SPECWEB (Now 2.2)

[Simon A]

Yes that would be a bug with the VST failing to launch. Can you reproduce constantly with a particular song or something?

Nah it just happened once. I hasn't again since.

 

[zeerround]

If folks are having crashing when the VST is running/visible, then yeah I could see why converting with AM would be more stable (as the VST doesn't get displayed).

Re levels in 1.4; The defaults were changed to normonly gains and use the PreZag VST. What this means is that the original stereo mix panning is determining the levels between the channels, so good results assume your 5.1 system is calibrated so that all channels are equally load at the listening position.

If that is not case, (or you simply don't like the results with normonly) feel free to change norm only to "yes" and put your zag channel to channel levels in.

The defaults were never meant to be for every system every song, just an attempt to be the best sounding (given a calibrated system) for most songs.

"DKA" like quality results are always going to come with effort and love applied per song, but since even I don't have that patience, I keep working to automate, etc.

I'm also always open to input on default settings too.

Re: VST crashes. I'm looking at another solution for VST hosting. Hopefully that will be more stable an also allow of users to create there own VST chain, for instance I know people that have Nugen ISL would probably want to use that instead of the free master 5.1 limiter.

I'll also look at a 1.4.1 where you can turn off the display of the VST (or, you can do it now buy using AudioMuxer).

Thanks for the feedback!

 

[zeerround]

Sorry I can't specifically help you with your problem, but generally, I had much better luck with 1.3 version finishing the upmix conversion than with my limited time with 1.4. Rarely ever had any problems with 1.3. This new version seems to run into roadblocks and the albums end up messing up. A few songs will go fine but not the entire list of songs.

Can you help me understand what errors you get and/or where in the process things "mess up"?

Thanks

 

[JerseyEd]

Can you help me understand what errors you get and/or where in the process things "mess up"?

Thanks

Hi Z,
The error I was getting was "error deleting VST level files - error #: +2.00". If I copied a complete stereo album into the desktop icon sometimes one or more songs would crash. When the conversion was done I could redo the ones that crashed and they would convert ok, so it doesn't seem to be specific songs. No songs crashed using AudioMuxer.

Another thing I wondered about is that one album I converted had a few mono (left and right channels were the same) bonus tracks. In 1.31 the conversion produced 5 pretty equal levels in the 5 main speakers, but 1.4 put all the sound in the center speaker while the other 4 speakers had barely audible levels. Is there something that needs to be done by me concerning mono files?

 

[zeerround]

Hi Z,
The error I was getting was "error deleting VST level files - error #: +2.00". If I copied a complete stereo album into the desktop icon sometimes one or more songs would crash. When the conversion was done I could redo the ones that crashed and they would convert ok, so it doesn't seem to be specific songs. No songs crashed using AudioMuxer.

Yes I remembered your errors. I wanted to see if other people see other things. I don't think it has anything to do with the audio. Something unstable in the vst hosting code, or my crappy windows gui (vs. pure c/c++) skills.

Another thing I wondered about is that one album I converted had a few mono (left and right channels were the same) bonus tracks. In 1.31 the conversion produced 5 pretty equal levels in the 5 main speakers, but 1.4 put all the sound in the center speaker while the other 4 speakers had barely audible levels. Is there something that needs to be done by me concerning mono files?

I wouldn't expect any version of Spec/SpecWeb to do anything with mono other than put it all in C. You could try prepossessing with a stereo synthesizer (overlapping comb filters) but you end up with a "phase pressure" kind of sound that gets annoying after a while.

There are some people doing some pretty cool stuff via spectral editing of mono songs, making them stereo, and I guess you could then use SpecWeb from there.

If you want me connect you to that community I probably can, or google "mixerrog". Talk about a labor of love, we're talking hours and hours per song as you have to tell the software note by note (really sound by sound)what's a guitar, what's a vocal, etc.

That method could also be used to go from stereo to surround, but again a very manual and labor intensive process.

 

[JerseyEd]

I wouldn't expect any version of Spec/SpecWeb to do anything with mono other than put it all in C. You could try prepossessing with a stereo synthesizer.

Thanks for the reply.
My Bad... I kept referring to the previous conversions I'm using for comparison as being done with 1.31. After further investigation I've discovered that they were actually done way back with 1.2. That was the version that gave me the 5 equal levels. Since I wasn't really looking to convert the mono tracks to "discrete" 5.1 I'm more than happy with my previous conversions. One thing I tried in my 1.4 tests was just using AudioMuxer to up the Bit Depth to 24 on the mono. This gave me left and right channels without center or rears that were still mono and fit in with the flow of the album.

I was also wondering about switching between the various SpecWeb versions. I have kept the original downloaded zip files for versions 1.0, 1.1, 1.2, 1.3, 1.31, and 1.4. If I wanted to use 1.2 for a few things, would I just uninstall 1.4 and reinstall 1.2 or would I get "newer version available" messages?

