Soundcard???

  • Thread starter Thread starter Tappy
  • Start date Start date
nice one ethan :D

have a look round ethans site, WELL worth the time :)...
 
Why would I be listening to it on loudspeakers?

I listened to it on several types of speakers and headphones. My BX8's are NOT exactly high end gear. But they showed that clip 2 has more high end / treble. Which for clip 2 sounds worse in this context. That does not mean that the X-Fi (clip 1) is better. Just that it's more pleasing in this context. Scratchy metal strings. Had the intent been a soundtrack to a horror flick, clip 2 would have won.

It is impressive how closely they are matched. And it's hard to tell a difference. But there is a difference. And had each device been allowed to run at their maximum capabilities, the difference would like be greater. Hence crippled. For example some think that the Sound Devices sounds the same recording at it's max or 24/96 (less than max). Where my Korg MR-1000 sounds noticeably better at DSD and converted to 24/96 instead of at 24/96 from the get go.

Of course we could argue semantics all day. Where the cables of the same make / brand, manufactured as part of the same lot with the same materials, and treated identically through their lifespans? Does the splitter favor one channel over the other. Does the combined impedence of splitting the signal make the converters respond differently than they otherwise might? Is that really the best mic + preamp to start with? Do you own stock in creative and would benefit by only running tests that make creative compare well against more expensive brands? yada yada yada...
 
EXACTLY....."yada yada yada". Anyone that is contemplating recording with a SB card vs. a 24/96 or 0404 could give a shit about the rest of it (for now). This is sounding more like a argument just for the sake of it. Ethan's test and clips are valid in this case.
 
clip 2 has more high end / treble.

This is why I'm always going on about the limitations of human hearing and perception. Attached are two screen-caps showing an FFT (spectrum) analysis of both files as created in Sound Forge using its highest resolution. As you can see, the spectral content is for all intents identical. There are tiny differences, but they're all down at -80 and below. So that doesn't account for hearing a difference.

I bet $100 that if I blind tested you here ten times in a row, half the time you'd pick g1.wav as brighter, and half the time you'd pick g2.wav. That's just the way our hearing works when differences are subtle or non-existent. I've seen and performed tests playing the same Wave file for someone a few times in a row, without them knowing it's the same file, and they report hearing small changes. This is also why some people claim to hear a change - always for the better, of course! - after replacing one competent AC power cord with another.

Do you own stock in creative and would benefit by only running tests that make creative compare well against more expensive brands?

LOL, no, I promise I have no vested interest in any of this. Other than learning The Truth (tm).

--Ethan
 

Attachments

  • g1.webp
    g1.webp
    20.8 KB · Views: 51
  • g2.webp
    g2.webp
    20.6 KB · Views: 48
Of course what I call high end presence might be something else. Like recorded detail. To much of it can confuse conventional equipment attempting to play it back in all it's glory. I should probably isolate a few bandwidths and compare to be sure. The clips as a whole are hard to distinquish, but isolated bandwidths should be more revealing.

I guess as far as scientific goes.

the amplify effect in audacity says 2.3dB for g1, and 2.1dB for g2. So g2 is louder by 0.2dB (in theory).

bandpass in sox. (14.3.0)

sox -v 0.9 input.wav output.wav bandpass -c 6k 10

afterwards amplify in audacity is 5.4 for g2 and 6.2 for g1. Less the .2dB from before and g2 is still .6dB louder in that bandwidth. Or would it be 1.0dB louder? Which verifies what I previously thought. More there in that annoying baby screaming frequency range from the Apogee(sp?).
 
bandpass in sox. (14.3.0)
sox -v 0.9 input.wav output.wav bandpass -c 6k 10

I have no idea what that's saying. But I do agree with this comment by you, which confirms my main point that converters don't vary nearly as much as the vendors would like us to believe:

The clips as a whole are hard to distinquish

--Ethan
 
Back
Top