QQ SpecWeb Helper App

QuadraphonicQuad

Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Wowowowowow!!!!!!!!! I'm completely blown away by this, and have been particularly lucky, hitting pure gold on my first attempt at an encode!

I have a copy of the first CD pressing of Kate Bush's "Hounds of Love" and always thought that side 2, collectively known as "The Ninth Wave" (if you didn't know) would be perfect for surround. Well, it turns out that Specweb/helper does a truly astonishing job of converting it. :) I'm amazed by how good it is and found myself shocked by where some of the parts were coming from!

Kate's vocal is mostly in the centre, but the effects, backing vocals and some of the key instrumentation end up almost perfectly positioned in the surround field. It's almost as if someone has discretely mixed it. For example, the "wake up" calls on Waking the Witch come discretely from alternate rear and front speakers and the chopped vocal is a complete mind-f*ck (as it should be.) But what surprised me most was Jig of Life where one of the fiddles is rear left and another is rear right, then the "I put this moment here..." vocal line has to be heard to be believed.

If you have a copy of this give it a go. I only used the default helper settings, though I trimmed up my rears by 1db on my streaming player. To make the piece continuous I edited ripped flacs of the whole of the second side together into one long track using Audacity, but this isn't essential.
 
attachment.php


I just did this one and it sounds great.

Dunno why I see two attached images :confused:
 

Attachments

  • mlmpp2.jpg
    mlmpp2.jpg
    64.3 KB
Wowowowowow!!!!!!!!! I'm completely blown away by this, and have been particularly lucky, hitting pure gold on my first attempt at an encode!

I have a copy of the first CD pressing of Kate Bush's "Hounds of Love" and always thought that side 2, collectively known as "The Ninth Wave" (if you didn't know) would be perfect for surround. Well, it turns out that Specweb/helper does a truly astonishing job of converting it. :) I'm amazed by how good it is and found myself shocked by where some of the parts were coming from!

Kate's vocal is mostly in the centre, but the effects, backing vocals and some of the key instrumentation end up almost perfectly positioned in the surround field. It's almost as if someone has discretely mixed it. For example, the "wake up" calls on Waking the Witch come discretely from alternate rear and front speakers and the chopped vocal is a complete mind-f*ck (as it should be.) But what surprised me most was Jig of Life where one of the fiddles is rear left and another is rear right, then the "I put this moment here..." vocal line has to be heard to be believed.

If you have a copy of this give it a go. I only used the default helper settings, though I trimmed up my rears by 1db on my streaming player. To make the piece continuous I edited ripped flacs of the whole of the second side together into one long track using Audacity, but this isn't essential.

Don't have that one but converting The Kick Inside now with hopefully the same results :banana:.
 
For my second attempt I gave Steve Hillage's "Green" a go and it's pretty mind blowing, too. :cool:

It up-mixes quite heavily to the rears, so I think I will re-do it with a 1.5db cut on them. It also puts more percussion in the rears than I would ideally like (anyone know if it's possible to "tame" that???) but overall it sounds amazing :banana: In particular Sea Nature, Ether Ships, UFO over Paris and Leylines to Glassdom sound super-trippy, with spacey synth arpeggios bouncing around the rears and Steve's guitar solos in the front.

I'm going to have to give some Ozric Tentacles a go now...
 
I hadn't upmixed anything for a while, well actually I needed to download the recent version as I had the first version. While in honesty, I didn't think Sabbath would upmix very well...and IMO, it didn't. Also, I've had this happen before with some other upmixed I tried, but in the Sabbath upmix it creates some odd "warbling" sound in the rear speakers. I think it's been described as swooshing. To me it sounds like warbling. I think I'll try something else that is "less dense" as Sabbath. :)
 
It also puts more percussion in the rears than I would ideally like (anyone know if it's possible to "tame" that???)

You guys should probably post this stuff in the SpecWeb thread, since SpecWeb is doing the actual upmixing...

Anyway, SpecWeb is spreading the stereo mix across the 5.1 sound space, so if something is heavy full right or left in the original stereo, it will be "heavy" in the rears in the upmix. Short of trading the rear channels with the fronts (outside of SpecWeb) you can't really get away from that (well, maybe try Diagonal in ArcTan, instead of Across) but you can play with the width controls (image, center and front) to shift things that aren't really hard left and right in the original. Also Slice, vs. ArcTan, is a completely different algorithm that might be better on some songs.

If you have 5.1 output on your computer you can use the SpecWeb Play icon to hear all this stuff in realtime, but you may have to edit the icon properties to change -P1 to -Px, where x is the device number of your 5.1 sound card. You get that number from the command line options that SpecWeb prints before the conversion starts. Drop a song on the the regular SpecWeb icon to see that stuff.

Oh Yeah QQHelper has a play button in it as well, but at the moment there isn't a way to use it if your 5.1 device isn't "1". That will be fixed in future versions (if not already?).
 
