Can You Hear the Difference between Uncompressed and MP3 Audio?

[Regeneration]

I can only distinguish between 24-bit FLAC in high bitrate vs MP3s.

Average 16-bit FLAC vs 320kbps MP3 is almost the same.

 

[Operandi]

Rubbish. What the tomato analogy has to do with it, makes no sense at all.

The fact is that:

1 Take a random WAV file
2 Losslessly compress it with FLAC
3 Decompress it with FLAC
4 Compare original file with the FLAC processed file: difference is zero and hence will sound exactly the same

It's as simple as that. That's precisely what lossless means. Please don't try to muddy the waters on something so basic and so settled. Google and Wikipedia are your friends if you're still unsure of this.

I really, really tried to make sense of that analogy... but yeah FLAC is just a storage container that compresses the PCM data which is what WAV is. Uncompressed its the same PCM data.

The only argument that I've ever heard that makes any kind of sense for WAV over FLAC is that decoding the file is more CPU intensive and puts more load on the powersupply components, creating more noise in the power delivery to rest of the system. Not my argument but its one that you'll hear if you ask.

 

[ChosenName]

Since we are in season. Say I take a tomato and split the seeds between two gardens in reasonably close proximity tended by separate people. To but a few intimately aware of them both will it be apparent one set of plants received better soil prep and nurturing. Across a wide range of uses they portray less unwanted flavoring or other characteristics. It is right that anyone should be happy with and freely mix the two.

They are highly similar. Not the same. Regardless, my intent here was to provide a basis of examination that furthered interest. Flac is a highly useful backend industrial streaming format. Meaning it can be quickly converted to any number of other formats or compressed for delivery. Portraying the term lossless as an absolute is a falsehood. It describes meeting a level of mathematic proof having no bearing on any other element that determines quality of what is stored. Both tomatoes proved insect resistant enough to ripen without flawed skin.

An interesting, but ultimately irrelevant, analogy.

As a lossless CODEC, by definition, outputs exactly the same samples as went in. The ones and zeros of the bitstream are no different from the original, unless, as indicated, something is broken.

Unless the quality of the equipment in some way, in the possibly (confirmation) biased opinion of the user in a sighted test, makes the ones more "oney" and the zeros more "zeroey"....

 

[qubit]

I really, really tried to make sense of that analogy... but yeah FLAC is just a storage container that compresses the PCM data which is what WAV is. Uncompressed its the same PCM data.

The only argument that I've ever heard that makes any kind of sense for WAV over FLAC is that decoding the file is more CPU intensive and puts more load on the powersupply components, creating more noise in the power delivery to rest of the system. Not my argument but its one that you'll hear if you ask.

It's true, the more compressed something is, the more processing power is required. And it doesn't matter if it's a zip file or audio, it's all the same bits and takes more out of the PSU as well. I wouldn't have thought it would be especially significant on a mains powered PC, but on anything battery powered, it could be quite significant.

 

[ChosenName]

It's true, the more compressed something is, the more processing power is required. And it doesn't matter if it's a zip file or audio, it's all the same bits and takes more out of the PSU as well. I wouldn't have thought it would be especially significant on a mains powered PC, but on anything battery powered, it could be quite significant.

FLAC decodes on my phone at over 5,000x realtime - which suggests that playing it in realtime uses a tiny fraction of the device's CPU capability.

 

[Ferather]

FLAC does need decoding, there could be something if the decoder is not programmed perfectly, or has bugs, else lossless is lossless WAV (PCM) or not.
You will also find lossless is a reference to digital audio, compressed or not, I have never seen lossless analogue advertised.

Anyone ever seen lossless DAC's? If so please list them, the DAC its self must be advertised as lossless.
Without a lossless certificate or similar the difference between DAC's will vary.

 

[ChosenName]

You will also find lossless is a reference to digital audio, compressed or not, I have never seen lossless analogue advertised.

