[MrProggie]

I use lossless vinyl. The lossless vinyls are only half the size of normal vinyls. 

 

[Anaxilus]

Quote:

Hehe. You people seriously shouldn't be worried about data corruption (except for CDs where scratches can be a problem, but that's unavoidable). Do you realize that if uncorrected data corruption was actually a problem, essentially all modern computer technology simply wouldn't work? Do you realize that Google would just collapse if computers randomly dropped/lost bits of data here and there? Do you realize that satellites would be falling from the sky right now if what you're worried about is true?
 
The last time I had a corrupted file was on a 3.5" floppy disk what was exposed to a magnet. Even if you did have a corrupted file, you'd know it immediately because it would sound like scrambled noise. So no worries about slowly degrading quality, that simply doesn't happen with digital (CD scratches is another matter though).
 
Anyway I personally find 128 kbps compressed audio to be about as good as I need so far, but I would not entirely rule out 320kbps and loss-less formats I suppose. If there is any difference though, it would be incredibly subtle and honestly probably not worth it.

This is a myth that all the digital word is uncompensated binary '1's and '0's.  Logic implies that an error would shut down a logical binary system but because such things are inevitable redundancies and compensation have been implemented.  Such techniques include Interpolation and Buffer strategies.  However, this does not guarantee integrity.  If you have a poor lossless rip that doesn't add up it via checksum it will still sound like the song but at that point you now have a lossy file. If that's not a problem then don't bother with lossless.  Archiving data that has lost its integrity is beyond my scope of comprehension other than a person being simply too cheap to buy enough storage.
 
 
 

 

[ac500]

> such things are inevitable redundancies and compensation have been implemented. 
 
Correct. However the word "compensation" is misleading. In the context of file storage (on a CD, flash drive, hard drive, etc.), errors are not "compensated" or "blurred over", they're fully corrected back to 100.00% accuracy.
 
> Such techniques include Interpolation and Buffer strategies. 
 
Yes, for live data streams. Static file storage is another matter entirely, and use different error correction schemes. Storage systems use mathematically modeled error correction code systems, which essentially self-repair your files if anything goes wrong. Thus, zero quality is lost in file storage, barring catastrophic failure (i.e. drop your computer out a window, shoot it with a shotgun, bake it in an oven by mistake, those sorts of everyday things).
 
If we're talking about streaming across live cables, that's another matter and depends on the protocol. However, without any exceptions I can think of off hand, everything going on in your computer is 100% loss free. Hard drives are the most risky because they wear out fast / less deterministically, and yet even they remain effectively loss free for quite a long time. As to exactly how they achieve this, well, it's complicated - but it's really quite amazing that we've achieved loss-less storage with such astoundingly low probability of failure.
 
To give perspective on this, if just ONE bit of your operating system software (which is stored identically to music) is flipped from a 1 to a 0 or vice versa, it's highly likely that your computer will simply crash. It will simply stop working. This is not an exaggeration. Think that over for a while before you get the crazy idea in your head that computers simply lose data all over the place.
 
P.S. Buffering in of itself has nothing to do with error correction or prediction, so I'd advise you not use that in your argument, or just please admit you're not a computer engineer and are unsure of what you're talking about..
 
>  However, this does not guarantee integrity. 
 
Nothing is 100% guaranteed, however modern computers are so close to 100% we shouldn't even be having this conversation.
 
> If you have a poor lossless rip that doesn't add up it via checksum it will still sound like the song but at that point you now have a lossy file.
 
Wrong. Even MORE wrong if we're talking about lossless compression as opposed to a raw data stream (the corruption of a compressed file would literally trash the entire file). I don't think you realize how precise and well designed it is how computers compress files in "lossy" formats - it doesn't happen because a few bits are omitted here and there. It's a very mathematicaloperation that is carefully designed to transform sample points into an alternate domain which can extract the "important" parts and omit the "inaudible" ones, to simplify the idea. Again, it does not happen by itself. Lossless compression also a very interesting field, which itself is usually either mathematically based (entropy coding) or sometimes resembling a sort of "DNA" like encoding of files into expansion instruction bits.
 
