[kiteki]

 
Do you know how lossless compression works kiteki?

 
zip/unzip.
 
 
Quote:

 
(I also have FLAC versions of 256K iTunes AAC downloads, but I'm fairly sure that isnt what happened here. OK, Im 80 per cent sure. Er, I have to go now !)

 
Haha, you can check with this - http://spectro.enpts.com/

 

[kiteki]

Yup. I'm not familiar with proper techniques / software for both aligning and volume matching, maybe someone more knowledgeable can point you in the right direction. All I could come up with, was to insert a 1 second sine wave in my recording for track alignment in Audacity (such samples are easy to identify and align), and using wavegain for volume matching (I found out that straight up normalization doesn't work reliably).

 
Thanks.
 
So... any thoughts on the differences in your suggested test, and mine, or do you think they're pretty much identical?

 

[skamp]

So... any thoughts on the differences in your suggested test, and mine, or do you think they're pretty much identical?

You can't conduct yours alone. Foobar ABX is a hell of a lot more convenient (and possibly more reliable).

 

[stv014]

Quote:

Yup. I'm not familiar with proper techniques / software for both aligning and volume matching, maybe someone more knowledgeable can point you in the right direction. All I could come up with, was to insert a 1 second sine wave in my recording for track alignment in Audacity (such samples are easy to identify and align), and using wavegain for volume matching (I found out that straight up normalization doesn't work reliably).

 
If someone uploads good quality recorded audio in 96/24 format, and the original source, I can create the ABX files.

 

[Tronz]

Quote:

I never wrote that it was audibly better, just measurably better.

Okay, but if you're going to be using the ES9018 DAC for music listening, I don't see the point in wasting so much money on specs that will never be fully utilized. It's like buying a Bugatti Veyron for driving to the supermarket. It's pointless. People are so caught up on having the most "technically" advanced DAC when they have little clue that they'll never take full use of it.

 

[maverickronin]

Quote:

zip/unzip.

 
So obviously there's a pretty big mistake in there somewhere.

 

[mikeaj]

Okay, but if you're going to be using the ES9018 DAC for music listening, I don't see the point in wasting so much money on specs that will never be fully utilized. It's like buying a Bugatti Veyron for driving to the supermarket. It's pointless. People are so caught up on having the most "technically" advanced DAC when they have little clue that they'll never take full use of it.

Next you're going to say that we don't need an amp with a 20V rms output to listen to HD 650s at 1.6V levels. But it's more powerful!!!1!11!1! :rolleyes:

 

[Tronz]

Quote:

Next you're going to say that we don't need an amp with a 20V rms output to listen to HD 650s at 1.6V levels. But it's more powerful!!!1!11!1!

Hey, if you want to lose your hearing, that's fine. 

 

[deadlylover]

Quote:

 
So obviously there's a pretty big mistake in there somewhere.

 
It happens man, sometimes, it just gets caught in the zipper. Ouch.
 
Quote:

 
Thanks.
 
So... any thoughts on the differences in your suggested test, and mine, or do you think they're pretty much identical?

 
I guess it depends on the quality of the A/D converter too if ya gonna record the outputs? Or maybe it wont matter, because it's constant.
 
I've always wanted to build an arduino based ABX comparator with built in passive preamp to level match difference sources (one that can be used alone, identical in function to the foobar one). I need to get around to doing that one day.

 

[stv014]

Quote:

Originally Posted by deadlylover /img/forum/go_quote.gif
 
I guess it depends on the quality of the A/D converter too if ya gonna record the outputs? Or maybe it wont matter, because it's constant.

 
Well, if anything, it can be used to test if the combination of D/A and A/D is transparent, with the hope that the A/D does not improve the quality. But when doing this type of recording, ground loops can be an issue if the D/A and A/D are separate devices but connected to/installed in the same PC.

 

[nick_charles]

Quote:

 
Well, if anything, it can be used to test if the combination of D/A and A/D is transparent, with the hope that the A/D does not improve the quality. But when doing this type of recording, ground loops can be an issue if the D/A and A/D are separate devices but connected to/installed in the same PC.

 
 I've done this kind of test a lot. I used the output from a CD player connected to  a couple of USB ADCs and recorded the results in Audacity. Since there is random error I did at least 10 tests for each combination and averaged the results. Cropping and aligning is royal pain but with a bit of practice you can get quite slick. Any added AD stage adds some noise but I was able to record samples that a high profile member here was incapable of DBTing from the original source files. I used this to compare analog cables and found that for properly terminated cables the differences (including between a $135 cable and a 77c cable) were so small as to be either triviial or obscured by a 16 bit digitization process...

 

[scud80]

if you get them aligned (*really* well) you could just difference them and actually hear what's not the same.  this is pretty interesting to do with mp3 compression since they're automatically aligned.

 

[skamp]

Nulling is useless for determining what you actually hear with normal playback. It's especially useless with perceptual codecs like MP3.

 

[Anaxilus]

Quote:

I used this to compare analog cables and found that for properly terminated cables the differences (including between a $135 cable and a 77c cable) were so small as to be either triviial or obscured by a 16 bit digitization process...

 
Interesting.  So what you are saying is there are demonstrable differences in cable based on your tests.

 

[nick_charles]

Quote:

 
Interesting.  So what you are saying is there are demonstrable differences in cable based on your tests.

 
Yes, but when I say these are small I really mean it - we are talking about a few 100ths or 1000ths of a db at a given frequency point. The overarching pattern is that all cables tested (apart from an unshielded silver cable which was noisy) are flat as pancakes in FR terms