AM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
Quote:
Originally Posted by AV_mike View Post
Hi again Andy,

Hopefully the attached files will assist in debug. The 'msow' file does not appear to have been created at this stage.
This time I set the sub LPF correction option first, then loaded Mains & Sub measurements - all went fine.
Then I tried to create a graph of the three 'mlp' measurements, C - S1 - S2, I was able to select the check boxes, but when clicking OK - up comes the exception.

Please let me know if I can provide any more info, regards, Mike.
Yikes. Thanks for the info, I'll have a look at it today.
andyc56 is offline  
post #69 of 797 Old 11-14-2015, 03:07 PM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
Hi Mike,

I found the bug and have a fixed .exe file for it. It's not a full installer, as this is an "in-between" version that has a partial implementation of some additions that I'm not finished with yet. I have disabled the menu options for all such features so they won't mess up normal program operation. You can just overwrite the .exe in the installed directory, which will take Admin privileges. When I get the new features finished, I'll make a new installer.

(Obsolete test version deleted)

One new feature that is implemented is the ability to define the AVR properties for sub distance increment and units, as well as sub gain increment (usually 0.5 dB). This is done in the Application Options, Hardware. It calculates AVR distance and gain settings, using the DSP to make up for the limited gain and distance step size resolution in the AVR. Here'a an example of the end of a filter report with AVR distance resolution configured for 0.1 feet and gain step of 0.5 dB.

Final gain and delay/distance settings:
Increase AVR sub out trim gain by 1.5 dB
Sub Channel 1 gain: -5.6301 dB
Sub Channel 2 gain: -0.0673176 dB
Sub Channel 3 gain: -14.0948 dB
Sub Channel 4 gain: -4.23466 dB
Decrease AVR sub out distance by 3 feet
Sub Channel 1 delay: 12.346 msec
Sub Channel 2 delay: 4.53837 msec
Sub Channel 3 delay: 4.46217 msec
Sub Channel 4 delay: 0.0462191 msec

In this case, Sub channel 2 didn't actually have a gain block in MSO., but the filter report adjusts it and the other gains for the resolution-induced error from the 0.5 dB step of the AVR. A similar situation occurs for the delay of Sub Channel 4, which has no delay block in MSO.

Last edited by andyc56; 11-18-2015 at 07:23 AM.
andyc56 is offline  
post #70 of 797 Old 11-15-2015, 08:24 AM
Senior Member
 
AV_mike's Avatar
 
Join Date: May 2009
Posts: 431
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 120 Post(s)
Liked: 106
^^^^^^^^^^^^^^
Wow!!!!
Thanks Andy, I'll install and give it a go.

Now that I've seen at first hand what can be achieved with well integrated multiple subs (in my initial case only two, but placement was far from ideal) I'm going to get my unused third sub back out of the cupboard and hook up all three. My attempts at manually aligning multiples did not go well - and the measure/adjust/measure/adjust routine can take ages. Much better to let the PC crunch though all the data.

Have you considered multi-threading support to increase the computation speed? Looking at Task Manager, it appears that only one main thread is operating.
Something else that may already be on your 'to-do' list, is there a way to stop identical or very similar filters from being generated (eg stacked filters) REW appears to be able to prevent this when creating EQ filters - maybe a visit to HTS Forum and a friendly chat with REW author John will provide some info.............................................. ...................

Keep up the good work - but please also spend some time on your own projects, regards, Mike.
AV_mike is offline  
post #71 of 797 Old 11-15-2015, 08:40 AM
Member
 
BUJonathan's Avatar
 
Join Date: Jan 2007
Posts: 124
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 37 Post(s)
Liked: 29
Quote:
Originally Posted by Jag768 View Post
What kind of speakers are those? DIY?
BUJonathan is offline  
post #72 of 797 Old 11-15-2015, 11:29 AM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
Quote:
Originally Posted by BUJonathan View Post
What kind of speakers are those? DIY?
There's some more info on his speakers in this thread. Very nice work, and beautiful room too. I believe there's a diyaudio.com thread on them as well.
andyc56 is offline  
post #73 of 797 Old 11-15-2015, 11:44 AM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
Quote:
Originally Posted by AV_mike View Post
Have you considered multi-threading support to increase the computation speed? Looking at Task Manager, it appears that only one main thread is operating.
At present, the optimizer only uses one thread, a worker thread launched from the UI thread (just to keep the UI responsive). This was actually my first multi-threaded program of any kind. Maybe at a later time I'll try to figure out how to get the optimization algorithm itself to use multiple threads.

