Can you tell the difference between 8bit and 16bit music?

[steve_1979]

I spotted this on that 'other' forum and thought it was interesting so posted it here aswell.

Some of the members on this forum claim that they can hear a difference between 16 bit and 24 bit music. But can you even tell the difference between 8 bit and 16 bit files? Take the test...

http://www.audiocheck.net/blindtests_16vs8bit.php

 

[chebby]

No way.

Anyone who is prepared to listen to ####ing 'Gangnam Style' ten times, obviously prefers testing to music!

Find me the same test with something I can listen to and I might take you up on it.

 

[Electro]

To be honest I could not tell the difference, both sounded equally carp !

I think more natural and detailed music would have been a help , I just got an overwhelming urge to turn it off regardless of the bit rate

 

[John Duncan]

I spotted this on that 'other' forum and thought it was interesting so posted it here aswell.

Some of the members on this forum claim that they can hear a difference between 16 bit and 24 bit music. But can you even tell the difference between 8 bit and 16 bit files? Take the test...

http://www.audiocheck.net/blindtests_16vs8bit.php

If the music's compressed to hell (as I'm sure it will be, being a radio pop song), nobody will be able to tell the difference anyway. Go off and create a better test with some real music.

 

[Paul.]

I couldn't bare to sit through that song 10 times, so I have constructed my own test. One of these is the original and the other is the 8 bit version...

http://www.youtube.com/watch?v=z9D71pQaTnc

http://www.youtube.com/watch?v=kL498EZnkm8

Quite close I think you will agree...

 

[davidzomec]

6 out of ten, so I would say I can't feel the difference. Test made with V-Moda M-100 through my computers 3.5 output. Does it make a differente the equipment I use?

 

[davidzomec]

{unpublished}duplicated post

 

[Paul.]

All jokes aside, this 8bit vesion of Butterflies and Hurricanes is not half bad...

http://www.youtube.com/watch?v=P_Kxla1yAzg

 

[MakkaPakka]

That song has almost no dynamic range (4db apparently) and is full of electronic noises rather than intsruments. I think it's been chosen deliberately to make the test as hard as possible. I haven't bothered trying it properly for that reason.

I can clearly hear the difference in sound on a Master System compared to a Mega-Drive so in answer to the thread titles - Yes, I can.

 

[NHL]

That song has almost no dynamic range (4db apparently) and is full of electronic noises rather than intsruments. I think it's been chosen deliberately to make the test as hard as possible. I haven't bothered trying it properly for that reason.

I can clearly hear the difference in sound on a Master System compared to a Mega-Drive so in answer to the thread titles - Yes, I can.

Was that a thought experiment?

 

[chebby]

This isn't a controlled test. I toggled all the answers (without playing the awful music) until I got 10/10. I could show the screen print here and pretend to have 'golden ears' or - conversely - jiggle my answers to get something to fit whatever 'agenda' I wanted it to.

I know what is fact (for me at least) in my real life listening.

256k AAC rips from CDs are almost (but not quite) good enough to play from my iPad through my system. Whereas 320K AAC VBR is spot-on and removes any - very occasional - roughness on explosive vocal transients like sudden laughter or shouts or screams.

So - despite the extra space it takes up - 320K AAC VBR is a minimum below which I won't go with CD rips because I don't want to be 'jarred' (even just occasionally) out of my concentration by distortion.

However, the same (or similar) BBC iPlayer radio content - 'real time' and 'catch-up' - does not produce such moments despite being (mostly) far below 320K and nor does Radio Paradise at 128K AAC.

As for 8 bit or 16bit, I await a test where something listenable is used and where you have to register and only get to do the test once..

 

[hone_u2]

1- 16bit

2,3,4,5,6,7- 8bit

8- 16bit

9,10- 16bit

10/10

I think this one was extremely difficult because of the lack of real instruments, noisy synth lines that I'm sure were clipping when they were mixed, and extremely brief vocal lines with no significant range... Therefore, there isn't really much to tell the difference with...

What the ear perceives in a song is I feel is largely if not wholly manipulated by the person at the arranging desk who arranges different tracks at different volume levels, adds panning for stereo effect, and plays with the dynamic range to prevent clipping and manipulates the phasing to prevent the cancellation of tracks...

When songs such as these have extremely low dynamic range (the difference between the quietest and the loudest sounds) there isn't really much to hold the comparison...

I once did the same tests with with 8, 16 and 24 with a lisa Batiashvili violin track... I was able to tell the difference! Though I do understand this isn't a case at all times, because of the way the track is mastered originally.

Sometimes though, with all this, whatever the difference, I have shockingly tend to prefer the lower quality one... For example the american idiot from HDTracks 192/24, and an AAC purchased from the iTunes store... I thought the soundstage was radically different but more deep in the AAC version... Really! You have to listen to that comparison! It's more than the Rumours by Fleetwood mac!

HAs anyone heard that album in high res and from the iTunes Store? If so which one was preferred?

