Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[BradC]

OK, I think I could do the same when my RME HDSP card will be installed. 
If the Plogue bidule uses the RME input channels  (which consist in a sweep coming from the REW output setting and has been sent to one RME output channel) and MC playback is directed to an other RME output with corrections made in the VST plugin?
Is it that simple?
I could use the MC EQ on top of the plogue bidule or it as to be the plogue bidule corrections only?

A bit confusing!  I will have to try it in real.

read the RME manual (it's quote long). It explains how to do the loopback in the mixer and set the mixer routing
The open source convolver has a VST version. Kelly industries have a free bass management VST and delay can be done natively in plogue bidule.

See here for a detailed explanation
http://www.martinloganowners.com/forum/showthread.php?7342-How-much-can-a-computer-do&p=82515

[Mitchco]

Hey Matt,

Thanks for this, it works perfectly. 

As an enhancement, it would be great to to be able to manually adjust the overall gain level similar to the replay gain slider in ConvolverVST.

Here is the issue.  In my case, using Audiolense generated filters, I lose about 10 dB or so of gain.  On most tunes, not a big deal, but with tunes that have a wide dynamic range like The Police Synchronicity or Dire Straits Brothers In Arms or Peter Gabriel Security, the gain is too low.  If I watch the peak level in Convolution, it hovers around 10% and sometimes hits 15% peak.  Even with my hottest mastered tune playing in my collection, I have not seen peaks over 60%.

When I used to use Convolver VST, I would take my hottest master and manually adjust the replay gain so that it was around 90% or so peak, but not clipping.

The output of my sound card goes into a passive preamp, so I can't make up the gain there.  I also have the normalize filter volume turned on.  I know I could enable the EQ or other VST plugin's that have gain controls, but...

Any chance an enhancement like that could be made?  I don't know if others are in a similar spot?

PS.  A second enhancement would have automatic filter switching when the sample rate changes between tunes.  This one probably a lot more work though...

Again, thanks for a great music player!

Cheers, Mitch

[Trumpetguy]

Mitchco, you can use Volume Levelling DSP plugin and add a fixed number of dBs. Normalize filter volume in Convolution plugin also works really well for me. Well, sometimes it creates short term clipping. Then you can see MC turning normalization level down slowly.

[Mitchco]

Mitchco, you can use Volume Levelling DSP plugin and add a fixed number of dBs. Normalize filter volume in Convolution plugin also works really well for me. Well, sometimes it creates short term clipping. Then you can see MC turning normalization level down slowly.

Thanks Trumpetguy.  I will give it a go.  I am also trying Blue Cat's Gain Suite, that is part of this free plug-in bundle: http://www.bluecataudio.com/Products/Bundle_FreewarePack/  Also includes a spectrum analyzer and interesting eq.

[Matt]

To adjust the volume of any or all channels, you might try DSP Studio > Parametric Equalizer > Adjust the volume.

Normally it's best to turn channels down instead of up, so you won't force clip protection to engage.  Using Internal Volume is also a good idea, since that gives you free headroom for convolution or volume adjustments.

[Trumpetguy]

Normally it's best to turn channels down instead of up, so you won't force clip protection to engage. 

Right. But MC's normalization does turn the channels up to meet some criterion for max 0dB? Which in turn sometimes leads to clip protection engaging. Or have I misunderstood something fundamental here?

Using Internal Volume is also a good idea, since that gives you free headroom for convolution or volume adjustments.

I have MC volume disabled since I use the Lynx mixer for volume adjustment (and the straight to power amps). Would setting the Lynx mixer to 0dB and using Internal Volume in MC give any improvement regarding headroom? I cannot see how, but if it is so, an explanation would be good.

Such a solution seems a bit risky, though, since other applications may pass through the Lynx driver at full 0dB signal strength.

[mojave]

Such a solution seems a bit risky, though, since other applications may pass through the Lynx driver at full 0dB signal strength.

I set my motherboard soundcard as the default audio device in Windows. This way no other applications can access my Steinberg UR824. I then use the loopback in JRiver if I want sound from another application.

A few months ago I was listening to music and noticed that some detail missing in the music. I checked my JRiver settings and couldn't find anything wrong. I finally realized that one of my kids had turned down the master volume control on my Steinberg using the big knob on the front. I turned it back up to 0 and turned down the internal volume and the detail was now present again. It was sort of a double blind test since I didn't even know I was being tested.