And lastly, usually files won't even be openable if the "checksum" is incorrect. If you lose data, chances are the file will not open at all. If by some chance you can open it (which might be the case for uncompressed raw files), is will not sound just like the song with just some lost quality. It will sound like horrible screeching or at best, scratching and clicking.
 
I'm sorry, but your idea of how digital data works is wrong. I could explain to any degree of detail you like, but please read up on some basic computer architecture first.

 

[Anaxilus]

Quote:

Yes, for live data streams. Static file storage is another matter entirely,
 
So when a CD is being ripped and streamed to your HDD that's not a data stream?  M'kay.  Your PC architecture is different than the rest of ours it seems.
 
P.S. Buffering in of itself has nothing to do with error correction or prediction, so I'd advise you not use that in your argument, or just please admit you're not a computer engineer and are unsure.
 
I didn't say that did I?  Notice the word 'compensation'.  That is certainly what buffering does.  Compensates for latencies in streaming data, etc.  I admit I'm not a CS but I'm not illiterate on the matter either having studied EST and building computers as a pre-teen for the last few decades.  I'm literate enough to know that not just any error will crash a satellite into the Earth too.  I presume you aren't an Aerospace engineer and were unsure.
 
Nothing is 100% guaranteed, however modern computers are so close to 100% we shouldn't even be having this conversation.
 
Then I guess you can stop having it then.  Let the rest of us who have been burned multiple times in the past by '100% computing perfection' care about that final fraction of a tenth of a percent.  It's funny how often that minute fraction rears it's head.
 
Wrong. First of all, usually files won't even be openable if the checksum is incorrect. If you lose data, chances are the file will not open at all. If by some chance you can open it, is will not sound just like the song but with lost quality. I'm sorry, but that's not how digital data works. I could explain to any degree of detail you like, but please read up on some basic computer architecture.
 
For the most part you are right.  The file won't usually open.  But since computers are so '100%' as you claim why even check the data?  You seem to be another expert or 'professional' that has everything figured out by proxy it would seem.  Or are you a student?  Too bad all that supreme knowledge can't make a perfect OS free of issues or hardware that can even perform to basic specs 100% of the time.  If you want to rip your music w/o any verification either by data comparison or listening because you have faith in binary perfection go for it.  Next time you go have a listen to that file you ripped months ago but never verified at least you'll know why it went wrong.  Or perhaps you'll blame Cosmic rays or that cheap Seagate HDD you bought on sale.  And I do read monthly periodicals on PC hardware and have for decades as well as being educated in PASCAL and EST at university w/ certs as an A+ technician and MCP/MCSE.  

You have an interesting perspective.  Corruption happens all the time.  Just not to the level you notice it all the time.  At least nowadays.  It's not enough to just flip a random 1 and 0 and crash a system.  It matters where the one and zero are.  If the corruption lies in an area infrequently or never accessed it will not commit the error as it was never asked to execute the command.  If this is wrong feel free to enlighten me.  I'd like to see something fail to execute w/o a command of execution.
 

 

[Head Injury]

Quote:

Then I guess you can stop having it then.  Let the rest of us who have been burned multiple times in the past by '100% computing perfection' care about that final fraction of a tenth of a percent.  It's funny how often that minute fraction rears it's head.

You've had a ripping error which resulted in distortion from a CD which was not scratched? He's not talking about scratched CDs.
 
The only time I've encountered something like that was with a brand new "OK Computer" CD. For some reason the last track kept hitting an error at the end. EAC corrected it and it was verified by AccurateRip. There's no distortion.

 

[Anaxilus]

Quote:

You've had a ripping error which resulted in distortion from a CD which was not scratched? He's not talking about scratched CDs.
 
