Hi, I've compiled here ver 3579 codec2 and FreeDV on my ASRock QC5000-ITX box. AMD A4-5000 APU with Radeon(TM) HD Graphics, 800MHz each of 4 CPUs
I cannot get a decode in mode 700D in any other than interleave 1. Files from: http://rowetel.com/downloads/codec2/700d_part4/ top PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 18695 alanb 20 0 1716200 69516 45048 R 93.4 0.9 12:14.09 freedv Do I need a faster box? Alan VK2ZIW On Sat, 5 May 2018 08:46:14 +0930, David Rowe wrote > I've now tested the interleaver with FreeDV GUI on 700D (SVN rev > 3579). The tx and rx must have the same interleaver setting, which > is set on Tools-Options. The number (1,2,4,8,16) is the interleaver > delay in 160ms modem frames, so 16(0.16) is 2.56 seconds (5.12 sec > total from tx to rx). > > Especially at low SNRs, there can be a big delay (10 seconds or more) > for the Interleaver to sync. The modem can also lock up on the > wrong frequency offset, leading to false syncs, hence the ReSync > button. I need to improve acquisition at some stage. This low SNR > world is challenging - the raw bit error rate is 10% at -2dB SNR. > > In terms of bit error rate, a 16 frame interleaver is worth 4dB in > simulated HF fast fading channels. Not sure how it will play out on > real channels for PTT radio. This is experimental radio, so lets > try and find out :-) > > Here are some test files to play into FreeDV: > > http://rowetel.com/downloads/codec2/700d_part4/ > > "fast" refers to a fading channel with 1Hz fade bandwidth, "intX" is > the interleaver depth. > > Notes: > > i) 700C doesn't maintain sync at these SNRs .... > ii) I can't actually hear the modem signal at -2dB > ii) You can make your own simulated files with: > > ./freedv_tx 700D ../../raw/ve9qrp.raw - --interleave 16 | > ./cohpsk_ch - - -24 --Fs 8000 --fast -f -20 | sox -t raw -r 8000 -s - > 2 - ~/Desktop/700D/ve9qrp_1600_int16_fast_snr2dB.wa > > Tweak the cohpsk_ch third argument "-24" (Noise spectral density > value) to get the reported SNR you want. > > -/- > > I'm not sure if my model for SSB is correct. I'm just calculating > the total signal power and adding calibrated noise to get the > desired SNR. I suspect the peaks are higher than they should be. In > a real SSB radio ALC will raise the mic gain of low level segments > of speech. The net effect will be reducing the peak/average power > ratio, taking the peaks down closer to the noise. Need to measure a > real SSB tx, and perhaps model ALC. > > - David > > ---------------------------------------------------------------------------- -- > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > Freetel-codec2 mailing list > Freetel-codec2@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/freetel-codec2 Alan Evil flourishes when good men do nothing. Consider the Christmas child. --------------------------------------------------------------------------- Alan Beard Unix Support Technician from 1984 to today 70 Wedmore Rd. Sun Solaris, AIX, HP/UX, Linux, SCO, MIPS Emu Heights N.S.W. 2750 Routers, terminal servers, printers, terminals etc.. +61 2 47353013 (h) Support Programming, shell scripting, "C", assembler 0414 353013 (mobile) After uni, electronics tech ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Freetel-codec2 mailing list Freetel-codec2@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freetel-codec2