Not sure that any analogue media are not prone to degradation with use.

 

[Ferather]

CD-DVD still degrade overtime, but you are right mostly, its nothing like VHS-Tape. Even floppy disks would decay fairly rapidly, get to disk 12 of Windows 95 install, bad disk QQ.

----

If we pulled away from storage media, and converted that back into transmission method (like media), where digital transmission goes further (less decay).
On the same note, the difference again in distance between conductive copper and optical is quite different, optical being in front.

Wired internet vs optical (fiber broadband), distance, speed and response time (ms).

 

[ChosenName]

CD-DVD still degrade overtime, but you are right mostly, its nothing like VHS-Tape. Even floppy disks would decay fairly rapidly, get to disk 12 of Windows 95 install, bad disk QQ.

Indeed - noting that a perfect backup copy of digital media can be made, even after some use, whereas the same cannot be said of analogue.

 

[qubit]

You will also find lossless is a reference to digital audio, compressed or not, I have never seen lossless analogue advertised.

Indeed, lossless analog is impossible.

For clarity, the physical representation of the bits is subject to "analog" effects and degradation, but as long as the bit is recognised correctly, then it's perfect. And of course, where the bit cannot be properly recognised is where error correction techniques come in.

 

[Ferather]

Indeed, you can still get clock jitter with digital (example HDMI, SPDIF), which influences audio if it happens. However, both HDMI and SPDIF PCM is rated lossless, so I guess not much of an issue these days.

 

[Operandi]

Indeed, lossless analog is impossible.

For clarity, the physical representation of the bits is subject to "analog" effects and degradation, but as long as the bit is recognised correctly, then it's perfect. And of course, where the bit cannot be properly recognised is where error correction techniques come in.

Yeah, lossless analog is not a thing. Every medium, transmission and reproduction method has noise and distortion in it. You can measure and compare how much noise and distortion is present in both and sort of compare them but its really kinda silly in my opinion.

Anyone ever seen lossless DAC's? If so please list them, the DAC its self must be advertised as lossless.
Without a lossless certificate or similar the difference between DAC's will vary.

Same thing with DACs. You can in theory get every bit from your PC, CD player, or streamer to the DAC in perfect timing but even that dosn't make the DAC lossless as the DAC has to do the (AC) analog conversion part and there is simply no perfect way to do that.

 

[Ferather]

That's exactly why DAC's should be rated, and also why I said it should be digital (in my case I said optical) right up to the speaker, and a PowerDAC right in front of the driver, less than and inch for example.
This eliminates as much analogue as possible, as lossless as possible, this way you can blame only the DAC, or the HDA device. A crap HDA device + HDMI does not mean lossless.

----

Another thing, which some of you might not like to hear, is also speaker power resistance, since analogue is power in the circuit, the speaker resists audio.

My Idea above counts as active speakers, opposed to passive speakers, and could potentially be 0 ohm (already matched).

The Complete Guide To Speaker Impedance (2Ω, 4Ω, 8Ω & More)
How to Measure Speaker Impedance (with Pictures)

Ohm Speakers | Custom Audiophile Speakers (scroll down)

----

PowerDAC-S - ECdesigns | Wadia Digital – Power DAC

https://www.techpowerup.com/forums/attachments/smart-speakers-png.234708/

 

[nomdeplume]

The fact is that:

1 Take a random WAV file
2 Losslessly compress it with FLAC
3 Decompress it with FLAC
4 Compare original file with the FLAC processed file: difference is zero and hence will sound exactly the same

Sir, I have randomly selected a portion of your response to deal with the logical fallacy being portrayed.

Lossless is a relative term. Seizing upon it as an absolute will garner no further responses. What is today reasonably low level criteria are being satisfied to meet the established technical conditions of lossless files. To be perfectly blunt, I doubt many could tell the difference between files using the same hardware in the same conditions to the same ends as are being depicted with regularity across this site. A fact I have until now politely dealt with by noting the production side of music currently favors that outcome. You don't unhear noise until you realize it is gone.

