hi
i m new here on this forum
i ve heard a little about the abx blind test
i must say that i don t know much about such things
my music is in flac format and i convert that flac in foobar to mp3 320kbps cbr,
can someone explain me how and where i can make a test where i can see if there s a diference in sound between
flac and mp3 320 cbr (converted form the same flac)
and i want to ask you something else: i have a lot music in mp3 320 from the net, is it possible that i test that music somehow if it is fake mp3 320kbps (someone could convert it from 128 to 320 and upload that on the net as 320kbps)
?
thanks in advance for answering me
You can use foobar2000 to perform ABX tests after downloading the ABX component. It works this way, you'll have to guess which song is which. Try it about 15 times. If the probability is low, then probably you can spot the difference.
And yes, you cannot know the past of downloaded songs.
i ve downloaded the foobar abx comparator (foo_abx.dll) and copied it in the foobar folder
where must i go now to make the test?
and you wrote: If the probability is low, then probably you can spot the difference,
you have thought if the propability of my guessing of the song i high, than there is a difference
and if the probabilitiy of guessing which is which song is low, than i can t hear the diference
am i right ?
(i write that because i don t speak and write english very good)
Select the two tracks (e.g. the uncompressed WAV and the compressed MP3), right-click, and choose the option to ABX.
Yes, the p-value is the probability of achieving the same results by selecting randomly/guessing. Thus, to be confident that you did identify a difference between the tracks, p must be low. The usual significance level is p<0.05; that is, a less than 5% chance of achieving the same test results by guessing. (This is standard throughout the sciences.) So, if you achieve a p-value of more than 0.05 (what do people do with a p of exactly 0.05, btw?), it means that your test has less power to indicate an actual difference.
i ve added two songs in foobar ,
first is in wav.format, the second is the same song converte in mp3 320kbps cbr,
i selected this two songs and and made the right click and i don t have the option abx test
what went wrong?
Do you have it installed?
http://www.foobar2000.org/components/view/foo_abx (http://www.foobar2000.org/components/view/foo_abx)
If so, your context menu preferences is set to show it only on shift+right click.
i think i found it
i selected those two songs in foobar (wav, mp3 320 converted from same wav)
i clicked on the right click-utilities-abx two tracks
(then after that i have chosen use dsp-current playback settings)
one question :can i test flac and mp3 320 or just wav and mp3 320
anything
Bear in mind, though, that comparing one lossy file to another is of little to no use. Also, you can find much more information on ABXing in previous topics.
i think now i understand it
i took two different songs so that i could understand the procedure
i took from billy holiday-you don t know what love is and from chris rea josephine
i select those 2 songs and right clicked it and than went to utilities abx two tracks-use dsp(current playback settings)
song a was billy holliday,song b was chris rea
of course i guessed everything correct with no mistake since it were two different songs
after i guessed the first time correct it has written down under the test score 1/1,probability that you have guessing 50%
, after i guessed the second time right, it wrote test score 2/2,probability that you have guessing 25% and so one...
when i scored the 11th time corect it wrote score 11/11 ,probabilitiy that you have guessing 0.0%-that means that i have scored all eleven tests correct and that there isn t a probability that i have randomly guessed anything of that am i right?
when i the 12th time purposely answered wrong to see what would happen it wrote score 11/12,propability of guessing 0.3%
…That’s a rather unorthodox way of visualising the procedure, but sure. In other words, if A and B were respectively the uncompressed and lossily encoded versions of the same song, and you achieved the same results, that would indicate to a high level of confidence that you were indeed hearing a difference. As I said, it is generally accepted that p-values of less than 0.05=5% are suggestive of a definite difference. In the case of lossy encoding, that would suggest that the particular track is not transparent (http://wiki.hydrogenaudio.org/index.php?title=Transparency) to you. (In other cases, it could suggest an audible difference introduced by a resampler, between two configurations of hardware, etc.)
i must say that i dont heard the difference between flac and 320 cbr
i must say that i have an awful sound card with very low quality so i can t hear the difference between anything
and i think that this test on my laptop(hp compac nc 8230) with such a sound card doesnt have any sense
it would have had sense if i had a quality sound card
i wanted to ask you about sound spectral analysis(i m not sure if its so called,i don t understand such things and can t remember)
a few months before on an other forum some guys recommended me that i make a spectral analysis
of the sound and that i will see after that if the sound is the same
does anyone know how to make that test (or a similar one)so that i can see grafically after that the difference
(i think that you can with that see if a mp3 320kbps is fake)?
i must say that i dont heard the difference between flac and 320 cbr
Most people cannot hear a difference for most songs with high-bitrate MP3s. That means that MP3 does its job: reducing filesize while staying transparent to the listener. Now you can try how low you can get with the (average!) bitrate before you notice a difference. Just repeat the test with a re-encoding of the FLAC with different settings (VBR V0, V1, V2, ...). Me as an untrained listener am able to ABX non-killer-samples only for bitrates
smaller than 128 kbit.
