-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The problem is not the sample frequency of the final audio mix. What i think is happens here is that most probably the Onyx and the debian recorder are not using the same sample rate at the time of recording, so the samples that are taken every second on the mixer (probably at a frequency of 48Khz?) are being spreaded over a 44.1Khz file, generating a 44.1Khz audio file that is longer in time, with a consequent change in pitch. Now, I have no information about what sample frequency the onyx 1640i is synched at, since the firewire module used by jackd does not "see" the onyx as an audio card, and doesn't seem to have the possibility to set the sample frequency of the mixer through jackctl. What i think it's happening is that jackd is simply setting its own sample rate, and not synching the mixer or being synched by the mixer. In pro-audio environments these kind of problems are normally avoided by using a word clock coax cable between the two devices, or by using AES-EBU connections (that carry the word clock on one pin, and therefore synch all devices in the chain automatically) but i ignore how this is achieved with the only use of firewire (no word clock coax connectors on the devices either here). The drivers for Mac OS and Windows provided with the onyx give at least the possibility to monitor and set the sample rate the onyx synchs to, so that an operator can at least choose safely at what sample rate to record (although this does not, in my experience of digital audio, give the guarantee of a perfect synch). Here i don't even get a debug line about the onyx synch to sample rate in the jackctl log file, so what i can come to is only assumptions. Playing mp3s (so 44.1Khz files) with mocp through the onyx by using jackctl has the effect of the songs improvisely changing pitch and speed, without actually having any change in what jackctl shows as a sample rate in its panel. This means obviously that the onyx is switching between different synchs without jackd noticing it. On 02/27/2015 05:06 PM, Chris Caudle wrote: >> Date: Thu, 26 Feb 2015 17:18:35 +0100 From: sub >> <subvertao@xxxxxxxxxxxxx> Message-ID: >> <54EF475B.8040601@xxxxxxxxxxxxx> i received a mail from the band >> in which they tell me that the mix is faster and at a higher >> pitch than the audio they recorded through a video camera at the >> concert. > > Video cameras record at 48k rate, so someone will need to rate > convert. Either the camera audio will need to be converted to 44.1k > or your audio will need to be converted to 48k. Which you choose > probably does not matter, but everyone involved needs to know and > agree on the chosen project sample rate. If the final output will > be a video then converting every thing to 48k is probably the > easiest choice. If the final output will be a CD (or audio only > files) then probably 44.1k is the easiest choice. If you will > have both a video and audio only files, my choice would be working > in 48k and then sample rate convert the final mix to 44.1k for CD. > -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQIcBAEBAgAGBQJU9OxdAAoJEHdKSBXNuVOyElcQAMOd4tYrZ3oQI9+rC6FtD6we 23VOUW/f357oJkS2V0fXaO0RYpMQUyKw4FvFL6FpN+ixoskUbSwdrwXHao7F04Fx 4OQrncVSLo4GmKDt/inTZ678ki/RXUEC2G3ToEZPL4x9bmaWN35YVVMmLsMh4fir MeWmUVDOuJQdhmWkcndqftkQ5SbWPmOvZ+uSmyu3tPM2x8H13iMMQhl9S83cGy2A ew+TKdD7Rzeu8uPgDrqgdKIjuX+DIs0aZsueL3EvycP7/Szdv++i6+3JOEQ1RV3/ sUZutqIOSte82CeG78tKNtro5ezpXJoMkR/tgBl4QQ9ZQc+rKwtUBzTB0uv7unRC ETOtzLUko4xAI9Cp+YlxdayTmo9S5cDazlKTkM6pIGOo6ZBnynVydIZsyOXgsOdt tSu9kDRQezS6x/GpTB5SOVZij0LnEfFF0RXkqLB3riNleYKMGNXec/1YJfW6rtb8 Eds2aB48vwAVdPzAF0AF7IWiuPfNwiUsHMFL92N585teNCqYu3Th7CujNKQllOtF AG0pdtIb8tpxTCoMuwm4LzQvv3hf4Dnl9f6zMjbZ2WUUqTVL3MTYqXtMde+VyJ9w dfafFWL9hlj6q7irv06bxDMVFNugLLL2jlLcphYIUTJIDjQyzblukZNQVtxoj5KH vVGx6W92QG8eo7ZRhB2S =TcU7 -----END PGP SIGNATURE----- _______________________________________________ Linux-audio-user mailing list Linux-audio-user@xxxxxxxxxxxxxxxxxxxx http://lists.linuxaudio.org/listinfo/linux-audio-user