Here is fun little audio project to survey the correctness of your response internally. For purposes of validation and proof of concept anyone can attempt.

1. Take a newly created WAV file.
2. Ask your favorite encoding software to preserve the original file and make a flac which you will then encode back to WAV
3. Continue with the resultant file until you have reached the 10th generation.
4. Play back as much of the original WAV file and as much of the last WAV file you made as it takes to make a decision if any differences exist and why.



@Ferather You are asking some important questions that lead away from popular opinion. Don't grow upset if you find answers don't immediatly make sense in relation to what you've experienced. Don't grow upset, grow wiser. As with any hobby or interest it is the people who make it enjoyable. Listen.

 

[R-T-B]

Since we are in season. Say I take a tomato and split the seeds between two gardens in reasonably close proximity tended by separate people. To but a few intimately aware of them both will it be apparent one set of plants received better soil prep and nurturing. Across a wide range of uses they portray less unwanted flavoring or other characteristics. It is right that anyone should be happy with and freely mix the two.

Except that flac is literally uncompressed pcm at the tail end. Your analogy makes no sense.

Lossless is a relative term.

It's not. If it was zip files, pngs, etc would not work.

Seizing upon it as an absolute will garner no further responses.

Good.

 

[Ferather]

Question: If HDMI and SPDIF, PCM transmission is considered lossless, and they both only do up to 24bit, its 32bit a white elephant?

 

[R-T-B]

Question: If HDMI and SPDIF, PCM transmission is considered lossless, and they both only do up to 24bit, its 32bit a white elephant?

I believe hdmi can transmit 32-bit but I have yet to see it used commonly.

 

[Operandi]

This is kinda going off topic but I think you are some misconceptions here and you missing the point a bit.

That's exactly why DAC's should be rated

Rate it how? You can look at a DAC's specs and see what formats its able to handle and the S/N ratio ratio, distortion figures, maybe a few other factors. On their own thats all pretty overrated though as even the what formats the DAC can handle is misleading as there isn't a DAC on the planet that can resolve 24 bits of resolution.

and also why I said it should be digital (in my case I said optical) right up to the speaker, and a PowerDAC right in front of the driver, less than and inch for example.
This eliminates as much analogue as possible, as lossless as possible, this way you can blame only the DAC, or the HDA device. A crap HDA device + HDMI does not mean lossless.

You mean an active speaker with a built in DAC like the KEF LS50 Wireless? Its been done before and in the case of the KEF the broad consensus is that the regular LS50 with its passive crossover sounds better. Class AB, and A amps are still generally considered the best sounding amplifers, class D is closing the gap when you get to into the stuff that Hypex and ICEpower are doing but if you look at the specs of the KEF its class AB on the tweeter and D on the woofer (thats done for a reason). You can also do an entire crossover network with DSP and they can sound good but the best speakers use passive resistors, capacitors, and inductors, and the better speakers use higher quality versions of those components that have better analog characteristics for their respective values. The limiting factor is not in analog components and it certainly isn't in the last few meter of wire between the amplifier and the speaker.

The thing to remember with all of this is that it starts analog and ends analog, being digital is only useful for preserving the signal along the way and as means to make (and preserve) perfect copies of that digitized analog signal. Digital audio has a lot of advantages but also introduces a shit load of problems which is why a DAC like the Topping E50 is nowhere near perfect despite what its specs and measurements may lead you to believe.

 

[lexluthermiester]

I believe hdmi can transmit 32-bit but I have yet to see it used commonly.

By default, HDMI uses MPEG audio compression. It can be configured to use PCM but comes at the cost of using bandwidth otherwise allocated to video data.

 

[Ferather]

Yes sorry for off-topic posts, well semi-related since playback equipment effects results. By rated DAC I meant rated lossless opposed to specs.