You guys should probably post this stuff in the SpecWeb thread, since SpecWeb is doing the actual upmixing...

Anyway, SpecWeb is spreading the stereo mix across the 5.1 sound space, so if something is heavy full right or left in the original stereo, it will be "heavy" in the rears in the upmix. Short of trading the rear channels with the fronts (outside of SpecWeb) you can't really get away from that (well, maybe try Diagonal in ArcTan, instead of Across) but you can play with the width controls (image, center and front) to shift things that aren't really hard left and right in the original. Also Slice, vs. ArcTan, is a completely different algorithm that might be better on some songs.

If you have 5.1 output on your computer you can use the SpecWeb Play icon to hear all this stuff in realtime, but you may have to edit the icon properties to change -P1 to -Px, where x is the device number of your 5.1 sound card. You get that number from the command line options that SpecWeb prints before the conversion starts. Drop a song on the the regular SpecWeb icon to see that stuff.

Oh Yeah QQHelper has a play button in it as well, but at the moment there isn't a way to use it if your 5.1 device isn't "1". That will be fixed in future versions (if not already?).

Good point about using the correct thread. Will do so from now on (Admins - please feel free to move my posts about what's worked, what hasn't.)

And thanks for the tips, too! :)

...unfortunately I don't have any 5.1 on my PC (yet) so have to encode and play on separate systems then go back and tweak. Slightly fiddly, but great fun!
 
I will see if I can move the SpecWeb related posts to the correct thread later today. Great to hear of people's experiences, especially from Neil who is a pro. audio engineer!

I hadn't upmixed anything for a while, well actually I needed to download the recent version as I had the first version. While in honesty, I didn't think Sabbath would upmix very well...and IMO, it didn't. Also, I've had this happen before with some other upmixed I tried, but in the Sabbath upmix it creates some odd "warbling" sound in the rear speakers. I think it's been described as swooshing. To me it sounds like warbling. I think I'll try something else that is "less dense" as Sabbath. :)

Gene. Have you tried using Helper to Oversample your tricky audio? It definitley removes a lot of the warbling/swooshing etc. also experiment with leaving the MCH at a higher sampled rate as that may affect the final sound. Just experiment a bit.

Neil, if you have better Oversampling software (Helper uses Sox) you may like to experiment with that by Oversampling your stereo files first then run them through SpecWeb (you can still use Helper but uncheck 'Use Oversampling') and report any differences you find. I've been thinking about trying another Oversampling tool in Helper, so love to hear your thoughts/experience.
 
... Oh Yeah QQHelper has a play button in it as well, but at the moment there isn't a way to use it if your 5.1 device isn't "1". That will be fixed in future versions (if not already?).

Helper 1.2 will add play device selection (coming soon)
 
I will see if I can move the SpecWeb related posts to the correct thread later today. Great to hear of people's experiences, especially from Neil who is a pro. audio engineer!

...sorry to disappoint but you may have your "Neil"s mixed up. Neil Wilkes is the pro. I'm just an amateur, albeit with some experience of writing, recording, mixing, engineering and more recently play testing and assisting with mixes, but, alas, I'm still very much in the amateur bracket.
 
Garry,

can I ask a question about workflow and the way the .ini files work, please?

I quite enjoy using the specweb gui just to try out flac rips of various tracks to see how they sound and tweak the .ini settings on a track by track basis without committing time to a full encode.

I would like to do this and save multiple .ini files in the format <trackname>.ini stored in the same directory as the source flac and then use the helper to convert each individual source/.ini combination in batch later. This seems to be implied by what you've said about how .ini files are handled "1.1 uses the 'current' SpecWeb.ini file for all SpecWeb's conversion settings, with the same rules of priority (see SpecWeb's manual). So you need to make your SpecWeb conversion settings in the appropriate SpecWeb.ini file, if you have multiple files." but doesn't actually seem to work that way in practice. according to the text log of what's happening it's picking up my .ini from the specweb starting directory, not the <trackname>.ini in the source directory. Bug, maybe?

I'm also thinking that, wrt this workflow, having an option to leave the over-sampled source in situ and use that in preference to the source if found may be useful, though it would mean changing the naming of the temporary file. Another advantage of this is that it may help if/when using an alternative means of oversampling? Just store the output oversampled file in the source directory using the correct naming convention.

Cheers, Neil
 
Garry,

can I ask a question about workflow and the way the .ini files work, please?

I quite enjoy using the specweb gui just to try out flac rips of various tracks to see how they sound and tweak the .ini settings on a track by track basis without committing time to a full encode.

