CamillaDSP - Cross-platform IIR and FIR engine for crossovers, room correction etc.

Jump to Latest
HenrikEnquist

HenrikEnquist

Member
Joined 2016
xorcz said:
I am just curious, how did you setup the chain - where do you resample and what formats.
Click to expand...
I just used the default settings of Fedora, camilla running at 44.1 kHz and playing 44.1 kHz material. I didn't try to optimize the pulse settings.

lykkedk said:
@Henrik...

Well i was storking your'e github and i see that you have a numpyless development in the github :)

Looking forward to see when it's eventually done.
You also mentioned some time ago, that you are developing a plugin without matplotlib also, is this part of this development (not to rush you at all, just curious) :up:

Good day out there.

Jesper.
Click to expand...
Matplotlib depends on numpy, so numpyless also means matplotlibless! I have all the plots working now, only needs a little cleanup and then it's ready. As a bonus this also made plotting much faster.
 
X

xorcz

Member
Joined 2020
HenrikEnquist said:
I just used the default settings of Fedora, camilla running at 44.1 kHz and playing 44.1 kHz material. I didn't try to optimize the pulse settings.
Click to expand...
Just an idea, in Fedora Chromium, Firefox and also Audacious (default configuration) output with format:
sample spec: float32le 2ch 44100Hz



Is it possible to set capture alsa loopback device also to FLOAT_LE? I tried to change in the asound.conf, but it changed only playback device.
 
X

xorcz

Member
Joined 2020
And one more question, I changed configuration so it is more similar to your setup, but tried again 192000 samplerate:


devices:
adjust_period: 10
capture:
channels: 2
device: hw:Loopback,1
format: S16LE
type: Alsa
capture_samplerate: 44100
chunksize: 4096
enable_rate_adjust: true
enable_resampling: true
playback:
channels: 8
device: hw:1,10,0
format: S32LE
type: Alsa
queuelimit: 100
resampler_type: BalancedAsync
samplerate: 192000
silence_threshold: -90
silence_timeout: 3
target_level: 0