There were some changes to internal volume in 17.0.148 that now make it even better. It is now possible to set the Reference Level and to cap the maximum internal volume:

2. Changed: Volume settings are available in Options > Audio > Volume (and also still available in Menu > Player > Volume or by clicking the icon next to the volume slider).
3. NEW: Added 'Options > Audio > Volume > Maximum volume' for enforcing a maximum level that Media Center will be capable of setting.
5. NEW: Added 'Options > Audio > Volume > Internal volume reference level' for specifying what volume level is shown as +0.0dB (all other volumes will be reported relative to that reference level).
6. Changed: The OSD and volume slider in Theater View use the same volume display text as other parts of the program (improves support for 'bitstreaming' display, decibels for internal volume, etc.).

[Matt]

I have MC volume disabled since I use the Lynx mixer for volume adjustment (and the straight to power amps). Would setting the Lynx mixer to 0dB and using Internal Volume in MC give any improvement regarding headroom? I cannot see how, but if it is so, an explanation would be good.

Using Internal Volume and enabling Volume Protection would be better.

I just added a little explanation on the Volume wiki:
http://wiki.jriver.com/index.php/Volume#Internal_Volume_Headroom

[Trumpetguy]

Now I have set up my system with Internal volume and Lynx mixer at 0dB. And - set default device to some audio device not connected to anything. I'll give it a try.

Is it correct that that the signal chain should be
Adjust x dB for Replay Gain
Adjust x dB for internal volume
Room Correction (reduces all channels by 10dB except LFE)
Convolution

?

Can I send the bill for new speakers to JRiver when they blow....;-) ?

[Matt]

Now I have set up my system with Internal volume and Lynx mixer at 0dB. And - set default device to some audio device not connected to anything. I'll give it a try.

Is it correct that that the signal chain should be
Adjust x dB for Replay Gain
Adjust x dB for internal volume
Room Correction (reduces all channels by 10dB except LFE)
Convolution

That seems good.

You probably know this, but for anyone else following along, the order of volume effects is not relevant:
http://wiki.jriver.com/index.php/Audio_Bitdepth#Bit-Perfect

Can I send the bill for new speakers to JRiver when they blow....;-) ?

I've used the program this way for a long time, and it works great.

Just be sure to select 'Volume Protection'.

[Trumpetguy]

That seems good.

You probably know this, but for anyone else following along, the order of volume effects is not relevant:
http://wiki.jriver.com/index.php/Audio_Bitdepth#Bit-Perfect

Thanks, I was wondering. I knew, but have forgotten, thanks for the link.

I've used the program this way for a long time, and it works great.
Just be sure to select 'Volume Protection'.

The way I have handled volume before have also worked fine, which explains why I have never bothered to make changes. Now I wathced a movie last night using Internal volume and it worked perfectly (as I would expect from MC). The nice thing is also that the volume bar appears on screen. Using the Lynx mixer I always had to bring it up on the screen to see the volume level.

The only negative thing is that volume lags behind the adjustment by a second or so. Is there a technical reason why, or is this a design choice? It feels a bit unnatural, and different from any volume control I have used before.



Volume protection is definitely selected.

[Matt]

The only negative thing is that volume lags behind the adjustment by a second or so. Is there a technical reason why, or is this a design choice?

My guess is your convolution filters add the one second delay.

Reducing hardware buffering might help a bit (Options > Audio > Output mode settings...).  With ASIO and a fast computer, it's fine to use a small value like 0.10 seconds of buffering and uncheck 'User large hardware buffers'.

[tiggerkater]

Hello to everybody,

I posted my question in the "media network" forum, but perhaps somebody here has an answer:

I want to use JRiver as a upnp server with enabled convolution. Via LAN I would like to send the convoluted stream to a network music renderer like a LinnDS or a Sonos streamer.

Right now, I am able to stream, but convolution does not work.

Is this setup possible with JRiver? Are there any other people out there trying to achieve this? With foobar, it is possible to send the convoluted stream to a renderer, so I thing it should be with JRiver, too. Right now, I did not figure out, how
(I would really like to convolve with JRiver, because of its 64bit convolution system - which should sound better than the foobar solution.)

Thanks and all the best

tiggerkater

[Matt]

Hi tiggerkater.  It's not currently possible to use the full DSP Studio with DLNA.

I agree it would be nice.

[tiggerkater]

Thansk Matt for your fast reply. I am sorry to read this, I hoped, that I was not able to configure it right. But it seems that JRiver has not implemented this feature so far.

