LEARN HOW TO USE BIDULE EXTENDED


In this guide we will teach you some things to make your layout more compact on
the screen and tell you more about how to work with a layout.
We also tell you about pre- and postprocessing, how to use the spreadsheet and outcome of several layouts.

CONTENT
1. Using groups
2. Are all VST's present ?
3. Pre- and Postprocessing
4. The Spreadsheet
5. 4.0, 5.0 or 5.1 ?

USING GROUPS

Let's go back to the layout of the teaching guide :



As you can see the layout is pretty crowded and we will teach you to compact
elements from the layout, so it's easier to "read".

Load the teach.bidule into Plogue Bidule , don't forget to set it "off" and
make sure it looks like the above picture.
What we're gonna do now is to make a group from the six gains before the
RMSBuddies. Set your cursor on the layout just slightly upper left from the
Gain Left Front entrance, hold your left mouse button and draw a square around
all the gains. They are now all selected. Instead of this, you can also select
the Gain Left, keep CTRL pressed and click on each other gain untill all 6 are
selected. They turn into yellow.





Place your cursor on one of the yellow gains, right-click and select from the
menu : groups - groups selected object(s) :



After that a new window comes up in which you need to set the amount of in-
and outputs first. Change the amount of samples to 6 and press OK.



The gains appear now as a group on your layout.



Reconnect all the connectors and change the name into "Gains":



Right Click on the Gains-group and select group - expand :



And you see the inside of the group. Connect all the connectors again



Put your cursor on the layout, right click, chose group - Parent group.
The last thing we need to do is to set the parameters for the complete group.
Right Click the group and chose group - parameters :



A new window appears :



First click on "Gain_SR" (it becomes blue), then click on "Amplitude" (becomes
blue) and then click on the "add" button. Your screen should look like this :



Notice the "Amplitude" is now no longer in the left part of the screen under
"Gain_SR", but moved to the right window. Repeat the process for all to get this



Now go to the right side of the window and place the items in the right order
with the up and down buttons (Select FL and use the up button)



We're almost there ! Last thing to do is to rename them. Select the first (FL)
line in the right window and click on "rename"



Give it the new name : Gain FL and press OK.



Repeat this for all and check with this picture. Everything ok ? Press "OK" !



Everything is done now. In the layout click on the Gains and you should see :



Now the layout looks far less crowded than before and filling in values in the
gains, is a lot easier, because you don't have to open every one separate.
Good time to save the teach.bidule and do a test run to see if everything works
as it should. Run it once without gains and once with gains and see if every
gain is indeed added to the original values.....
A nice thing to do now, is to select the group on the layout, right-click, goto
"groups" and "Save selected group". On the question that appears, answer "no".
When you look in the groups list, you will see there is a new "Gains" group
that can be used in other layouts. One important last thing about it : if you
used it and filled in values, don't forget that when you want to save the layout
values are saved with it !! So always make sure before you save a layout, that
every value is back to it's original ! >> All gains on zero, no name in file
player and recorder.... In other words : always save as if it hasn't been used
before. Never save it with values.....

ARE ALL VST'S PRESENT ?

In every layout of a method you will see we placed the name of the VST's that
must be present to let the layout work, on the left side of the screen with the
"mute" symbol next to them. This instance of the VST's is not for the working of
the layout, but only to show it's presence. When the VST is present in the VST-
folder of Bidule, it's color is red. When it's not present, it will be coloured
red. At the start of every bidule we tell you which software and VST's you need,
by loding the layout you will notice immediately if everything is present. As an
example we will now add this to the teach.bidule. As there's only one VST present
namely the HNM filter, we will place an empty instance of that VST at the left
side of our layout :



What we didn't show you yet and what will come in handy now, is how to put a VST
"on" or "off" or "mute" it. For it's presence we mute it. Right-click on the
HNM filter instance on the left, chose processing mode - mute



and next to the VST you will see a red circle appear :



Save the layout now. If you would remove the HNM filter now from the VST folder
of Plogue, you will see that the next time you load the layout, the HNM filter
changed in color to red. And RED here, means you don't have the VST in it's
folder ! You can try this yourself with the teach.bidule.......
Because we showed you also how to work with groups, imagine a VST is built in a
group, then this is a handy way to show if the VST in the group is present. Some
layouts can contain more than 5 groups with built in VST's. To see if you have
all the right VST's, you would have to open every group and search for it. Now
you will see immediately if everything needed is present !

