As a matter of policy, I won't moderate this thread (since I'm an active participant), but if I could, I would tell several people to cut out the sniping and stay on topic.
But I can't, so I won't.😉
But I can't, so I won't.😉
Rip a track from a CD to a wav file, encode it to 256 kB/s mp3 then decompress it to another wav. Then load the two in this program and let's see if you can hear a difference:Andre Visser said:
http://ff123.net/abchr/abchr.html
(I suggest you use the ABC+hidden reference test instead of the ABX)
Obviously you need to hear a difference in enough of a fraction of the trials for the result to be meaningfull. The program will calculate that.
abzug said:
What will you say if I tell you that I don't want to listen to copied CD's because even on high quality CDR's and low speed recording, it doesn't sound as good as the original?
Not on CD. Rip to your computer, and play from there using the program. Use the digital output on your computer.
By the way, generally CDRs have lower jitter than pressed CDs. Then again, maybe you like the sound of jitter. It was found by Benchmark that when they added artificially jitter, some people preferred it as they perceived it as extra microdetail.
By the way, generally CDRs have lower jitter than pressed CDs. Then again, maybe you like the sound of jitter. It was found by Benchmark that when they added artificially jitter, some people preferred it as they perceived it as extra microdetail.
Therein lies the problem.Andre Visser said:
If the rest of the world is stupid enough to accept the reduced quality coming from lossy compressed music, then you won't be able to buy any new music on lossless compressed or uncompressed formats.
Unless the industry tries to use 24bit/192ks/S compressed down to a data rate similar to uncompressed 16bit/44ks/S (CD standard data rate). This may give a quality that will satisfy most of us that object to MP3s.
abzug said:
By digital output, do you mean spdif?
I've tried that once and wasn't impressed with the results.
With a computer it's possible to create a setup that guarantees bit-perfect audio and no interface jitter. In such a setup the computer is a perfect transport in an absolute sense, and then any issue can only come from further down in the chain--DAC etc. The simplest option for this is a high end DAC that has word clock out, and a high end sound card that has a word clock in, and using a player that bypasses any operating system sound processing. Then the interface becomes purely digital with no analog timing information impacting the audio. It is analogous to transferring a music file from one digital device to another digital device, or the digital data from the digital mixer in the studio to the machine that makes the glass master.
AndrewT said:
I agree with that, to me there is not much wrong with Redbook CD's, a few good recordings proof that, however the recording quality on, may I say most, CD's are pathetic. I will always welcome the higher resolution but that won't help much if recorded poorly.
Abzug,abzug said:
could you give us a the run down on how to achieve this with the standard on board sound card?
Preferably with XP sp2 or sp3.
But Vista if that's the only way, or do we need to go Apple?
AndrewT said:
I don’t' think too many fans of "classical" music will accept mp3 quality for in-house listening. Also I suspect that as the cost of memory and bandwidth decreases there will be little incentive to use lossy technologies.
But I do love my mp3 player as a way to make good use of time when stuck in queue etc--sweet 🙂
Jakob2 said:
Yes, after reading your post I looked up that article and read it.
He made quite a few astonishing claims indeed, yet did not provide any evidence to support any of them. Quite a few of those claims and results would each merit their own individual article in the AES Journal with test setups, experimental data results, statistical analysis and so forth. Note that the AES conference proceedings (as this was) and AES preprints are not peer-reviewed as the AES Journal articles are. I'll look further to see if he subsequently published any such papers.
After reading the article, I ended up with a very negative view of the author.
I'm quite certain no one here can hear the difference between 320 kb/s encoded MP3 and uncompressed in a blind test--I would bet good money on it. I would also guess anyone is most likely to fail for 256 as well.AndrewT said:
PCM is linear, but the ear isn't. That means that 24 bit linear PCM wastes a lot. Not only does it cover much more than the dynamic range of hearing (which is 120 dB from threshold of hearing to threshold of pain), but also ignores the basic fact that hearing is logarithmic and while you can hear differences between levels close together in one part of the range, they have to be spaced further apart to hear a difference in a different part of the range. Instead of linear PCM, an encoding should be used where the difference between levels produced by consecutive code words is just below threshold of detectability for that section of the range, in a setup where minimum and maximum levels are scaled to match the ear's 120 dB range.
That's like asking how to achieve good analog audio with a crappy turntable. Onboard sound cards are crap, and even add-in internal ones are suboptimal because there's too much interference inside the PC case. You need to use a DAC that has an asynchronous connection to the PC. You can either buy a sound card that has word clock input to which it can synchronize, then use the DAC of your choice (any quality DAC would have word clock out), or a USB DAC that uses specifically an asynchronous USB mode, and the only quality DACs that have this option are Wavelength Audio's more recent USB models (and I have a prototype of my own but that's not finished yet).AndrewT said:
On the software side, use foobar with the KS or ASIO plugins to bypass Windows processing. There's an alternative with Vista posted in another thread but I forget what it's called.
dwk123 said:
If you forgive my late stumbling in, but I agree that this is key.
Without meaning disrespect, everybody who says: "I trust my ears" is making a judgement error of truly heroic proportions.
Problem is, that unlesss you are seriously digging into this, it is pretty much impossible to even consider that what you hear is not what it really is. Time and again it has been proven that you CANNOT trust any of your senses to give you a clear, real, objective view of what's out there. The point is that you don't 'hear' what your ears hear; you hear what pops up in your conciousness after those air vibrations that get in your ears are heaviliy processed by that machine we call mind/brain.
It's impossible to overestimate the amount of manipulation that goes on. The brain/mind is not in the least interested to give you an accurate (sound) picture. It is only interested to preserve your ego, to make sure you feel great, to make sure you feel like the world is as you expect it to be.
There is such a wealth of well documented scientific research out there on these subjects, but most are not even aware of the issues, totally ignorant about it. So if someone says: "maybe it is not the way you think you heard it" the immediate reaction is defensive, as if it is something culpable. And it is understandable: you hear what you hear, right? You see what you see, right? But it has nothing to do with hallucination, because what we hear, what we see, is real enough for us. But it aint necessarily so. No, it is a totally wrong-placed trust in your perceptions.
The only way to circumvent that devious manipulation in your mind is to prevent it from getting any clues. And I mean ANY clues. As an example of our extreme sensitivity, one research has shown that the effects of medicine tests were skewed when the person giving out the medicine knew which were the placebo's or not: his bodylanguage gave him away. The test subjects were totally unaware of it, yet it influenced the results.
One important method to shut those clues out is called (double) blind testing.
Jan Didden
Trust your ears, folks! Ignore those who hear no difference. They are only a waste of time and enjoyment stiflers.
abzug said:
I thought sure I could in one of my own pieces. I seem to have lost the wav file I'll make another and try it as soon as I get time.
- Status
- Not open for further replies.