Forums


some feature suggestions
#81
Thanks again for great updates and fixes. You rock. I had to do a review at KVR, I don't think I've ever bothered to do one for audio software, even though I've been working with them for 20 years or more..
Reply
#82
(10-31-2022, 06:50 AM)balanced Wrote: Thanks again for great updates and fixes. You rock. I had to do a review at KVR, I don't think I've ever bothered to do one for audio software, even though I've been working with them for 20 years or more..

Thank you! Smile
Reply
#83
Perhaps you are following this already, but here it is nonetheless. Open source plugin format CLAP seems to get supported by more plugins and hosts. I'm not a developer, but as I understand there are unofficial developer tools for creating CLAP plugins already. If this format will get more widely used and official developer tools released then support for it at some point would be great. There's a site that tracks software supporting CLAP:

https://clapdb.tech
Reply
#84
Also, how hard do you think it would be to integrate support for multichannel audio files and channels in Soudop at some point? Right now I'm working with a surround master by using Reaper and I was thinking that it would be quite a bit easier to use Soundop for this. It would perhaps broaden the user base too.

In addition to support for multi channel file editing this would surely require at least some channel mapping/remapping and metering tools. I realize this is not a single or easy feature and it might need considerable work and planning. However if basic multi channel functionality would be implemented then there are also external plugins which would help users in multichannel work. Just a thought.
Reply
#85
(11-03-2022, 01:43 AM)balanced Wrote: Perhaps you are following this already, but here it is nonetheless. Open source plugin format CLAP seems to get supported by more plugins and hosts. I'm not a developer, but as I understand there are unofficial developer tools for creating CLAP plugins already. If this format will get more widely used and official developer tools released then support for it at some point would be great. There's a site that tracks software supporting CLAP:

https://clapdb.tech

Thank you.
Reply
#86
(11-03-2022, 02:16 AM)balanced Wrote: Also, how hard do you think it would be to integrate support for multichannel audio files and channels in Soudop at some point? Right now I'm working with a surround master by using Reaper and I was thinking that it would be quite a bit easier to use Soundop for this. It would perhaps broaden the user base too.

In addition to support for multi channel file editing this would surely require at least some channel mapping/remapping and metering tools. I realize this is not a single or easy feature and it might need considerable work and planning. However if basic multi channel functionality would be implemented then there are also external plugins which would help users in multichannel work. Just a thought.

Thanks for the suggestion.

We have considered adding multi-channel support in the future. But we are not sure how many users will need it. And whether to support it only in the audio file editor or support it both in the audio file editor and multi-track editor.
Reply
#87
Hi again. I was playing with the new options for frequency analysis and comparing it to the "go to" analysis plugin I use. Here are suggestions for further improvement:

-set the default spectrum tilt to 4.5 so it would more closely resemble the perceived spectrum (hearing-oriented) than measured. Someone might argue that 3 would be a valid setting too. Voxengo span uses 4.5 as default and I have found it to be a very good reference value.

-It would be useful to make distinction between monitoring peak and average levels, at least when thinking about how you setup or design the measurement tool. As I understand this tool now shows peaks with adjustable drop rate? In some measurement tools you have a setting to monitor peaks which are then left to display permanently. It can be achieved with your tool by setting drop rate to 0. Often it's more useful however to monitor some kind of average value. Some analyzers have just one knob for adjusting rise and drop rates or setting the time period over which the value is calculated (if I understand right the way they work). Would it be possible to monitor average signal values per frequency just by adding a rise rate too? I'm asking because I'm not entirely sure how the visualizations are calculated in Span for example, but i suspect it works close to this way. When for example setting the rise and drop rates to extreme values (or extremely small in your analyzer's case) you should see a slowly forming frequency spectrum that just barely evolves over time. Finding the right rise/drop value can be a question of personal preference, but using slower or faster rates can be used for understanding different aspects of the signal too. I would for example like to use also values between 0 and 1 for setting the rise/drop rate of your analyzer in order to achieve a very slow averaging display but I suspect using decimals is not possible now.

-I have been trying the "scan selection" option you have but I'm not completely sure if it produces an analysis that represents peaks or averages? Which is it? I couldn't find this from manual. I believe seeing average representation of frequency spectrum over a selected time period is the more useful option, although seeing the peaks could be useful too.

-You could separate the peaks and averaging display just by showing a separate peak line over the average line. This could be used in non-real time "scan selection" and real time mode. Peaks could be reset either by restarting the playback or by pressing a button in real time mode?

-you might want to adjust the way spectrum is spread out / visualized: 0-100 Hz region should have a lot more width on screen than 10kHz-20kHz region. The lower the frequency the wider it should be spread on the screen. Then you can see whats happening with lower frequencies. You don't really often need to make precise corrections in 12-13 kHz region for example (if ever really), but you often might need to do that for 30-40Hz region. This is common practice so you could just see how this works in Span or some other measuring tool.

