[Skarecrow77]

Quote:

The chances of you losing more than one drive at a time are 1 in a number I can't even fathom.  It's not a back up really, but it is a very safe method of storage.  

Actually, the chances are a lot higher than you think.
 
The MTBF rating of a drive is not determined by taking a couple drives, running them until they fail, and then taking the average of how long it took for them to fail. No, they'll take an absurd number of drives, run them for a short period of time (relativly speaking), and however many failed in that batch is factored against how many total "drive hours" were run for the entire batch, and the MTBF is calculated for that.
 
What does that mean? It means that drive failures are not random and you can't take the MTBF as an estimate of how long your batch of drives will last. Stuff wears out. Especially stuff under heavy use. Drives from the same batch VERY often die en mass in a short period of time. Since most people don't build raid arrays with drives they purchased months apart, but instead buy 3-5 drive simultaniously and then build from that all at once, it means that once one drive goes, you've got a fairly short period of time (months if you're lucky, weeks most of the time, days or even hours if you're unlucky) before another drive from the same batch fails. If you buy 4 tires at the same time, and drive until one of them bursts from normal use (as opposed to a manufacturing defect), what are the chances the other 3 are in good shape? very slim.
 
To give you an example. I've got a server in the server room with 16 drives in it in multiple RAID arrays (Three raid 6s and a raid 5). I've had two fail in the past 2 months. It's not uncommon at all, and is actually quite a well known problem.
 
RAID arrays buy you time before complete data loss, but just time. Take it from me. Once that first drive goes, don't just replace it, get a spare. In fact, if you can swing it, get a spare -now- so it's not from the same batch as the next spare you buy.

 

[ramicio]

Take it from me, too.  I bought 2x 300 GB drives at the same time and RAID 0'd them and they both started having errors around the same time.  Sucks it's all I've got and some of my shows now have missing chunks.

 

[EddieE]

Different rigs?
 
Maybe, but I have encountered people online with some seriously expensive kit and a lot of experience state there is no audible difference expect when compression errors are themselves further compressed. So maybe, but I have reason to doubt it.
 
I am more bothered by when you say people have "different ears".  You are continuing to reaffirm a theme expressed by many others on this thread who have said they can hear a difference - the notion that if you can hear a difference between FLAC and LAME320 it is because they are superior in their ability to listen to and appreciate music.
 
Now while there is actually no proof for this idea, there is plenty of proof, in journals left right and centre, that when you put a concept like that into play (and is is very much at play at this forum) that people will want to hear a difference and therefore people will convince themselves there is a difference.
 
Were you never told the story of the Emporer's New Clothes as a child?
 
Every time someone says any of the things which have been said on this thread by non-MP3 users it puts pressure on people to be able to hear:
 
"If you can't hear the difference between mp3 of any bit rate and FLAC I would suggest that you are probably deaf and wasting your money on expensive headphones"
 
"With all the bands I know to have good sound, it's easy to tell the difference between lossy and lossless files." 
 
"I can't see an excuse for using MP3 other than for use with a small portable.  Ripping CD's in MP3 makes zero sense."
 
Really man, you could put a guy in a room and ask him if he likes music. When he says yes, say "great" and then tell him he's here for a test and its a test to see how well versed people are at appreciating and understanding the qualities in music. Explain to him there are two versions of the same track - (although they are in fact the exact same file) - and tell him the first is compressed and the second is lossless. He goes into a room full of actors playing the parts of other testees and a guy with a clip board tells them they'll hear it three times and just to discuss what they thought after each. The tracks get played through high end speakers and all the actors enthusiastically agree the second sounds better - even if they guy didn't think it the first time the second time he'd think he did and the third he'd be joining in with all the actors on the "shimmering highs" and "tight punchy lows" he can definitely hear.
 
He wouldn't be lying either - he'd be actually hearing it - that's the way perception works. From everything I have read about other similar tests that would work 100% of the time.
 
Now this is true even if a difference can be heard with high end rigs. Even if there is actually is an audible difference between LAME320 and FLAC - that test would still work.
 
So that does shift the burden of proof somewhat.
 
So again, I am not in any way saying I am definitely right, and am open to being proved wrong -
 
But people writing in and anecdoatal manner about how "of course!" they can hear the differences means absolutely nothing to me, and would never play any part in my decision making on this issue. 
 
Show me the tests.

 

[ramicio]

No one cares about your hippie stories and your thoughts that everyones ears are those of old men.  How are you bothered by people having different ears?  Every single bit about the ear makes sound different, even the outside, having earrings or not.  It's like saying everyone has the same sense of taste because everyone has a tongue.

 

[EddieE]

It's not a hippy story mate, it's basic psychology.
 
Whether or not there are differences between FLAC and LAME320  - the way the ability to hear that alleged difference is held up as a sign of superiority here - people would convince themselves they heard one even if they didn't.
 
That means I'm not going to listen to anecdotes to make my mind up on this one - only actual test results.
 
And human ears are all basically the same, different people have different amounts of damage yeah but they all work the same (my ears are fine by the way).
 
When people say "people have different ears" they mean in metaphorically - its an expression to say "everyone has different tastes in what sounds good and bad" to them. 

 

[Skylab]

That's not entirely true.  It is absolutely the case that you can learn to hear things that you might not hear without being taught.

 

[QbertEnhanced]

I'll come right out and say it
On a $16k pair of Canton Reference 3.2 speakers, I can't hear a darned difference between 320kbps and FLAC.
And I'm happy for it!  I have my CD's for archive, no need to archive digitally as well.

