Lines Matching refs:phase
173 * ProcessL() calls ProcessBase() with TFUNC = InterpNull, for fixed/locked phase.
174 * Process() calls ProcessBase() with TFUNC = InterpCompute, for interpolated phase.
282 * @param phase is the fractional distance between input frames for interpolation:
283 * phase >= 0 && phase < phaseWrapLimit. It can be thought of as a rational fraction
284 * of phase/phaseWrapLimit.
289 * @param coefShift gives the bit alignment of the polyphase index in the phase parameter.
314 * In between calls to filterCoefficient, the phase is incremented by phaseIncrement, where
317 * The filter polyphase index is given by indexP = phase >> coefShift. Due to
323 * lerpP = phase << 32 - coefShift >> 1; // for 32 bit unsigned phase multiply
324 * lerpP = phase << 32 - coefShift >> 17; // for 16 bit unsigned phase multiply
328 * lerpP = phase << sizeof(phase)*8 - coefShift
329 * >> (sizeof(phase)-sizeof(*coefs))*8 + 1;
331 * For floating point, lerpP is the fractional phase scaled to [0.0, 1.0):
333 * lerpP = (phase << 32 - coefShift) / (1 << 32); // floating point equivalent
339 const uint32_t phase, const uint32_t phaseWrapLimit,
351 uint32_t indexP = phase >> coefShift;
352 uint32_t indexN = (phaseWrapLimit - phase) >> coefShift;
364 uint32_t indexP = phase >> coefShift;
365 uint32_t indexN = (phaseWrapLimit - phase - 1) >> coefShift; // one's complement.
384 static const TC scale = 1. / (65536. * 65536.); // scale phase bits to [0.0, 1.0)
385 TC lerpP = TC(phase << (sizeof(phase)*8 - coefShift)) * scale;
390 uint32_t lerpP = phase << (sizeof(phase)*8 - coefShift)
391 >> ((sizeof(phase)-sizeof(*coefs))*8 + 1);