The only time I've encountered something like that was with a brand new "OK Computer" CD. For some reason the last track kept hitting an error at the end. EAC corrected it and it was verified by AccurateRip. There's no distortion.

Most cases were the song playing then stopping and causing the program to crash.  Only on those songs and w/o scratched discs.  Could have been any number of causes.  If and when you correct it yes it plays fine.  Granted most issues occurred years ago but you develop phobias based on past experiences.  I have an extremely low level of trust in engineers, manufacturers, just about everything.  Too many lies and too many profits out of my pocket. 
 
Ultimately just do whatever you are comfortable with.  
 

 

[ac500]

> You have an interesting perspective.
 
At risk of sounding harsh, this is as much a "perspective" as it is a "perspective" that santa clause is not real.
 
>  I have an extremely low level of trust in engineers, manufacturers, just about everything.  Too many lies and too many profits out of my pocket. 
 
That explains a lot. I'd love to know though, why don't you trust engineers? Literally, we are the people who build the computer you are typing on right now. Engineers aren't generally the ones to put out lies, blame the marketing / management for that. Engineers just make stuff that works. 
 
Think of it like this. Say you make a gourmet dinner, very complex. Someone says, "Hey it tastes like you have _____ in here." You know very well you didn't put that ingredient in, yet this ignorant person who knows nothing of the inner workings of the meal distrusts you "I don't trust that lying chef.". Don't you see how absurd that is?
 
> For the most part you are right.  The file won't usually open.  But since computers are so '100%' as you claim why even check the data? 
 
Data format is often checked so you don't accidentally open, for example, an image file, and interpret it as a sound file. That would sound terrible / screeching sounds. These guards are also in place in case, by chance some user inadvertently opens a sound file in a text editor and saves it (thus messing it up, obviously).
 
>  Next time you go have a listen to that file you ripped months ago but never verified at least you'll know why it went wrong.
 
You seriously think that a file left alone for a long time, or repeatedly copied around in a digital system, will deteriorate? I have no words to describe the ignorance of that. The whole point of digital is that it is lossless. All I can say, is ask any expert if your sound files will sound worse over time, and if you find one that doesn't break out laughing, thinking you're joking, I'll be impressed.
 
Also you questioned where I'm getting this info: I'm a professional computer engineer (I work at one one of the worlds best known chip companies; we make the processors that make your computer actually do stuff ). But seriously, it doesn't take a computer engineering degree to tell you that audio files don't degrade over time.
 
Listen, it's okay that you were mislead here, everyone get misinformed. I'm certainly guilty of many misunderstandings myself. But please, don't take this argument as an excuse to solidify your "opinion" in the face of the real truth that is out there. 
 
You mentioned you have some programming experience in Pascal. I'd recommend implementing an MP3 or Vorbis decoder some time, and you'll learn a lot (I did). At the very least, you'll see how I mean the quality can't simply degrade over time.
 

 

[Anaxilus]

Quote:

 
At risk of sounding harsh, this is as much a "perspective" as it is a "perspective" that santa clause is not real.
 
You didn't answer my question so I assume my statement was correct. 
 
That explains a lot. I'd love to know though, why don't you trust engineers? Literally, we are the people who build the computer you are typing on right now. Engineers aren't generally the ones to put out lies, blame the marketing / management for that. Engineers just make stuff that works.
 
Until it doesn't work.  If you can't even think of one failed engineering design or project that explains your 'perspective' ill-informed as it might be.  No offense but your view of engineering perfection is the fairy tale here.  Maybe engineering flaws are more prevalent in the Mech E world I'm more familiar than they are in your binary verse.  But it really doesn't matter at this point as we are way off topic and getting rather surreal.   
 
Think of it like this. Say you make a gourmet dinner, very complex. Someone says, "Hey it tastes like you have _____ in here." You know very well you didn't put that ingredient in, yet this ignorant person who knows nothing of the inner workings of the meal distrusts you "I don't trust that lying chef.". Don't you see how absurd that is?
 