I would like to do this and save multiple .ini files in the format <trackname>.ini stored in the same directory as the source flac and then use the helper to convert each individual source/.ini combination in batch later. This seems to be implied by what you've said about how .ini files are handled "1.1 uses the 'current' SpecWeb.ini file for all SpecWeb's conversion settings, with the same rules of priority (see SpecWeb's manual). So you need to make your SpecWeb conversion settings in the appropriate SpecWeb.ini file, if you have multiple files." but doesn't actually seem to work that way in practice. according to the text log of what's happening it's picking up my .ini from the specweb starting directory, not the <trackname>.ini in the source directory. Bug, maybe?

I'm also thinking that, wrt this workflow, having an option to leave the over-sampled source in situ and use that in preference to the source if found may be useful, though it would mean changing the naming of the temporary file. Another advantage of this is that it may help if/when using an alternative means of oversampling? Just store the output oversampled file in the source directory using the correct naming convention.

Cheers, Neil

Hi Neil

Using SpecWeb Play to tweak settings is a great way to tune your results. That was a good idea implemented by zeerround.

Your proposed method to save the ini to same name as the stereo should work, but I realise now that Helper actually renames the over sampled file so, of course, SpecWeb does not 'see' the ini. So I need to do a fix for this scenario. I will do that within 24hrs.

Your suggestion about leaving any over sampled stereo file (currently a temp file only that's deleted) could be done easily with a new option. In fact, the solution to this and the first (ini) issue, will be to do all the Oversampling in a sub-folder of the source folder, use the same file names for over sampled stereo as the original stereo (now in sub folder) and copy any ini files to the Oversample folde, now SpecWeb works as designed.

I can add an option to 'delete temporary oversampled files' (default 'checked'), you can uncheck to keep all your over sampled stereo files (If required)

I'm planning to add a new 'Helper log' to record all Helper SpecWeb conversions and store the ini settings that were used for each track individually. Users will have a new view to see all their conversions and the ini settings used, and rerun a conversion with those settings or with new settings. That will be done in Helper 1.2, maybe ready by the weekend...

THX
Garry
 
I just updated the Helper installer to fix a couple of issues and add requested functionality:

Added:
Click on any queue columns to sort the queue
Now support Drag and Drop stereo audio files onto the queue
Added a new option to allow users to keep their oversampled stereo files (if required) - See Preferences - Oversampling section - Delete is the default)

Fixes:
Conversion with Oversampling in previous versions did not honor SpecWeb's ini (settings) file priority & ignored 'song title'.ini files (fixed)
Helper's 'Play' command had same ini issue as above
The INI file Helper did not allow gains less than -6db or greater than +6db (now increased range to -9 to 9db - except for LFE and C where min is -110db)

Latest version can be downloaded from the link on first post.
 
I just want to give a huge shout out of thanks to HomerJAU aka Garry for his sterling work on the QQ Helper and Zeerround for bringing SpecWeb to QQ.

I have been trying out (and suggesting) a few changes for Garry over the last week and today I used the Helper and SpecWeb to make a huge and very impressive sounding surround version of Eloy's "Planets" album, with very little effort at all. :)

There are still a few niggles and tweaks to be done to the software, but I'm finding that the Helper makes the whole business of over-sampling (for quality) then editing/tweaking album and song specific .ini files and rendering surround versions very much easier than it would otherwise be.

This whole Specweb project is as close to the real deal of DIY surround as most of us are likely to get and I have found the results to be very satisfying and surprisingly easy to get great results with a little practice. I think that their efforts deserves our support and gratitude. How lucky we are to have such talented guys as Garry and Zeerround making it possible to "home-brew" exceedingly convincing surround versions of our favorite albums.

So thanks, guys! :)
 
I updated the first post for Helper 1.2 release (new link and info)

The new Parallel Task options are in Preferences.

With a complete SpecWeb conversion of a CD album (11 tracks in queue) my timing tests:
Old Conversion method as per Helper 1.1 (with oversampling): 7.00 minutes
New Parallel option with 4 concurrent tasks (with oversampling): 3.55 minutes (3 ½ minutes)
New with 6 concurrent tasks (with oversampling): 3.52 minutes

So there’s a point of diminishing return. My test PC is a quad core i7 with 16GB RAM and SATA 3 SSD drives for TEMP and Source/Output files, so others may not get similar results.

I suggest user try varying these settings to optimise for their PC hardware.
 
I think this info in the top post is outdated:

To update Helper: (if you already have installed any previous version of Helper): Please delete all previous QQ SpecWeb Helper folders and files. They are no longer needed and follow the download and install instructions above.

I think now you have to uninstall it from add remove programs, before you can install the new version. Otherwise it tells you "You can't install from here because it's already installed from another location".
 
I think this info in the top post is outdated:

I think now you have to uninstall it from add remove programs, before you can install the new version. Otherwise it tells you "You can't install from here because it's already installed from another location".

Thanks for the reminder. I've just updated the first post regarding installing new versions of Helper.
 
Back
Top