It is crashing:
Code: 
ct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camilladsp] Read config file Some("/opt/camilladsp/configs/config_alsa.yml")
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camilladsp] Config is valid
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::socketserver] Start websocket server on 127.0.0.1:1234
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camilladsp] Wait for config
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camilladsp] Config ready
Oct 29 23:22:20 localhost camilladsp[6574]: Buffer frames 2048
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::filters] Build new pipeline
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::filters] Build from config
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.99277506546254 a2=0.9928623492296514 b0=0.00002182094177779448 b1=0.00004364188355558896 b2=0.00002182094177779448
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.982767874875442 a2=0.9828547203265044 b0=0.00002171136276561948 b1=0.00004342272553123896 b2=0.00002171136276561948
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::filters] Build from config
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::alsadevice] Creating resampler
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.99277506546254 a2=0.9928623492296514 b0=0.00002182094177779448 b1=0.00004364188355558896 b2=0.00002182094177779448
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::audiodevice] Creating asynchronous resampler with parameters: InterpolationParameters { sinc_len: 128, f_cutoff: 0.92591465, oversampling_factor: 1024, interpolation: Linear, window: Blackman2 }
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG rubato] Create new SincFixedOut, ratio: 4.35374149659864, chunk_size: 4096, channels: 2, parameters: InterpolationParameters { sinc_len: 128, f_cutoff: 0.92591465, oversampling_factor: 1024, interpolation: Linear, window: Blackman2 }
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG rubato] sinc_len rounded up to 128
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.982767874875442 a2=0.9828547203265044 b0=0.00002171136276561948 b1=0.00004342272553123896 b2=0.00002171136276561948
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::filters] Build from config
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.8802660261071684 a2=0.8870400687502007 b0=0.0016935106607580248 b1=0.0033870213215160495 b2=0.0016935106607580248
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.8802660261071684 a2=0.8870400687502007 b0=0.0016935106607580248 b1=0.0033870213215160495 b2=0.0016935106607580248
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.9809331968875066 a2=0.981113259870915 b0=0.9905116141896054 b1=-1.9810232283792109 b2=0.9905116141896054
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.9809331968875066 a2=0.981113259870915 b0=0.9905116141896054 b1=-1.9810232283792109 b2=0.9905116141896054
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::filters] Build from config
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.8802660261071684 a2=0.8870400687502007 b0=0.0016935106607580248 b1=0.0033870213215160495 b2=0.0016935106607580248
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.8802660261071684 a2=0.8870400687502007 b0=0.0016935106607580248 b1=0.0033870213215160495 b2=0.0016935106607580248
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.9809331968875066 a2=0.981113259870915 b0=0.9905116141896054 b1=-1.9810232283792109 b2=0.9905116141896054
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.9809331968875066 a2=0.981113259870915 b0=0.9905116141896054 b1=-1.9810232283792109 b2=0.9905116141896054
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::filters] Build from config
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.8802660261071684 a2=0.8870400687502007 b0=0.9418265237143423 b1=-1.8836530474286846 b2=0.9418265237143423
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.8802660261071684 a2=0.8870400687502007 b0=0.9418265237143423 b1=-1.8836530474286846 b2=0.9418265237143423
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::filters] Build from config
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.8802660261071684 a2=0.8870400687502007 b0=0.9418265237143423 b1=-1.8836530474286846 b2=0.9418265237143423
