192 kbs and 320 kbs, is there really a difference?
Dec 24, 2008 at 6:25 PM Post #121 of 372
Quote:

Originally Posted by xz123 /img/forum/go_quote.gif
So... who can prove it? Like, with a blind test?

Look at hydrogenaudio, for example. Except for rare cases ("problem samples" etc.) the consensus there is (afaik) that with current codecs even 128 is in most cases transparent.



Did you try it yourself?

For me, 128 isn't tranparent. Even on a crummy built-in sound card and a CMOY amp I can ABX 128 vs. Flac easily. I didn't notice artifacts, but the 128 is flat, less air or soundstage. ( Now, 192 vs. Flac is much harder for me - I'll have to try that one again later... )

[Edit, I'd better qualify that, 128 CBR is easy. 128 VBR not nearly so easy, but still discernable.. I don't think I want to even try 198 VBR now.]

It bothers me that hydrogen audio dosn't take into account the type of equipment being used. Claiming transparency based on tests with "stock" sound card setup isn't the same as if it was acheived with high quality DAC and headphones.
 
Dec 25, 2008 at 3:50 PM Post #123 of 372
Quote:

Originally Posted by powertoold /img/forum/go_quote.gif
I think the best way to go if you're worried about disk space is 160kbps VBR. I don't think you would be able to tell much difference from 320 or FLAC.


Depends who you ask, and which gear he/she listed to.
Cause some of us can hear an audible difference between 320Kbps lossy and lossless. Hence 160Kbps would be even less transparent.

To each their own of course.
smile.gif
 
Dec 26, 2008 at 12:56 AM Post #124 of 372
I just did a back and forth clicking on Funkytown by Lipps, one 128 and the other 192, about 4 or 5 seconds of the beginning of each(iTunes).
Easily distinguished the difference, no problem. Didn't have to listen hard or strain to hear it either, and I really don't think I have all that good hearing!
I really don't think my equipment is anywhere near top of the line either.
I'm gonna do a 192/320 comparo now.
 
Dec 26, 2008 at 4:20 AM Post #125 of 372
Quote:

Originally Posted by digger945 /img/forum/go_quote.gif
I just did a back and forth clicking on Funkytown by Lipps, one 128 and the other 192, about 4 or 5 seconds of the beginning of each(iTunes).
Easily distinguished the difference, no problem. Didn't have to listen hard or strain to hear it either, and I really don't think I have all that good hearing!
I really don't think my equipment is anywhere near top of the line either.
I'm gonna do a 192/320 comparo now.



Is this an ABX or just a back and forth comparison?
 
Dec 26, 2008 at 9:28 AM Post #127 of 372
digger945. Using an ABX application I presume?
Some claim to hear an audible difference, so its no surprise if you do.
 
Dec 26, 2008 at 9:33 AM Post #128 of 372
Quote:

Originally Posted by b0dhi /img/forum/go_quote.gif
Is this an ABX or just a back and forth comparison?


Quote:

Originally Posted by krmathis /img/forum/go_quote.gif
digger945. Using an ABX application I presume?
Some claim to hear an audible difference, so its no surprise if you do.



He says he did this via iTunes, so it's really not a valid testing method to go by.
wink.gif


Quote:

Originally Posted by digger945 /img/forum/go_quote.gif
I just did a back and forth clicking on Funkytown by Lipps, one 128 and the other 192, about 4 or 5 seconds of the beginning of each(iTunes).
Easily distinguished the difference, no problem. Didn't have to listen hard or strain to hear it either, and I really don't think I have all that good hearing!
I really don't think my equipment is anywhere near top of the line either.
I'm gonna do a 192/320 comparo now.



 
Dec 26, 2008 at 10:43 AM Post #129 of 372
Between 128 and 192 I think the difference is big enough to tell just by comparing ad hoc, but for the 192/320 test I suggest using an ABX tool (ideally the Java one from the Public ABX thread, since it's very well designed).

My own experience is that for extremely subtle differences, such as differences between high bitrate VBR and lossless, the differences are well below the threshold of suggestibility. Even for samples I can tell apart with p < 0.001, it's just as easily possible to literally hear what I expect to hear. You definitely need an objective feedback mechanism such as ABX to train yourself to work with these subtleties IMO.

I think a lot of people become disheartened when they can't tell the difference in an ABX at first and then give up. The brain isn't used to doing such a precise analysis of the audio coming from the ears, so it takes some training and a lot of concentration to get it happening. With some effort, though, I think most people can do it.
 
Dec 26, 2008 at 11:22 AM Post #130 of 372
Quote:

Originally Posted by TheMarchingMule /img/forum/go_quote.gif
He says he did this via iTunes, so it's really not a valid testing method to go by.
wink.gif



Ooops, I missed that text in his post.
redface.gif
 
Dec 26, 2008 at 12:42 PM Post #131 of 372
^Yea I wasn't trying to do some formal test, I just noticed that I had the same song on iTunes in both 128 and 192 formats, so I just clicked on one and listened for about five seconds, then clicked on the other one and listened for about five seconds, back and forth, over and over.
The difference is very noticable, with the 192 track having much more information, echo, decay, you can hear the cow bell much differently.

Just curious, but why exactly is it that iTunes is not a valid testing method?
If not then what is?
 
Dec 26, 2008 at 1:07 PM Post #133 of 372
Quote:

Originally Posted by digger945 /img/forum/go_quote.gif
What is an ABX application?


An application allowing you to perform a blind test.
For audio you input sample A (the lossless source) and sample B (the lossy file). Then you play back the stream, determining if sample X (random A and B) equals A or B. Then make a score for you, based on how you pick the correct one and hence how likely it is that you guess the answer.

Read more: ABX test - Wikipedia, the free encyclopedia

Depending on you OS you may want to check out these applications:
http://www.rarewares.org/files/other...-java-0.5b.zip
ABC/HR Audio Comparison Utility
 
Dec 26, 2008 at 2:21 PM Post #134 of 372
Ahhh I see. Thanks for the link.
That would work well if the person being tested were able to have control over the player so they could switch back and forth quickly.
I would imagine that the differences are much more apparent with music that the listener is intimitely familiar with.

Good reading.
 
Dec 26, 2008 at 3:10 PM Post #135 of 372
Quote:

Originally Posted by digger945 /img/forum/go_quote.gif
Ahhh I see. Thanks for the link.
That would work well if the person being tested were able to have control over the player so they could switch back and forth quickly.
I would imagine that the differences are much more apparent with music that the listener is intimitely familiar with.
.



The only one I've used is winabx. You can set begin and end marks to compare a short passage, and certainly if you set the test up yourself you can use familiar music.

If you are training yourself to tell the difference you might want to start with some classic cases known to expose problems like "castinets"

Here are a bunch of samples considered "obvious," with descriptions of the artifact and the sample available in "bad" (perhaps Xing 128kb/s), "good"(lame insane), and lossless.

Artifact Training Page
 

Users who are viewing this thread

Back
Top