# Basic info <!--- Add commit SHA used to reproduce --> - Float reference: - Encoder (float): ivas-float-update - Decoder (float): ivas-float-update - Fixed point: - Encoder (fixed): - - Decoder (fixed): ivas-main # Bug description It was observed that the dirac renderer azimuth and elevation angles compted in ivas_dirac_dec_compute_directional_responses_fx() often are off by one degree because of the lack of rounding and usage of extract_h instead of round_fx. Since the angle computation occurs only few times per channel, it is not really expensive to use round_fx instead of extract_h and it is suggested to prefer round_fx in that case. # Ways to reproduce [stvOMASA_2ISM_2MASA2TC48c+10dB.wav_OMASA_2Dir2TC_2ISM_at_br_sw_techs_13_2_to_512_kbps_start_48_kbps_48kHz_in_48kHz_out_7_1_out_JBM_Prof_5.netsimout](/uploads/09ff4234f4da5dd891f0f80736d37c35/stvOMASA_2ISM_2MASA2TC48c+10dB.wav_OMASA_2Dir2TC_2ISM_at_br_sw_techs_13_2_to_512_kbps_start_48_kbps_48kHz_in_48kHz_out_7_1_out_JBM_Prof_5.netsimout) While decoding the attached bit stream, the angles computed in ivas_qmetadata_to_dirac_fx() differ by one degree causing the directional responses vectors to be different between float and basop. command line: `IVAS_dec -q -fr 20 -Tracefile stvOMASA_2ISM_2MASA2TC48c+10dB.wav_OMASA_2Dir2TC_2ISM_at_br_sw_techs_13_2_to_512_kbps_start_48_kbps_48kHz_in_48kHz_out_7_1_out_JBM_Prof_5.dectrace -VOIP 7_1 48 stvOMASA_2ISM_2MASA2TC48c+10dB.wav_OMASA_2Dir2TC_2ISM_at_br_sw_techs_13_2_to_512_kbps_start_48_kbps_48kHz_in_48kHz_out_7_1_out_JBM_Prof_5.netsimout out.wav` <!--- Below are labels that will be added but are not shown in description. This is a template to help fill them. Add further information to the first row and remove and add labels as necessary. -->