PRE - AND POSTPROCESSING


Nowadays stereo CD's are mixed very different from CD's mixed in early stages
when CD's were just on the market. Nowadays the recording are a lot louder but
also better. In the first days a lot were just bad sounding. You also see this a
lot with cheap compilations where the songs don't sound as good as they can.
For improving the sound of the stereo, we use preprocessing.
There are also layouts that give a good separation, but not an optimal sound.
This is expecially the case when we use VST's to separate as much as possible
and this separation has a nasty side effect : artefacts. They destroy a bit the
original sound and must be postprocessed to get back as much as possible the same
quality of the original sound.
There are several ways of pre- and postprocessing of which we will discuss some
preprocessing tools here and the postprocessing ones as soon as we show the first
layout that separates enormous but also has a negative influence on the sound.

- The most important thing to make sure your surroundresults are good, is to feed
your layout with the best possible sound. As almost every layout works better
with higher resolutions, this is one of the things to change first...
In the teach.bidule we feed it with a 16 bits 44.1 Khz wav file. It will already
be a big improvement to feed it with 32 flaoting bits instead of the 16 bits and
save the 6 mono wavs also in 32 bits float or in 24 bits.
To change the 16 bits to 32 bits float, you need a soundeditor like :
- Soundforge : good program, among the best
- Adobe Audition : good, but sometimes using it's own way
- R8Brain (pro) : not a soundeditor, but very useful for changing bitdepths
- Several freeware editors, as long as they are capable of changing the bitdepth

Saving can now be done in 32 bits float too, because the DTS encoder accepts 44.1
Khz files in 16, 24 or 32 bits float. You will notice a better sound than when
doing the same album in 16 bits.
Another aspect as discussed on many audio forums is dithering. The best possible
sound is achieved by dithering the outcome to 24 bits.
In other words : input 32 bits float, output 24 bits. Dither ITU is a group that
you can download here, Dither ITU unpack and place in the group folder of Plogue.
Open the teach.bidule in Plogue Bidule, right click on the layout and chose from
groups - Dither ITU :



Check the settings :



And remember to save in future in 24 bits !



Now is a good moment to save the teach.bidule again !

Till now we only discussed making surround for CD's in 44.1 Khz, but for DVD or
DVDA you need other bitrates. For DVD we need 48 Khz and for the best quality on
DVDA we need 96 Khz. DVDA can also be filled with 48 Khz and even if you want
with 44.1 Khz. But 96 Khz gives the best result. Now upsampling your stereo wav
from 16 bits 44.1 Khz to 24 bits 96 Khz, doesn't give you any improvement in the
quality of the sound. It only takes more place. BUT...... in a lot of methods
working with 96 Khz gives you a better output than working with 44.1 Khz.
To upsample from 44.1 Khz to 48 or 96 Khz, we need again a soundeditor :
- 1. Soundforge. This one gives the best upsampling.
- 2. Magix Samplitude. Hardly any difference with Soundforge.
- 3. R8Brain Pro. A little less than the two others.
- And again several freeware soundeditors, but the quality of their upsampling
is questionable. The commercial tools are definitely a better choice here !
When you upsampled your stereo to 96 Khz and 32 bits float, you also have to
change a setting in Plogue Bidule. Goto edit - preferences



From there goto DSP and change 44.100 into 96000 and press "OK"



Save the layout again, but don't forget : these settings are fixed. So the next
time you want to do a conversion in 44.100 Hz, you first need to change the
preferences back to 44.100 Hz !!