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::biquad] a1=-1.8802660261071684 a2=0.8870400687502007 b0=0.9418265237143423 b1=-1.8836530474286846 b2=0.9418265237143423
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::processing] build filters, waiting to start processing loop
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG rubato::sinc] Generate sincs, length: 128, oversampling: 1024, normalized by: Some(1.0800131580457402)
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::alsadevice] Opened audio device "hw:Loopback,1" with parameters: HwParams { channels: Ok(2), rate: "Ok(44100) Hz", format: Ok(S16LE), access: Ok(RWInterleaved), period_size: "Ok(512) frames", buffer_size: "Ok(4096) frames" }, SwParams(avail_min: Ok(512) frames, start_threshold: Ok(0) frames, stop_threshold: Ok(4096) frames)
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camilladsp] Capture thread ready to start
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::alsadevice] Opened audio device "hw:1,10,0" with parameters: HwParams { channels: Ok(8), rate: "Ok(192000) Hz", format: Ok(S32LE), access: Ok(RWInterleaved), period_size: "Ok(1024) frames", buffer_size: "Ok(8192) frames" }, SwParams(avail_min: Ok(1024) frames, start_threshold: Ok(3072) frames, stop_threshold: Ok(8192) frames)
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camilladsp] Playback thread ready to start
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camilladsp] Both capture and playback ready, release barrier
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::alsadevice] Starting playback loop
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::alsadevice] Starting captureloop
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z INFO  camillalib::alsadevice] Capture device supports rate adjust
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z WARN  camillalib::alsadevice] Async resampler not needed since capture device supports rate adjust. Switch to Sync type to save CPU time.
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::alsadevice] Capture buffer too small, extending
Oct 29 23:22:20 localhost audit[6574]: ANOM_ABEND auid=4294967295 uid=0 gid=0 ses=4294967295 subj=system_u:system_r:unconfined_service_t:s0 pid=6574 comm="AlsaCapture" exe="/opt/camilladsp/camilladsp" sig=6 res=1
Oct 29 23:22:20 localhost camilladsp[6574]: memory allocation of 18446744073709547452 bytes failed
Oct 29 23:22:20 localhost systemd[1]: camilladsp.service: Main process exited, code=killed, status=6/ABRT
Oct 29 23:22:20 localhost systemd[1]: camilladsp.service: Failed with result 'signal'.
Oct 29 23:22:20 localhost audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=camilladsp comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=failed'
Oct 29 23:22:20 localhost systemd-coredump[6580]: Process 6574 (camilladsp) of user 0 dumped core.#012#012Stack trace of thread 6578:#012#0  0x00007f0fdf950bc5 raise (libc.so.6 + 0x3dbc5)#012#1  0x00007f0fdf9398a4 abort (libc.so.6 + 0x268a4)#012#2  0x000055c08c4b1347 _ZN3std3sys4unix14abort_internal17h0f898796c2c7a3a0E (camilladsp + 0x2f1347)#012#3  0x000055c08c4a9e16 _ZN3std7process5abort17h09286292ab5084e4E (camilladsp + 0x2e9e16)#012#4  0x000055c08c4ace0e rust_oom (camilladsp + 0x2ece0e)#012#5  0x000055c08c4c4aa7 _ZN5alloc5alloc18handle_alloc_error17h7945fa3f20835640E (camilladsp + 0x304aa7)#012#6  0x000055c08c2d2859 _ZN10camillalib10alsadevice18capture_loop_bytes17h73f84cbba1b0430dE (camilladsp + 0x112859)#012#7  0x000055c08c33e281 _ZN3std10sys_common9backtrace28__rust_begin_short_backtrace17h58b7d2b40a33b7e1E (camilladsp + 0x17e281)#012#8  0x000055c08c354e05 _ZN4core3ops8function6FnOnce40call_once$u7b$$u7b$vtable.shim$u7d$$u7d$17h286d547bb7dadcc6E (camilladsp + 0x194e05)#012#9  0x000055c08c4b0d1a _ZN3std3sys4unix6thread6Thread3new12thread_start17h5ad4ddffe24373a8E (camilladsp + 0x2f0d1a)#012#10 0x00007f0fdfb053f9 start_thread (libpthread.so.0 + 0x93f9)#012#11 0x00007f0fdfa14b03 __clone (libc.so.6 + 0x101b03)#012#012Stack trace of thread 6574:#012#0  0x00007f0fdfb0b9e8 pthread_cond_timedwait@@GLIBC_2.3.2 (libpthread.so.0 + 0xf9e8)#012#1  0x000055c08c4afae8 _ZN3std3sys4unix7condvar7Condvar12wait_timeout17h179794ff6e1c6db4E (camilladsp + 0x2efae8)#012#2  0x000055c08c49f45f _ZN3std6thread12park_timeout17h1e113769b9fc6214E (camilladsp + 0x2df45f)#012#3  0x000055c08c4aa037 _ZN3std4sync4mpsc8blocking9WaitToken14wait_max_until17ha87d9e217eab05d0E (camilladsp + 0x2ea037)#012#4  0x000055c08c21a840 _ZN3std4sync4mpsc6shared15Packet$LT$T$GT$4recv17had750f2886547df0E (camilladsp + 0x5a840)#012#5  0x000055c08c22333d _ZN3std4sync4mpsc17Receiver$LT$T$GT$12recv_timeout17hdbb738c746756b05E (camilladsp + 0x6333d)#012#6  0x000055c08c20fd6d _ZN10camilladsp3run17hc91f4069af2db3feE (camilladsp + 0x4fd6d)#012#7  0x000055c08c213a97 _ZN10camilladsp4main17h6aca6b1442d367aaE (camilladsp + 0x53a97)#012#8  0x000055c08c215e63 _ZN3std10sys_common9backtrace28__rust_begin_short_backtrace17hd62180310806cf24E (camilladsp + 0x55e63)#012#9  0x000055c08c215e79 _ZN3std2rt10lang_start28_$u7b$$u7b$closure$u7d$$u7d$17hd6392be1a4ffc669E.llvm.9561805414456027721 (camilladsp + 0x55e79)#012#10 0x000055c08c4ae011 _ZN3std2rt19lang_start_internal17h73711f37ecfcb277E (camilladsp + 0x2ee011)#012#11 0x000055c08c214992 main (camilladsp + 0x54992)#012#12 0x00007f0fdf93b1a2 __libc_start_main (libc.so.6 + 0x281a2)#012#13 0x000055c08c20b0ae _start (camilladsp + 0x4b0ae)#012#012Stack trace of thread 6575:#012#0  0x00007f0fdfa16550 accept4 (libc.so.6 + 0x103550)#012#1  0x000055c08c4a8ba4 _ZN3std3net3tcp11TcpListener6accept17hf3e2def4282f28a7E (camilladsp + 0x2e8ba4)#012#2  0x000055c08c4a8d36 _ZN82_$LT$std..net..tcp..Incoming$u20$as$u20$core..iter..traits..iterator..Iterator$GT$4next17hbeb3837f47974a2bE (camilladsp + 0x2e8d36)#012#3  0x000055c08c33da49 _ZN3std10sys_common9backtrace28__rust_begin_short_backtrace17h3f08367dc390a1f0E (camilladsp + 0x17da49)#012#4  0x000055c08c3550ff _ZN4core3ops8function6FnOnce40call_once$u7b$$u7b$vtable.shim$u7d$$u7d$17h5279fb5ef45d6e50E (camilladsp + 0x1950ff)#012#5  0x000055c08c4b0d1a _ZN3std3sys4unix6thread6Thread3new12thread_start17h5ad4ddffe24373a8E (camilladsp + 0x2f0d1a)#012#6  0x00007f0fdfb053f9 start_thread (libpthread.so.0 + 0x93f9)#012#7  0x00007f0fdfa14b03 __clone (libc.so.6 + 0x101b03)#012#012Stack trace of thread 6576:#012#0  0x00007f0fdfb0b6c2 pthread_cond_wait@@GLIBC_2.3.2 (libpthread.so.0 + 0xf6c2)#012#1  0x000055c08c49f0d3 _ZN3std6thread4park17hc96bf027914d3461E (camilladsp + 0x2df0d3)#012#2  0x000055c08c4a9f22 _ZN3std4sync4mpsc8blocking9WaitToken4wait17h5c4c3964dcd70993E (camilladsp + 0x2e9f22)#012#3  0x000055c08c34db00 _ZN3std4sync4mpsc4sync4wait17h4dfc36e7d98b6f00E (camilladsp + 0x18db00)#012#4  0x000055c08c34b19f _ZN3std4sync4mpsc4sync15Packet$LT$T$GT$4recv17h88740c28850c2e29E (camilladsp + 0x18b19f)#012#5  0x000055c08c352ec9 _ZN3std4sync4mpsc17Receiver$LT$T$GT$4recv17h88e73e39572bda4bE (camilladsp + 0x192ec9)#012#6  0x000055c08c33a570 _ZN3std10sys_common9backtrace28__rust_begin_short_backtrace17h2112ec3fb3284dfbE (camilladsp + 0x17a570)#012#7  0x000055c08c354c95 _ZN4core3ops8function6FnOnce40call_once$u7b$$u7b$vtable.shim$u7d$$u7d$17h213c7e082add7064E (camilladsp + 0x194c95)#012#8  0x000055c08c4b0d1a _ZN3std3sys4unix6thread6Thread3new12thread_start17h5ad4ddffe24373a8E (camilladsp + 0x2f0d1a)#012#9  0x00007f0fdfb053f9 start_thread (libpthread.so.0 + 0x93f9)#012#10 0x00007f0fdfa14b03 __clone (libc.so.6 + 0x101b03)#012#012Stack trace of thread 6577:#012#0  0x00007f0fdfb0b6c2 pthread_cond_wait@@GLIBC_2.3.2 (libpthread.so.0 + 0xf6c2)#012#1  0x000055c08c49f0d3 _ZN3std6thread4park17hc96bf027914d3461E (camilladsp + 0x2df0d3)#012#2  0x000055c08c4a9f22 _ZN3std4sync4mpsc8blocking9WaitToken4wait17h5c4c3964dcd70993E (camilladsp + 0x2e9f22)#012#3  0x000055c08c34db00 _ZN3std4sync4mpsc4sync4wait17h4dfc36e7d98b6f00E (camilladsp + 0x18db00)#012#4  0x000055c08c34b19f _ZN3std4sync4mpsc4sync15Packet$LT$T$GT$4recv17h88740c28850c2e29E (camilladsp + 0x18b19f)#012#5  0x000055c08c352ec9 _ZN3std4sync4mpsc17Receiver$LT$T$GT$4recv17h88e73e39572bda4bE (camilladsp + 0x192ec9)#012#6  0x000055c08c2cff92 _ZN10camillalib10alsadevice19playback_loop_bytes17h270d9f7cd99f17e8E (camilladsp + 0x10ff92)#012#7  0x000055c08c33cd1c _ZN3std10sys_common9backtrace28__rust_begin_short_backtrace17h35237c0523ddd84dE (camilladsp + 0x17cd1c)#012#8  0x000055c08c355297 _ZN4core3ops8function6FnOnce40call_once$u7b$$u7b$vtable.shim$u7d$$u7d$17h6d5d1963a1202281E (camilladsp + 0x195297)#012#9  0x000055c08c4b0d1a _ZN3std3sys4unix6thread6Thread3new12thread_start17h5ad4ddffe24373a8E (camilladsp + 0x2f0d1a)#012#10 0x00007f0fdfb053f9 start_thread (libpthread.so.0 + 0x93f9)#012#11 0x00007f0fdfa14b03 __clone (libc.so.6 + 0x101b03)
 