Perhaps it will be, soon?!

all the best

tiggerkater

[MartinG]

Just to make a tiny refresher in this discussion:
+1
support of different filters for different samplerates

Martin

[Hordor]

Just adding +1 to support of different filters for different samplerates.

Would be nice to be able to switch between to setups for comparing filters via hot button.

Uwe

[v_erich]

+1
No Upsanpling should be possible.

Regards,
Erich

[TheLion]

Matt,

my newest set of filters use 66k taps for all 7 speaker channels and 131k taps just for subwoofer/LFE. I just want to make sure that mixed filter resolution doesn't cause a problem with your convolution engine. Is this a problem?? Thanks!

[TheLion]

Matt,

I guess I just answered my own question -> when using a filter with different filter resolution for the channels (like mixed 66k and 131k taps) the convolution engine doesn't compensate for this - the result is that the channels with 131k taps play delayed to the channels with 66k taps. So I will not use mixed filter resolution.

[Matt]

Matt,

I guess I just answered my own question -> when using a filter with different filter resolution for the channels (like mixed 66k and 131k taps) the convolution engine doesn't compensate for this - the result is that the channels with 131k taps play delayed to the channels with 66k taps. So I will not use mixed filter resolution.

The engine should work fine with mixed length filters.

But if you have mixed delay filters, it will make a mess.  I suppose there could be an option to try to figure this out automatically, but that seems like a job for the program creating the filters.

[markuspac]

Hello Matt,

I would like to report on a strange effect that occur if using delay-definition in convolver config-file.

By reason that I wasn’t pleased using the JRiver convolver comparing to ConvolverVST in the same environment I’ve done some measurements to find out the cause.

For that purpose I’ve created a 3-way stereo Xover with lowered levels for bass (-6dB) and highs (-3dB). With this Xover I’ve carried out a measurement using the JRiver convolver and summing up the output of the 3 Xover channels to one channel using the mixing function of my soundcard (RME AES32). The result of this measurement and for comparison the computed signal (which is expected) see first attachment. The result appears correct.

Next I’ve used the same configuration as before but applied a delay definition in the convolver config file. The parameters were “0 0 22.68 22.68 45.35 45.35” which should delay the channels 3/4 by 1.000 samples and the channels 5/6 by 2.000 samples (with regard to channels 1/2 at 44.100 Hz). The result of this measurement and for comparison the computed signal can be seen in the second attachment. In contrast to the expectation the convolution result for the 1st (-1dB) and 3rd (+0,3dB) Xover channel has a different level. The delay itself was applied correctly.

Do you have any idea concerning the reason or a solution for that phenomenon/problem.

kind regards
Markus

[Matt]

@markuspac

Welcome.

Is there any chance the level differences are from normalization?  Would you be willing to uncheck 'Normalize filter volume' and test again?

Thanks.

[markuspac]

Hello Matt,

thanks for your reply. I’ve supposed also that this effect could be caused by normalization. However the “normalize file volume” was already unchecked and also “flat line overflows” was chosen. As another try I’ve reduced the replay-gain by 6 dB (using “volume leveling” DSP function). But at the end the measured curve looks exactly as posted before. Seems to be tricky.

regards
Markus

[TheLion]

Markus, that's an interesting find. Does Uli have any idea what could be wrong?

I noticed a rather strange behavior with delays and volume level/gain using "room correction" in MC. When you set the distances (=inverse delay) it also changes the volume level of a given channel - these two parameters interact. The more distance the more volume level a channel gets by default. The logic behind this seems to be that identical speakers will be volume matched just by setting the distances - more or less because room interaction/placement cannot be considered.

Matt, could this logic be present with delays in convolution config files as well? This would certainly make no sense.

On another matter: I will change my setup to fully active 3-way fronts. Therefor I will use up to 16 output channels in the future (7.2 with 9 channels active XO for the fronts). You have added support for such high channel numbers some time ago (I think Bernt wanted it for Audiolense support) which is great. My problem is that when using the parametric EQ and/or Room Correction (to e.g. set relative volume levels) we don't have access to any of the additional output channels. It would be great if this could be changed - Parametric EQ filters can be used for a large variety of tasks (calibration, testing,...). So when selecting 12,16,24 or 32 channels in output format it would be great if they show at least in PEQ (and probably Room Correction).

Thank you very much!

  

[Matt]

Markus, could you email your two sets of configuration files to matt at jriver dot com?  I'll take a look in the debugger and see if anything jumps out at me.

Thanks.

