Date: Thu, 21 Mar 2013 08:16:05 +0100
From: Ralf Mardorf <ralf.mardorf@alice-dsl.net>
To: freebsd-multimedia@freebsd.org
Subject: Re: [Bulk] Re: snd_envy24{,ht} recording support?
Message-ID: <1363850165.576.45.camel@archlinux>
In-Reply-To: <1363779850.586.86.camel@archlinux>
References: <alpine.BSF.2.00.1303130213430.74881@freefall.freebsd.org> <20130313215327.d4d84624ddc07ca07b83a879@yamagi.org> <CAO%2BPfDfxK_CL8DNH2zLWWQZFV7CYD2WTABHm1%2B3_Tv%2BxxKa%2BYw@mail.gmail.com> <20130320085207.b21d890c59babbeea0431da9@yamagi.org> <1363767755.586.32.camel@archlinux> <CAFHbX1JBj8fjCaxsKUZWGSofXvEM6ecOF-JE5GqP7d95d%2Bz6jw@mail.gmail.com> <1363779850.586.86.camel@archlinux>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 2013-03-20 at 12:44 +0100, Ralf Mardorf wrote: > I don't like consumer SPDIF, but this is OT now [snip] > Yes, if SPDIF should work, it should be without loss, even for those > on-board devices. There anyway could be quality issues [snip] You even can't discuss this issue among professional audio engineers, so it for sure is useless to discuss it on a mailing list for an OS, that's not known for it's audio production abilities. I had loss when coping from one consumer DAT 48KHz recorder to another consumer DAT 48KHz recorder using SPDIF, when I was a professional audio engineer, so you can assume that the cable wasn't defect. Theoretically there shouldn't be loss, imagine it would be executable binaries and not music that digital copied would have loss. Those binaries would be borked and you couldn't run them anymore. But as Lev already explained _you never know what brilliant idea they implement to a consumer device_, e.g. at lest resampling is needed if there are different sample rates. Even if it is done correctly, to lower the sample rate is audible, but if you simply use the signal from _good_ analog IOs and you connect it to a good consumer HiFi amp (you don't need high end or professional studio gear) than the sound quality will be much better, than 44.1 KHz are able to provide. Note that regarding to matching filters most devices perform best at 48 KHz, even if they should provide higher sample rates. Regarding to sound quality professional studios use 48 KHz only, because higher sample rates even with matching filters aren't needed for the human ear. Professionals use higher sample rates, if they need lower latency on stage. To choose 48 KHz as a default is a good choice. It's also better to produce using 48 KHz and to resample to 44.1 KHz later, than to produce with 44.1 KHZ. The 44.1KHz were a compromise, when they invented the CD, that's not needed nowadays. Sample rates higher than 48KHz only make sense regarding to latency, not regarding to audio quality. It's likely that even very expensive gear will perform less good regarding to sound quality at higher sample rates, caused by filters that match to 48 KHz. Regards, Ralf
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1363850165.576.45.camel>