swan: New sound core #2401
Closed
swan: New sound core #2401
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Now outputs 24000 Hz digital samples, matching real hardware (as opposed to 3072000 Hz samples). It has also been rewritten from scratch to match recent research and make the code significantly more readable.
Don't worry, seems like VGM export is totally broken anyway. |
|
Ah! In that case, I'll wait for further feedback. |
|
Works great now! |
tildearrow
requested changes
Mar 11, 2025
|
If you're willing to make both cores available, that's perfectly fine by me. However, I don't think the old core exposes internal speaker output, so that functionality won't work on it. |
tildearrow
requested changes
Mar 11, 2025
|
Merged #2401 into master. Sadly, GitHub has failed to detect merge... |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I've come across a recent push to optimize Furnace's sound drivers with
acquireDirect(). I've also spent months researching the WonderSwan console and noticed that Furnace's sound core uses the old implementation from Mednafen, which does not incorporate recent discoveries. In particular, it was found that the WonderSwan has a fully digital output pipeline of 16-bit 24000 Hz stereo samples (the headphone adapter has its own DAC onboard!), and not an analog (here: 3072000 Hz) output like other sound chips typical of its era.As such, I've rewritten the WonderSwan sound driver to, in a hopefully clean and readable manner, implement the sound system as is currently understood. It also provides a very preliminary hook for Hyper Voice support (which can pretty much be used as a direct headphone-only 16-bit 24000 Hz stereo DAC channel; one can use sound DMA to drive it, but Furnace doesn't really appear to concern itself with emulating DMA limitations at this time).
I've reverted the
acquireDirectchange and some of the optimizations as I question their usefulness (relative to code clarity) when processing a few hundred samples per call as opposed to a few dozen thousand; I'm ready to change it back or optimize it further in case benchmarks show that was a bad call.I've also taken the opportunity to add support for the WonderSwan's internal speaker output. It's 8-bit mono as opposed to 16-bit stereo, and requires manually controlling the loudness for optimum performance. By default, I've set it to minimum loudness (let's call it 100%), which should be ideal for most tracks, but may lead to quiet audio on real hardware if only one channel is playing at a time. Some artists may wish to use the newly added
Headphone outputtoggle in the chip manager, and adjust with the newly proposed20xxeffect if so.TODO:
It appears one of the changes breaks VGM export, and I'm really not sure which one yet...Apparently not any of mine - it was an upstream issue.Feedback, benchmark and testing welcome!