Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Rex_Nex]

I giggled when I heard police sirens in the RHumphries thunderstorm file. Broke immersion, but I can always make myself believe my dorf was arrested.

[Thundercraft]

I giggled when I heard police sirens in the RHumphries thunderstorm file. Broke immersion, but I can always make myself believe my dorf was arrested.

That would definitely be immersion breaking. But you must have very good ears, because I've listened to that file several times and I can't hear any sirens. Could you specify at how many minutes and seconds the sirens start? (I should be able to edit that out, if I could find it...)

While on this subject, I'm curious to know if there are any other sounds or music that seem to break immersion. (Though, with the music, I'd say what breaks immersion is subject to interpretation.)

Edit: Ah, turning up the volume I think I can very faintly hear sirens starting at 1:53 and continuing to about 2:00. Does that sound right?

BTW: zwei, why are the following two weather sounds so long?:

2523__RHumphries__rbh_thunder_storm.mp3
22606__martypinso__DMP013016_HEAVYSNOWSTORM.mp3

Do they both really need to be about 3 minutes and 30 seconds to fit a particular design scheme? (I think I'd prefer a short sound just long enough to make the player aware that a weather change occured, leaving the player to enjoy more background music.)

[Rainseeker]

Why do we have to break the weather and music up?  Can't they overlap?  I'd enjoy that more, I think.

[Rex_Nex]

Yes, that sounds completely correct. They are faint, but I usually have the volume cranked up, and it really bugs me hearing that.

I have to go with Rainseeker. Having a very light and somewhat more quiet thunder/rain overlapping the music would be perfect.

Other then then some small things that really can't be helped with the way SoundSense currently works (In particular, music still playing after DF closes,) I am enjoying this project a lot. There's only so much of Toady's guitar I can take.

[Thundercraft]

zwei, can you recall if there are already some GNU GPL 2.0 or 3.0 licensed sound FX or music in the Sound Pack? I need to know if GPL-licensed stuff can be included, or if the license itself might conflict in some way with some of the Creative Commons licensed material in the pack.

From what I've read and understood, GPL licenses are a bit similar to the "share-alike" clause of CC-BY-SA and CC-BY-NC-SA, but are more restrictive. (Though, I could me mistaken.) Also, I've read that it's not possible for GPL-licensed software projects to use CC-BY-NC or CC-BY-NC-SA licensed media (sounds/music/videos/images) because GPL has to allow commercial distribution. (GPL even seems to conflict with CC-Sampling-Plus for similar reasons.) But I'm not sure if something similar could apply to using GPL stuff in a Sound Pack with NC stuff. I tried asking on the #GNU IRC chat (freenode), but didn't get a response. Perhaps I should ask on the Creative Commons forum?

Edit:
On another subject, I assume that if more than one sound file is listed together in an entry in the .xml, then SoundSense randomly picks one of them. Correct? Example:

Code: [Select]

<sound logPattern="The (.+) is knocked over!" concurency="4" timeout="500">
<soundFile fileName="73583__J.mp3" weight="400" />
<soundFile fileName="43601__FreqMan__object_falls_3_.mp3" />
<soundFile fileName="43602__FreqMan__object_falls_4_.mp3" />
<soundFile fileName="43603__FreqMan__object_falls_5_.mp3" />
</sound>
If that's the case, is it possible to get SoundSense to play two sound files in a row for the same logPattern ? Or, even better, is there a way to intentionally get a sound file to repeat a certain number of times? I might be able to cut down on file size and have more options in sound fx combinations if that's possible.

[zwei]

...

I do not recall any gnu/gpl licensed stuff and ss itself is currently has no license.

---

Yes, of those four sounds, one is picked randomly with first one being more likely to be chosen than others (weight attribute).

You can allow more sounds to be played for same pattern by giving first one haltOnMatch="false" attribute, they will start at same time. Example is in seasons.

Currently, you can not loop something set number of times or delay sound execution.