[markuspac]

Hello Matt,

thanks for your willingness to follow-up this topic. I've send you my config-files as requested.

In addition to the already reported tests I've carried out today a convolution test using filters with intrinsic applied delays and with no use of delay definition in the convolver configuration file. Interestingly the (negative) effect is the same as if using the usual filter and applying the delays in convolver configuration file. Really strange!

kind regards
Markus

[markuspac]

Hello Matt,

I’ve to apologize. Please do not continue investigating my issue at the moment. I’ve made some more double-checks with ConvolverVST and found out that the problem persists even with my existing ConvolverVST environment. I’ll give you an update if I’ve found out something.

regards
Markus

[eddyshere]

much of the community that I mention are not aware of what J River really is over another player like wmp. I think a bit of marketing over that way wouldnt be a bad idea. The community is slowly coming round to the idea that computers can play quality audio, but theres a whole lot of marketing thats battling the change (cables for example) who pay big money for a good magazine review. Ill admit that at one point I was one of them.

Ive been on a mission to build the most acurate/best audio system for 15 years, (thats how I got here) most of my efforts are in building speakers and amplifiers though. I have Ł50k in drivers ready to build my ultimate active speaker. (it is a private cinema though)

 Proper EQ combined with JMLC profile 90 deg corner horns on beryllium compression drivers, and active crossovers = no early reflections, and eq'ed late ones + sub 0.2% distortion at reference level across the whole spectrum. Its audio nervana if I can do it.

+1

[Waveformfidelity]

Markus,

The convolution engine is carrying out exactly what you specified:

“0 0 22.68 22.68 45.35 45.35”

woofer: 0ms delay added

mid:  22.68ms relative to woofer

tweeter:  44.35ms relative to woofer

This is clearly seen in time domain of your posted pics.

I emulate this in Cool Edit with Linkwitz-Riley 24dB/octave crossover with high pass /low pass attenuation: see attachment

[Waveformfidelity]

...the way the shelving displays depends on window size and location relative to the low pass and high pass components of the summation. 

Here is same time domain as above with shifted FFT window and new spectrum with spectrum from above pic:

[Matt]

I'd like to ask for recommendations on how to best support convolution where you have a different file for each sample rate.

Should we just support some naming convention, like adding " (44.1 KHz)", " (48 KHz)", etc. to the end of the filter files?  Or should the configuration file contain a list of filters?  Or something completely different?

Thanks for any help.

[markuspac]

Hello Matt,

the apparently problem of the JRiver convolution engine, if applying delays, could be clarified in the meantime. Uli (audiovero) gave me the hint that the effect only occurs in processing the pulse result of the logsweep-measurement. Like "Waveformfidelity" wrote it's a matter of the window size. After applying the windowing and cutting off most of the samples ahead the pulse except of the last 12.000 (resp. 6.000) the effect in frequency domain can be observed. If using 64k (resp. 32k) windowing the frequency response looks as it should. Summing up your JRiver convolution engine works absolutely correct.

regards
Markus

[lasker98]

Hi Matt,

When first starting with Audiolense filters, I was using the Convolver VST plugin, since this was before MC implemented native convolution. If I remember correctly, Convolver VST used a config file that was created at the time the DSP filters were created. If filters were created for multiple sample rates, ie; 44.1k and 96k for example, then the config file would be pointing to the location of those filters on the hard drive. I'm clueless on the technicalites of how this all works but I thought the intent was that Convolver VST plugin in MC would then use this config file to load the appropriate filter based on the sample rate of the file being played. This obviously never worked this way in MC, and as a newbie to all this I posted this question in the forums back in January this year:

"General / Media Center 17 / ConvolverVST Warning Message  on: January 01, 2012, 01:14:15 pm 
Hello,

I'm using convolvervst plugin with audiolense DSP filters. I'm trying to have the configuration file automatically switch for different sample rates. In my random playlists I may have a mix of 16-44.1, 24-88.2 and 24-96 files.
I seem to have the config file setup correctly to do this but the problem I have is when a song with a different file sample rate comes up, I get a pop up window that says "ConvoverVST Warning   Filter sample rate (4410096000Hz) is different from input sample rate (96000Hz)".
This would be the warning in an example where a 44.1k sample rate song was playing and the next song up was 96k. J River playback now stops and waits for the warning popup window to be closed. If not closed manually, after what seems like a set time (1 minute?, not sure) the window closes itself and playback stops completely.
Is there a way to have J River ignore that warning and play through? I use JR17 on a headless computer which makes it impractical to have to manually acknowledge and close the warning popup. If I'm logged into the computer remotely, I can close the popup warning when it shows up and playback continues onto the next song, automatically switching to the correct sample rate. I don't really see the purpose of that warning."