My other idea was to take the .exe and .ini of each version and load them in the bin folder, renaming each as SpecWeb1.2.exe, SpecWeb1.2.ini, etc. Then if I wanted to use 1.2 I would remove the 1.2 from the .exe and .ini so that version would be executed (did I explain that ok?), or does each version have to use the .dll's, etc that came with each specific version? If that's the case, a variation would be to add the version numbers to multiple bin folders, again removing the version number for the bin folder I want to use.

These are the kind of things I think of when I'm stuck at home in a blizzard!!! :couch

 

[zeerround]

Thanks for the reply.
My Bad... I kept referring to the previous conversions I'm using for comparison as being done with 1.31. After further investigation I've discovered that they were actually done way back with 1.2. That was the version that gave me the 5 equal levels. Since I wasn't really looking to convert the mono tracks to "discrete" 5.1 I'm more than happy with my previous conversions. One thing I tried in my 1.4 tests was just using AudioMuxer to up the Bit Depth to 24 on the mono. This gave me left and right channels without center or rears that were still mono and fit in with the flow of the album.

@:

The core stereo to surround algorithms haven't changed since 1.0. The changes have been bug fixes or different features, different default settings. So, using older settings in newer versions (for the most part) should yield similar results.

The only way to get any sound in other than center, with mono input, is via blends or adjacent speaker. In that case you are just spreading the mono sound around.

I was also wondering about switching between the various SpecWeb versions. I have kept the original downloaded zip files for versions 1.0, 1.1, 1.2, 1.3, 1.31, and 1.4. If I wanted to use 1.2 for a few things, would I just uninstall 1.4 and reinstall 1.2 or would I get "newer version available" messages?

My other idea was to take the .exe and .ini of each version and load them in the bin folder, renaming each as SpecWeb1.2.exe, SpecWeb1.2.ini, etc. Then if I wanted to use 1.2 I would remove the 1.2 from the .exe and .ini so that version would be executed (did I explain that ok?), or does each version have to use the .dll's, etc that came with each specific version? If that's the case, a variation would be to add the version numbers to multiple bin folders, again removing the version number for the bin folder I want to use.

These are the kind of things I think of when I'm stuck at home in a blizzard!!! :couch

Just run them each out of their own folders, and or make shortcuts for each and rename and put those somewhere handy.

Note that the various helpers are only going to work with the version they were made for.

Don't expect AudioMuxer to work with anything other than 1.4, for instance. You may get some results, but there are definitely things that don't work as they are supposed to.

Again, if you want results like 1.2 whatever, you can likely get them by editing a 1.4 ini file to have the settings from 1.2.

That's a generalization, but mostly accurate.

Oh, and yeah, you are going to get the version popups if you use older versions. You can add the -q flag (quiet) in your shortcut (or when it stops to ask you for parameters) to stop popups in general.

 

[himey]

Can you help me understand what errors you get and/or where in the process things "mess up"?

Thanks

One problem I am getting with 1.4 is an output device error. Never got that before using 1.3.

Using QQ SpecWeb helper for both versions.

Solved problem by not listening to music (foobar) during conversions, and I now am using Audiomuxer. I am 1 for 1 doing full albums, as long as I don't open and play tracks simultaneously in Foobar2000.

Version 1.3 and QQ helper doesn't have that limitation for whatever reason. Not a big deal.

 

[himey]

https://en.wikipedia.org/wiki/Lux_(album)

This turned out nicely using 1.4 version + Audiomuxer without any setting changes.

 

[JerseyEd]

@:

Just run them each out of their own folders, and or make shortcuts for each and rename and put those somewhere handy.

Thanks for your patience and answers. I now have enough info to keep me busy testing different situations for a while, inbetween shoveling snow. :xp:

 

[himey]

Thanks for your patience and answers. I now have enough info to keep me busy testing different situations for a while, inbetween shoveling snow. :xp:

Are you using a Gui/front end (qq SpecHelper or Audiomuxer)?

I am curious about the resample toggle in Audiomuxer...

For 44.1/88 conversions I am using 176. For 48/96 I am selecting 192. QQ helper seems to automatically upsample. Have you decided on one selection or another?

I have a few albums going through the process now and all is good with 1.4+Audiomuxer.

Random Access Memory, Homework, and Discovery are nearly finished.

Still curious why the last version (1.3), you could load up finished songs in foobar, and listen, with more songs still processing, and not get an error.

 

[zeerround]

Are you using a Gui/front end (qq SpecHelper or Audiomuxer)?
Still curious why the last version (1.3), you could load up finished songs in foobar, and listen, with more songs still processing, and not get an error.

I'm not understanding exactly what you are doing and which program is given an error.

