NastyDLA – technical architecture

simplified technical architecture

Internally, NastyDLA consists of quite a bunch of DSP processing building blocks which as a whole are summing up to an authentic signal path simulation of it’s analog models. The blocks and the according signal flow are shown in the diagram above. Basic signal flow goes from left to right except the feedback path which goes in the opposite direction.

With NastyDLA, signal path coloration already starts in the input stage which provides a complete model of both, frequency and phase response as well as dynamic saturation. It’s located in the dry path but all nonlinear processing and coloring can be disabled on demand so it remains as a simple input volume control then. But while switched in, the input stage can greatly contribute on getting the processed signal to fit right into a mix.

The pre-delay block changes the timing between the following dry and wet signal paths and can be positive or negative which means that the wet signal can be delayed related to the dry path but the opposite as well.

The chorus algorithm is hard-wired and is the very first block in the wet signal chain. Internally it computes four audio lanes and depending on the operation mode of the plug-in (mono, dual-mono/stereo) they are dynamically configured to compute the appropriate audio output. Rate and amount can be adjusted to obtain a broad effects spectrum ranging from subtle and smooth chorusing up to flanger like effects.

Next in the chain is the delay algorithm itself which offers seven different configurations (two mono modes, dual-mono, ping-pong, a cross feedback delay and two retro modes) and contains additional modulation options. Given the modulations, static audio delay imaging can be avoided and even tape-like timing imbalances can be obtained. This is also the place where a sophisticated tape hiss simulation takes place which basically implements a colored and animated noise model. With the two retro modes the dry signal is routed to the left audio output channel and the wet is to the right, being a reminiscent to the analog originals.

Serious amounts of additional phase distortion can be introduced with a dedicated phase section behind the delay block. This can be used for an increased dispersion of re-occurring delays in between the feedback path and contributes to a more “reverb-ish” sound and also the typical screaming sound when the units feedback is driven right into self oscillation.

NastyDLA contains a comprehensive and detailed modeled dynamics section. It consists of a compressor/saturator tandem where not only static waveshaping is applied but subtle frequency dependent compression as well. The basic technology for this is taken from the already released and award-winning FerricTDS tape dynamics simulator. The compressors sidechain can be switched from the processed feedback path back to the dry signal to provide a slight audio ducking effect. Also, the saturator itself is not static at all but aware of dynamics too. Altogether an authentic and consistent saturation experience can be achieved.

Last in the feedback loop resides EQ and filtering. While the frequency adjustable high-pass is a standard DSP 12dB/octave filter, the low-pass filter is a custom design which highly contributes to the overall specific sound especially when feedback is used. Most digital tape delay emulations just offer plain standard DSP resonant low-pass filters where increased resonance is utilized to achieve screaming effects. The obtained results sounds harsh and plastic quite often. NastyDLA avoids this and the desired and typical feedback driven sound is created solely based on smooth filtering, dynamic saturation and massive phase distortion in the feedback path. Instead of a conventional EQ, a simple but effective “niveau” leveling filter is implemented which makes it dead easy to change the overall audio tonality. It’s center frequency adopts automatically to the actual frequencies of the LP/HP filters.

All that meticulous modeling entirely the way through the whole signal path comes to a price at the end: the computational cost. Nonetheless, decision was made to have a no compromise technical design for NastyDLA available and not just yet another flat and digital sounding delay.

Internally, NastyDLA consists of quite a bunch of DSP processing building blocks which as a whole are summing up to an authentic signal path simulation of it’s analog models. The blocks and the according signal flow are shown in the diagram above. Basic signal flow goes from left to right except the feedback path which goes in the opposite direction.

With NastyDLA, signal path coloration already starts in the input stage which provides a complete model of both, frequency and phase response as well as dynamic saturation. It’s located in the dry path but all nonlinear processing and coloring can be disabled on demand so it remains as a simple input volume control then. But while switched in, the input stage can greatly contribute on getting the processed signal to fit right into a mix.

The pre-delay block changes the timing between the following dry and wet signal paths and can be positive or negative which means that the wet signal can be delayed related to the dry path but the opposite as well.

The chorus algorithm is hard wired and is the very first block in the wet signal chain. Internally it computes four audio lanes and depending on the operation mode of the plug-in (mono, dual-mono/stereo) they are dynamically configured to compute the appropriate audio output. Rate and amount can be adjusted to obtain a broad effects spectrum ranging from subtle and smooth chorusing up to flanger like effects.

Next in the chain is the delay algorithm itself which offers seven different configurations (two mono modes, dual-mono, ping-pong, a cross feedback delay and two retro modes) and contains additional modulation options. Given the modulations, static audio delay imaging can be avoided and even tape-like timing imbalances can be obtained. This is also the place where a sophisticated tape hiss simulation takes place which basically implements a colored and animated noise model. With the two retro modes the dry signal is routed to the left audio output channel and the wet is to the right, being a reminiscent to the analog originals.

Serious amounts of additional phase distortion can be introduced with a dedicated phase section behind the delay block. This can be used for an increased dispersion of re-occurring delays in between the feedback path and contributes to a more “reverb-ish” sound and also the typical screaming sound when the units feedback is driven right into self oscillation.

NastyDLA contains a comprehensive and detailed modeled dynamics section. It consists of a compressor/saturator tandem where not only static waveshaping is applied but subtle frequency dependent compression as well. The basic technology for this is taken from the already released and award winning FerricTDS tape dynamics simulator. The compressors sidechain can be switched from the processed feedback path back to the dry signal to provide a slight audio ducking effect. Also, the saturator itself is not static at all but aware of dynamics too. Altogether an authentic and consistent saturation experience can be achieved.

Last in the feedback loop resides EQ and filtering. While the frequency adjustable high-pass is a standard DSP 12dB/octave filter, the low-pass filter is a custom design which highly contributes to the overall specific sound especially when feedback is used. Most digital tape delay emulations just offer plain standard DSP resonant lowpass filters where increased resonance is utilized to achieve screaming effects. The obtained results sounds harsh and plastic quite often. NastyDLA avoids this and the desired and typical feedback driven sound is created solely based on smooth filtering, dynamic saturation and massive phase distortion in the feedback path. Instead of a conventional EQ, a simple but effective “niveau” leveling filter is implemented which makes it dead easy to change the overall audio tonality. It’s center frequency adopts automatically to the actual frequencies of the LP/HP filters.

All that meticulous modeling entirely the way through the whole signal path comes to a price at the end: the computational cost. Nonetheless, decision was made to have a no compromise technical design for NastyDLA available and not just yet another flat and digital sounding delay.