-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Fons Adriaensen a écrit : Thanks a lot! It's been some time I was waiting for such a meter! Of course the adress is : http://www.kokkinizita.net/linuxaudio/downloads/ - -- Raphaël Doursenaud _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user

Under the K-20 system, it seems the maximum SPL when monitoring with two speakers would be a scary 106 dB(C)

I think the K-14 system might be more to my own taste. I believe it keeps the 83 dB(C) reference level, but with peaks up to 100 dB(C) in a two-speaker setup. Sustained exposure for 2 hours even at the maximum level should still be safe as per the OSHA limits, but of course that level should rarely be reached anyway.

I think the K-14 system might be more to my own taste.

Hello all, The first release of jkmeter is now available in the usual place: <www.kokkizinita.net/linuxaudio/downloads> Also updates of japa and jnoise (mostly maintenance, but see also below). From the README: jkmeter-0.1.0 (03/08/2008) -------------------------- Jkmeter is a horizontal or vertical bargraph level meter based on the ideas of mastering guru Bob Katz. See <http://www.digido.com/bob-katz/index.php> and follow the links on 'level practices'. This is the type of meter you want for live recording, mixing and mastering. It probably makes no sense to use it on all tracks of a DAW, where keeping digital level within limits is the main purpose of metering. This release implements the K-20 meter. Future releases will include the K-14 meter as well. A K-meter displays both the true RMS level and the digital peak level. The ballistics as defined by Bob Katz are somewhat ambiguous. In this implementation the RMS meter is about 15% faster than an VU, but without the overshoot. This provides a good indication of subjective loudness. Instead of providing extra gain for the RMS level, the K-meter displays it on the same scale as the digital peak level, but puts the '0dB' mark and the color change well below the OdB full scale level. For the K-20 meter it is 20dB down, for the K-14 this is (surprise !) 14 dB. To use the meter as envisaged by Bob Katz, you should have a fixed monitoring level, adjusted so that pink noise indicating 0dB on the meter corresponds to 83 dB(C) (from each speaker) as indicated by an sound level meter. Note the (C) - not (A) - weighting. As of release 0.4.0, both japa and jnoise provide a pink noise source at exactly this level. The current release does not include the 22kHz lowpass filter required for frequencies such as 96kHz and higher. Enjoy ! -- FA Laboratorio di Acustica ed Elettroacustica Parma, Italia O tu, che porte, correndo si ? E guerra e morte ! _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user

Fons Adriaensen wrote: Hi, I was a little sceptical at first, but having read the article I think it's a great idea. My main concern was the magic number '83 dB(C)', which has somewhat vague, historic origin, but also I thought I would go deaf at that reference level. With two speakers, I guess that would be 86 dB(C). Quite interestingly, extrapolating the OSHA noise exposure limits down to that level, a person could endure that for up to 16 hours a day without hearing loss. That would make it, effectively, the loudest safe level for someone who mixes music all day, and so making the fullest use of the ear's comfortable dynamic range. (OSHA limits are detailed on http://www.elvex.com/facts07.htm ) As an experiment, I tried calibrating both of my monitor speakers with -20 dBFS pink noise to 83 dB(C). Sound at that volume doesn't feel comfortable to me. Perhaps I misunderstood something? Maybe I'm just over-sensitive to that volume, perhaps because I have neighbours here, don't have a house to myself, etc. Under the K-20 system, it seems the maximum SPL when monitoring with two speakers would be a scary 106 dB(C). Those monitoring levels would force engineers to keep loudness down, or risk going deaf otherwise. Presumably that would serve the intended purpose of the system, of avoiding excessive loudness of recordings, but I think it carries some risk. I think the K-14 system might be more to my own taste. I believe it keeps the 83 dB(C) reference level, but with peaks up to 100 dB(C) in a two-speaker setup. Sustained exposure for 2 hours even at the maximum level should still be safe as per the OSHA limits, but of course that level should rarely be reached anyway. I'd very much like to hear how a K-20, K-14, or K-12 recording compares to mainstream pop/rock/metal recordings. Perhaps the high reference level would feel more comfortable with program material that isn't so heavily compressed. I'm also curious to know what sort of monitoring levels other people have been using. I'd estimate that in the past I've been down to something like 60 dB as a reference level, and resorting to headphones when I felt more scrutiny was needed. This unfortunately seems necessary here because I have other people living around me. I'm looking forward to trying out the jkmeter implementation. And I imagine a K-14 and even K-12 version would be very useful someday. Regards, -- Steven Chamberlain steven(a)pyro.eu.org _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user

I've been using the excellent Jkmeter for a few weeks. Recently I gt hold of a SPL meter so using Jnoise I was able to calibrate the speakers volume to the suggested 83db(C). Similar to what Steven mentioned in his email, to me that sound level felt very uncomfortable (way too high) so following my perception (and without using Jkmeter) I started mixing everything way too low (I confirmed this after playing one of my pieces at a electroacoustic music concert setup). So I'm also wondering, did I misunderstood something or I'm just too sensitive? Is my SPL meter not working (it's a new and made by a Galaxy Audio)?

Does the existent room noise (computer fans, street noise leaking through the window, measuring around 50dBC) significantly affects the measurement?

Yes, the nostalgia meters look nice and warm in these cold digital times... I'd be happy to add an XXL version, if someone can tell me how to make a PNG with an anti-aliased alpha mask using cairo.