To be fair, I wouldn’t call it lossy. It’s not bit perfect, it’s unnecessarily big and it doesn’t add any extra audio information, but it doesn’t remove any either. It should be sonically transparent. A kind of “lock in” of the up-sampling and bit reduction DACs do during playback if you like.
Sorry, I can’t let this go. No disrespect intended.
You are correct when you speak of bit perfect file decompression. This is very important.
But you are overlooking the fact that compressed files need to be decompressed. And that introduces noise. Yes, and this is important, digital files can carry noise.
Having a galvanically isolated DAC helps but noise and jitter introduced during decompression is a thing. More current has to flow through circuits, heating them up, requiring more cooling and when the fans kick in they create RF noise that is picked up by other components on the motherboard. In a very noisy system the noise can even get picked-up by nearby devices. That can affect timing which can have an impact on your bit-perfect file.
Whenever possible, current conventional wisdom is to avoid compressed file formats in order to minimize the work required by your streaming device, whether that be an audiophile streamer or a computer. The more simple the electronic path, the less opportunity to impact the file.
FLAC does not meet this requirement. I’m happy that it works for you but please stop trying to ram it down everyone’s throat.
For me, and likely some others here, we are not interested in using FLAC.
1 Like
Nonsense and more nonsense. How on earth can a digital file with correct bits carry noise?
3 Likes
PS over promising and under delivering, no!
Enough of this nonsense. When you get an engineering degree, come back let us know what you have learned.
If life has taught me one thing, it is that things are not always what the seem. Bits, ones and zeros are not always reproduced exactly bit perfectly. Shocking because protocols are in place to catch and fix these things. And newer technology handles this much better than technology of old.
However timing of those bits is equally, and some might argue more, important than a missing bit here and there. Clearly you do not understand that having a bit perfect file is only part of the problem. You assume that what you can’t see must be perfect. An oversimplification used to teach children. I had hoped you were not a child. Apparently I was wrong.
This is taking a turn in a nasty direction where I’m sure we’ll end up with pistols at 10 paces if we don’t stop now.
You have your opinion. I disagree with it. I think it’s grossly oversimplified. Let’s leave it there and call it a day.
How can one rationally deal with so much nonsense? The author clearly has no understanding at all of how things work…
You, sir, are entitled to your beliefs and opinions, as are all the little ones to their belief in the tooth fairy and Father Christmas. But you will be called out and confronted, and the best one could hope for is that the thread will be shut down ASAP.
Naaah, you are clearly wrong. ‘Conventional’ wisdom is that FLAC is the most common format for lossless PCM audio compression, widely employed in the audio industry to great success. Again, your understanding of how things works is greatly flawed.
I am out of this thread, feeling no desire to further confront nonsensical fake knowledge.
7 Likes
It is I who am calling you out. Your lack of understanding of things technical is disappointing.
Let me give it one last try. You are pinning your opinion on the fact that a bit perfect file is the same as the original file. Are we agreed so far?
The problem was never the file. It is the fact that decompressing the file requires work (in the physics sense) that uses electrical energy. The cleaner the path, meaning the less work, i.e., less energy required, is better. The moment you introduce any extra work, like decompression, in to the system you introduce noise in a variety of ways. You can’t avoid it. However you may not notice it for a variety of reasons. The most likely is that your system is not sufficiently resolving for you the hear the differences. For example, when I’m listening to music in my car or through Apple AirPods Pro (v2), I won’t notice noise very much. However when I’m listening to the same file through my DAC, tube amp and high end headphones, I’m likely to hear the difference. Context is everything. I’d prefer to think your system is less resolving and you haven’t noticed the difference. The alternative is that you’re an idiot.
I’m sure the readers of this forum have drawn their own conclusion and are too polite to tell us. And that’s ok. 
Have a great weekend.
@David_Gow, please keep things civil, and if wish to respond to something by disagreeing with it, that’s fine. However, remember to criticize ideas, not people.
2 Likes
Sure, I meant lossy as in loses the original bit-perfect data. Anyway, after ripping to FLAC the DSD can be created any time, so surely is what one should do, as you already wrote previously
1 Like
While noise-into-the-DAC is already questionable even during FLAC playback, it is completely wrong that noise gets introduced into files during CD ripping and therefore is entirely off topic in the context of the thread. When you rip the CD bit-perfectly to FLAC, the resulting PCM data after decoding is 100% indistinguishable from the PCM that was on the CD, so once you have the FLACs you can later convert them to DSD files at any time.
If you think you can distinguish the PCM data in the ripped FLAC from the PCM data on the CD, I invite you to prove it and I guarantee you a Nobel price in physics or mathematics.
Don’t know about your tube amp, but these are usually of worse signal to noise and distortion characteristics - not the best resolving in technical terms, although you might prefer the added signature.
Just saying!
You seem to imply that this even does audibly impact streaming over Ethernet to an RAAT enabled DAC.
Why hasn’t any proponent of said narrative ever been able to show factual proof?
On the other hand, there is testament available online, showing no measurable, let alone audible, consequence when “stressing” the computing side of things.