lykkedk

lykkedk

Member
Joined 2002
HenrikEnquist said:
I just used the default settings of Fedora, camilla running at 44.1 kHz and playing 44.1 kHz material. I didn't try to optimize the pulse settings.


Matplotlib depends on numpy, so numpyless also means matplotlibless! I have all the plots working now, only needs a little cleanup and then it's ready. As a bonus this also made plotting much faster.
Click to expand...

This is really great news Henrik.

Tell me if you need some external testing :p

Jesper.
 
P

phofman

Member
Joined 2005
Perhaps this?

Code: 
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::alsadevice] Capture buffer too small, extending
Oct 29 23:22:20 localhost audit[6574]: ANOM_ABEND auid=4294967295 uid=0 gid=0 ses=4294967295 subj=system_u:system_r:unconfined_service_t:s0 pid=6574 comm="AlsaCapture" exe="/opt/camilladsp/camilladsp" sig=6 res=1
Oct 29 23:22:20 localhost camilladsp[6574]: memory allocation of 18446744073709547452 bytes failed
 
HenrikEnquist

HenrikEnquist

Member
Joined 2016
phofman said:
Perhaps this?

Code: 
Oct 29 23:22:20 localhost camilladsp[6574]: [2020-10-29T22:22:20Z DEBUG camillalib::alsadevice] Capture buffer too small, extending
Oct 29 23:22:20 localhost audit[6574]: ANOM_ABEND auid=4294967295 uid=0 gid=0 ses=4294967295 subj=system_u:system_r:unconfined_service_t:s0 pid=6574 comm="AlsaCapture" exe="/opt/camilladsp/camilladsp" sig=6 res=1
Oct 29 23:22:20 localhost camilladsp[6574]: memory allocation of 18446744073709547452 bytes failed
Click to expand...
Yes and there should be a few trace level messages before. But it doesn't matter, I figured it out :) There will be a fixed version soon.
 
HenrikEnquist

HenrikEnquist

Member
Joined 2016
v0.4.0-beta5 is up. Changes:
New features:

  • Add loglevel option.
  • Use local time instead of UTC in log messages.
Bugfixes:

  • Fix buffer size handling for alsa capture (the bug reported by xorcz).
This will probably be the last beta before a proper 0.4.0 release. But I'm getting a bit tired of the leading zero, maybe it's time to bump it to 1.0.0...
 
