What I meant is yeah, you are right about that, but no, lossless formats aren’t called lossless because they don’t lose anything to the original, they’re called lossless because, after compressing and decompressing, you get the exact same file that you initially compressed.
Another commenter on this post explained it really well.
They’re deemed “lossless” because there are no data losses - the word actually comes from the broader domain of data handling, specifically Compression were for certain things - like images, audio and video - there are compression algorithms that lose some information (lossy) and those which don’t (lossless), for example JPEG vs PNG.
However data integrity is not at all what your average “audiophile” would be talking about when they say there are audio losses, so when commenting on what an non-techie “audiophile” wrote people here used that “losslessness” from the data domain to make claims in a context which is broader that merelly the area were the problem of data integrity applies and were it’s insuficient to disprove the claims of said “audiophile”.
By your definition, PNG isn’t lossless because it’s not an exact representation of every single photon of a picture that was taken. You’d need infinity pixels in order to be completely faithful to the “analog” thing that you’re trying to picture, in the same way you’d need infinity points to completely translate an analog wave to digital.
When you compress anything with FLAC, you will get the exact same thing you compressed out, so there is no data loss.
Of course, that wave which you compress will not be faithful to the analog thing, but that’s just a limitation of digital computers.
A PNG is indeed an imperfect representation of reality. Are you claiming that the lossness in the data domain of the compression algorithm in a PNG means its contents are a perfect representation of reality?!
(Funnilly enough, the imperfections in the data contained on a PNG are noticeable for some and the lower the “sampling rate” - i.e. number of pixels, bits per pixel - the easier it is to spot, same as audio)
As I’ve been trying to explain in my last posts, a non-Techie “audophile” when they claim FLAC is not lossless aren’t likely to be talking about it’s technical characteristics in the data domain (i.e. that data that you take out of a FLAC file is exactly the same as it goes in) but that its contents don’t sound the same as the original performance (or, most likely, a recording made via an entirelly analog pathway, such as in an LP).
Is it really that hard to grasp the concept that the word “lossless” means different things for a Technical person with a background in digital audio processing and a non-Technical person who simply compares the results of a full analog recording and reproduction pathway with those of a digital one which include a FLAC file and spots the differences?
This feels like me trying to explain to Junior Developers that the Users are indeed right and so are the Developers - they’re just reading different meanings for the same word and, no, you can’t expect non-Techie people to know the ins and outs of Technical terms and no they’re not lusers because of it. Maybe the “audiphile” was indeed wrong and hence “Confidently Incorrect”, but maybe he was just using lossless in a broader sense of “nothing lost” like a normal person does, whilst the other one was using the technical meaning of it (no data loss) so they were talking past each other - that snippet is too short to make a call on that.
So yeah, I stand by my point that this is the kind of Dunning-Krugger shit junior techies put out before they learn that most people don’t have the very same strictly defined technical terms on their minds as the junior techies do.
Not really infinite points since energy is quantized. In a crazy particle physics sense analogue data is effectively the same as digital, when resolutions match.
I’m not a Sound Engineer, I’m an Electronics Engineer - we’re the ones who had to find the right balance between fidelity, bit error rates, data rates and even circuit price when designing the digital audio sampling systems that capture from the analog world the digital data which the Sound Engineers use to work their magic: so I’m quite familiar with the limits of analog to digital conversion and that’s what I’m pointing out.
As it so happens I also took Compression and Cryptography in my degree and am quite familiar with where the term “lossless” comes from, especially since I took that elective at the time when the first lossy compression algorithms were starting to come out (specifically wavelet encoding as used in JPEG and MPEG) so people had to start talking about “lossless” compression algorithms with regards to the kind of algorithms what until then had just been called compression algorithms (because until then there were no compression algorithms with loss since the idea of losing anything when compressing data was considered crazy until it turns out you could do it and save tons of space if it was for stuff like image and audio because of the limitations of human senses - essentially in the specific case of things meant to be received by human senses, if you could deceive the human senses then the loss was acceptable, whilst in a general data sense losing data in compression was unacceptable).
My expertise is even higher up the Tech stack than the people who to me sound like Junior Devs making fun of lusers because they were using technical terms to mean something else, even while the Junior Devs themselves have yet to learn enough to understand the scope of usage and full implications for those technical terms (or the simple reality that non-Techies don’t have the same interpretation of technical terms as domain experts and instead interprete those things by analogy)
STFU and stop dropping your resume. Nobody gives a shit, and I can tell you I’m FAR more knowledgeable than you.
You literally don’t understand the difference between a sensor, data, compression, or anything else. You don’t understand energy, physics, or the underlying concepts.
You are not as informed as you personally believe.
