[krmathis]

^ Indeed!
With the difference that on HA.org you need to back up your statement with some ABX result, while here its enough just to claim it. In the end its all down to your set of ears, the music played and the gear its played on. And of course which codec/encoder are used.

 

[ExplosiveDuck]

Yeah - I need to do some ABX testing on myself. I did some iTunes testing, with the two songs in the playlist and it on shuffle with my eyes closed and I feel like I heard the difference if I could really concentrate. I could have been chance though, but I "guessed" it over 10 times. (Probability of that isn't even too insane...)

Anyways, the only difference I could sense was the absolute slightest difference in soundstage from a set of hand drums. (I only listened critically to the first 3 seconds of it.)

Like, I'm talking about, the SMALLEST, and I would definitely not know it was there if I wasn't looking for it / didn't have something to compare it to.

I was comparing .FLAC to 192kbps .mp3 -- Meh.

I don't care if I can't hear the difference and point it out between the two, I'm still going to try and get all my music in lossless and listen in lossless. I'm one of those believers in the whole "Just because you can't hear it doesn't mean it doesn't effect your brain."

And people always talk about the placebo effect as if it's some negative thing. If you have an ego and you care about it in terms of being able to hear hi-fi audio, then yeah it's a negative, but if a placebo effect makes me enjoy my music more, then I couldn't care less that it's just a placebo.

Oh, and the testing I did was just through iTunes and a pair of HD280's plugged into the headphone jack.

I would like to be able to tell the difference between the two, and I feel like if anyone trained themself to they could, but I'm not going to risk ruining my non-lossless listening experience just to be able to pass an ABX test. I really couldn't care less.

And if listening to a 192kbps mp3 doesn't "sound" any different to my ears than listening to a lossless, I really don't care, because no one will be able to convince me that it doesn't -feel- different.

Laying down knowing I'm wearing a nice pair of amp'd headphones with a good source makes me enjoy certain music better, and that won't change regardless of the "truth" behind what I'm hearing.

But yeah, I'm going to have a friend test me through my normal little audio chain and i'll post the results. I probably won't be able to tell a difference.

And most importantly, just because someone can't hear the additional quality / data in the lossless file doesn't mean that it isn't affecting the way they're enjoying / "hearing" their music. This is key when criticizing hi-fidelity audio. I mean, look at that paper that someone posted, it was an interesting read and shows that it still changes our brain activity.

Here it is again so you don't have to dig for it:

Inaudible High-Frequency Sounds Affect Brain Activity: Hypersonic Effect -- Oohashi et al. 83 (6): 3548 -- Journal of Neurophysiology

Later!

 

[endless402]

i cant tell the difference on headphones but i can definently tell the difference on my speakers

 

[haloxt]

That was a real breath of fresh air, ExplosiveDuck, thanks. Something all people who think there's no difference between between what can't be heard and what can't be heard but removed should read up on Tomatis' audio/vocal therapy which by the way agrees with that study ExplosiveDuck linked when it said excessive ultrasounds are good and excessive infrasounds bad.

 

[Mayzei]

ExplosiveDuck you legend, that's what i would post, if i had the time and could sit at the computer that long. It's just the material things i like such as a gold plated, diamond headphone or interconect. I couldn't care if i don't hear a difference, im enjoying myself.

xx

 

[Ham Sandwich]

Quote:

Originally Posted by krmathis /img/forum/go_quote.gif ^ Indeed! With the difference that on HA.org you need to back up your statement with some ABX result, while here its enough just to claim it. In the end its all down to your set of ears, the music played and the gear its played on. And of course which codec/encoder are used.

And what clip you extract from the music to base your ABX test on.

Lets say your goal is to determine if you can hear the difference between a 320 kbs LAME MP3 file and a FLAC file in symphonic classical music. You scan the music for the one cymbal hit in the entire symphony and extract a 5 second clip that you use for the ABX testing. You successfully identify preecho in the MP3 clip and are able to score 100% in the ABX test. With those results you can claim that 320 kbs LAME is not transparent for symphonic classical music.

Bah! All that tells you is that the cymbal crashes are not transparent. What about the other 50 minutes of music? If the ABX test sample isn't how you actually listen to music then what is it telling you?

 

[ExplosiveDuck]

Well, Ham Sandwich, if you can hear a difference in one part of a song, then you can hear a difference in the song. If you can hear preecho on a cymbal crash, then you're listening to two "different" sounds from the same song.

What happens when you take a classical track that has 40 cymbal crashes in it, and you can hear a preecho every time in 320 and not in lossless?

 

[Ham Sandwich]

Quote:

Originally Posted by ExplosiveDuck /img/forum/go_quote.gif Well, Ham Sandwich, if you can hear a difference in one part of a song, then you can hear a difference in the song.

Technically true.
But such a test still tells me nothing about whether the artifact can be heard during normal careful listening of the entire symphony.

When it gets to the point that you have to train yourself to be able to identify the artifacts and listen specifically for those artifacts to the exclusion of the music, is it really an issue for normal listening where the goal is to enjoy the music?

 

[Suntory_Times]

Quote:

Originally Posted by Ham Sandwich /img/forum/go_quote.gif Technically true. But such a test still tells me nothing about whether the artifact can be heard during normal careful listening of the entire symphony. When it gets to the point that you have to train yourself to be able to identify the artifacts and listen specifically for those artifacts to the exclusion of the music, is it really an issue for normal listening where the goal is to enjoy the music?