-it would make the tool more usable if one could adjust the smoothing value (dB/oct.) for frequencies so that they would not appear as precise as now, especially for higher frequencies. This is because it's very hard to perceive the overall balance of frequencies if the high frequencies show as a jungle of sharp spikes. Selectable values from no smoothing and 1/48 oct smoothing all the way to 1/1 oct smoothing can be used in order to tame the visualizing. Some of the more used values are 1/24, 1/12 and 1/6 oct smoothing, but freely adjusting this value gives you a lot more options to monitor the spectral balance.

-One very useful thing that I use in REW measurement software but I don't remember seeing in plugins, is variable smoothing of frequencies. This means using a variable dB/oct smoothing value that doesn't affect the lowest frequencies (so you still see them as precisely as possible) and starts to get smoother the higher you get to the frequencies. This would be very useful when setup correctly and produces a very readable and informative visualization of frequencies. For example using 1/2 or 1/3 oct smoothing for the 10kHz and up region and lessening gradually it to no smoothing at all for 80 or 100Hz and below might produce a very readable visualization

-Given the amount of options needed when setting up the frequency analysis tool for different tasks it would be useful to have a simple preset system. So that you could save all of the used values, name the preset and recall it from a drop down menu inside the tool from example. Feel free to contact me if you want to enhance the tool with these options and would want to have someone to make some basic presets for it.

Cheers Smile
Reply
#88
(11-15-2022, 12:44 AM)balanced Wrote: Hi again. I was playing with the new options for frequency analysis and comparing it to the "go to" analysis plugin I use. Here are suggestions for further improvement:

-set the default spectrum tilt to 4.5 so it would more closely resemble the perceived spectrum (hearing-oriented) than measured. Someone might argue that 3 would be a valid setting too. Voxengo span uses 4.5 as default and I have found it to be a very good reference value.

-It would be useful to make distinction between monitoring peak and average levels, at least when thinking about how you setup or design the measurement tool. As I understand this tool now shows peaks with adjustable drop rate? In some measurement tools you have a setting to monitor peaks which are then left to display permanently. It can be achieved with your tool by setting drop rate to 0. Often it's more useful however to monitor some kind of average value. Some analyzers have just one knob for adjusting rise and drop rates or setting the time period over which the value is calculated (if I understand right the way they work). Would it be possible to monitor average signal values per frequency just by adding a rise rate too? I'm asking because I'm not entirely sure how the visualizations are calculated in Span for example, but i suspect it works close to this way. When for example setting the rise and drop rates to extreme values (or extremely small in your analyzer's case) you should see a slowly forming frequency spectrum that just barely evolves over time. Finding the right rise/drop value can be a question of personal preference, but using slower or faster rates can be used for understanding different aspects of the signal too. I would for example like to use also values between 0 and 1 for setting the rise/drop rate of your analyzer in order to achieve a very slow averaging display but I suspect using decimals is not possible now.

-I have been trying the "scan selection" option you have but I'm not completely sure if it produces an analysis that represents peaks or averages? Which is it? I couldn't find this from manual. I believe seeing average representation of frequency spectrum over a selected time period is the more useful option, although seeing the peaks could be useful too.

-You could separate the peaks and averaging display just by showing a separate peak line over the average line. This could be used in non-real time "scan selection" and real time mode. Peaks could be reset either by restarting the playback or by pressing a button in real time mode?

-you might want to adjust the way spectrum is spread out / visualized: 0-100 Hz region should have a lot more width on screen than 10kHz-20kHz region. The lower the frequency the wider it should be spread on the screen. Then you can see whats happening with lower frequencies. You don't really often need to make precise corrections in 12-13 kHz region for example (if ever really), but you often might need to do that for 30-40Hz region. This is common practice so you could just see how this works in Span or some other measuring tool.

-it would make the tool more usable if one could adjust the smoothing value (dB/oct.) for frequencies so that they would not appear as precise as now, especially for higher frequencies. This is because it's very hard to perceive the overall balance of frequencies if the high frequencies show as a jungle of sharp spikes. Selectable values from no smoothing and 1/48 oct smoothing all the way to 1/1 oct smoothing can be used in order to tame the visualizing. Some of the more used values are 1/24, 1/12 and 1/6 oct smoothing, but freely adjusting this value gives you a lot more options to monitor the spectral balance.

-One very useful thing that I use in REW measurement software but I don't remember seeing in plugins, is variable smoothing of frequencies. This means using a variable dB/oct smoothing value that doesn't affect the lowest frequencies (so you still see them as precisely as possible) and starts to get smoother the higher you get to the frequencies. This would be very useful when setup correctly and produces a very readable and informative visualization of frequencies. For example using 1/2 or 1/3 oct smoothing for the 10kHz and up region and lessening gradually it to no smoothing at all for 80 or 100Hz and below might produce a very readable visualization

-Given the amount of options needed when setting up the frequency analysis tool for different tasks it would be useful to have a simple preset system. So that you could save all of the used values, name the preset and recall it from a drop down menu inside the tool from example. Feel free to contact me if you want to enhance the tool with these options and would want to have someone to make some basic presets for it.

