the optimal soundstracks for tbs animations

hi guys,
i’ve read the manual sound section, checked the faqs and forum on the sound topics back in last two years entries but i still can’t find the answer: what is the soundtrack preference to use for toonboom animations?
all i’ve found is the suggestion for .wav files to be at 16bit sampling rate.

i ask because i used a 16bit .mp3 file for my animation, streamed it and the result was pretty crappy.
the event sound of it sounds ok, the streamed version contains some kind of knocking every few seconds (the source file is fine). it’s necessary for me to maintain the soundtrack streamed, though.

is it a tbs sound compression issue (streaming fits the sound to the images flow, sometimes even dropping sound frames to catch up with the visuals)?

should i convert the file to an .aiff or .wav?
what is the workaround for this?

what is the best sound file format for the animations in general? let’s put the size issue aside, i’m referring to the sound quality.

thanx in advance.
cheers,
rob

Rob,
The Flash player is the mechanism that is playing your movie all Toon Boom does is create a SWF that meets the Flash standards. Now Toon Boom might effect your file size and it might create real processor hog graphics so that I can’t address. But the Flash player does not drop sound in favor of visuals, it is just the opposite. It subdivides the sound across the frames to spread it out and then if it can’t keep up with the sound it starts cutting off frame rendering of graphics on those frames that have too large a graphics load. So there are several methods of dealing with this. First the smaller the sound file the smaller the segments that get subdivided for each frame, that’s why it is recommended to go with a lower bit rate because it reduces the sound file size.

Additionally, the frame rate of your published movie is important. The more frames per second the more you can subdivide the sound data in Flash. So if you animate on the 2’s or the 3’s and use hold frames then half or two thirds of your frames have no graphics rendering load and your weaker processing clients will be able to devote more to sound processing.

The last tip is to minimize the graphics per frame so avoid excessive vectors or excessive bitmaps when ever possible.-JK

jk,
thanx for your input, as analytical as usual from you :wink:
i’m sure you’d make a good animation teacher, too.
the information about dropping of the sound frames by streamed soundtracks comes afair from tb team. i can’t recall now where i’ve found it, but i’ll look for this. → update: it’s in the v3 manual, page 227.

my problem is the following:
i prepare a long term animation project for, well, a kind of a music video for a classical music piece by beethoven. the soundtrack is almost 7 minutes long and i don’t have much graphics yet, except for a pretty well functioning rough storyboard.

i scanned all the images (84 pics) and imported them into tbs’ image element.
i set the frame rate at 12 fps, although i prefer 25 (we’re using pal in europe), but it would be too much work and too much data for the film, so i think the cut was necessary.

i extended the storyboard pictures frame numbers to cover the whole soundtrack (i’ve described this process in the storyboard thread in flashfilmmaker forum as well). so i have the full animatic of the movie. each pic appears for a roughly 5 seconds on the screen, so there is not much graphics information to be computed.

the sound (the only sound element in the animation) is streamed and this appears to be a big problem.
the event sound (non-streamed) plays back well, as soon as i stream the audio, which is pretty necessary for this purpose, it causes a really bad quality.
i tried a flash output, and an avi and a qt as well. either compressed or non-compressed, all result in the same bad output. the soundtrack contains some kind of knocking every few seconds and the sound floats occassionally as well.

i currently don’t have any explanation for this behaviour of the program. i tried this on two computers to no avail. the quality of the streamed sound remains not acceptable.
(the source is a .mp3 with a pretty good sound).

the next step would have to be for me to try to convert the sound file to any other format and to try the outputs again…
but maybe there is a workaround for this?
cheers,
rob

Rob,
Please re-read the manual. It says exactly the same thing I wrote. Flash gives preference to sound over visuals and the so called dropped frames are in fact un-rendered or under rendered visuals. This is due to the size of the visual rendering load on the Flash player and the client system. That is the reason to increase the FPS and insert “hold” frames that require no rendering. It gives you more subdivisions of your sound data and allows the Flash player to have time to deal with both sound and visuals. Also if you are using large images that may be a significant problem if they are so large per frame that they are increasing the render loading. 12 Frames per sec and large graphical rendering load are choking down the Flash player. -JK

jk,
i wasn’t saying you’re wrong, i was only sayin’ i’ve found this expression on the page 227, event and streamed sounds > synchronization, first point: ‘…the player drops frames if necessary…’
to be frank, neither did i need the sound information up to this situation nor have i pondered on this topic.

but now it seems to become an issue.
what i see ( actually: hear :wink: ) is a bad sound of the streamed animation and a clean sound of the event sound (e.g. non-streamed), although the film remains the same.

and it doesn’t matter if it plays back in the flash or qt or real player. they all deliver the same.
don’t get me wrong: as long as the sound is ok, i don’t even care if the players drop the frames or not.
i’m simply a bit angry that further sound files i tested today, if streamed, all wind up in just the same bad quality.
if only i knew the reason for this…
cheers,
rob

update:
my test machines were:
a p3, 750mhz, 256mb desktop
a 2.5ghz, 256mb laptop

tb guys:
1. is the memory size for the computing of a streamed sound compared with computing of an event sound an issue?
2. what is the purpose of the .wff sound file being generated after the saving of the streamed soundtrack?

ok, i was able to additionally test the tbs sound export performance on the desktop-pc of my kids (it was a hard time, the mechine is usually available only for 10 minutes in two months time due to permanent gaming or occasionally learning).

the result is devastating: the same old spoiled streamed soundtrack issue (the computer is an athlon 1.8ghz with a 512mb ram).

what is going on with the streamed sound files in tbs export?
do i miss anything or it’s just a feature, not a bug?
can anybody confirm or contradict the bad sound quality situation of an inserted long .mp3 file for streaming export?
thanx in advance.
cheers,
rob

anybody any experience with streamed .mp3 music files?
thanx.
cheers,
rob

completing the excellent research done on this topic by marczero in the general section (‘lip-sync drifting…’ post) there’s an easy way to change the soundtrack from stereo to mono in a free audio tool (audacity, to be found here:
http://audacity.sourceforge.net/).

the above mentioned method is here:
http://www.julianyap.com/wiki/Convert_Stereo_to_Mono_with_Audacity
cheers,
rob

hi guys, i read this post wishing i had been able to give my two bits worth closer to the time of post.

i too have found streamed sounds very disappointing. there is no clarity, top end is gone, even sounds like bitrate is lower. in my latest project (a video clip of sorts) this is very frustrating as i want good quality sound, but it just wont work on event sounds as the video falls behind and things arent synchronised anymore. i have tried a program that converts flash to DVD which seemed to work well on a project with lots of short sounds and dialogue but on a long sound like a soundtrack its no good :frowning:

there must be a way to do this… cartoons on tv have excellent sound, how do they achieve that?

hi angrypetal,

there is a separate thread with a kind of a solution to this, but i can remember it:
in order to stream your sound check the ‘streamed’ box in the sound editor, but don’t select any of the soundtracks into the timeline ! just check the box and leave the editor.

the thus exported audio sounds fine.
i hope it helps.
cheers,
rob