|
|
Posted by PTravel on 10/05/27 11:52
"Richard Crowley" <richard.7.crowley@intel.com> wrote in message
news:e8tvfj$1mb$1@news01.intel.com...
> "PTravel" wrote ...
>> "Martin Heffels" wrote ...
>>> Data drop-out is the equivalent of generation-loss in digital
>>> duplication.
>>
>> No, it isn't. Generational loss is inevitable and progressive when
>> duping analog video. Drop out is caused by physical failure of the tape
>> and is neither inevitable nor likely for D-25.
>
> Small, correctable, dropouts likely occur more frequently in
> D-25 than you think. But since they are automatically detected
> and corrected, you don't know they are happening. The only
> time we *see* it is when it gets so bad, the ECC can't handle it.
But that's the whole point -- if it's corrected, it's not a drop out, and
there's no data loss. If a particular section of tape exhibits flaking, but
not severe enough so that the ECC can handle it, there's no drop out, and a
copy made from that tape will be a bit-for-bit copy of what's supposed to be
on the tape. There's no generational loss, because, by definition, the data
was corrected.
We're talking apples and oranges here. On the one hand, there's everyone's
DV deck that can invent data if it's missing, and on the other hand there's
ECC which can _repair_ data, i.e. restore it to its original state, if
there's an error. The former will result in generational loss (to the
extent that it happens). The latter will not.
>
>> This thread appears to have been an old one that someone decided to
>> resurrect.
>
> It appears to have started only two days ago. (8-Jul)
I must have missed it, then.
>
>> The question appears to be, "Is there generational loss when duping
>> digital video?" The answer is, "No," not, "Yes if you do it 100 times,
>> as another poster has claimed."
>
> Well, there is *NO* generational loss when copying a digital
> data.
That's what I said. The OP made a later post in which he was talking about
the relative loss between, IIRC, SuperBeta and digital. SuperBeta, as good
as it is, will result in generational loss because it's analog -- the loss
isn't the result of any failure in the video chain, but of the inherent
imprecision in trying to match analog levels. To the extent that there is
loss in digital video, it is the result of failure somewhere in the
process -- either flaking off the binder, or Martin's flying particles and
interference.
> However some digital media have less robust ECC than
> others. Data media (hard drives, floppies, computer backup
> tape, CD-ROM, DVD-ROM, etc.) use very robust ECC
> mechanisms which ensure 99.999999% dependable bit-for bit
> reliability. OTOH, digital audio/video media formats (like audio
> CDs, DVDs, DV tape, etc.) use less robust ECC because a
> few bits of error here or there don't mean the same thing in an
> audio or video signal that they do in a computer executable file
> or your bank's accounting data.
>
> Technicaly, you could make a valid argument for either "yes"
> or "no" as an answer to answer this question.
The big difference, though, whereas generational loss is inevitable with
analog, loss due to drop outs or other uncorrectable errors is not
inevitable with digital.
>
>> There is no recompression at all when making a digital dupe -- whether
>> D-25, mpeg, divx or whatever,
>
> Absolutely, and one of the huge advantages of digital over
> analog.
>
>> as when any digital data, you can copy the file ad infinitum with no loss
>> whatsoever.
>
> Not as valid to try to apply computer-type data handling
> reliability to digital audio/video media. They use different
> levels of ECC for their different requirements and their
> different economic markets, etc.
Okay, this gets into statistics. If you agree that, for a given digital
tape, drop out is not inevitable, then it is solely a question of the
uncorrectable error rate for a particular digital tape medium which, I would
think, is incredibly low. If you can copy a 60 minute tape to a computer
and then back to another tape without an error, it is no more likely that,
if you repeat the cycle, an error will be introduced, i.e. the statisical
likelihood that you'll experience data loss, whether it's one dupe or 100,
remains the same (just like the roulette wheel that has hit red 20 times in
a row -- it is not more likely to hit black than red on the 21st spin).
>
>> By "loss," I assume you may "drop out." Because drop out is caused by a
>> physical failure of the tape, you're not going to lose just one pixel on
>> one frame -- drop out on digital tape is rather obvious because it effecs
>> a block of data.
>
> But frequently that block of data can be completely and
> accurately reconstituted by the ECC. This likely happens
> hundreds of times an hour in what we think are "perfect"
> DV-25 tape playbacks.
And if it is completely and accurately reconstituted by the ECC, then there
has been no data loss, and the copy made from the tape is bit-for-bit
accurate.
>
>> I use an old TRV-20 consumer camcorder for transferring my miniDV to my
>> computer (I fried the 1394 port on my VX2000 a couple of yearas ago).
>> This camera puts out what is one the tape, and doesn't detect or fix drop
>> outs.
>
> It most certainly does both detect and fix dropouts.
> Else it wouldn't even work.
Again, we need to define terms. My camera does ECC (as does any D-25
device). It does not synthesize video data from prior frames to make a drop
out look unnoticeable, as due the digital decks that other posters have
described.
> I suspect that all of you who
> think that digital is so pristine would be horrified if you had
> a peek "under the hood". :-) Take a look at a service manual
> for any DV camcorder if you need to be reminded of the
> analog nature of tape (even digital tape).
I don't think it's pristine. I do think that the ECC for D-25 is
sufficiently robust so that, by using quality tapes only once, and
maintaining my camcorder, I can produce digital transfers to and from the
computer without any dataloss.
This was a discussion about generational data loss, not error correcting
algorithms.
>
> I work side-by side with hundreds of engineers who design the
> most advanced and complex digital chips in our known universe.
> But almost none of them think in the "digital" domain. Their jobs
> involve dealing with the real analog world and making it look
> "digital" at the interface where the pins connect to the motherboard.
> The people who design magnetic and optical digital media have
> the same task of making the fundamentaly analog world behave
> like "digital".
>
>> Sorry, you're wrong. Hard disks use ECC, just like tape.
>
> They both use ECC, but the ECC used on hard drives is much
> more robust and reliable than what is used on DV tape. (And it
> takes more space away from storage the capacity).
It also has to be more robust because (1) hard drives are constantly
re-written, (2) data is packed more closely on hard drives, and (3) data is
read more quickly on hard drives. The ECC on miniDV is sufficiently robust
that the likelihood of losing any data on a transfer is very, very low.
> The expectation
> for data is that you will get one uncorrectable error in several
> trillion bits whereas most of us likely have seen and/or heard
> uncorrectable errors in digital audio/video media in recent memory.
As I said, they happen. Just not often enough so that they should be
considered a source of generational loss.
>
>
Navigation:
[Reply to this message]
|