Jump to content

doosh

BI Developer
  • Content Count

    7
  • Joined

  • Last visited

  • Medals

Community Reputation

23 Excellent

4 Followers

About doosh

  • Rank
    BI Developer

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Hello guys, thank you for all your valuable commets and suggestions! We are continuously improving our audio engine and pushing it's capabilities further! I would like to share a short list of upcoming audio features/improvements, which have been implemented and are currently under the scope: SoundSet technology (with all related features) can be used within vehicle configuration speed of sound simulation for vehicle loops (sound controllers/vehicle position history) improvements of surround_panner playTrigger parameter for SoundSets several new sound controllers and simple expressions functions Proper documentation will appear on Community Wiki soon! Currently, the only vehicle using soundSets is Wipeout, so please be patient about this, after we'll fine-tune everything, your feedback will be very welcomed! What I'd like you to ask for is feedback on newly remixed close shot sound of MX, which is present in actual Dev. Thank you and stay tuned! _d
  2. As you might have gathered from a few tidbits of information dropped here recently, there are a few essential changes coming to A3's audio. Although public progress might have seemed a bit slow since our recent Audio Roadmap, in truth, we've been working hard to preparing the ground work for all future audio impovements, which basically means the low-level audio engine was completely refactored to support all the new features. Together with engine refactoring, we considered it necessary to make a 'cleansweep' of vanilla sound data, so let me announce this first: Reorganization of sound data (Sounds_F_*): - There is a new structure of addons containing sounds (some of them were alredy there, some are new): Sounds_F_arsenal, Sounds_F_environment, Sounds_F_characters, Sounds_F_sfx, Sounds_F_vehicles - These addons correspond with subfolders within the Sounds_F folder - Remaining folders (air, ambient, missions, structures and weapons) will be DELETED, but samples from there were already copied to appropriate folders (arsenal, environment, characters, sfx, vehicles) - Addons Sounds_F_EPB, Sounds_F_EPC and Sounds_F_Bootcamp will be removed and samples from there will be also relocated to appropriate folders. - Sounds_F addon will remain, but it will include configuration only. - Sounds within the DLCs will remain untouched. (Sounds_F_Mark, Sounds_F_Heli, Sounds_F_Kart) Second reorganization was performed on Radio Protocol Data (Dubbing_Radio_F_*): - Sets of samples for specific speakers are no longer organized according to Voice Type, but speaker's spoken/native language and class name - This means all the samples were moved to different folders; Dubbing_Radio_F_data addon was split into several smaller addons based on speaker's spoken/native language. Please note, deleting of obsolete folders/addons/samples will happen with the release of Eden Update, due early next year. In general, things are currently in 'beta' state, so we are finalizing and testing them before announcement. Considering our plans mentioned in the latest Audio Roadmap, it was important to start from the most general improvements, which relate to the unification and customization of sound configuration, together with low-level audio features. Here is quick overview: Unification / Customization: - There is new class called cfgSoundGlobals, which will be continuosly filled (during further development) by previously hardcoded parameters (like general attenuation curves, etc.). - The biggest change is brand new sound configuration architecture, which - together with old features like sample arrays - brings really exciting possibilities! For now, let's mention synchronized multilayering, and custom amplitude attenuation curves in two layers (which can serve for the definition of range-based samples, for example). Low-level audio features: We have implemented a first iteration of handling stereo samples we had in plan: Stereo Emitter. Simple explanation: - Anywhere the engine detects a stereo sample as a source, it creates two virtual speakers in the position of sound source. - While you are approaching sound source location, the engine 'spreads' those speakers (moves them to different sides), so you can experience a stereo effect at closer range. - This approach is appropriate for situations where sound is produced by something clearly localizable (vehicles, explosion, etc.); for reflections and similar sounds, a slightly different mechanism will be used and announced later. - Currently, there are two parameters defined in the MultiChannelEmitter class in cfgSoundGlobals set to temporary values (for testing purposes) - Once we release the new configuration system, it will be set to more appropriate general default values. - Custom values for specific sound will be possible within new configuration system, of course! More details coming soon!
  3. Although we've been checking it continuously, it's been a while since we directly replied to this thread. After some intense development with Marksmen DLC - implementing all those new features and samples - we're now summarizing all your feedback, creating prioritized list of upcoming fixes and investigating new ideas and suggestions. Thank you all for your feedback! Originally, I started to write replies to specific comments, but it turned out to be so long that it became an Audio OPREP.:) It will be a bit more technical and will also include several answers to some general questions (which you are posting to megagoth's ask-bohemia-thread). For now, though, I want to assume you that the Marksmen DLC features are just the beginning - there's a huge backlog of improvements we'd love to implement in time. The new audio features are, let's say, in their first form, which we decided to publish to improve the audio experience as soon as possible. Thank you again for your support!
  4. Thank you for your feedback. This is the first approach which we consider to be an improvement of the existing state, because it affects all sounds not only weapons, where there was nothing before. You are right with the complexity of attenuation in real world, later there is a possibility to implement more sophisticated processing (http://www.persona.uk.com/barnfield/Core_docs/G/G7.pdf) (at least bass frequencies attenuation could happen sooner). Currently, LP cutoff frequency is changing with distance non-linearly, and there is one more important thing to mention: as we are splitting "tails" and the actual "shot" sound, tails will be the exception for this attenuation filter, which will preserve all frequencies.
  5. Hello LAxemann, yes, we decided to improve/redesign distance attenuation feature (air absorption). The idea of custom classes for specific distances (which anyway worked for weapons only) was abandoned, we will use dynamic EQ (LP) instead - for ALL positional sounds (with exceptions). One motivation is that a dynamic filter gives us an opportunity to take other aspects (than the distance) into calculation, like occlusions. Backward compatibility will be preserved (weaponSoundEffect parameter will remain functional) and there will be the possibility to turn this feature off by script command.
  6. Hi, 1st/3rd person samples differentiation is possible only for the vehicles and their loop sounds by using camPos controller in the volume calculation. There is also new playerPos controller which returns int according to the player's position within the vehicle. Currently there is no engine support for doing similar thing within the weapons (shooting). Anyway, this feature is very closely linked to your second question, the possibility to define distance-based samples for shooting (or for the same event generally). It's true, it was under consideration for a long time. But simulation of distance shooting sound is more complicated issue than simple ballance between close shooting samples and the tail samples (resp. crossfading close/distant samples). "The tail" in reality is generated by the environment, which could be so variable. Forest, town, hills, or the interior - all that produces different tails to the basic close-range shooting sound. Yes, I am pointing to different feature which is missing in the game, which would change the overall sound experience. Talking about shooting, it would generate different sound basically for each shot. Using virtually positioned stereo samples with wet/dry parameter of environmental effects based on distance in combination with eq could work much better than simple distant samples.
  7. Thanks guys for all of your feedback! Yes, there is lot of work to do. Soundscape was a bit underestimated aspect of the game, so let's change it together. The list of fixes and features we would love to implement is quite long (and yes, there are improvements for occlusion!) Most of missing sounds from the list needs engine implementation and the list basically fits our actual plan. We will start (already did) with the vehicles. Thank's again, we'll keep you updated.
×