4027 replies to this topic

[toxie]

When using -minimized command line param I can't see the progress bar when loading from frontend but can't use Q button to pause the VPX while playing

When not using -minimized I see the progress bar window when loading a table but can use Q for pauing the gameplay....

Can We have a param for just hiding the progress bar window ? something line -NoLoadingBar or something ? That would allow to have no progress bar and still be able to pause vpx...

I added a new paramater "-ExtMinimized" which does that. In addition, it will directly leave also the editor when quitting from the pause menu.

The VP installer will also use that then from now on when installing the registry entry for playing (double clicking a table).

 

See https://github.com/v.../runs/716664509

Edited by toxie, 04 April 2021 - 01:10 PM.

[toxie]

When loading a table in PinupPopper, the Visual Pinball Toolbar always pops up

 

Is this still the case nowadays?
 

The first small thing, when you reimport a sound, say for example, a flipper sound (I really like the f1 flipper sounds and usually replace my tables with them) when I change them, the new sound no longer plays. On the last official build, when you changed any sounds, after you hit ok, the sound you selected played. This was nice, because it reaffirmed you hit he correct audio file.

 

I did this intentionally as i thought it may be nerve-wrecking, especially when nowadays importing music files (i.e. much longer sounds).

So instead of having to press stop then to end the music, i decided to make it the other way round and so now one needs to press play instead.

[toxie]

2. The sound manager exports the OGG file with an unrecognized format.

 

Fixed with https://github.com/v.../runs/716755115

Thanks!

[toxie]

and for everybody without github login: https://github.com/v...7.0-127-c9c069f

[jpsalas]

Yes, toxie, the OGG files plays fine, and the volume is right. Also the repeat function of PlaySound works fine, thanks I'll remove my workarounds and I'll upload my last tables again, but only a few had OGG music in them

[remdwaas1986]

wow toxie what a hard work! ogg will be less space consuming then wav!

[wiesshund]

Thanks (as usual ) for the excessive testing!
 
As for the 3D position stuff: njk to the rescue, please. I have no clue about that (plus no setup to test this).
 
Stereo Panning: Unfortunately we're stuck in legacy hell here, as this is on a log scale since it was implemented back then.  :/
And so we cannot change it, unless we want to break existing tables.  :/
If you look at the example table, there is for example a ^10 scaling involved when panning, due to this, to make the setting more sane.

 
I assume you are talking about
 

Function AudioFade(tableobj) ' Fades between front and back of the table (for surround systems or 2x2 speakers, etc), depending on the Y position on the table. "table1" is the name of the table
	Dim tmp
    tmp = tableobj.y * 2 / table1.height-1
    If tmp > 0 Then
		AudioFade = Csng(tmp ^10)
    Else
        AudioFade = Csng(-((- tmp) ^10) )
    End If
End Function

Function AudioPan(tableobj) ' Calculates the pan for a tableobj based on the X position on the table. "table1" is the name of the table
    Dim tmp
    tmp = tableobj.x * 2 / table1.width-1
    If tmp > 0 Then
        AudioPan = Csng(tmp ^10)
    Else
        AudioPan = Csng(-((- tmp) ^10) )
    End If
End Function

?
 
The thing is, fade has some semblance of functioning
you can at least get something to play at the front of table, middle of table or rear of table
I did not test exactly how fine it will move incrementally
 
But Pan seems to be a wee bit broken
basically 0 Pan is center and right speaker
 
Using the example table
 

' PlaySound soundname, loopcount, volume, AudioPan, randompitch, pitch, useexisting, restart, AudioFade
	If keycode = LeftFlipperKey Then
        LeftFlipper.TimerEnabled = True 'This line is only for ninuzzu's flipper shadows!
		LeftFlipper.RotateToEnd
		PlaySound "fx_flipperup",0,1,0.00,0,0,0,1,-1.00
	End If

	If keycode = RightFlipperKey Then
        RightFlipper.TimerEnabled = True 'This line is only for ninuzzu's flipper shadows!
		RightFlipper.RotateToEnd
		PlaySound "bumper mono",0,1,0.00,0,0,0,1,1.00
	End If

The expected outcome is that one flipper plays in the middle of the table, but at the back, and the other plays in the middle, at the bottom
The received outcome is
one flipper plays in the middle at the table front
the other flipper plays out the right and center at the table rear
 
Now changing this to

		PlaySound "bumper mono",0,1,0.00,0,0,0,1,0.00

 
Results in sound in middle of table, between the front and rear
but panning to the rear left hard, nothing from the rear(rear of table) right side at all