This may give you a bit more of an idea of the problem that occurred when trying to use config files and  different sample rate filters with Convolver VST and MC. The idea of the config file seems sound, but for whatever reason it didn't work.

Thanks,

Bill

[hermannreuter]

I am using JRiver for some years and Audiolense for generating convolution filters since half a year. My audio system and Audiolense is 2 channels stereo with no active crossovers etc. Recently I started to notice no more sound difference between convolution activated or deactivated. I did a room measurement with Carma 3.0 playing the frequency sweep through JRiver with convolution on and off - the graphs look the same. The DSP-windows for convolution shows: Processing 0 paths and speed factors from 110x to 220x depending on the source resolution and sampling frequency. I recall the effects of convolution were very distinct to hear some weeks ago.
What may be wrong or what may I do I wrong?

[Mikkel]

I am using JRiver for some years and Audiolense for generating convolution filters since half a year. My audio system and Audiolense is 2 channels stereo with no active crossovers etc. Recently I started to notice no more sound difference between convolution activated or deactivated. I did a room measurement with Carma 3.0 playing the frequency sweep through JRiver with convolution on and off - the graphs look the same. The DSP-windows for convolution shows: Processing 0 paths and speed factors from 110x to 220x depending on the source resolution and sampling frequency. I recall the effects of convolution were very distinct to hear some weeks ago.
What may be wrong or what may I do I wrong?

I had the same problem. Uninstalling JRiver and reinstalling it did the trick. Somewhere inbetween all the updates something had messed up the software.

[hermannreuter]

Thank you very much. I'll will give that a try tonight.

[AudioVero]

I'd like to ask for recommendations on how to best support convolution where you have a different file for each sample rate.

Should we just support some naming convention, like adding " (44.1 KHz)", " (48 KHz)", etc. to the end of the filter files?  Or should the configuration file contain a list of filters?  Or something completely different?

Thanks for any help.

Hello Matt,

a full convolution setup can be quite complex. Imagine a multi-channel multi-way setup including some additional crosstalk mix (e.g. adding some channels to a subwoofer output but also crosstalk cancellation). You may like to create a wav-file including all the filters for a given samplerate. But this also may become difficult, as the sequence of channels in a mutli-channel wav file does not necessarily correspond with the sequence of filters/outputs.

Thus it is IMHO best to handle each filter in a separate file in native format, e.g. double format, or as wav file (double precision is also possible).

By default Acourate creates crossovers and correction filters as dbl-files (but wav is also possible). For stereo applications the names are given as XO1L44.dbl, XO1R44.dbl, XO2L44.dbl, XO2R44.dbl, XO3L44.dbl ... and Cor1L44.dbl, Cor1R44.dbl, Cor2L44.dbl ...
For multichannel applications it makes sense to use corresponding abbreviations like C=Center, BL=Back left, BR=Back right, LF=low frequency, SL=Side left, TL=Top left ...
So a filter name like Cor1BC44.dbl can easily be identified as a filter for a back center channel.

So we have some classification:

a)
XO = crossover filter
Cor = correction filter
XTC = crosstalk filter

b)
1, 2, 3, ... -way identification, e.g. 1=bass driver, 2=midrange driver, 3=tweeter in a 3-way setup

c)
L, R, C, LF, BL, BR, SL, SR ... = multichannel identification

d)
44, 48, 88, 96, 176, 192, 352, 384 = samplerate

Especially d) can be used for an automatic search of filters depending on samplerate.

Of course it is not mandatory to follow such a specification of filenames. If MC will interprete a configuration file then a user can arbitrarily defines his own filenames. Then it is the user's responsibility to define the names, to set up the config file and to assure that the filters are available.

Configuration files as XML have become quite fashioned today. BUT IMO these files are difficult for human editing. So a file like the VST Convolver config file or an old-fashioned ini-file is better to read, edit and understand.

BR
Uli

[Matt]

44, 48, 88, 96, 176, 192, 352, 384 = samplerate

Especially d) can be used for an automatic search of filters depending on samplerate.

Should the configuration file call out each file for each sample rate?

Or should there be some generic naming scheme so the configuration file says:
XO1R[SampleRate].wav and we convert it to XO1R44.wav, XO1R48.wav, etc.