After all, flac decompression is not handled by the endpoint anyway - it’s fed LPCM data.
And it’s galvanically isolated on top of things.
Your claimed timing issues can be revealed in the analog output of DACs via jitter measurements, and any current state of the art DAC easily keeps those artifacts orders of magnitude below audibility - so not an issue to be worried about.
Instead of spreading unsubstantiated claims, bring forward factual evidence.
2 Likes
Hey guys.
I’m sorry if I offended anyone. That’s not my intent. I replied to someone asking about file formats and as I use the same software I thought it would be helpful to share my thoughts on the file format I choose to use. Then someone kept coming back over and over to espouse the benefits of FLAC. I over reacted and I’m sorry. FLAC will never be my first choice and as I clearly state, if it’s yours that’s ok.
As for calling in to question my technical knowledge, well I readily confess I might be a little rusty but suggesting that tube amps are inferior is just silly. And you do understand that square waves are never perfectly square, right? A lot depends on the quality of the components and the design.
For the record my signal path is: DSD up to 512 and PCM files connected via a windows laptop running ASIO USB drivers → Holo Audio May KTE DAC → Woo Audio WA33 (standard with JPS wiring, fully class A amp) via balanced cables → Audeze LCD 4z using twin XLR Kimber Axios copper headphone cable. I consider this to be a very resolving system and I consider myself lucky to be able to listen to music through it.
Enjoy your music in whatever way works for you and have a great Sunday evening.
It may well be that DSD into a DSD DAC sounds better in some implementations. I, for one, have no opinion on that, but again, it’s irrelevant in this thread.
The question was whether to rip CDs to WAV or DSD and there is only one technically and logically correct answer, FLAC. The CD contains PCM samples, FLAC stores them bit-perfectly at a smaller size than WAV and saves checksums in the file, and you can create DSD files from that at any later point, which are 100% indistinguishable from files that are DSD-converted immediately from the PCM during the ripping.
1 Like
This old “your system is less resolving” Is just an ad hominem attack directed at gear rather than a person (ad machinem?). It’s not a valid argument. It’s the same as “I’m smarter so I’m righter.”
Decoding FLAC at the core does not create more noise at a separate RAAT endpoint. The same identical PCM stream enters the endpoint regardless of the codec or wrapper.
No intent to insult the poster but misinformation in this area is rampant and begs comment. In no way is it conventional wisdom to avoid decoding at the core.
With due respect there’s almost no salvageable truth from this word casserole. Just a caution to the newbie…this isn’t far off from sacrificing a goat to stop the volcanic eruption…
Use a separate endpoint as Roon recommends.
5 Likes
Even if the DAC was connected directly to the core, it would still not be a problem. I measured the output of a USB DAC connected to a machine with both low and high CPU and didn’t find any differences.
1 Like
We’ll, during a move I was temporarily using a first generation ifi DAC connected directly to a very noisy core with a bunch of fans and spinning hard disks. I could hear drives click through the DAC and my phone ring when I had it charging from a USB port. So it’s not like every DAC will be immune to everything on a PC.
So the possibility of noise over USB is there. And it didn’t take much in terms of a resolving system to hear it. But I agree it’s theoretical with decent gear and pure fantasy over network connected RAAT endpoints. As though the endpoint knew what a file started as…
2 Likes
If you think about this logically, referring to a configuration where the endpoint is separate from the Core, there is nothing special about what is happening on the core. It’s just another machine on the network generating noise and the emergence of the PCM stream from that machine is irrelevant to the theoretical impact of noise on some other device on the network.
If referring to electrical noise generated by computation, it’s not as though that noise associates itself with the digital PCM stream leaving the network adapter and heading to the RAAT endpoint - it’s not its shadow or something. So that noise could just as easily be generated by another computer operating a web browser, or a bitcoin miner, whatever, even your cable modem, and even equipment upstream at your ISP!
Point being, there’s a difference between worrying about noise from devices on your network (which is theoretical relative to its impact on a DAC) and some form of data damage occurring on your core because it is decoding a compressed file. The former is theoretically possible but practically unlikely and networks have a ton of technology to eliminate it; the latter is just fantasy, and if you are a believer in this, only addressing the electrical noise your core makes, without addressing every device on your network, is equivalent to attempting to sound-proof your front yard.
2 Likes
Except they are. They intentionally add distortion for a distinct sound signature. As with anything, some like this, some don’t. Neither preference is wrong but to think a tube amp performs as well as any decent solid state amp is just silly. I have a tube amp and (very transparent) solid state amp in my primary listening location so I get to choose what sound signature I’m in the mood for (or if I’m just in the mood to see the glowing tubes and vu meters).
2 Likes
They don’t add distortion intentionally. They add distortion irrevocably.
I have a PrimaLuna system feeding Tannoys that I wouldn’t change for any solid state preamp/amplifier.
3 Likes
Somebody should ask for their money back for that engineering degree…
Tubes may sound quite pleasing and I would not argue with someone who prefers that sound aesthetically, but claiming that a tube system could even theoretically be more resolving than any but the cheapest and intentionally mis-engineered solid state one is… just ridiculous.
6 Likes