aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Use a constant value for the post-position paddingChris Robinson2015-10-153-36/+26
|
* Store the source's previous samples with the voiceChris Robinson2015-10-154-92/+49
| | | | | | This helps avoid different results when looping is toggled within a couple samples of the loop point, or when a processed buffer is removed while the source is only a couple samples into the next buffer.
* Fix absolute path detection on WindowsChris Robinson2015-10-141-2/+4
|
* Silence implicit double-to-float cast MSVC warningChris Robinson2015-10-141-1/+1
|
* Properly limit the calculated source offset componentsChris Robinson2015-10-141-4/+8
|
* Print the wave type being playedChris Robinson2015-10-141-2/+14
|
* Make the tonegen sampling rate match the device rate by defaultChris Robinson2015-10-141-18/+14
|
* Add a tone generator test programChris Robinson2015-10-132-0/+269
| | | | | | Currently used to test the general output, particularly the resampler, by checking the results with a spectrum analyzer and/or oscilloscope (for example using PulseAudio's "Monitor of ..." devices feeding an external app).
* Shut GCC upChris Robinson2015-10-131-1/+1
|
* Rename a couple HRTF Mode combo box entriesChris Robinson2015-10-131-2/+2
|
* Properly apply fractional source offsets when a user offset is setChris Robinson2015-10-131-22/+22
|
* Fix HRTF group box sizeChris Robinson2015-10-121-1/+1
|
* Replace the resample_fir6 declaration with resample_fir8Chris Robinson2015-10-121-1/+1
|
* Remove 96khz from the combo boxChris Robinson2015-10-121-5/+0
| | | | | | It can still be entered in manually, as can any value between 8khz and 192khz, but having it selectable can create the mistaken idea that selecting it is better than picking 44.1khz or 48khz.
* Reorganize the HRTF group boxChris Robinson2015-10-123-65/+54
| | | | | The three radio buttons are moved to a combo box, which gives a bit more width for the file list contents.
* Only show certain options that are available at build-timeChris Robinson2015-10-112-11/+90
|
* Reformat the CPU Extensions' checkboxesChris Robinson2015-10-111-6/+6
|
* Avoid multiple sin, cos, and sqrt calls for filter calculationsChris Robinson2015-10-111-31/+35
|
* Update changelogChris Robinson2015-10-111-3/+1
|
* Replace the sinc6 resampler with sinc8, and make SSE versionsChris Robinson2015-10-118-23/+177
|
* Move the FIR4 from SSE2 to SSE3Chris Robinson2015-10-1112-80/+173
| | | | | SSE3 can avoid the slow _MM_TRANSPOSE_PS4 call thanks to the inclusion of horizontal adds.
* Slightly improve the FIR4 SSE resamplersChris Robinson2015-10-112-7/+6
|
* Use doubles to calculate the Lanczos coefficientsChris Robinson2015-10-091-20/+20
|
* Update changelogChris Robinson2015-10-081-2/+2
|
* Allow apps to request a specific HRTFChris Robinson2015-10-072-17/+50
|
* Use the enumerated HRTF list for selecting an HRTFChris Robinson2015-10-064-148/+57
| | | | Also report the proper specifier of the one currently in use.
* Store the Hrtf struct with the corresponding HrtfEntryChris Robinson2015-10-062-16/+83
| | | | | Also store the filename with the Hrtf struct so it can be reused for multiple HrtfEntry objects.
* Try to use the full path for the CWD instead of .Chris Robinson2015-10-061-8/+26
|
* Enumerate and list HRTFs per-deviceChris Robinson2015-10-065-4/+131
|
* Skip empty data directory entriesChris Robinson2015-10-041-3/+6
|
* Implement SearchDataFiles for WindowsChris Robinson2015-10-042-11/+298
|
* Add a function to get a list of data filesChris Robinson2015-10-036-4/+217
| | | | | | The method takes a marked-up filename (e.g. may include %r for a sample rate, %% for %, etc), and returns a vector of strings of found filenames that match. It will search the CWD, the local, and global data directories, in that order.
* Add methods to enumerate and query device HRTFsChris Robinson2015-10-032-36/+91
| | | | Currently just returns a dummy entry.
* Combine two arraysChris Robinson2015-10-011-13/+11
|
* Move the resampler stuff to mixer.c where it's usedChris Robinson2015-10-015-54/+53
|
* Search more aggressively for libs needed by SDL_soundChris Robinson2015-09-301-169/+218
|
* Revert "Only use SDL1 with SDL_sound"Chris Robinson2015-09-301-23/+18
| | | | This reverts commit 8883ce1cef86c07d6eaa7e0eff2e132e181184f0.
* Avoid double-checks for the stepping mixer loopsChris Robinson2015-09-303-8/+16
|
* Replace separate vaddq_f32/vmulq_f32 calls with a vmlaq_f32Chris Robinson2015-09-301-1/+1
|
* Remove the --make-tab option from makehrtfChris Robinson2015-09-291-93/+1
| | | | We haven't used a built-in HRTF table for a while now.
* Implement a 6-point sinc-lanczos filterChris Robinson2015-09-2913-26/+63
|
* Merge pull request #21 from aaronmjacobs/masterkcat2015-09-291-1/+1
|\ | | | | Fix resample_fir4 link error
| * Fix resample_fir4 link errorAaron Jacobs2015-09-291-1/+1
|/
* Only use SDL1 with SDL_soundChris Robinson2015-09-281-18/+23
|
* Auto-replace the "cubic" resampler with "sinc4"Chris Robinson2015-09-283-3/+8
|
* Update changelogChris Robinson2015-09-271-1/+1
|
* Replace the cubic resampler with a 4-point sinc/lanczos filterChris Robinson2015-09-2711-39/+52
|
* Don't keep selecting the mixer to useChris Robinson2015-09-275-36/+34
|
* Increase the max pitch to 255Chris Robinson2015-09-263-4/+4
| | | | | | | Note that this is the multiple above the device sample rate, rather than the source property limit. It could theoretically be increased to 511 by testing against UINT_MAX instead of INT_MAX, since the increment and positions are using unsigned integers. I'm just being paranoid about overflows.
* Use the macros to allocate the wrapper backendsChris Robinson2015-09-241-6/+2
|