And: The equipment is
not the bottleneck when trying to ABX lossy files. Laptop + Headphones are a perfect setup.
can anyone help me with the spectral analisys of the sound?
has anyone tried it
whats v0,v1,v2 and how much each of them does have a bitrate?
do they have 320kbps?
in which format(mp3,wma or?) are the v0,v1,v2
whats the difference beetween v0,v1,v2 and mp3 320-cbr
http://wiki.hydrogenaudio.org/index.php?title=LAME (http://wiki.hydrogenaudio.org/index.php?title=LAME)
BTW, it would be good to determine who you want to listen to for advice. Most people here will tell you that you use your ears to listen to music, not your eyes.
A picture will not give you definitive information about how something sounds to your ears.
then i will cancel all this about the spectral analysis
i m here on this forum because i wanna now buy a high quality hifi system (denon or yamaha or something third)
and high quality headphones (denon ahd2000 or sennheiser hd 598) and i want to listen to music on that hifi via usb
(mp3 320kbps-cbr) stick
all my music albums are in mp3 320kbps cbr and im afraid if 320kbps cbr (via usb stick) won t have sufficient
sound quality for quality hifi all in one system and quality headphones
so i asked many questions about 320 kbps that i don t make a mistake and buy a good hifi and quality headphones
and that the music wont be good enough to drive this hifi and headphones
i would like that someone gives me an advice for that so that i can be sure that i can buy quality hifi and phones without worryng if i will get a good sound
(sory for bad english , i don t speak very well, but i hope you have understanded me)
then i will cancel all this about the spectral analysis
Thank you!
After a few recent topic by those refusing to listen, my fuse has gotten a bit short on the subject.
I'm about at the point where I might start aggressively binning posts.
Regarding the hardware issue, if your soundcard and/or listening environment (assuming your headphones have poor isolation) are particularly noisy then the ability to hear lossy artifacts may be compromised, otherwise, a poor system (high distortion, uneven frequency response, etc.) can actually cause artifacts to be more easily revealed.
i converted a flac file to ogg 45kbps (when i clicked on that converted song in ogg it was written type:ogg flac,ogg vorbis,speex) in foobar and for me there was no diference beetween flac and ogg 45kbps
the both songs where the same
i dont know whats the problem,my hearing is great,
the converteg song in ogg has had 736kb and it sounded the same like the flac file which has more than 20 mb
how can it sound the same?????
now im really confuzed and dont understand nothing anymore
my hearing is great
We get this here all the time. I think it's akin to almost everyone claiming they're above average drivers.
...i dont know whats the problem,my hearing is great,
the converteg song in ogg has had 736kb and it sounded the same like the flac file which has more than 20 mb
how can it sound the same?????
now im really confuzed and dont understand nothing anymore
Welcome to the club. We are the 99%!
Your hearing is just normal, just like almost anybody else (myself included.)
I'd also do a frequency hearing test. I thought I had good upper frequency hearing until I found out I can't actually hear anything above around 15.5KHz
does this mean now as i cant on abx test on my laptop hear the diference between flac and mp3 320,
that when i will listen on my (all in one high quality) hifi system music from audio cd(flac) and the same music via usb stick(mp3 320 cbr ripped from the same flac) on the same hifi that i wont notice the difference in sound or?
i didnt heard for the frequency test
can you explain me the procedure and tell me where i can test it
I am confident that your 320kbps mp3s will sound great on your HiFi.
I am also confident that VBR mp3s (e.g. -V2) will sound equally great.
(...)is it possible that i test that music somehow if it is fake mp3 320kbps (someone could convert it from 128 to 320 and upload that on the net as 320kbps)?
There are some indications you can look after, particularly a cut in high frequencies (above 15 or 18kHz).
Sometimes people record in 320kbps from internet radio stations that broadcast in 128kbps or lower and that apply some additional tools that deteriorate sound quality (compressors/limiters are heavily used by radio stations). You can additionally look after a low peak volume (lower than ca. -0.5dB, e.g. -5dB) to detect some of those recordings.
But if you really can't differentiate at all between flac and 45kbps Ogg with headphones, all of this shouldn't bother you too much. Just a hint on this: try some different tracks and positions, try to concentrate, e.g. on highs, on the stability of the room, on unnatural sounding instruments or on artifacts. Use acceptable headphones, listen in a quiet room. Repeat it later before you get tired.
Just a hint on this: try some different tracks and positions, try to concentrate, e.g. on highs, on the stability of the room, on unnatural sounding instruments or on artifacts. Use acceptable headphones, listen in a quiet room. Repeat it later before you get tired.
Another hint: Don't train yourself to identify the weaknesses of lossy encoding and just enjoy the music (unless you really want to)