Specifications
& Features.

Input Formats.

File types	.wav, .flac, .aif, .aiff, .m4a, .dsf, .wv
Channels	Stereo (2-channel)
PCM rates	44.1 kHz to 3.072 MHz
DSD rates	DSD64 to DSD2048

Output Formats.

PCM files	.wav, .wav (RF64), .wv, .flac
DSD files	.dsf or .wv (WavPack)
PCM rates	Up to 3.072MHz
DSD rates	Up to DSD2048
Direction	Upsampling and downsampling
Bit depths	12- to 32-bit fixed-point, 64-bit floating-point

High Precision.

64-bit	Normal
107-bit	Chill
128-bit	Insane
256-bit	Ludicrous

Processing precision is separate from output bit depth. Optimized math libraries were developed with ADVANPIX.

Internal processing precision is separate from the output bit depth your DAC receives and does not increase output file size. PGGB also offers 107-bit processing, alongside 64-, 128- and 256-bit options.

Precision determines how accurately PGGB can calculate reconstructed samples when upsampling or downsampling. It sets the upper limit on numerical reconstruction accuracy, the dynamic range of the reconstructed signal, and how low the quantization noise can go.
Higher precision helps PGGB preserve very small signal changes and reproduce transient timing accurately. Learn more about how PGGB uses precision ↗

New in V9.

Engine	Precision optimized upsampling, on by default Read about the PGGB engine ↗
PGGB-Live	Real-time playback to a DAC or a network renderer (beta)
Live sources	Your files, another player through the PGGB Bridge, or UPnP
PGGB-Remote	Playback control from a browser on your local network
PGGB-ARC	Crossfeed and head measurements for PCM and DSD; room effect for offline processing
FLAC output	Compression now supported up to 768 kHz / 32 bits
Licensing	Self-service, with your email and an activation code
Interface	Dedicated Offline / Live selector; Light and Dark themes, with the initial theme chosen to match your operating system
Live calibration	Basic estimate or complete input-rate sweep, with saved results and no silent output-rate reduction
Playback	Seeking for seekable tracks, tagged queue rows and WavPack file input

Every V8 license works in V9. See What’s new in V9

DSD Reconstruction Accuracy.

Quantization noise (the error introduced when representing audio at the output bit depth) in the audible band

DSD2048	−300 to −800 dB
DSD1024	−290 to −755 dB
DSD512	−275 to −750 dB
DSD256	−225 to −665 dB
DSD128	−200 to −650 dB
DSD64	−150 to −395 dB

Ranges cover 5th- through 25th-order modulators. Six orders are available in three groups: Low Noise, Very Low Noise and Ultra Low Noise. These figures describe numerical processing, not the physical noise floor of a DAC. Full comparison

PCM Reconstruction Accuracy.

Numerical quantization-noise floor in the audible band, by processing precision. These figures do not describe the physical noise floor of a DAC.

256-bit	−740 dB
128-bit	−540 dB
107-bit	−500 dB
64-bit	−360 dB

Small signal accuracy at 256-bit precision

−301 dB signal	Amplitude error: 1e-14 dB; phase error: 1e-28 degrees
−651 dB signal	Amplitude error: 1e-5 dB; phase error: 1e-5 degrees

Algorithms & Controls.

• DSD or PCM resampling modes
• Conversion within or between the 44.1 kHz and 48 kHz sample-rate families
• Adaptive noise shaper and DSD modulators
• Adaptive per-track algorithm optimization
• Inter-sample over detection
• ReplayGain normalization
• Parametric & convolution EQ (107–256 bit)
• High-frequency filters for noise from analog-to-digital conversion (ADC)
• Track trimming and silence adjustment

Other Features.

• Built-in DAC presets for many popular DACs
• Real-time CPU and memory tracking
• Batch process albums in multiple folders
• Combine continuous recording tracks using combine.json in the album folder, with track ranges such as ["1-6", "7-16"]. With the Precision optimized engine, combining tracks no longer changes sound quality.
• Transfer of common metadata tags and cover art
• Multi-core parallel processing
• Dark and Light themes (follows system)
• Tooltips and built-in Help
• Plenty of HHGTTG references