Or should the configuration say:
XO1R.wav and we automatically search for files with 44, 48, etc. at the end?

[mojave]

Why not just have 4-6 slots to load config files and allow the user to assign the sample rate to each file?

 

[AudioVero]

Take a given config file (VST Convolver), e.g.:

44100 2 2 0
0 0
0 0
C:\Filter\Cor1L44.dbl
0
0.0
0.0
C:\Filter\CorRL44.dbl
0
1.0
1.0

You can simply create your own internal configs by stripping 44 from the filenames and adding 48, 88, 96 ...
Then check the specified folder for the availability of the filter files.
In case you do not find a complete file set for a samplerate you may issue a warning and apply a filter samplerate conversion as already carried out.

Uli

[Matt]

Why not just have 4-6 slots to load config files and allow the user to assign the sample rate to each file?

I'm trying to keep it easy for a user.  Having several settings files and piles of config files to handle different sample rates seems complex.

[Matt]

You can simply create your own internal configs by stripping 44 from the filenames and adding 48, 88, 96 ...
Then check the specified folder for the availability of the filter files.
In case you do not find a complete file set for a samplerate you may issue a warning and apply a filter samplerate conversion as already carried out.

That's easy to code, but a little cryptic.

Is '44' at the end of the filename the magic marker?  What if it's 48, or 44.1, or 44kHz at the end?  Would it be better to use a marker that's more explicit like [SampleRate]?

[AudioVero]

That's easy to code, but a little cryptic.

Is '44' at the end of the filename the magic marker?  What if it's 48, or 44.1, or 44kHz at the end?  Would it be better to use a marker that's more explicit like [SampleRate]?

Anyway there must be a syntax for a config file. So it does not really matter what definition you like to use.
If you select a file like Cor1L44.dbl then you can copy and paste the filename. In case of Cor1L[samplerate].dbl you must edit it. So IMO it is more user-friendly to just use 44.
The application can also deny a config file in case of 48 or 44.1. Whereas 44kHz can be switched to 48kHz or 88kHz ... by a simple string replace.

Finally it is up to you how much effort you like to take for coding a user-friendly and fault tolerant application.

Uli

[Mitchco]

There is a config file specification at: http://convolver.sourceforge.net/config.html  If you scroll down to the end, there is a filter list specification that says:

"Filter list

In situations where you are using several different formats (eg, stereo 44.1kHz and 5.1 48kHz) it is convenient to be able to switch automatically between filters.

So a config file can also comprise of a list of filter specification config file names of the type described above; and WAV impulse response filenames one per line.  The filter used will then be the first to match the current sound source (in terms of number of input and output channels and sample rate).  This allows you to play both stereo and 5.1 sources, say, without having to change the config file."

Not sure if that helps or is the desired approach.

[nostro66]

Still no support for switching of filters. Please guys, JRiver is the best of audiophile players, and internal convoler is so great feature...
Thank you!

[Mitchco]

Still no support for switching of filters. Please guys, JRiver is the best of audiophile players, and internal convoler is so great feature...
Thank you!

My understanding is that this is implemented in version 18.  Maybe Matt can comment if that is correct...

[Matt]

Automatic switching has not been implemented in v18 yet. 

I want to add it. 

I asked for advice about how to best add it above, and the answer wasn't too definitive.  Maybe we just need to do something / anything, but I was hoping for an example of what works elsewhere.

If someone has a filter set that does rate switching with Convolver, please mail it to matt at jriver dot com, and we'll try to support the same format.

[mojave]

You could have a drop down list and let the user select whether they use Audiolense, Accourate, or another program. Audiolense files end with 441, 48, 882, or 96 after the last space. A user can also create custom sample rates such as 176.2 or 192 which I assume would look like 1762 or 192. It looks like Accourate uses 44, 48, 88, 96, 176, 192, 352, 384 as the marker. After the first convolution file is selected, JRiver could use the same filename with the matching sample rate at the end like AudioVero mentioned earlier.

[mojave]

If you want to make it easier, I'm sure Audiolense users wouldn't mind renaming their files to end with 44, 88, and 176 instead of 441, 882, and 1764. Then Audiolense and Accourate files would end with the same sample rate designation. As far as that goes, I could rename the files to 44.cfg, 48.cfg, 88.cfg and 96.cfg to make it really easy. Instead of using file names to differentiate filter sets, I could just use folders.

p.s. I purchased Audiolense XO last week.