Reply to thread

Message: [QUOTE="Bombastus, post: 90986, member: 2178"] CarbonDSP Biased models two cascaded gain stages, a moving grid-bias point, power-supply sag, and separate input and output transformer saturation. Those stages are not presented as one generic warmth control. They react at different points in the signal path, so changing one can alter the way the next stage is driven. That matters because a fixed waveshaper can produce different distortion as level rises without having an operating point that moves over time. Biased is built around [B]signal-dependent tube bias movement[/B], where CarbonDSP says the grid-bias point shifts with the incoming signal. A quiet phrase and a hard transient can therefore meet the same knob settings and leave the modeled valve in different conditions. The practical consequence is simple. Gain staging becomes part of the sound rather than housekeeping around it, because Drive, Stage 2, Bias, and Sag determine how strongly later parts of the model are excited. The plugin can stay fairly restrained on lower-level material, then become denser or less linear when the source pushes harder. [HEADING=2]Dynamic grid bias changes more than clipping[/HEADING] Grid bias sets a valve's operating point by controlling the voltage relationship between its grid and cathode. In real valve circuits, strong drive can create grid current or alter average current through bias networks, temporarily shifting that operating point. The result is not merely a taller input hitting the same static curve harder. CarbonDSP uses that behavior as a defining distinction for Biased. Its Bias control sets the starting condition, while the modeled bias point can move under signal rather than remaining frozen there. That helps explain why level matching before the plugin matters when you compare settings, since a louder source is capable of changing the model's state as well as producing more instantaneous distortion. This is also why bias and Drive are not interchangeable controls. Drive decides how forcefully the signal reaches the valve stages, while bias changes the condition those stages operate around. You can push harder without choosing the same clipping balance, or move the bias while keeping the incoming level comparatively steady. The [B][URL='https://goldmidi.com/community/threads/carbondsp-introduced-four-audio-plugins.76243/']first CarbonDSP plugin lineup[/URL][/B] introduced that moving-bias behavior alongside three processors built for very different jobs. Biased goes further than the launch description by exposing Stage 2 separately, so the first stage can reshape what reaches the second rather than behaving like one oversized saturation block. [HEADING=2]Sag is a different kind of movement[/HEADING] Power-supply sag comes from voltage dropping when a supply faces greater current demand. In valve hardware that can reduce available headroom and create a softer, compressed response until the supply recovers. It may sound related to bias movement because both are level-dependent, but the electrical cause is different. Biased therefore gives Sag its own control instead of folding the effect into Drive. More sag can make harder passages compress and recover differently without requiring the bias setting itself to move in the same way. That separation is useful on drums, bass, or dense buses where you may want transient softening without simply adding another layer of clipping. CarbonDSP's preset redesign makes the distinction unusually concrete. Its current library includes patches built around running the valve colder, letting the supply sag more heavily, or leaning on the transformer stages without valve drive. The developer also level-matched the presets against untreated music, which reduces the usual temptation to mistake a louder patch for a better one. [HEADING=2]Transformer color does not need valve distortion[/HEADING] Biased places independent Input Iron and Output Iron stages around the valve path. That means transformer coloration can happen before the tubes are driven or after the valve stages have already changed the waveform. Order matters because the first nonlinear or tonal stage changes what every later stage receives. Transformers also become more difficult to keep linear as signal level rises and frequency falls. That is why bass-heavy material can reveal transformer saturation before a midrange test tone at the same nominal level does. Input Iron can therefore change what reaches the valve on a kick or bass part, while Output Iron works on the result after the tube stages. CarbonDSP's own measurements add a useful wrinkle. In its control-interaction testing, the Iron coloration and capacitor Age control changed tone with essentially no added distortion at ordinary test settings, while Drive and Grit consumed most of the distortion budget. Age behaves mainly as a low-pass coloration rather than another clipping stage. Oversampling matters most once those nonlinear stages are pushed. CarbonDSP measured its 16x mode as substantially lowering alias energy on driven and extreme settings, while clean settings already sat near the measurement floor at 4x, 8x, and 16x. That makes the highest quality mode more sensible for hard saturation or offline rendering than for every lightly colored track in a busy session. [/QUOTE]

Name