Another aspect that can improve your surround is when there is a big difference
between the left and right channel of the original stereo. When you listen to
the stereo and the sound seems far louder in the left or right channel, you can
correct this a bit by adding a gain to the stereo and make both channels a bit
more even in RMS. In the conversionprocess this will give you most of the times
a better result. To make some kind of provision for that, replace the original
Stereo gain with two separate gains, one for the Stereo Left and one for the
Stereo Right channel. The original stereo gain is meant as a pregain when in
the conversion one of the channels clips. Lowering that gain will prevent that
clipping. When we use 2 separate gains, we need to take that into account too !
Let's make this more clear with an example : Left : RMS -21 Db Peak -4 Db
Right : RMS -19 Db Peak -2 Db
Now you can add 2 Db gain to the left channel, so both are -19/-2. But when this
leads to clipping during the conversion, you have to make them lower, buy adding
a negative gain of for instance 3 Db. So both channels will be -22/-5 Db.
After what you learned, you must be able to make it like this :



Don't forget to adjust the Freeze_RMS in the parameters !

Preprocessing to improve the sound of the original stereo CD is only necessary
when it's indeed bad sounding. There are several ways to improve the sound and
what we use to do so are mastering tools. Their is an enormous amount of VST's
that promise you the best sound, but we limit ourself to the most common :
1. BBE Sonic Maximizer : one of the best and most used. (commercial)
2. Yamaha Final Master : very good too (commercial)
3. Izotope ozone : very good too (commercial)
4. Stereo Enhancement : This is a group instead of a VST and is freeware. But
it gives excellent results. You can download it here : Stereo Enhancement

Let's try the freeware example. Download it and place it in the groups folder
after unpacking. If Plogue Bidule is still loaded : goto edit - scan groups.
Let's add it in the teach.bidule. Of course we can test it on a stereo file and
you'll be amazed how much difference this one makes. Rightclick in the layout,
goto groups and chose stereo enhancement. Connect it right between fileplayer
and the Gain Stereo Left and Right :



Goto "edit" and put offline processing off. Add an output device to the outpins
of the stereo enhancer. Doubleclick on it and play with the settings untill you
get the best result.



While your playing with the settings, select the "Bypass" box, to hear the
difference between the original stereo and the enhanced one.
BTW. every instance in a layout has a mode that contains the following set-
tings : Processing, Mute and Bypass. When you rightclick on an instance and go
to "processing mode" you will also see "Use Fades". This one is always checked
and is best left that way.
- processing : the instance is working
- mute : the process is working but doesn't produce output
- bypass : the process is bypassed.
The bypass can be useful on layouts with a lot of options, where you want to
exclude one or more. In the example with Stereo Enhancement, it will not be used
when you put it on bypass. The advantage is that when bypassing certain VST's or
groups, a slow layout becomes faster and not always everything is needed. When
for instance you have a perfect sounding stereo, enhancement is not needed and
can be put on bypass, leaving a faster running bidule.

POSTPROCESSING

As told in the Introduction, there are two different kind of surrounds on which
we concentrate : Separation and Soundfield. Soundfield methods always end in 6
mono's that form together the soundfield. Those methods are mostly based on
Ambisonics and the 6 mono's NEVER need postprocessing. When you postprocess an
Ambisonics sound, you destroy the Ambisonics soundfield !!
Separation methods use VST's that not only separate, but also degrades the sound
by adding artefacts. And example of an artefact is a small part of the voice
in the rears instead of in the center only. These artefacts can be improved or
repaired in a lot of ways. It can be done with several commercial tools, but
also by something simple like up- and downsampling. Imagine you want to make a
DTS cd in 44.1 Khz. Take your original stereo, upsample it to 32 bits float,
88.2 or 96 Khz. Process the file in the layout to get six 24/96 mono wavs and
downsample these back to 24/44.1 . A lot of the artefacts are minimized this way.
The commercial tools we use mostly are the same as those for the stereo files.
The difference is that now we need a tool for postprocessing with 6 channel file
players and recorders. To every channel you must connect one of these VST's .
So actually the process comes in two steps : a normal recording layout and one
that does the postprocessing. The postprocessing layout must have a fileplayer
with 6 channels and a filerecorder with 6 channels. In the first layout instead
of recording multiple files, we record one 6 channel 24/96 wav.