Not sure how usefull layering/repeating stuff is going to be. repeating SFX is not really usefull and music/background is better of looping randomly.

---

I like overlapping weather/music, but guess that is a preference. Weather tracks are a bit longish, but they can be broken up (and sirens edited out - ooops!) - It would add a but of variety for rain begining anyway.

[Thundercraft]

I do not recall any gnu/gpl licensed stuff and ss itself is currently has no license.

Being concerned about the licensing, I asked about possible license conflicts (in distributing GPL-licensed sound FX or music with files using other licenses) on the #cc channel in IRC and someone replied with, "IANAL but I don't see how there is any cross-license compatibility problem". Though, he did seem to suggest that there could be a possibility of "violating the terms of the gpl" for all he knew about it. (They know much more about CC than GNU GPL.)

Also, I asked on the #GNU channel (also on irc.freenode.net) and when I mentioned that the sound files are distributed in a Sound Pack download separate from the program, someone told me:

It seems like you're describing a collection of sound files that are simply being distributed along side other files. I see no problem with that, similar to how you can combine GPL and non-GPL files onto a CD without everything on the CD being under the GPL.

That said, I do not plan to use much, if any GPL-licensed material. So far, there is only one set of GPL sound files I'm considering (for Giant Cave Spider and another spider). And I've already received a response from the source with special permission to use these even if there was a GPL license issue.

You can allow more sounds to be played for same pattern by giving first one haltOnMatch="false" attribute, they will start at same time.

They will start at the same time? Hmm... That's not exactly what I need. Could a new function be made to play files consecutively? I'd find that really useful!

Example: I want to introduce sounds for Lions reaching adulthood. For the regular, untrained Lions, that could be a long, low growl. For the hunting and war trained Lions it could be a short growl, immediately followed by a classic roar. Depending on the licenses and permissions, I might be able to append the roar onto the end of the growl to get them both to play on the same logPattern. But what if the growl sound has one license (say, CC-BY-Share-Alike) and the roar has a different license (say, CC-Sampling-Plus)? They would have to remain separate files and I would need some way of telling SoundSense to first play the short growl file and then play the roar file.

Currently, you can not loop something set number of times or delay sound execution. Not sure how usefull layering/repeating stuff is going to be. repeating SFX is not really usefull and music/background is better of looping randomly.

Delaying the sound execution might remedy the problem in the example I gave above. If I could use haltOnMatch="false" to get both the growl and roar to play, but then insert something extra to give the roar a specific delay... problem solved. (But, it might be easier to have a way to get one file to start playing immediately after another has finished.)

As for looping something, I see that as a way to save on file size. For instance, if a sound effect would work best by having an extra long growl or a series of chirps or grunts (or a hammer banging metal, laughs, etc...), one way is to copy and past in a sound editor to get a short sound FX to repeat several times. But another is to just loop the same file. Or maybe the growl or chirps were already long enough, but space could be saved by chopping it into a small file and having it repeat a few times?

Also, in situations where both a short and long version of the same (loopable) sound FX is desired (or maybe several lengths), this currently has to be implemented as separate files. But with a repeat function a single file could fulfill several sound FX lengths.

Further, I've thought about repeating rather short, loopable pieces of music to make them play longer. There are some songs and pieces of songs I have found that seem fitting enough for DF, but are too short. Again, I could just use a sound editor with copy 'n paste, but...

On the other hand, the savings in extra files and file size probably isn't that much (especially since the files in the current Sound Pack are not designed to be loopable). And coding a loop or repeat function would probably take away from more important things.

I like overlapping weather/music, but guess that is a preference. Weather tracks are a bit longish, but they can be broken up (and sirens edited out - ooops!) - It would add a but of variety for rain begining anyway.

I've changed my mind and the overlapping does not bother me now. But if you want to break them up, that's fine too.

[Rainseeker]

You know what would make this mod even more useful?  It would be cool if it could somehow be easily bundled with the Noob pack or another mod pack.