 

and changing to this

		PlaySound "bumper mono",0,1,0.00,0,0,0,1,-0.01

Which is only the smallest amount away from the center

Results in the sound, which is still panned dead center (0.00)

jumping to the right rear of the table, and also fully to the rear of the table, not a peep from the front of the table

 

In fact this

		PlaySound "bumper mono",0,1,0.00,0,0,0,1,-0.001

Also results in a full swing to the rear of the table
So fading wise, basically there are 3 positions with no in between
REAR MIDDLE or FRONT

 

And as soon as the sound goes full REAR, it pans to the RIGHT REAR
And when it is in the MIDDLE, it pans to the LEFT REAR

and with a panning value of 0 the entire time

(We will call fade FRONT MIDDLE and REAR and pan LEFT CENTER and RIGHT)

 

Now, lets go to the other end of the table

 

There is no effective panning when you come to the front of the table

 

This
 

' PlaySound soundname, loopcount, volume, AudioPan, randompitch, pitch, useexisting, restart, AudioFade
	If keycode = LeftFlipperKey Then
        LeftFlipper.TimerEnabled = True 'This line is only for ninuzzu's flipper shadows!
		LeftFlipper.RotateToEnd
		PlaySound "fx_flipperup",0,1,-1.00,0,0,0,1,1.00
	End If

	If keycode = RightFlipperKey Then
        RightFlipper.TimerEnabled = True 'This line is only for ninuzzu's flipper shadows!
		RightFlipper.RotateToEnd
		PlaySound "bumper mono",0,1,1.00,0,0,0,1,1.00
		'PlaySoundAt "bumper mono", REAR
	End If

Should result in one sound coming out hard panned right, the other hard panned left
but it does not, it kind of comes out in the CENTER, with maybe the very smallest of hint of being to one side or the other.

 

Now to have this test have the widest of meaning, it is done in a 5.1 setup
because this is what the typical desktop setup would have if it is not stereo, as well as a good number of cabs, especially mini cabs
where they have the backglass speakers for backglass and table rear and a pair of exciters for the table front

But the results are the same in 7.1, with some added weirdness

 

And this was used to certify that the audio setup is configured and functioning properly
5.1 audio test

 

I do not know what the deal is with the front of table (rear channels) panning, or not really panning i should say
I do suspect what some of the deal is though with the rear of table(front channels) panning strangely on their own when centered
and that is, that vpx is getting hung up on a center channel, which it should be ignoring

 

In 5.1 the only channels that should be valid for table audio are front left, front right, side left, side right (rear left and rear right in gamer terms)

In 7.1 the only channels that should be valid for table audio are side left, side right, rear left, rear right

 

NJK can hopefully shed some light on that

 

I am not sure what to suspect on the apparent lack of pan/fade gradients

 

set a flipper to playsoundatball
then roll a ball up and down in a track in the dead center of the table (A ramp works nicely)
you will see what i mean

 

PS, i did try with WAV and OGG to see if there was any difference, but there was not in this respect

Edited by wiesshund, 04 April 2021 - 08:31 PM.

[toxie]

wow toxie what a hard work! ogg will be less space consuming then wav!

Yes, and actually the final fixes thankfully took less time than expected. Apparently sometimes it can really help to let bugs rest for a bit, and then investigate again after a while. 

@Wiesshund: As said, unfortunately njk will need to help here (i hope). And (at least) the pan is unfortunately really whacky 'by design', as it works on this logarithmic scale, so most of the effect only happens very close to the extreme values.

 

For fade i actually don't know the behavior, so if it works on log scale or linear.  :/

Plus for pan i don't know how it actually (should) react on 3D audio setups, as DJRobX did most of that work back then.

[wiesshund]

@Toxie
 
On the other hand
I just want to mention that now
If one had the knowing
and the knowing led to doing
 
a multi track ogg will play
 
On the playfield(table)

In the manner it was encoded
 
take this MULTI TRACK ogg
and assign it to a table object, flipper or what ever, play at ball etc doesnt matter
 
And it probably should not do that, because the table should only know about 4 speakers

[wiesshund]

 

wow toxie what a hard work! ogg will be less space consuming then wav!

 

@Wiesshund: As said, unfortunately njk will need to help here (i hope). And (at least) the pan is unfortunately really whacky 'by design', as it works on this logarithmic scale, so most of the effect only happens very close to the extreme values.

 

 

Well, so far, i have yet to find a value small enough and i am out to 0.000000 places
it hits the magic number and flips
 

Now this would not be the end of the world entirely, IF it would stop deciding that the center channel is a valid location for the table audio to play though
but it still does not pan at all when the sounds location is at the front of the table.

