Using FL Studio. -

The Audio Settings page contains options and settings concerning the live sound (wave) output from FL Studio. The options change depending if ASIO or Direct Sound (standard) drivers are selected in the Output selector.

Output - Select the 'live' output device to be used by FL Studio from the list of loaded soundcard drivers (DirectSound and ASIO driver standards are supported). If you have more than one soundcard installed, this control can be used to switch between them. Select ASIO drivers if possible, they are identified by word 'ASIO' somewhere in the name. ASIO (Audio Stream Input/Output) drivers allow the soundcard to communicate with the host computer with high efficiency, so that the audio stream is processed with much lower delays than with standard audio drivers. ASIO generally allows lower latencies with less CPU load when compared to Direct Sound drivers. If your soundcard does not 'natively' support ASIO, there is a 3rd party soundcard driver 'work-around' you can try at www.asio4all.com. Please be aware that you use this ASIO driver at your own risk. If possible, avoid drivers which have 'emulated' in the name, as they have worse performance than DirectSound or ASIO drivers.

Auto close device - Releases soundcard output when FL Studio looses focus, so other applications may use the same output.

ASIO Drivers:

Visible only when using ASIO driver.

Buffer Length - Shows information about the buffer latency the ASIO device will use and is a 'read-only' property. As the buffer is increased the delay between playing a MIDI keyboard, tweaking a knob or making changes in FL Studio and the sound response increases. The aim is to minimize the buffer size without causing buffer underruns (described below). The delay is equal to the buffer setting, shown in ms.
Clock Source - Some audio cards might provide external clock source which can fix sync/output problems. However for most cards work properly with the default "Internal" source selected.
CPU Limit - Due to architectural specifics of ASIO drivers, high CPU usage in some projects might render the host computer system non-responsive. Enabling this option allows FL Studio do detect CPU peaks and prevent such system lockups.
Show ASIO Panel - Opens the ASIO driver settings panel, use this to change latency settings. Settings between 1-4 ms without underruns are 'cutting edge', 5-10 ms are excellent and 11-20 ms are good.
Priority - Sets the priority of the audio mixing thread. Higher = more CPU devoted to the audio mixing thread, but increases the risk of lockups/freezing when CPU demands become high. Lower = greater risk of buffer underruns. Adjust this (in combination with the buffer settings) if you have problems with lockups and or buffer underruns.
Safe overloads - Off: The audio mixing thread is given a very high priority, so that the GUI doesn't cause hiccups in the audio engine. When the audio mixing thread is using all the CPU, it may leave nothing to the Graphical User Interface (GUI), which will then appear frozen. On (default): 'Safe overloads' adapts the mixer priority when CPU overloads occur, leaving a little CPU to run the GUI, so that you can sill interact with FL and minimize the CPU usage.
Underruns - Shows the total underrun count with the current settings. An underrun is counted when the temporary store that holds audio prior to output to your soundcard runs out of data (a click or pop is usually heard), it means the CPU didn't process information fast enough. Testing to reduce underruns should be carried out with a typical project (song) playing. There are a number of ways of reducing underruns as described below. After each change, if the underrun count stops increasing, try to reduce the Buffer length setting further. Your goal is to find the shortest setting with no new underruns:

First, a reminder that as the Buffer length is increased, underruns decrease, but the delay between playing a MIDI keyboard, tweaking a knob and the response of FL Studio also increases. The aim is to minimize the buffer size without causing buffer underruns. For ASIO drivers, settings of 1-4 ms are 'cutting edge', 5-10 ms are 'excellent' and 11-20 ms are 'good'.
Make sure the Mixer Interpolation is set to Linear and the Sample rate is 48,000 Hz or less.
Increase the audio thread 'Priority' setting to 'Highest'.
Turn the 'Safe overloads' switch off.
Download the latest ASIO drivers from your soundcard manufacturer. We recommend sticking with the native ASIO drivers and only trying alternatives if you experience problems with them.
In some cases the 3rd party www.asio4all.com drivers outperform native ASIO drivers and may resolve underrun issues. Please be aware that you use this ASIO driver at your own risk.
Decrease polyphony of the instrument channels.
Turn off 'Keep on disk' for Sampler and Audio-Clip channels. This loads samples into memory which is faster.
Record mixer channels to audio and disable the instruments feeding those mixer channels.
Note: If your Buffer length setting is greater than 50 ms and your CPU usage meter peaks over 80%, it may be simply be your computer is not fast enough to play the project. Welcome to the never ending cycle of PC upgrades!

Standard Drivers:

Visible only when using Standard drivers (DirectSound, WDM, Primary etc).

