[andY.fRa]

The Topic is my question, what are the best settings with my Aureon which has the Envy24HT.

andy

 

[Glassman]

first flash the card for Audiotrak Prodigy 7.1, do a search here, you'll find answer to any and all of your questions..

use ASIO plugin for foobar, set output to 24bit padded to 32bit and that's it.. you can experiment with some DSP plugins like Crossfeed for headphones or SSRC resampling..

 

[andY.fRa]

well i did a search in the forums and couldn't find the mentioned thread, am i stupid ?

something else, i only use the aureon's analog output

and the audiotrak is a 7.1 mine is 5.1, is this firmware working with my soundcard?

 

[Glassman]

the cards are identical, you're just missing analog stuff for 7/8 channels, that's all.. Audiotrak drivers are much better than Terratec's.. try searching for 'flash' and 'prodigy', it must work..

 

[andY.fRa]

well i really cannot find detailed infos about flashing my card, but what i'd like to know, in which way are the audiotrak drivers better, is the sound better or what?

 

[MiChael.]

My friend tried Prodigy firmware and drivers for his Aureon Sky 5.1 and at the end he turned back to Terratec firmware and drivers, actually Space 7.1 stuff. He thought Audiotrak drivers are weird, and those shouldn't improve soundquality.

I'd choose 32bit fixed-point over 24bit padded to 32bit, cause it's more efficient. Turn off dithering, it's not needed for over 16bit quantisation. You can try SSRC at 88.2khz or 96khz too. I'd turn of replaygain too, cause it breaks dynamics a bit because of lower signal, but still remember to use peak information.

You'll find ASIO component from http://www3.cypress.ne.jp/otachan/ and there's too higher quality upsampler component than foobar's own, imho it gives better bass.

edit: Shibatchi Otachan's upsampler at http://www3.cypress.ne.jp/otachan/secret/foo_secret.cgi

 

[Mr.Radar]

Quote:

Originally Posted by MiChael. I'd choose 32bit fixed-point over 24bit padded to 32bit, cause it's more efficient.

But there are no soundcards that can truely do floating point so it's always converted to integer before it reaches the DAC.

 

[MiChael.]

Quote:

Originally Posted by Mr.Radar But there are no soundcards that can truely do floating point so it's always converted to integer before it reaches the DAC.

I wasn't talking about floating point, I was talking about fixed-point. IMHO floating point have no use for playback, it's more for editing. Floating point have more range when fixed-point have more accuracy, maybe floating point could have some advantages if the listener would use heavy collection of DSPs, but I don't see any other ways to exploit it.

IIRC 32bit fixed-point is actually 24bit, but last 8bits are ignored by filling it with zeros to get lower cpu usage.
... but the difference might not be big, but anyway I prefer 32bit fixed.

 

[Mr.Radar]

Quote:

Originally Posted by MiChael. IIRC 32bit fixed-point is actually 24bit, but last 8bits are ignored by filling it with zeros to get lower cpu usage.

Actually, that's a perfect description of how 24-bit padded to 32-bit works. 32-bit fixed uses all the bits (including the last 8).

 

[MiChael.]

Quote:

Originally Posted by Mr.Radar Actually, that's a perfect description of how 24-bit padded to 32-bit works. 32-bit fixed uses all the bits (including the last 8).

As far as I know, there's no real 32bit mode, 24bit is the highest... hmm, I'm a bit confused how 32bit fixed differs from padded. And I'm wondering what 32bit fixed is, cause there's 3-byte packed PCM and 4-byte PCM. IIRC PCI channel handles data in 32 bit chunks.
Anyway the difference is probably quite unsubstantial. And this is already offtopic, sorry about it