Now i can make it do 100% exactly what i want, but i am thinking that 3d encoding audio is not in most table authors lists of "Things i really wanna do"

not to mention, that totally prevents someone like thalamas from adjusting a table to better suit his SSF set up

 

 

 

For fade i actually don't know the behavior, so if it works on log scale or linear.

 

Best i can tell, fade has 3 positions
Very rear of table
Dead middle of table
Very front of table

Which explains why when thalamus hits a bumper that is not way high up on the table
his front exciters start vibrating a lot, both left and right
Because if said objects fade is within a very large chunk of the table, say over a 3rd if the total table length, it goes to middle fade
which makes it play out front and rear of the table equally, and the front has very little if any Pan, so both his front exciters will be buzzing, even if said bumper and slammed up against the left wall of the table.
Not so detrimental for someone with acoustic speakers, by kind of a terminal illness for someone with tactile transducers

 

 

 

Plus for pan i don't know how it actually (should) react on 3D audio setups, as DJRobX did most of that work back then.

 Yea, i tried to ask him about the audio once, as you directed me to him to show him something i found, but he kind of did not want to talk about it, and basically said

the audio was done and he didnt want to work on it any more.

 

So, hopefully NJK will maybe find this info useful, at least to pin down behaviors to look for.
I wish i could be more help, because i know what it should do, and i know what it should look for etc
but i dont know how to turn that into C++, nor do i understand the present source code to even look at it and say oh here, this thingy is wrong, needs changed.

Edited by wiesshund, 04 April 2021 - 09:49 PM.

[toxie]

That's alright. We also desperately need ppl like you that do all the excessive testing! So thanks again!

[Thalamus]

Wiesshund : You are taking about OGG files placed on the table now aren't you ? Since my wife is in bed, I can't really test it very well before tomorrow, so, all I did for r127 was to start one of the favorite tables STTNG and let the balls get ejected. I can clearly hear them being ejected and shot all the way up the table via the wire ramps.

 

And - by the way - it is Thalamus - not Thalamas

Edited by Thalamus, 04 April 2021 - 10:15 PM.

[toxie]

Oh, one more thing though: Are these findings of yours new to VPX 10.7, or did some/all already exist in 10.6.X?

[wiesshund]

Wiesshund : You are taking about OGG files placed on the table now aren't you ? Since my wife is in bed, I can't really test it very well before tomorrow, so, all I did for r127 was to start one of the favorite tables STTNG and let the balls get ejected. I can clearly hear them being ejected and shot all the way up the table via the wire ramps.

 

No and yes

 

Yes, you can place an ogg on the table now
and yes a mono or stereo ogg will play the same as a wav file, same behavior

 

but no the tests above were done with wav audio, just the wav files in the example table

 

And if you test the multi chan ogg i linked above on the table
turn down your speakers.

 

The audio gain problem is fixed, so that wont blow you up
But that multi track ogg does have an LFE channel
and i do not know how your particular set up is set to redirect those.
Maybe it plays no place for you, or maybe it shakes the entire house and wakes the wife and everyone else for a 3 mile radius

 

 

 

And - by the way - it is Thalamus - not Thalamas 

 

Yes well, auto correct keeps trying to change that to Phallus

which is more embarrassing than typing thalamas

 

 

Oh, one more thing though: Are these findings of yours new to VPX 10.7, or did some/all already exist in 10.6.X?

 

"Seems" to exist in 10.6.1 also

 

I say seems because the updated dll's are present here, and i do not know if they would have any effect here or not.
and i do not want to guess

 

To say for certain, i would have to set up a virgin install with just the 1.6.0 installer and 10.6.1 stand alone update.

[njk70]

When I get my cabinet back together and I stop my table downloading binge I'll take another look at this. There was a little too much information in too many permutations to take in all at once here.

 

I'll make a few comments.

 

My changes should only be triggered in 7.1 enhanced mode (but I was thinking of a change that would have gone in the other 7.1 mode that I will make sure I didn't actually make).

 

My changes are intended for just WAV files, but I think I applied some to the non-WAV code path in 10.7 and it is COMPLETELY untested and maybe I should put in a PR to remove it until I can actually try it out.

 

The logarithmic scaling is a pain, perhaps a second api with a non logarithmic scale could be added for tables that want to use it. The vpx code undoes this scaling (as it did before my changes) and then my fixes apply a different attempt to enhance the table exciter positioning rather than letting window's surround sound algorithms have complete control over it (which doesn't work well because ssf is not really surround sound). 

 

I haven't tried anything that goes through the BASS api path (ogg files in any format).

 

My initial feeling was that ogg files should be fully encoded for what you want to play in each channel and you should keep using wav files for positioned sound effects. But that was probably a little simplistic on my part but we will probably have an issue with people trying to position multichannel audio files which is a no go.