Buffer Length - This slider controls the audio buffer latency. As the buffer is increased the delay between playing a MIDI keyboard, tweaking a knob or making changes in FL Studio and the sound response increases. The aim is to minimize the buffer size without causing buffer underruns (described below). The delay is equal to the buffer setting, shown in ms. Setting between 5-10 ms without underruns are 'cutting edge', 11-20 ms are excellent and 21-50 ms are good.
Use Polling - Polling is a technique for managing DirectSound's audio buffer, which usually allows much smaller buffer without underruns. On some PC-s, however, it can have the opposite effect.
Use Hardware Buffer - Uses the hardware audio buffer of DirectSound enabled sound cards.
Priority - Sets the priority of the audio mixing thread. Higher = more CPU devoted to the audio mixing thread, but increases the risk of lockups/freezing when CPU demands become high. Lower = greater risk of buffer underruns. Adjust this (in combination with the buffer settings) if you have problems with lockups and or buffer underruns.
Safe overloads - Off: The audio mixing thread is given a very high priority, so that the GUI doesn't cause hiccups in the audio engine. When the audio mixing thread is using all the CPU, it may leave nothing to the Graphical User Interface (GUI), which will then appear frozen. On (default): 'Safe overloads' adapts the mixer priority when CPU overloads occur, leaving a little CPU to run the GUI, so that you can sill interact with FL and minimize the CPU usage.
Underruns - Shows the total underrun count with the current settings. An underrun is counted when the temporary store that holds audio prior to output to your soundcard runs out of data (a click or pop is usually heard), it means the CPU didn't process information fast enough. Testing to reduce underruns should be carried out with a typical project (song) playing. There are a number of ways of reducing underruns as described below. After each change, if the underrun count stops increasing, try to reduce the Buffer length setting further. Your goal is to find the shortest setting with no new underruns:

1. First, a reminder that as the Buffer length is increased, underruns decrease, but the delay between playing a MIDI keyboard, tweaking a knob and the response of FL Studio also increases. The aim is to minimize the buffer size without causing buffer underruns. For standard drivers, settings of 5-10 ms are 'cutting edge', 11-20 ms are 'excellent' and 21-50 ms are 'good'.
2. Make sure the Mixer Interpolation is set to Linear and the Sample rate is 48000 Hz or less.
3. Increase the audio thread 'Priority' setting to 'Highest'.
4. Turn the 'Safe overloads' switch off.
5. Try each of the 4 possible combinations of the 'Use polling' and 'Use hardware buffer' switches. 4 combinations? Two switches with two states (on/off) mean 4 possible combinations, try them all.
6. Switch to ASIO mode (if supported by your soundcard). There is also 3rd party 'work-around' at a www.asio4all.com that allows many non native ASIO soundcards to operate in ASIO mode. Please be aware that you use this ASIO driver at your own risk.
7. Decrease polyphony of the instrument channels.
8. Turn off 'Keep on disk' for Sampler and Audio-Clip channels. This loads samples into memory which is faster.
9. Record mixer channels to audio and disable the instruments feeding those mixer channels.
10. Note: If your Buffer length setting is greater than 100 ms and your CPU usage meter peaks over 80%, it may be simply be your computer is not fast enough to play the project. Welcome to the never ending cycle of PC upgrades!
Auto Close Device - Releases the wave output device when FL Studio loses focus, so other applications may use the same output.

Plugin output:

VSTi/DXi connection plugin

ReWire client

Slave Tempo - When turned on, FL Studio will synchronize with the tempo of the host.

Record Automation - When turned on, remote control messages (MIDI) from the host will be recorded during recording sessions.

Mixer:

Sample Rate - Sets the sample play-back rate used by the mixer. Where possible use the default sample rate of 44100Hz. Many Creative brand cards (the Audigy series for example) have a minimum sample rate of 48000Hz. In this case, please be aware that some older plugins may not perform correctly (usually tuning related issues) although the majority of plugins today are multi-rate compatible.

Interpolation - Sets the live playback sample interpolation method for sampler channels (as opposed to the same settings found in the rendering/exporting audio dialog). NOTE: The interpolation used for exporting audio files is set in the export dialog, so you do not need to modify the Audio Settings interpolation method once set.

Interpolation is the process of smoothly creating changes in sample data when the system is called to ‘invent’ intermediate volume levels between any two known sample points. This happens when samples are transposed from their original pitch, so the benefits of higher quality interpolation will only be audible for transposed samples. FL Studio provides several methods -

Linear interpolation provides the lowest CPU hit with basic linear averaging between samples, however this may result in aliasing (high frequency noises) when samples are transposed far from their original pitch. We recommend linear settings for most live mixing situations.
6-point Hermite is the fastest curve interpolation method and as it provides superior quality to linear interpolation. If you have a fast PC you may like to try this method during critical mixing sessions. However it will use more CPU than linear.
64, 128, 256, 512-point Sinc methods provide, increasingly, the highest quality interpolation, but they are also very cpu intensive. Anything above 64-point Sync is not suitable for live-playback (perhaps one day when we have 8-core 5 GHz CPUs). So why are these methods available? So that if someone requires the highest quality live interpolation they can have it.

Reset Plugins on Transport - Resets all plugins when changing song position, starting playback etc. Uncheck for faster response when changing song position.

Use Mixer as Playback Position - Enable this option if you experience unstable position indicators with WDM drivers (usually under Windows 2000/XP). If you don't have any problems leave this option disabled as it reduces position indicators' refresh rate with large audio buffers.

Preview Mixer Track - Select the mixer track to be used for metronome clicks, audio previews from the Browser and Wave Editor, etc. By default these sounds are sent to the Master Track by (select none /"--"/ to send to the Master Track).