What I meant is yeah, you are right about that, but no, lossless formats aren’t called lossless because they don’t lose anything to the original, they’re called lossless because, after compressing and decompressing, you get the exact same file that you initially compressed.
Another commenter on this post explained it really well.
They’re deemed “lossless” because there are no data losses - the word actually comes from the broader domain of data handling, specifically Compression were for certain things - like images, audio and video - there are compression algorithms that lose some information (lossy) and those which don’t (lossless), for example JPEG vs PNG.
However data integrity is not at all what your average “audiophile” would be talking about when they say there are audio losses, so when commenting on what an non-techie “audiophile” wrote people here used that “losslessness” from the data domain to make claims in a context which is broader that merelly the area were the problem of data integrity applies and were it’s insuficient to disprove the claims of said “audiophile”.
By your definition, PNG isn’t lossless because it’s not an exact representation of every single photon of a picture that was taken. You’d need infinity pixels in order to be completely faithful to the “analog” thing that you’re trying to picture, in the same way you’d need infinity points to completely translate an analog wave to digital.
When you compress anything with FLAC, you will get the exact same thing you compressed out, so there is no data loss.
Of course, that wave which you compress will not be faithful to the analog thing, but that’s just a limitation of digital computers.
A PNG is indeed an imperfect representation of reality. Are you claiming that the lossness in the data domain of the compression algorithm in a PNG means its contents are a perfect representation of reality?!
(Funnilly enough, the imperfections in the data contained on a PNG are noticeable for some and the lower the “sampling rate” - i.e. number of pixels, bits per pixel - the easier it is to spot, same as audio)
As I’ve been trying to explain in my last posts, a non-Techie “audophile” when they claim FLAC is not lossless aren’t likely to be talking about it’s technical characteristics in the data domain (i.e. that data that you take out of a FLAC file is exactly the same as it goes in) but that its contents don’t sound the same as the original performance (or, most likely, a recording made via an entirelly analog pathway, such as in an LP).
Is it really that hard to grasp the concept that the word “lossless” means different things for a Technical person with a background in digital audio processing and a non-Technical person who simply compares the results of a full analog recording and reproduction pathway with those of a digital one which include a FLAC file and spots the differences?
This feels like me trying to explain to Junior Developers that the Users are indeed right and so are the Developers - they’re just reading different meanings for the same word and, no, you can’t expect non-Techie people to know the ins and outs of Technical terms and no they’re not lusers because of it. Maybe the “audiphile” was indeed wrong and hence “Confidently Incorrect”, but maybe he was just using lossless in a broader sense of “nothing lost” like a normal person does, whilst the other one was using the technical meaning of it (no data loss) so they were talking past each other - that snippet is too short to make a call on that.
So yeah, I stand by my point that this is the kind of Dunning-Krugger shit junior techies put out before they learn that most people don’t have the very same strictly defined technical terms on their minds as the junior techies do.
Not really infinite points since energy is quantized. In a crazy particle physics sense analogue data is effectively the same as digital, when resolutions match.
Fake it 'till you make it is not applicable to scientific or technical discussions.
Nice content-free slogan.
I’m not a Sound Engineer, I’m an Electronics Engineer - we’re the ones who had to find the right balance between fidelity, bit error rates, data rates and even circuit price when designing the digital audio sampling systems that capture from the analog world the digital data which the Sound Engineers use to work their magic: so I’m quite familiar with the limits of analog to digital conversion and that’s what I’m pointing out.
As it so happens I also took Compression and Cryptography in my degree and am quite familiar with where the term “lossless” comes from, especially since I took that elective at the time when the first lossy compression algorithms were starting to come out (specifically wavelet encoding as used in JPEG and MPEG) so people had to start talking about “lossless” compression algorithms with regards to the kind of algorithms what until then had just been called compression algorithms (because until then there were no compression algorithms with loss since the idea of losing anything when compressing data was considered crazy until it turns out you could do it and save tons of space if it was for stuff like image and audio because of the limitations of human senses - essentially in the specific case of things meant to be received by human senses, if you could deceive the human senses then the loss was acceptable, whilst in a general data sense losing data in compression was unacceptable).
My expertise is even higher up the Tech stack than the people who to me sound like Junior Devs making fun of lusers because they were using technical terms to mean something else, even while the Junior Devs themselves have yet to learn enough to understand the scope of usage and full implications for those technical terms (or the simple reality that non-Techies don’t have the same interpretation of technical terms as domain experts and instead interprete those things by analogy)
STFU and stop dropping your resume. Nobody gives a shit, and I can tell you I’m FAR more knowledgeable than you.
You literally don’t understand the difference between a sensor, data, compression, or anything else. You don’t understand energy, physics, or the underlying concepts.
You are not as informed as you personally believe.
Literally STFU