Actually, how I see your argument is you are the Chef.  You make a meal that tastes like crap and blame the diner for not liking it.  But that can't be because you are the Chef.  You are the 'professional'.  What does this normal person know about food?  You see this in a lot in bad restaurants.  I would also recommend you stay away from the cooking analogies unless I'm not the only one w/ books from Robuchon and Ducasse on their shelves.     
 
You seriously think that a file left alone for a long time will deteriorate? I have no words to describe the ignorance of that.
 
Define 'left alone'.  I've experienced iTunes wiping my libraries at least twice in my lifetime.  Go ahead and explain ignorance to all the people on these threads.
 
All I can say, is as any expert if your sound files will sound worse over time, and if you find one that doesn't break out laughing, thinking you're joking, I'll be impressed.....But seriously, it doesn't take a computer engineering degree to tell you that audio files don't degrade over time.
 
I'm not sure where you get this sound quality degradation over time argument from.  I don't recall making that point but you seem to have imagined it somewhere.
 
Also you questioned where I'm getting this info: I'm a professional computer engineer (I work at one one of the worlds best known chip companies; we make the processors that make your computer actually do stuff ).
 
Like these?  Now you can talk about degradation.  
 
"The flaw means it may degrade with use over a period of months or years, slowing down the transfer of data to and from the computer's hard drives and DVD drives."
 
How about these?  Or these?  Well, they've improved them w/ these.  Not to pick on Apple too much but they are ever present and much lauded.  Every company has ENGINEERING problems because Engineers and Scientists are humans and not gods.    
 
Listen, it's okay that you were mislead here, everyone get misinformed. I'm certainly guilty of many misunderstandings myself. But please, don't take this argument as an excuse to solidify your "opinion" in the face of the real truth that is out there.
 
What exactly was your overarching universal truth again?   

 
 

 

[mralexosborn]

Whatever sounds good. The end. **** all of you audiophiles. XD

 

[koolkat]

Unless you're the type who torrents them, or downloads them. Sometimes they convert files from aac or mp3 to flac, gotta look out for those.
 
Quote:

Good grief, I thought I was paranoid. 
To the OP, grab whatever lossless file format takes your fancy and convert your music. If you find another lossless file format you'd prefer later, convert it into that from your lossless files.
To clarify koolkat, lossless compression always retains all of the details of the original file.

 

 

[Redcarmoose]

Buy CDs.

 

[ac500]

Anaxilus: I really don't have the time to correct each and everything you say. Doing so would would take pages, so don't mistake my not addressing one of your statements as me conceding to it (quite the contrary).
 
I'm not going to continue this debate though, it's about as productive as arguing that the world is semi-spherical to a flat-worlder who doesn't believe science. Like I said, ask any expert if you don't believe me. Beyond that if you still think those 'magic computers' built by those 'lying engineers' lose audio quality in files over time, then you're hopeless; at some point when it becomes willfully ignorant person vs. the world of experts, it truly is hopeless stubbornness. 

 

[Anaxilus]

Quote:

Beyond that if you still think those 'magic computers' built by those 'lying engineers' lose audio quality in files over time, then you're hopeless. At some point when it becomes willfully ignorant person vs. the world of experts, it truly is hopeless stubbornness. 

Thanks for not reading anything I said.  Cheers.  Sorry to have kept you from Moore's Law.
 

 

[ac500]

If you agree that audio files on your computer do not degrade at all when transferred around your computer digitally and over time, then we have no disagreement.

 

[khaos974]

You should just encode everything in FLAC, am md5 and a cheksum number is calculated for the raw pcm data during the encoding, checking for integrity afterwards is as trivial as right click > verify integrity.

@ac500, most of the time a corrupted audiofile will open without major issues, audio is separated in small chunks before encoding anyway, if the corrupted bit is not in the header, there just will be a something odd at the specific chunk in question.