Cheers Smile

Hi,

Thank you very much for the suggestions.

For the default spectrum tilt, we prefer to set it to 0 for non-professional users.

As for frequency scaling, you can choose in the shortcut menu. And you can adjust the visible frequency range with the mouse. Mouse wheel to zoom. Hold the middle button and drag or hold the Ctrl key + left button and drag to select visible range.

   

The scanning selection calculates the average spectrum.

We will check other feature requests and try to implement them in future versions.

Best regards.
Reply
#89
(11-15-2022, 12:44 AM)balanced Wrote: Hi again. I was playing with the new options for frequency analysis and comparing it to the "go to" analysis plugin I use. Here are suggestions for further improvement:

-set the default spectrum tilt to 4.5 so it would more closely resemble the perceived spectrum (hearing-oriented) than measured. Someone might argue that 3 would be a valid setting too. Voxengo span uses 4.5 as default and I have found it to be a very good reference value.

-It would be useful to make distinction between monitoring peak and average levels, at least when thinking about how you setup or design the measurement tool. As I understand this tool now shows peaks with adjustable drop rate? In some measurement tools you have a setting to monitor peaks which are then left to display permanently. It can be achieved with your tool by setting drop rate to 0. Often it's more useful however to monitor some kind of average value. Some analyzers have just one knob for adjusting rise and drop rates or setting the time period over which the value is calculated (if I understand right the way they work). Would it be possible to monitor average signal values per frequency just by adding a rise rate too? I'm asking because I'm not entirely sure how the visualizations are calculated in Span for example, but i suspect it works close to this way. When for example setting the rise and drop rates to extreme values (or extremely small in your analyzer's case) you should see a slowly forming frequency spectrum that just barely evolves over time. Finding the right rise/drop value can be a question of personal preference, but using slower or faster rates can be used for understanding different aspects of the signal too. I would for example like to use also values between 0 and 1 for setting the rise/drop rate of your analyzer in order to achieve a very slow averaging display but I suspect using decimals is not possible now.

-I have been trying the "scan selection" option you have but I'm not completely sure if it produces an analysis that represents peaks or averages? Which is it? I couldn't find this from manual. I believe seeing average representation of frequency spectrum over a selected time period is the more useful option, although seeing the peaks could be useful too.

-You could separate the peaks and averaging display just by showing a separate peak line over the average line. This could be used in non-real time "scan selection" and real time mode. Peaks could be reset either by restarting the playback or by pressing a button in real time mode?

-you might want to adjust the way spectrum is spread out / visualized: 0-100 Hz region should have a lot more width on screen than 10kHz-20kHz region. The lower the frequency the wider it should be spread on the screen. Then you can see whats happening with lower frequencies. You don't really often need to make precise corrections in 12-13 kHz region for example (if ever really), but you often might need to do that for 30-40Hz region. This is common practice so you could just see how this works in Span or some other measuring tool.

-it would make the tool more usable if one could adjust the smoothing value (dB/oct.) for frequencies so that they would not appear as precise as now, especially for higher frequencies. This is because it's very hard to perceive the overall balance of frequencies if the high frequencies show as a jungle of sharp spikes. Selectable values from no smoothing and 1/48 oct smoothing all the way to 1/1 oct smoothing can be used in order to tame the visualizing. Some of the more used values are 1/24, 1/12 and 1/6 oct smoothing, but freely adjusting this value gives you a lot more options to monitor the spectral balance.

-One very useful thing that I use in REW measurement software but I don't remember seeing in plugins, is variable smoothing of frequencies. This means using a variable dB/oct smoothing value that doesn't affect the lowest frequencies (so you still see them as precisely as possible) and starts to get smoother the higher you get to the frequencies. This would be very useful when setup correctly and produces a very readable and informative visualization of frequencies. For example using 1/2 or 1/3 oct smoothing for the 10kHz and up region and lessening gradually it to no smoothing at all for 80 or 100Hz and below might produce a very readable visualization

-Given the amount of options needed when setting up the frequency analysis tool for different tasks it would be useful to have a simple preset system. So that you could save all of the used values, name the preset and recall it from a drop down menu inside the tool from example. Feel free to contact me if you want to enhance the tool with these options and would want to have someone to make some basic presets for it.

Cheers Smile

Hi,

We have released a new version that supports smoothness, averaging, restarting, and minimum frequency in the Frequency analysis panel. And Presets are also supported. Please check it out.

   

Best regards.
Reply
#90
It's hard to describe how grateful this makes me feel. Tonight I have been searching for a standalone audio file analyzer that would allow me to do some of these things as I'm about to finish a project and I need to see the spectrum better. Real time meters won't do the trick as I'm working on a 30 minute file. I really did not find such software. Perhaps Process.audio Decibel would have done some of that but I do not want to use iLok or be connected to web at all times for a software to work.

Then I decided to check your version history and there it was. You make my day quite literally. Cheers & kudos! This makes Soundop much more useful for mastering purposes.
Reply