You're an a site dedicated to how you listen to your music. So for various reason many like to be able to do this. I personally don't now much about mp3 compression etc, as I always use the actual cd or vinyl and play straight from that. Though an unqualified opinion, I would say if the compression causes there to be faults that you can hear, then there are faults throughout the entire song as a result (which I would say is bad). Just remember, at the end of the day the most important thing is usually to enjoy your music. Whatever optimizes youre enjoyment is something you should consider doing, anything that doesn't should be avoided. Hence it's all arbitrary when it comes to compression and is likely to be argued as long as music is still played on DAP's and similiar devices.

 

[dharma]

For those who are interested to adopt lossless audio format, but do'nt want to change their equipment (for now):

MP3HD
...'Thomson, the firm that helped invent MP3 in the first place, has come up with a solution - MP3HD, and you won't need to bin your existing kit as the file format will still play on normal MP3 players.

MP3HD is a lossless format, which means it delivers a perfect digital copy of the original audio - but it manages to do it more efficiently than WAV or AIFF files, which can be massive.

At the highest quality setting MP3HD files are around 2/3rds of the size of a WAV file, so you can stuff more CD-quality files into your portable player or hard disk.

Of course, lossless formats already exist in the form of FLAC, WMA Lossless, Apple Lossless and several others, but support for those formats is fairly limited: you can't play an Apple Lossless file on a Sony smartphone, or a FLAC file in an iPod'...


on plus side:
...'MP3HD has been designed to solve that problem, and it works on anything that can play MP3s. If the program or device supports MP3HD you'll get full CD quality; if it doesn't, you get a 320Kbps MP3'...

on minus side (at least this moment):
...'So what can you actually play MP3HD on? So far the answer is WinAmp, WinAmp and WinAmp. At the time of writing there isn't a single hardware player that supports MP3HD, and the software is limited to a solitary WinAmp plugin and an MP3HD command line encoder'...

...'MP3HD is also up against several other lossless formats: TTA, WavPack, Monkey's Audio, WMA Lossless, FLAC and Apple Lossless.

Of those formats, three in particular are going to make MP3HD's life difficult: FLAC, which is the format of choice for audiophiles on Bittorrent; WMA Lossless, which is baked into Windows Media Player; and Apple Lossless, which works on iPods, in iTunes and in AirTunes.

For MP3HD to take off Thomson needs to persuade FLAC users to switch, Windows Media Player users to install yet another codec (when it arrives) and Apple to add MP3HD support to iPods and iPhones. If that doesn't happen'...

 

[ILikeMusic]

Quote:

Originally Posted by dharma /img/forum/go_quote.gif on plus side: ...'MP3HD has been designed to solve that problem, and it works on anything that can play MP3s. If the program or device supports MP3HD you'll get full CD quality; if it doesn't, you get a 320Kbps MP3'...

Yes, you get a 320kbps (or whatever you chose when encoding) MP3 but with a file size equaling lossless. If you're willing to accept a lossless filesize why not just go with FLAC (or Apple Lossless or whatever) and be done with it? MP3 HD seems like a solution in search of a problem.

 

[dharma]

We can say it another way: MP3 seems like a solution in search of a problem

and all other lossy formats too!!!

 

[logwed]

FLAC ftw!!!!!

 

[krmathis]

Quote:

Originally Posted by ExplosiveDuck /img/forum/go_quote.gif Well, Ham Sandwich, if you can hear a difference in one part of a song, then you can hear a difference in the song. If you can hear preecho on a cymbal crash, then you're listening to two "different" sounds from the same song. What happens when you take a classical track that has 40 cymbal crashes in it, and you can hear a preecho every time in 320 and not in lossless?

^ Exactly!
Even the slightest artifact on a single cymbal crash in a whole song mean that you're not listening to the same audio data. Aka -> you hear a difference.

Of course one can decide to still use lossy encoded audio, but he/she may run into samples where its not transparent.

 

[Ham Sandwich]

Quote:

Originally Posted by krmathis /img/forum/go_quote.gif ^ Exactly! Even the slightest artifact on a single cymbal crash in a whole song mean that you're not listening to the same audio data. Aka -> you hear a difference. Of course one can decide to still use lossy encoded audio, but he/she may run into samples where its not transparent.

The intent of my comment was more about how ABX listening tests can become completely artificial to the actual music.

Curiosity got the better of me and I just did a quick listening test of my own to find out what pre-echo sounds like. I tried the ElectroDrip sample here and the florida_seq sample here. I tried them both at 112 kbps CBR and -V0 VBR using LAME. I couldn't hear a pre-echo in the ElectroDrip sample even at 112kpbs. I did hear a reasonably obvious pre-echo in the florida_seq sample at 112 kbps but was unable to hear any pre-echo at -V0 VBR.

Both samples are electronic sounds. Anyone know of a good pre-echo listening sample that has natural cymbals and/or natural percussion?

I now have an idea what pre-echo sounds like. I also realize that hearing pre-echo at -V0 is going to be very difficult for me if I can even manage to hear it at all. My hearing isn't that great to begin with.