The people at Microsoft didn't think there should be a difference playing through default sound vs kernel streaming either. If you can't perceive a difference there, will you want to be talking about that, too?
Due to the fact that the flac file goes through your processor, the result has a noisy edge added. You can say that theoretically flac is meant to be the same, but remember, it's designed to fool you. Uncompressed is the one that is theoretically perfect.
MP3s are designed to "fool" listeners as it's lossy compression and the designers attempted to strip out data that supposedly people can't hear to make smaller files.
Flac isn't lossy and, regardless of the amount of compression used, it's bit perfect, hence, no designs to fool anyone.
Adding the various OS and kernels / drivers / latencies / jitter etc for playback all into the mix is opening another level of things that make huge difference. That's what makes digital audio playback so interesting, and way more difficult to do it right...
I agree with 7ryder that flac isn't designed to fool anyone, it's just a format that saves space and bandwidth at the cost of some processing. If the processing is done correctly, there should be no audible impact at the end of the chain.
But it is designed to fool you. Uncompressed is the original, flac is designed to be smaller without losing any data. Whether or not sounding no different is possible while having to make a superspeed chip work on it is what you're telling me about your own flac experience. I'm not arguing about all the bits still being there, otherwise, yes I would be saying that mp3's obviously don't tell us much of what goes on, on top of chip noise.
Yeah, unfortunately this is sort of what I was expecting. The usual jump for “I hear A and B” to “and it is because of reason C”. Just like that, bam, categorical statement.
I can’t say that you don’t hear any difference, but I can say that you’re searching for the reason for any difference you might be hearing in the wrong place.
I don’t know why you’d be hearing such a difference, but compression and processing isn’t it, it’s just not the way this works, at all. That much is clear for anyone with basic knowledge on how computers, processing and digital works.
Then I'd say read the FLAC into for instance Adobe Audition and convert it back to WAV... the data is the same, just like unzipping a zipfile.
FLAC is lossless so there is no regret needed for that way of ripping, it can still be undone...
The benefit is with wav, perfectly lossless doesn't mean there isn't noise added. A flac file is a wav file that has to be unpacked to wav to play. It takes a processor operation, first.
Playing a WAV requires multiple processor actions including identifying addresses, looking up blocks, instructions to read, interrupts, as well as reading into a buffer, a movement across a bus, reading from a buffer, movement across a different bus etc. There are other parallel workloads such as network interrupts, screen refreshes etc.
If you think that the workload of conversation makes a net difference to the total computer noise, you demonstrate a shortfall in your knowledge of how computers work. I'm sure you believe are hearing what you are hearing, but you aren't.
Because it is not an audio related operation, it doesn't care about your frequency or time domain and it doesn't produce distortion products that are not accounted for. It is a lossless, bit perfect operation, that hardly registers in activity monitors of a normal system. It was designed to be that way, it is not an accident, or something after the fact, some miraculous thing that we found about are are still trying to tweak around for best outcome because we don't know the rules of how it operates. If you want to discuss things further you're going to have to read some books on computer and software engineering (not blog posts from overconfident keyboard experts that somehow know everything about everything), go back to your experiments and try to figure out what is happening. Assuming you'll still hear a difference, you may have noise on your I/O bus, you may have shitty codecs or drivers. You may have 101 things (other than the normal and unavoidable bias) that make you hear something. But compression isn't it.
It's a problem of knowing what you don't know at this stage.
Because it is not an audio related operation, it doesn't care about your frequency or time domain and it doesn't produce distortion products that are not accounted for. It is a lossless, bit perfect operation, that hardly registers in activity monitors of a normal system. It was designed to be that way, it is not an accident, or something after the fact, some miraculous thing that we found about are are still trying to tweak around for best outcome because we don't know the rules of how it operates. If you want to discuss things further you're going to have to read some books on computer and software engineering (not blog posts from overconfident keyboard experts that somehow know everything about everything), go back to your experiments and try to figure out what is happening. Assuming you'll still hear a difference, you may have noise on your I/O bus, you may have shitty codecs or drivers. You may have 101 things (other than the normal and unavoidable bias) that make you hear something. But compression isn't it.
It's a problem of knowing what you don't know at this stage.
A new operation happens, playing flac. If you can't hear it, your gear must be too noisy to notice. flac will never be better, you can only hope you can't tell it was flac.
Hardly. I’m always padding my statements ver carefully to never dismiss actually hearing something different. What is important is the attribution of that difference to a specific effect. In this case it is just wrong. We know that of all of the moving pieces in this process, compression is not the one to introduce a difference, by the very way in how it works for this specific case. It’s just barking up the wrong tree, unnecessary. IMO if you will…
A new operation happens, playing flac. If you can't hear it, your gear must be too noisy to notice. flac will never be better, you can only hope you can't tell it was flac.
I know this looks like a super smart thing to say, but I don’t think it is. Besides being useless.
On the other hand you’re probably the first person to find out about this, you can apply for an award or something.
Noisy flacs…
I’m assuming all the extra processing on the MQA steps also introduced a lot of extra noise right?
I know this looks like a super smart thing to say, but I don’t think it is. Besides being useless.
On the other hand you’re probably the first person to find out about this, you can apply for an award or something.
Noisy flacs…
I’m assuming all the extra processing on the MQA steps also introduced a lot of extra noise right?
Of course it would, having to go through an extra cheap chip. Switching from Tidal to Qobuz immediately proves mqa is a hoax.
Yes, I hear the processor bite, others do too. I don't know why I'm bothering in this thread. You don't have to hear it, if you don't want, don't worry. You didn't have to be able to tell mp3 wasn't all there, for lots of people, either. ymmv
Of course it would, having to go through an extra cheap chip. Switching from Tidal to Qobuz immediately proves mqa is a hoax.
Yes, I hear the processor bite, others do too. I don't know why I'm bothering in this thread. You don't have to hear it, if you don't want, don't worry. You didn't have to be able to tell mp3 wasn't all there, for lots of people, either. ymmv
So now comes the question that I posed in the beginning: how do you know the difference you’re hearing is process noise (whatever that is) and not the actual difference in the file (in the case of mqa)? Do you see my point? You can’t, you literally don’t have enough variables for the problem you state, you can’t know what you don’t know in an underdetermined system. I don’t doubt you hear a difference, but stating you know where that difference comes from where clearly you can’t explains a lot about the state of digital audio discussions…
Btw, I hear plenty of differences between some things. Most of those, I’m simply humble enough to say ‘I don’t know why that is’ instead of making up a whole new branch of reality. That doesn’t preclude me from asking, and I’m sure as hell not going to be making categorical statements about them.
So now comes the question that I posed in the beginning: how do you know the difference you’re hearing is process noise (whatever that is) and not the actual difference in the file (in the case of mqa)? Do you see my point? You can’t, you literally don’t have enough variables for the problem you state, you can’t know what you don’t know in an underdetermined system. I don’t doubt you hear a difference, but stating you know where that difference comes from where clearly you can’t explains a lot about the state of digital audio discussions…
Btw, I hear plenty of differences between some things. Most of those, I’m simply humble enough to say ‘I don’t know why that is’ instead of making up a whole new branch of reality. That doesn’t preclude me from asking, and I’m sure as hell not going to be making categorical statements about them.
I am comparing a wav file to a file converted to flac. I'm not even getting into mqa, which sounds like there is no extra info there from 44.1. I don't understand why I must be making something up, to hear noise introduced by playing back flac.