So do not select the "multiple files" in the recorder. It's easier when you save
your recording with a name that indicates that it is a 6-channel file : like
01mch.wav. Now we replace the recording bidule with the postprocessing bidule
and use that one for the final step. In this example we will use the Sonic BBE
Maximizer version D82. This is commercial software and can be purchased here :
Sonic Maximizer
The layout can be downloaded here : 6mch to 6 mono wavs
Make sure BBE Sonic Maximizer D82 is installed correctly in the right folder and
load the layout :



Click on the SM fronts and this appears :



On the right side you see "menu". Chose this and then "load presets", "05. mix
mastering", "mix enhancer 1". You can chose between a lot of presets, just make
sure it sounds good and is not "overdone". You can monitor the sound and use the
knobs to finetune. Load the xxx-mch.wav, run the layout, take care of the gains
and when recording select "create multiple files" !. Now again you end up with
six mono wavs, but in 96 Khz. The last step is to downsample these to 44.1 Khz.
All these last steps have been discussed before ;-)

THE SPREADSHEET


Already mentioned a few times, but now we're gonna use it......
First download it and unpack it to your desktop. Because you will use this a lot,
that's a good place. Of course you need MS Office or an Excel compatible program.
Download it here : Surround Gains Spreadsheet
Place it on your desktop and open it : Because it takes a lot of space on screen, we do it in several steps. First we
show you the left side, because this is where you fill in values.....



Open the teach.bidule in Plogue and chose 2 songs from the same album to convert
from which we're going to fill in the gains into the spreadsheet. Take good care
that you don't have to fill in the minus sign ! Run the first song :



Now fill in the values for the first song in the spreadsheet :



Do the same for the second song :



On the right side you see the corrected values that you need to fill in in the
gains in the layout for each song. BUT....... we want the difference in loud-
ness on the stereo CD to be the same in surround. So first we need to measure
the average RMS of the stereo. You can do this by using the Stereo Buddy and
add the two gains and divide them by 2 : 20.37+20.74=41.11/2=20.55 for the
second song, or you can use a soundeditor like Soundforge, goto normalize and
let it calculate the average RMS. For the first song it's : -19.3 Db and for
the second -17.3 . So there is a difference of 2 Db in loudness. Fill these
values into the spreadsheet and this time with a minus sign :



As you can see both songs calculated to an average RMS of -20 Db without the
values from the original stereo, but after filling these in, the first song
becomes -22 Db and the secon -20 Db, meaning in surround the difference is now
the same as in stereo. When we take a look at the right side, you see the
corrected values in red. Now start processing both songs with the values from
the spreadsheets : song 1 :



Run the process to record 6 mono wavs. Load song 2 and fill in the values and
record it. Encode both two DTS WAV and listen on the HTS.

4.0 , 5.0 OR 5.1 ?


There has always been and will always be a lot of discussion if you should add
a LFE channel to your mixes. Purists say no, because your receiver or decoder
will take care of the bass and send it to the subwoofer. Also it's impossible
to tell at which crossover frequency your HTS is setup, so the best thing to
do is to leave it completly out of the mix. And LFE is for the very low stomach
effects, that you hear in movies but not (much) in music.
In some cases it can be useful to convert to 4.0 or quad for a better experience.
The other camp always encodes to 5.1, because not every receiver or build in
encoder does a proper job. A lot of cheap DVDA players for instance lack a good
bassmanagement, so you better add a good bass channel.
The question is what you should do........
IMO the first thing to do is to measure ! Measure how much bass is really present
in the original stereo first and depending on that result we chose if we add a
/1 channel or not. For measuring the bass we have a layout ready that you can
download here : LFE Measurer
Unpack the archive and place the bidule into the layouts folder. Run it :



Load a stereo file in the player and compare the RMS of the stereo with the RMS
of all crossovers, and you will notice that almost every song has a far lower
bass tthan you would expect. With "Year of the cat" again as example, you can
see there is a difference of about 2.5 Db at 200 Hz lowering to a difference
of 5.5 Db at 80 Hz.



So you can conclude that the adviced setting of making the LFE 4 Db lower than
the fronts, is indeed reasonable. If you want you can add an extra instance
with values of 60 and 40 Hz. The layout is so simple and with all explanation
you already got, this will be a simple process. But you can be assured that
it certainly will not be louder at those lower frequencies !
In this case you can chose for 5.0 if your carrier will be a DTS CD where your
bassmanagement in the receiver or decoder in your dvd-player will do it's job.
When you want to have your surround on DVD or DVDA I would use the 4 Db setting.

BACK