 

[ramicio]

No, I mean literally, everyone has different ears.  Does every single person's ear consist of the same number or atoms, cells, what have you?  Are the internals and externals EXACTLY the same shape?  NO.  Everyone has different voices because everything that makes a voice such as vocal cords, throat shapes and sizes, and every single cavity in the head, makes a voice unique.  So the same would happen with hearing.

 

[VulgarDisplay]

Quote:

 
Quote:

The chances of you losing more than one drive at a time are 1 in a number I can't even fathom.  It's not a back up really, but it is a very safe method of storage.  

The data is safe if you are only concerned about losing data due to a drive failure.  The data is not safe if you consider the many other ways data can be lost.  How do you recover if you run a buggy media player that corrupts the tags in every audio file it loads?  How do you recover from user error that manages to damage or corrupt a file?  How do you recover if an electrical surge wipes out the entire RAID array or the cat knocks the RAID box over and crashes several drives?  Not to mention fire or theft.
 
RAID is a very poor backup.

Anything can happen.  I could drop dead right now and I wouldn't be worried about my music being backed up.  Nothing is certain in life.  RAID just buys you time to save your data which is what it is designed for.  
 
As for people saying that buying HD's at the same time will result in their failure within a short amount of time I'll believe it when I see it.  I've been building PC's for the last 10 years and I haven't had a single HD fail yet.  I will eventually be moving on to better and faster storage media like flash drives soon, and then my track record with HD's will have been perfect for the time I used them.  With any electronic there is a chance for random failure.  The chances of 2 HD's failing at the same time, or before you can replace the damaged disk are extraordinarily low.  You could say that some catastrophic event could destroy the data, but some catastrophe could just as easily destroy any other back up you have too. 

 

[Deep Funk]

When you know certain tracks and albums good enough you will hear differences, subtle though. I have the same thing with 320 kbps WMAs and CDs. It sometimes depends on the headphones, otherwise the set up. Anyway, for the Clip + I am going for FLAC from now on. 
 
Ramicio, your argument may be valid but will lead us to hairsplitting arguments. I am out for I have to do some FLAC-testing.

 

[EddieE]

Skylab,
When a person, any person, does an ABX test it is an incredibly focussed experience with one goal - to discern a difference between two versions of identical passages of music.
 
This is completely outside of the realm of training your hearing to notice details - if there is a difference every human being will hear it, unless their hearing is significantly damaged.
 
Perhaps the untrained ear will think the compressed track sounds better and the lossless worse, who knows, but you cannot train your hearing to actually become physically better and all healthy ears pick up the same vibrations.
 
Maybe someone is going to argue the differences lie in those extremes at the borders of hearing where some people's frequency range goes further than others - well the borders of our frequency range tend to be so weak you need an isolated environment to hear anything there, so any music would make that unhearable and irrelevant.
 
Anyway, there's no point getting sucked in so I'll leave you guys to it.

 

[Ham Sandwich]

 
Quote:

Anything can happen.  I could drop dead right now and I wouldn't be worried about my music being backed up.  Nothing is certain in life.  RAID just buys you time to save your data which is what it is designed for.  
 
As for people saying that buying HD's at the same time will result in their failure within a short amount of time I'll believe it when I see it.  I've been building PC's for the last 10 years and I haven't had a single HD fail yet.  I will eventually be moving on to better and faster storage media like flash drives soon, and then my track record with HD's will have been perfect for the time I used them.  With any electronic there is a chance for random failure.  The chances of 2 HD's failing at the same time, or before you can replace the damaged disk are extraordinarily low.  You could say that some catastrophic event could destroy the data, but some catastrophe could just as easily destroy any other back up you have too. 

RAID doesn't always buy you time.  There have been cases of firmware bugs in drives that would cause corruption if the drives were used in a RAID setup.  If you were unfortunate enough to have used those drives in a RAID you would find your data would corrupt eventually and not be recoverable.  RAID is not as reliable as you think it is.
 
The way to protect against catastrophe is to keep an off-site backup.  Friends and family are strangely very happy to keep a backup of your music files.  Then you don't have to worry about fire, accidents, user error or other catastrophe.
 
I've invested a lot of time in ripping, tagging, finding cover art, and other details to risk losing it.  Along with digital photos and scans of old family photos.

 

[Skylab]

EddieE, you can train yourself to hear coding artifacts. That will impact one persons abilities to discern a lossy versus lossless codec in an ABX test.

 

[pp312]

Just putting aside whether people can or can't hear differences, there's an argument that says, "If you've got a choice, why not choose lossless anyway, since storage is cheap?" Good question, but many of us don't listen at our computers; we want to transfer our files to other playback devices in other parts of the house. I use an LG DVD recorder with a hard drive and USB input set up with my stereo system to which I transfer all my files, but it only accepts MP3 or WMA. Linked to a portable DVD player screen it works great, so long as I'm satisfied with MP3, and I am (with the exception that MP3 puts a gap where a new track is listed but there's no actual gap--my biggest complaint about MP3!). I was just looking at a very nice full size media player on Ebay, retail $995, but it only converts to MP3. Away from the computer MP3 is often the only choice.
 
Oh, and Waiter! I'll have eggontoast with a Ham Sandwich, thanks, and no Vulgar Display.

 

[Kibble Fat]

I was thinking about doing a RAID1 setup.  I would keep one at home safe and secure on my desk and take the other with me on trips, to school, etc.  I thought the point of having such a setup was that as soon as one died, you could quickly acquire a new drive and make a copy: no harm done.  The probability of the other drive failing in such a short period of time (1-3 days) would be almost zero.