HenrikEnquist

HenrikEnquist

Member
Joined 2016
W

wineds

Member
Joined 2006
HenrikEnquist said:
The new gui is also up now!
It can be used without numpy and matplotlib. The plots are now generated in the browser, and as a bonus plotting is also much faster.


To run it you need:
Backend v0.5.0: Release v0.5.0: Merge pull request #6 from HEnquist/develop * HEnquist/camillagui-backend * GitHub
pycamilladsp v0.4.1: Release v0.4.1: Merge pull request #8 from HEnquist/develop * HEnquist/pycamilladsp * GitHub
pycamilladsp-plot v0.4.0: Release v0.4.0 * HEnquist/pycamilladsp-plot * GitHub
Click to expand...

There still seems to be a matplotlib dependency in plot_filters.py line 4 of pycamilladsp-plot v0.4.0 viz :

import matplotlib

BTW the new plots with impulse and phase are very nice! Plotting takes about 7 seconds though.
 
Last edited:
HenrikEnquist

HenrikEnquist

Member
Joined 2016
wineds said:
There still seems to be a matplotlib dependency in plot_filters.py line 4 of pycamilladsp-plot v0.4.0 viz :

import matplotlib

BTW the new plots with impulse and phase are very nice! Plotting takes about 7 seconds though.
Click to expand...
Oh right, I'll fix that. It still uses matplotlib for the command line tool, but it should only import it when running that.

7 seconds is annoyingly slow. What hardware was that done on, and how long was the filter?
 
T

torgeirs

Member
Joined 2011
Thank you, Henrik for your answers in post #1226 CamillaDSP - Cross-platform IIR and FIR engine for crossovers, room correction etc.
I accept that windowing is needed both in time and frequency domain in real world signal prosessing:yes:

Does anyone know if polyphase techniques can be used when a FIR filter with FIR sampling freq = M x signal sampling freq (like FIR at 192k fs and signal 48k fs, M = 4)
Thinking og page 23 in this:
http://www.ws.binghamton.edu/fowler.../EE521_files/IV-05 Polyphase Filters_2003.pdf

To Implement Polyphase Decimation
•Chop up filter into M sub-filters
•Chop up signal into M sub-signals
•Filter each sub-signal w/ a sub-filter
•Add outputs point-by-point

My naive approach is then to use signal as lovest freq sub-signal and pad the rest sub-signals with zeros.

Quess the sub-filters must be combined by last bulletpoint?
But the signal can always be downsampled to signal fs afterwards.

Then just one FIR filter is needed for all (FIR fs)/M signal sample rates
 
Last edited:
T

torgeirs

Member
Joined 2011
And when I'm at it:
FIR filter that has only unity gain and 0 phase in upper half of spectrum. Can polyphase be used to simplify filtering of upper sub-signal?

Think alot of FIR filters out there do nothing to the signal over 24k Hz and 96 kHz sampling frequency is used. (Disregarding windowing :cool:)
0Hz - 24 kHz: Filter is changing signal
24kHz - 48 kHz: Filter is not changing signal, CPU power wasted
 
Last edited:
HenrikEnquist

HenrikEnquist

Member
Joined 2016
torgeirs said:
And the magic lies in the summing of the subfilters (last bullet point) This is where the anti aliasing occurs as described here for quadrature version:
Polyphase quadrature filter - Wikipedia
And it also shows that this tecnique is widely used in audio. (Subband coding and other)
Click to expand...
I started looking a little at polyphase filters when I started on the resampler. I actually use a simple form in the asynchronous resampler. But doing more general filtering seems a bit tricky and I would need to do a lot of reading, so I didn't continue along that path. It's interesting though and I would like to dig into it. If I only had more time...
 
You must log in or register to reply here.
Top Bottom