 

I'll reread this thread later to see if I can pick out what was relevent to WAV files as at least I understand how that works right now.

 

Right now I am trying to fix my cabinet. I keep losing small parts whenever I take things off it. 

... it would be a great help to have a test table or two for ssf.  I'd like to be able roll a ball around the entire extremes of a table and hear a constant positioned tone (not the variable ball rolling sound) and maybe another table of just bumpers positioned in a grid (3 x 3) that could be triggered to test bumper positional sounds.

[toxie]

Oh, one more thing though: Are these findings of yours new to VPX 10.7, or did some/all already exist in 10.6.X?

 

"Seems" to exist in 10.6.1 also

 

I say seems because the updated dll's are present here, and i do not know if they would have any effect here or not.
and i do not want to guess

 

To say for certain, i would have to set up a virgin install with just the 1.6.0 installer and 10.6.1 stand alone update.

 

No, that's fine. dll's will not affect the .wav behavior whatsoever.

[studlygoorite]

I have been following and trying to learn but haven't a friggin clue about most of the discussion here. What I can tell you is my system has never sounded so good with 10.6.2 and the newest 10.7. Whatever you guys are doing is working and I for one, as well as many others I'm sure, appreciate everyone's time and effort.

[wiesshund]

 

 

My initial feeling was that ogg files should be fully encoded for what you want to play in each channel 

 

And you know what, that does in fact work perfect, like 100% perfect.

You plot the position of your sounds in the 3d sound scape, and then encode them, and they play at the exact position you put them
the table does not need to do anything but say play sound abc.ogg
And if abc.ogg is a coin hitting the floor off the left side of the table, that is exactly where the coin sound comes from

 

Couple problems though

I think most table authors do not have access to a decent software package to allow them to create the sounds like that

It is not terribly intutive to do in audacity, and i cant share pro tools

 

And it is un-adjustable
Thalamus's exciter set up for example, is going to barf at it, because it was based on an acoustic set up
and he has tactile transducers, so he needs some way to modify that which does not translate properly from air transmitted to material induced vibration
(as does anyone else with the same kind of set up)

 

And it is channel dependent

if you do a 7,1 encoded sound, to allow the table sound to dodge the backglass
it does not translate to 5.1 (because of how we are using the channels)
table front and table rear will be merged (because 7.1 to 5.1 merges the side and rear surrounds)

 

and if you do a 5.1 encoded file, and run it back on a 7.1 channel set up
you have rear of table in backglass, and front of table in the rear of table exciters/speakers
and nothing in the rear of table speakers at all.

 

But, it does in fact work absolutely 100% correctly, for the machine you put it together on.

I do not think you were actually suggesting doing that, just putting it out there in case someone else is reading

[Thalamus]

wiesshund : So, when you say

 

'I think most table authors do not have access to a decent software package to allow them to create the sounds like that It is not terribly intutive to do in audacity, and i cant share pro tools'

 

Why is it then a problem ??

 

If some table author added something that was multichannel to the table area. I'm pretty sure that I would delete it right away as no sound from the table should come from more than one place at any given time.

 

I'm sure that it something I don't get here ?

Edited by Thalamus, 05 April 2021 - 12:23 AM.

[njk70]

@Thalamus I think the idea of multichannel on the playfield is that it was recorded/encoded for the sound to come out at the correct positional spot. For a table sound that should just be the rl,rr,sl,sr channels so a bumper sound might (simplistically) play at 85% volume sl, 15% sr, 5% rl, 2% rr for a bumper in the upper left corner of the playfield... rather than playin a mono track and having vpx/windows attempt to position it based on 3d coordinates.   I don't imagine many people would go to the effort to create such a recording. (we would make no effort to adjust or keep it off the backglass, its entirely up to the sample author to use the appropriate playfield channels).

 

My view is that all playfield samples should be mono track and anything else should be rejected (I believe stereo sample have been accepted in the existing system, not sure if it like just uses the left channel or something).  Multichannel samples can go to the backglass device which would play on whatever channels the ogg file was configured on and would be untouched by vpx for positioning. I'm not sure what else we can do that makes sense but I am certainly open to ideas.

 

I think wiesshund wants playfield samples that go full surround sound for audio surround sound rather than ssf support. I am not sure how we can have both without some enhancement to the sound manager to use different samples for different audio modes.

 

I'm not fully following this thread, just taking a break as I wire my flipper buttons into the new holes I drilled.  For the record, I hate adjusting leaf switches, especially double ones.

Edited by njk70, 05 April 2021 - 12:52 AM.