Quote:
Originally Posted by AV_mike View Post
Something else that may already be on your 'to-do' list, is there a way to stop identical or very similar filters from being generated (eg stacked filters) REW appears to be able to prevent this when creating EQ filters - maybe a visit to HTS Forum and a friendly chat with REW author John will provide some info.
There's no way to do this automatically. One can manually define the min and max center frequencies of e.g. PEQs to be in mutually exclusive ranges, but this does restrict the set of potential solutions. REW optimizes only a single response, not the magnitude of a complex summation of multiple responses, so it can use a special-purpose optimizer designed specifically for the problem at hand. When dealing with summations of multiple responses at multiple listening positions, one needs a general-purpose optimizer. Such optimizers have no knowledge of the problem they're solving. In this case, the optimizer generates trial solution vectors according to its algorithm and passes them to a provided objective function to evaluate the error to be minimized. All the optimizer knows about the vectors it's generating is the minimum and maximum allowable values of the parameters at each position of the vector. That's it.
andyc56 is offline  
post #74 of 797 Old 11-16-2015, 12:32 AM
AVS Forum Special Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 9,947
Mentioned: 115 Post(s)
Tagged: 0 Thread(s)
Quoted: 4254 Post(s)
Liked: 1654
Andy,

Would it be possible to add gain and delay blocks to the main channels?

Larger max/min bounds of filter settings would be helpful.

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole

Last edited by markus767; 11-16-2015 at 12:39 AM.
markus767 is offline  
post #75 of 797 Old 11-16-2015, 09:52 AM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
Quote:
Originally Posted by markus767 View Post
Would it be possible to add gain and delay blocks to the main channels?
I've been thinking about this, and I came up with an idea, though I'm still not completely certain it will work in all cases.
  • If there are N main filter channels, allow one each gain and delay block in N-1 of them
  • No shared delays or gain blocks allowed in main channels
  • This disallows gain and delay blocks in the main channel completely when there is only one main channel

With this setup, there's still a way to fool the program though. You could:
  • Define two main channels (A and B), with one of them (A) having both a gain and delay block and the other (B) not having either
  • Define all the measurement groups to be optimized such that all the main measurements in them are associated with channel A and none with channel B

This would lead to an uncontrolled system gain and delay shift, but maybe I could detect this situation when the optimization is started and disallow running it.

Quote:
Originally Posted by markus767 View Post
Larger max/min bounds of filter settings would be helpful.
Can you be more specific? Which filter types? Which parameters, and what do you propose that these new limits should be?

I had someone ask me to allow the PEQ center frequency to be as high as 400 Hz. He wanted to use the PEQ to kill a peak in the passband response of his horn sub. But this causes a problem. Let's say the optimization had an upper frequency limit of 200 Hz. If the optimizer is trying to change the parameters of, say, a PEQ filter assigned to a main channel, it could end up causing all kinds of bad effects up around 400 Hz. The optimizer would in general not be able to detect these effects if the optimization frequency range had a maximum of 200 Hz, because the flatness error above 200 Hz would not be known or taken into account.

The idea I thought of was to have "non-optimizable filters". These would be filters that could never be configured to have any of their parameters be optimizable. So their use would be limited to user adjustments via the Properties grid.
andyc56 is offline  
post #76 of 797 Old 11-16-2015, 10:55 AM
AVS Forum Special Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 9,947
Mentioned: 115 Post(s)
Tagged: 0 Thread(s)
Quoted: 4254 Post(s)
Liked: 1654
After thinking more about this I believe having gain and delay blocks in the main channels is a bad idea. I wanted to be lazy with measurements but this would do more harm than good.

The bounds of gain should be increased though. I had measurements that were not level aligned so having large variable gain could put such responses into the ballpark for optimization.
Maybe having the option to manually shift responses by a fixed amount is a good idea? Such a feature would be particulary useful for the target curve in order to align it to the level of the measurements.