[nenjin]

Would that really serve much beyond how it's already incorporated though? Has its own installer, a separate .exe that needs to run....I thought the Lazy Newb Pack Utility also could launch SS from its own loader as well.

[zwei]

You know what would make this mod even more useful?  It would be cool if it could somehow be easily bundled with the Noob pack or another mod pack.

It can be done, but remember that current soundpack is quite large. Lazy Newb Pack did not include it for this reason, but there are no technical reasons i am aware of that prevent this.

Suppose we could make "light" version of soundpack that would contain clue-sfx for important events and soundsense itself will loose several megs once i get rid of useless text-to-speech library (it works and can echo log lines, but it is not used for anything because voice sucks )

...

So it seems GNU is resolved - soundpack it goign to be unaffected by it.

Having delay option sounds like it is going to be usefull for sound overlays and with it it should be easy to simulate repeats / consecutive sounds and is trivial to add, so i will add that first and ponder about some better scheme to overlay sounds.

[Rainseeker]

I think a good way to do it would be to work out with the noob pak people a way to incorporate a skeleton file system that you could download and unzip Soundsense to.  At that point maybe they could use their program to allow players to manage the sounds if they want?

[Thundercraft]

You know what would make this mod even more useful?  It would be cool if it could somehow be easily bundled with the Noob pack or another mod pack.

It can be done, but remember that current soundpack is quite large. Lazy Newb Pack did not include it for this reason, but there are no technical reasons i am aware of that prevent this.

Suppose we could make "light" version of soundpack that would contain clue-sfx for important events and soundsense itself will loose several megs once i get rid of useless text-to-speech library (it works and can echo log lines, but it is not used for anything because voice sucks )

Personally, I do not think of this as a good idea. And I doubt LucasUP wants to incorporate SoundSense into the Lazy Newb Pack. Besides, on the previous poll for the Lazy Newb Pack thread the majority voted for a small pack with only the most important utilities. Don't get me wrong: I love SoundSense and I think it adds a lot to the DF experience! But it's not for everyone.

Looking at it another way, the current Lazy Newb Pack is 22.39 Mb and merely adding the SoundSense runtime package (with hardly any sounds) would increase that by 9.23 Mb... which is an increase of over 40%. If you add both the SoundSense package and the current SoundPack, that would push it to almost 143 Mb, which is about a 637% increase.

Also, both the Lazy Newb Pack and the SoundSense project has been getting rather frequent updates recently. I'm sure many newbie and first-time players would complain about having to download over 130 Mb (being generous here) for each new update or just to play the game to see if they like it.

As for the suggestion of getting SoundSense incorporated into a game package (such as Lazy Newb Pack) by stripping the sounds down to the basics: Wouldn't that cripple the SoundSense experience? I'm worried that it would only give newbies a taste of what the full SoundSense sound package can do and they may or may not ever learn what they're missing. They might be rather disappointed in a "lite" version and hence could be more prone to not trying out the "full"... Just a thought.

[Rainseeker]

See my previous post.   

I am not suggesting to add anything other than a file folder system to allow easy install by downloading SoundSense to, or a "Newb Friendly" version.  There may need to be collaboration, and instructions within the Newb Pack on how to download and install the sounds, but it's do-able and would be nice to include.

[zwei]

Well, I have stripped sound SoundSense to 0.5MB by removing text to speech engine and small refactoring:

http://df.zweistein.cz/soundsense/soundSense_24_108.zip

(also added delay attribute to sound tag - <sound logPattern="Dummy5" delay="1000"> is one seccond delay).

This is small enough to be bundled with something and it can be reduced by aditional 114Kb if you remove dummy/example pack.

[zwei]

New soundpack finished uploading: http://df.zweistein.cz/soundsense/soundpack.zip

Some small changes amd poorly constructed danger room will now sound like it should (instead of implaled death messages not producing any sound).

Weee!