Can you provide more details?

Thanks

 

[himey]

I'm not understanding exactly what you are doing and which program is given an error.

Can you provide more details?

Thanks

Sorry. If you reread my posts there are no errors unless I am playing back finished songs from unfinished conversions in the Cue (qq helper or Audiomuxer). Only using 1.4. Seems 1.4 needs output device while 1.3 doesn't.

Just a minor problem on my pc which isn't a big deal as I mentioned in my other post.

If I play files in foobar while conversions are still running, I get an error using 1.4 and not using 1.3.

 

[zeerround]

Sorry. If you reread my posts there are no errors unless I am playing back finished songs from unfinished conversions in the Cue (qq helper or Audiomuxer). Only using 1.4. Seems 1.4 needs output device while 1.3 doesn't.

Just a minor problem on my pc which isn't a big deal as I mentioned in my other post.

If I play files in foobar while conversions are still running, I get an error using 1.4 and not using 1.3.

So you are saying foobar2000 is reporting the device in use, while SpecWeb 1.4 is running?

And that didn't happen with 1.3?

What is your default sound device?

What device (and driver, MME, Direct, ASIO) are you using in foobar2000, that it is complaining about?

What happens if you change your ini setting for useprezagvst to no (or can you play via foobar when a VST is not in use but other parts of SpecWeb are running)?

 

[JerseyEd]

Are you using a Gui/front end (qq SpecHelper or Audiomuxer)?

Traditionally I've just been copying the files into SpecWeb and letting it rip using the defaults, not using SpecHelper or AudioMuxer. I've tried changing the levels on one of my tests using the SpecWeb Play, but can't really tell what's happening in real time because the computer I'm using doesn't have surround capabilities. I saved the .ini of the change but have no idea what to do with it . My main object right now is to try files in each of the versions and see what differences they produce.

 

[zeerround]

I saved the .ini of the change but have no idea what to do with it .

Ini files can be in four different places:

1) In the same directory as SpecWeb.exe. This will affect SpecWeb globally, unless one of the below ini files is present

2) In the “starting directory” of the SpecWeb command. If you are running SpecWeb from a shortcut, the “starting directory” is controlled by the shortcut properties (Right Click  Properties). If you are running SpecWeb from a command prompt, the starting directory is the directory you were in when you invoked SpecWeb. If you are running SpecWeb as a foobar2000 converter, the starting directory is the directory containing the file foobar2000 is converting.

SpecWeb will look for a file named “SpecWeb.ini” in the starting directory. If no other ini files are found in the below cases, a SpecWeb.ini file in the starting directory will control settings (unless overridden by command line options or during live playback).

3) A file named “SpecWeb.ini” in the same directory as the input file being converted. If no other ini files are found in the below case, a SpecWeb.ini file in the input file directory will control settings (unless overridden by command line options or during live playback).

4) A file with the same name as the input file, except with an “.ini” extension, in the same directory as the
input file will control settings (unless overridden by command line options or during live playback).

 

[himey]

So you are saying foobar2000 is reporting the device in use, while SpecWeb 1.4 is running?

And that didn't happen with 1.3?

What is your default sound device?

What device (and driver, MME, Direct, ASIO) are you using in foobar2000, that it is complaining about?

What happens if you change your ini setting for useprezagvst to no (or can you play via foobar when a VST is not in use but other parts of SpecWeb are running)?

SpecWeb stops and says device in use. Never happens with 1.3, only 1.4. Sorry I don't have the answers to all your questions right now. I have just stopped using foobar while 1.4 is running. Device is WASAPI in Foobar. I will check if VST is the problem next round of converting.

 

[deepsky4565]

Would it be possible to use this with JRiver Media Center for live upmixing? It can use VST plugins. Thanks.

 

[zeerround]

Would it be possible to use this with JRiver Media Center for live upmixing? It can use VST plugins. Thanks.

Specweb is not a vst itself.

However, you can run SPEC inside the Plogue Bidule VST, and run that inside of JRiver.

You will need to buy a copy of Plogue Bidule. You can test with a demo version of PB for ~3months (depending on when the last beta version was posted).

My stuff and instructions are here:

www.surroundbyus.com/sbu/download/Spec 4.6 for Plogue VST.zip

PB Demo stuff is here:

http://www.plogue.com/bidule/latest/

If you decide to go that way we should start another thread (under Stereo to Surround Upmixing) to cover SPEC inside PB VST.

OR, if you read through the SpecWeb manual there are instructions for how to use SpecWeb live. Depending on your sound hardware (or if you buy a virtual sound cable) you may be able to play from J-River trough SpecWeb in live mode.

SpecWeb can play live from one sound device as an input to a multichannel sound device output.

 

[deepsky4565]

Thanks for that detailed reply. I'm just starting to play around with it.