 

[Petherick]

A genuine question here - if you post them on youtube, does the original bit rate make any difference??

Anyway to answer the original question, yes I can.

 

[andyjm]

Many years ago a well known national broadcaster ran a series of tests to confirm the appropriate bit depth for its new distribution system to FM transmitter sites. The test used a switchbox where the number of bits of resolution could be selected. The box contained an A2D converter followed by an D2A converter. Different programme material was either played direct, or through the box. The test was run and the number of bits was adjusted until the audience were unable to tell which was the original analogue signal, and which was digital.

I participated in a re-run of the tests a few years later. This was with a decent full range analogue tape source, decent amplification and speakers in an acoustically treated listening room. The most telling material was the decay of a single piano note. At low bit depths, a distint 'tzzz' could be heard just as the note died.

A few of the group I was with could detect 12 bits vs. analogue. No one could detect the difference between 13 bits and analogue.

 

[proffski]

That song has almost no dynamic range (4db apparently) and is full of electronic noises rather than intsruments. I think it's been chosen deliberately to make the test as hard as possible. I haven't bothered trying it properly for that reason.

I can clearly hear the difference in sound on a Master System compared to a Mega-Drive so in answer to the thread titles - Yes, I can.

Afreed, and it is NOT music!

Try the test on massed strings, choir or even solo violin with piano... Then it would be easy and one of the reasons why I only use DAB for spoken voice. Using compressed electronic garbage is neither a subjective or scientific test.

 

[oldric_naubhoff]

I'm affraid most outraged with "Gangnam Style" people failed to spot the preamble for the test. So here it is again for you:

This provocative test serves two purposes: to raise the awareness of what has happened to our music dynamic range lately, and to question the need for a 24-bit digital audio as a delivery format in such a context.

Differences between 8-bit audio files and 16-bit are audible though - and are demonstrated here - but only marginally when it comes to most of the music that people are listening to nowadays.

 

[John Duncan]

Zackly. But on the whole 24 bit audio is not used as a delivery method in that context, so it's a bit of a pointless test...

 

[oldric_naubhoff]

Zackly. But on the whole 24 bit audio is not used as a delivery method in that context, so it's a bit of a pointless test...

I don't understand your point :?

the way I see it this test was to prove most of todays music (as exampled by "Gangnam Style") is so dynamically compressed so you'd struggle to distinguish 8bit from 16bit version. in that context why bother with pushing on 24bit resolution?

 

[andyjm]

Differences between 8-bit audio files and 16-bit are audible though - and are demonstrated here - but only marginally when it comes to most of the music that people are listening to nowadays.

Even 8 bits are wasted on half the compressed rubbish my kids listen to. Maybe 6 bits would be sufficient.

 

[John Duncan]

My point was that *a lot* of today's music (particularly that sold through iTunes to a younger market) is appallingly mastered, to suit radio etc consumption. In THAT context 24 bit really is pointless. However, that does not mean that all music is appallingly mastered, and for some music and consumers, 24 bit may very well be a valid delivery medium.

In other words, for the music I listen to, and the things I listen to it on, it is irrelevant, but it may not be for others - this test does not prove or disprove the latter.

 

[John Duncan]

the way I see it this test was to prove most of todays music (as exampled by "Gangnam Style")

Further, one track (that happened to sell millions of copies) does not equal "most music".

 

[hone_u2]

the way I see it this test was to prove most of todays music (as exampled by "Gangnam Style")

Further, one track (that happened to sell millions of copies) does not equal "most music".

As an example, i think if you were to take the songs on the Billboard hot 100 at the moment, they would all have really poor dynamic ranges, and would be terribly compressed.

 

[MakkaPakka]

Was that a thought experiment?

No, the Master System incapable of reproducing the famous 'Sega' speech sample that the Mega Drive version of Sonic The Hedgehog had.

Clearly conclusive proof that 16-bit is better than 8-bit without having to endure Gangnam Style.

 

[AlbaBrown]

I haven't bothered with this test. No doubt it will spark off more (tedious) debate about who can and cannot hear the difference between lossy/lossless or different bit/sample rates.

So. to add to the tedium....

With well produced recordings, the people who can hear the difference generally have well sorted, and genuinely well balanced hifi systems.

Those that can't hear the difference generally have a PC/Mac/phone/tablet as a source directly connected, or a system that features mediocre products (whether they gained "5 Stars" or not) or bad system synergy. Or they have B&W 805s speakers on the end of the system (ho ho ho). Of course I would hate to single out B&W. KEF, Dali, Dynaudio, Monitor Audio, ATC and the like have all produced products that measure well but stink when it comes to playing a tune!

Or that's at least my experience. If anyone disagrees, shoot me.

Funny how we all forget the lectures from manufacturers years ago about keeping electrically "dirty' products away from sensitive hifi, and it's all in the name of convenience

 

[SpursGator]

So. to add to the tedium....

Top post!

So this test uses 'Gangham Style?' Are we sure this isn't meant to be tongue and cheek?