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
markus767 is offline  
post #77 of 797 Old 11-16-2015, 11:57 AM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
Quote:
Originally Posted by markus767 View Post
After thinking more about this I believe having gain and delay blocks in the main channels is a bad idea. I wanted to be lazy with measurements but this would do more harm than good.

The bounds of gain should be increased though. I had measurements that were not level aligned so having large variable gain could put such responses into the ballpark for optimization.
Maybe having the option to manually shift responses by a fixed amount is a good idea? Such a feature would be particulary useful for the target curve in order to align it to the level of the measurements.
So, should I make it, say +/- 30 dB adjustment, or do you need more than that?

The target curve is internally a delta from the reference value, and it gets offset to a value of 0 dB at the highest frequency. To visually align the target curve to a reference trace, you can use the display offset in the graph trace options (as the target curve is just another type of trace). I guess I could put a spin button in that edit control to allow tuning it. Pressing Apply allows tweaking the offset without closing the dialog.
andyc56 is offline  
post #78 of 797 Old 11-16-2015, 12:58 PM
AVS Forum Special Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 9,947
Mentioned: 115 Post(s)
Tagged: 0 Thread(s)
Quoted: 4254 Post(s)
Liked: 1654
Quote:
Originally Posted by andyc56 View Post
So, should I make it, say +/- 30 dB adjustment, or do you need more than that?
That should be enough.

Quote:
Originally Posted by andyc56 View Post
The target curve is internally a delta from the reference value, and it gets offset to a value of 0 dB at the highest frequency. To visually align the target curve to a reference trace, you can use the display offset in the graph trace options (as the target curve is just another type of trace). I guess I could put a spin button in that edit control to allow tuning it. Pressing Apply allows tweaking the offset without closing the dialog.
are you saying the absolute level of the target curve shouldn't affect the shape of the optimization result? That's not what I'm seeing.

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
markus767 is offline  
post #79 of 797 Old 11-16-2015, 01:06 PM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
Quote:
Originally Posted by markus767 View Post
are you saying the absolute level of the target curve shouldn't affect the shape of the optimization result?
Yes. That's the way I intended it to work

Quote:
Originally Posted by markus767 View Post
That's not what I'm seeing.
Hmmm. I'll go back and double-check the code just to make sure. Maybe it's a bug?
andyc56 is offline  
post #80 of 797 Old 11-16-2015, 01:17 PM
AVS Forum Special Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 9,947
Mentioned: 115 Post(s)
Tagged: 0 Thread(s)
Quoted: 4254 Post(s)
Liked: 1654
Sorry Andy, my bad. I got confused with settings and loaded responses. The results are now as expected.

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
markus767 is offline  
post #81 of 797 Old 11-16-2015, 01:26 PM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
Quote:
Originally Posted by markus767 View Post
Sorry Andy, my bad. I got confused with settings and loaded responses. The results are now as expected.
No problemo. I just double-checked the code. Before the optimization starts, the code determines the indices in the shared frequency array of the highest optimization frequency and the highest reference frequency. The frequency array is then resized to max(iLastRef, iLastOpt) + 1 in length. This resized array is passed to the configuration object, which interpolates the target curve (using log or linear interpolation), passing the optimizer a vector of interpolated target curve values. But before passing it back, it takes the highest-frequency curve value in the vector and subtracts it from all the elements, forcing the curve value to 0 dB at the max(highest_ref_freq, highest_opt_freq).

When optimizing, this curve (in dB) is subtracted from the combined response data at each position, and this difference is optimized to flat.
andyc56 is offline  
post #82 of 797 Old 11-17-2015, 10:56 AM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
Quote:
Originally Posted by markus767 View Post
The Graph Properties panel can't be resized to see all of the left column:
Hi Markus,

Could you check out this latest version to see if it fixes the too-narrow tree view?

(Obsolete test version deleted)

The ratio of the tree view width to the width of the property pages was system-dependent. I now scale the pixel width of the tree views based on the pixel dimensions of the system font, so it is adaptive now. I also made the tree views wider on several other property sheets. Later I'll make the property sheets resizable.

If this works I'll build a new installer and put the new version up.

