aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Move the AL_EFFECT_NULL state into a separate fileChris Robinson2013-05-234-96/+107
|
* Reword a couple cmake option namesChris Robinson2013-05-231-4/+4
|
* Don't explicitly check for fesetroundChris Robinson2013-05-222-6/+0
| | | | It's C99 standard with fenv.h, which we already check for.
* Use fegetenv/fesetenv to backup and restore the FPU stateChris Robinson2013-05-222-4/+8
|
* Merge the fesetround FPU handler with the assembly oneChris Robinson2013-05-221-22/+23
| | | | | | The assembly is only needed for SSE since we can't access the intrinsics without the -msse switch, which itself would cause SSE code to be generated elsewhere automtically.
* Check for SSE2 and set the denormals-are-zero bit for mixing if availableChris Robinson2013-05-225-8/+22
|
* Set the correct SSE bits for round-to-zeroChris Robinson2013-05-221-1/+1
|
* Use restrict instead of RESTRICTChris Robinson2013-05-2220-103/+99
|
* Recognize DSSPEAKER_5POINT1_SURROUND (and 7POINT1) with DSoundChris Robinson2013-05-221-4/+11
|
* Remove an unnecessary check for vsnprintfChris Robinson2013-05-221-10/+0
|
* Avoid using a temp buffer for al_printChris Robinson2013-05-221-12/+5
| | | | | | | | It's now using two *printf calls, which unfortuantely means there could be a race between the two and cause the message to break up if something else tries to print to the same file. This shouldn't really be a big deal since al_print isn't used that often, and it now allows for lines of practically unlimited length.
* Add casts to silence MSVCChris Robinson2013-05-222-2/+2
|
* Rename DELETE to DELETE_OBJChris Robinson2013-05-213-9/+8
| | | | Because Windows.
* Add a DELETE macro to help destroy objectsChris Robinson2013-05-213-10/+10
|
* Rename the effect state's Destroy method to DestructChris Robinson2013-05-2111-26/+28
|
* Use factories to create and destroy effect statesChris Robinson2013-05-2112-143/+513
|
* Remove some unused codeChris Robinson2013-05-211-5/+2
|
* Use a properly-defined history for the FILTER structChris Robinson2013-05-217-33/+28
|
* Avoid storing the device frequency in the equalizerChris Robinson2013-05-211-18/+19
|
* Auto-generate wrappers to upcast objects before calling user methodsChris Robinson2013-05-2110-87/+58
|
* Use macros to help define vtables for effect statesChris Robinson2013-05-2110-166/+168
|
* Avoid storing the device frequency in the distortion stateChris Robinson2013-05-211-7/+5
|
* The effect state being destroyed can't be NULLChris Robinson2013-05-214-28/+22
|
* Rename some inheritance macrosChris Robinson2013-05-219-76/+76
|
* Pre-scale the chorus and flanger delays to be in samplesChris Robinson2013-05-212-36/+22
|
* Avoid storing some chorus and flanger properties in the effect stateChris Robinson2013-05-212-52/+32
|
* Process 64 samples at a time for chorus and flanger effectsChris Robinson2013-05-202-96/+130
|
* Process 64 samples at a time for some effectsChris Robinson2013-05-205-117/+201
| | | | | This should help with the non-interleaved samples of the output, and allow skipping channels that don't contribute to the output.
* Split flanger delay calculation into separate methodsChris Robinson2013-05-201-69/+60
|
* Split chorus delay calculation into separate methodsChris Robinson2013-05-201-37/+44
|
* Fix possible overflow when converting float to intChris Robinson2013-05-201-2/+2
| | | | | Same as with the mixer, we can only use 25 bits of precision from floats.
* Use some macros to help with deriving typesChris Robinson2013-05-209-93/+85
|
* Remove an unused macroChris Robinson2013-05-201-2/+0
|
* Clear all the old channel gains to 0 before setting themChris Robinson2013-05-202-1/+5
|
* Fix a potential leak when mmdevapi fails to openChris Robinson2013-05-191-0/+3
|
* Avoid recalculating the same coefficientChris Robinson2013-05-191-11/+9
|
* Implement distortion and equalizer effectsChris Robinson2013-05-1810-15/+919
| | | | Code provided by Mike Gorchak
* Modulator fixesChris Robinson2013-05-181-6/+5
| | | | | | | | Modulator functions return 0..1 instead of -1..+1 Apply high-pass filter before modulation Increase fractional precision to 24 bits from 16 Thanks to Mike Gorchak
* Fix overflow issues when converting float to intChris Robinson2013-04-221-7/+10
| | | | | | | | 32-bit floats can only reach between -16777215...+16777215 (25 bits) before it starts losing whole-number accuracy. Some compiler optimizations may break trying to multiply a float by 2147483647.0 when the result is given right to a float parameter, causing it instead to multiply by the nearest representable float value, 2147483648.0.
* Simplify al_try codeChris Robinson2013-03-247-65/+41
|
* QSA is always build-time linkedChris Robinson2013-03-241-6/+2
|
* Reorder some switch casesChris Robinson2013-03-241-34/+32
|
* Add a missing cast to silence MSVCChris Robinson2013-03-221-1/+1
|
* Document QSA backend options, and move PulseAudio's upChris Robinson2013-03-221-17/+31
|
* Add win64build to .gitignoreChris Robinson2013-03-191-0/+1
|
* Avoid an unnecessary size variableChris Robinson2013-03-191-6/+5
|
* Avoid explicitly checking each effect to createChris Robinson2013-03-191-53/+40
|
* Fix MMDevApi backend option textChris Robinson2013-03-191-4/+4
|
* Report the actual ALSA error if setting the buffer metrics failsChris Robinson2013-03-181-2/+2
|
* Use less math to clamp floats to -1...+1Chris Robinson2013-03-171-3/+3
| | | | | The previous code could have issues as precision lowers. This should hopefully work better while only using one if check instead of two.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-14 19:50:46 +0000