 

[R-T-B]

By default, HDMI uses MPEG audio compression. It can be configured to use PCM but comes at the cost of using bandwidth otherwise allocated to video data.

What? No PCM mode comes at the cost of channels if anything. It never cuts into video data, they are on completely seperate wires...

 

[Operandi]

Yes sorry for off-topic posts, well semi-related since playback equipment effects results. By rated DAC I meant rated lossless opposed to specs.

Thats not how that works though either. You can have a FLAC file or an MP3, both will get decoded into PCM by your PC or streaming device and then sent to your DAC.

 

[Ferather]

Don't think the MPEG is right, EDID defines LPCM, but the pins, I just double checked here. I remember doing something with the EDID on a converter and got some bandwidth issue, might of been the cable.

@Operandi, no I mean the audio coming out of the DAC is lossless, opposed to lossless in, but lossy out, hence rated a lossless DAC.

 

[qubit]

Sir, I have randomly selected a portion of your response to deal with the logical fallacy being portrayed.

Lossless is a relative term. Seizing upon it as an absolute will garner no further responses. What is today reasonably low level criteria are being satisfied to meet the established technical conditions of lossless files. To be perfectly blunt, I doubt many could tell the difference between files using the same hardware in the same conditions to the same ends as are being depicted with regularity across this site. A fact I have until now politely dealt with by noting the production side of music currently favors that outcome. You don't unhear noise until you realize it is gone.

Here is fun little audio project to survey the correctness of your response internally. For purposes of validation and proof of concept anyone can attempt.

1. Take a newly created WAV file.
2. Ask your favorite encoding software to preserve the original file and make a flac which you will then encode back to WAV
3. Continue with the resultant file until you have reached the 10th generation.
4. Play back as much of the original WAV file and as much of the last WAV file you made as it takes to make a decision if any differences exist and why.

Well, it would have been better if you'd have quoted my whole post as that last sentence is born from the previous paragraphs. Why are you afraid to Google it and find out the truth about lossless encoding? Also, note how all the other members are saying the same as me, in their own ways and from different angles, but the bottom line is that a FLAC file is an identical copy of the original WAV when expanded back out again, which you have not addressed. Why is that so hard for you to understand? Or are you just trolling to get a rise out of the members? On something so obvious, this possibility sounds quite plausible.

Assuming you're not trolling, then that's a cute little gauntlet you've thrown down there and valid, too. I might just take you up on it. However, I've not done any audio processing for over a decade, so what would be a good app to do this with, the free Audacity maybe?

If you heard a difference when you tried it, then it wasn't a lossless copy, end of story. Perhaps you converted to 320Kbps MP3 and didn't realise it? The first few copies would sound fantastic, but by the 10th copy would most likely sound quite degraded.

Of course, we both know what happens when I come back to you and say that there was no difference after 10, 15, 20, 30 copies or more. You'll come up with some other excuse to say that the process somehow produces different files and maybe come up with some other crazy analogy like a different vegetable or something to wind everybody up and you'll keep going round in circles.

If you really won't reply further, then that may probably be for the best.

 

[Dr. Dro]

Except that flac is literally uncompressed pcm at the tail end. Your analogy makes no sense.

I agree with you, they're greatly exaggerating, but if one wants to be exceptionally picky about data integrity, even raw PCM actually has a small degree of quantization noise, it's just that the data resolution is so insanely high that it's - IMHO - biologically imperceptible, unless you're Superman.

This article is amazing and it covers the "loss" from format conversion (mostly centered around DSD which normally uses 1-bit samples at a rate of 2.82 MHz instead of the Red Book CD's 16-bit samples at a 44.1 kHz rate), but honestly, it's nothing that anyone should be knocking themselves over unless they're at the mastering stage and absolute perfection must be achieved.

DSD vs. PCM: Myth vs. Truth

DSD vs. PCM: Myth vs Truth

www.mojo-audio.com