# Basic info <!--- Add commit SHA used to reproduce --> - Float reference: - Encoder (float): 2e062fa279923ebc9b39600e34d44c36cc145de2 - Decoder (float): 2e062fa279923ebc9b39600e34d44c36cc145de2 - Fixed point: - Decoder (fixed): 4c82f1d24d39d0296b18d775f18a006f4c7d024b # Bug description There is a somewhat high (18.4) MLD score for McMASA BASOP decoder with rate switching and FER involved. Looking at spectrogram does not show any clear reason. Time-domain plot indicates a couple difference spikes but it is hard to say if these are the main reason. Plots below.   # Ways to reproduce Use this rate switch profile [mcmasa_7_1_4_techs_start_48k_20frames.bin](/uploads/4535fc5195aca6995ddaee4f4bfc3ff5/mcmasa_7_1_4_techs_start_48k_20frames.bin) <!-- Commandline or script --> ```bash IVAS_cod -no_delay_cmp -mc 7_1_4 -max_band wb mcmasa_7_1_4_techs_start_48k_20frames.bin 48 ltv48_MC714.wav test_0010.bs eid-xor -fer -vbr -bs g192 -ep g192 test_0010.bs ep_10pct.g192 test_0010.bs_eid IVAS_dec -no_delay_cmp BINAURAL 48 test_0010.bs_eid 0010_ltv48_MC714.wav_mcmasa_7_1_4_techs_start_48k_20frames_48-48___BINAURAL-no_delay_cmp.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. -->