Last edited by andyc56; 11-18-2015 at 07:23 AM.
andyc56 is offline  
post #83 of 797 Old 11-17-2015, 11:38 AM
AVS Forum Special Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 9,947
Mentioned: 115 Post(s)
Tagged: 0 Thread(s)
Quoted: 4254 Post(s)
Liked: 1654
It did improve:
Attached Thumbnails
Click image for larger version

Name:	Screen Shot 2015-11-17 at 20.36.17.png
Views:	130
Size:	57.1 KB
ID:	1063706  

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
markus767 is offline  
post #84 of 797 Old 11-17-2015, 02:20 PM
AVS Forum Special Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 9,947
Mentioned: 115 Post(s)
Tagged: 0 Thread(s)
Quoted: 4254 Post(s)
Liked: 1654
Andy

Any chance you could add mic calibration file import? Exporting a .frd from FuzzMeasure does not contain the mic calibration file adjustments for example.

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
markus767 is offline  
post #85 of 797 Old 11-17-2015, 06:44 PM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
Quote:
Originally Posted by markus767 View Post
It did improve:
Still a problem though. Please try this newer version:
(Obsolete test version deleted)
If I did it right, the ratio of the pixel width of the property sheet on the right to the tree view on the left should be 2.2 instead of the 3.6 value of your example above.

Quote:
Originally Posted by markus767 View Post
Any chance you could add mic calibration file import? Exporting a .frd from FuzzMeasure does not contain the mic calibration file adjustments for example.
For the release after this coming one, I can add that and dialog resizing. I need to work on my sub project! I have four of the Dayton RSS460HO-18" in 4 cu. ft. boxes that have been half-finished for months.

Last edited by andyc56; 11-18-2015 at 07:23 AM.
andyc56 is offline  
post #86 of 797 Old 11-17-2015, 11:35 PM
AVS Forum Special Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 9,947
Mentioned: 115 Post(s)
Tagged: 0 Thread(s)
Quoted: 4254 Post(s)
Liked: 1654
Now it's fixed. Thanks Andy.

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
markus767 is offline  
post #87 of 797 Old 11-18-2015, 07:43 AM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
I put a new installer up for version 1.01. It has the following new features and bug fixes:
  • Fixed bug of system dependency in width of tree views of property sheet dialogs containing them (thanks Markus!)
  • Fixed bug in which a crash occurred in certain situations when MSO is configured to compensate sub measurements for sub LPF response (thanks AV_mike!)
  • Fixed bug in which shared filter biquad information was not showing up in the filter reports (only individual channels)
  • Fixed bug in which the project options weren't being set back to defaults when File, Close was invoked
  • Added the ability to have plot-only measurement groups (when you want to plot combined responses but not optimize them)
  • Allow specifying whether you are using an AVR, and if so, what the sub out gain step, distance step and distance units are
  • If "using AVR" is chosen, the filter reports say how much to adjust the AVR distance and gain, using the chosen distance units
  • Changed allowable gain limits from +/- 15 dB to +/- 30 dB per request of Markus
  • Added "Rearrange gains" and "Rearrange delays" features for sub channels. These will be explained in a later post.

The download link is the usual: http://diy-audio-engineering.org/web_dl/apps/mso.zip

Last edited by andyc56; 11-18-2015 at 08:25 AM.
andyc56 is offline  
post #88 of 797 Old 11-19-2015, 01:16 AM
AVS Forum Special Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 9,947
Mentioned: 115 Post(s)
Tagged: 0 Thread(s)
Quoted: 4254 Post(s)
Liked: 1654
Quote:
Originally Posted by andyc56 View Post
404?

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
markus767 is offline  
post #89 of 797 Old 11-19-2015, 06:27 AM
AVS Forum Special Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 1,191
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 408
Quote:
Originally Posted by markus767 View Post
404?
That's strange. I re-uploaded it. Should be working now.
andyc56 is offline  
post #90 of 797 Old 11-19-2015, 06:36 AM
AVS Forum Special Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 9,947
Mentioned: 115 Post(s)
Tagged: 0 Thread(s)
Quoted: 4254 Post(s)
Liked: 1654
Quote:
Originally Posted by andyc56 View Post
That's strange. I re-uploaded it. Should be working now.
Yes, download working again.

I don't see "Measurement Associations" anymore in the tree view although they are usable to the optimization code.

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
markus767 is offline  
Sponsored Links
Advertisement
 
Reply Subwoofers, Bass, and Transducers

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