Going from VHS to DVD is a waste of time!!!!!

[Guest]

Archived from groups: rec.video.desktop (More info?)

Going from VHS to DVD is a waste of time! I just recently realized it. Just like many of you,
I'd spend hours and days trying to convert some VHS footage to DVD. But the end result
was always worse than original, no matter what you'd do. The idea is to have the end result
identical to source. It will never happen. Why? Just think about it: first you digitize the video.
When you digitize the video, there's a certain degree of degradation. Then whatever codec
you use Huffyuv or DV would compress your footage. And then you further compress it
to DVD compliant MPEG 2 file. Now the most interesting thing is when you have your DVD
ready. You play your DVD and you convert the digital signal *BACK* to analog to display
on TV. So basically what you have is Analog->Digital->Compression->Compression->Analog.
And this is just as some claim to preserve a footage 'cause VHS deteriorates. But VHS only
deteriorates when you use it. When you just keep them in cool storage, nothing will happen
to them. So I found a good solution. Just copy VHS to another VHS or S-VHS. You have the
master copy which you store in your storage. To playback just use the 1st generation copy. Yes
there's a quality loss, but it is a tiny one. You don't go analog->digital->analog. You just go
analog->analog. The results are much better. I don't know about you, but I'm off DVDR market.

--Leonid

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

Leonid Makarovsky wrote:
> Going from VHS to DVD is a waste of time! I just recently realized it. Just like many of you,
> I'd spend hours and days trying to convert some VHS footage to DVD. But the end result
> was always worse than original, no matter what you'd do. The idea is to have the end result
> identical to source. It will never happen. Why? Just think about it: first you digitize the video.
> When you digitize the video, there's a certain degree of degradation. Then whatever codec
> you use Huffyuv or DV would compress your footage. And then you further compress it
> to DVD compliant MPEG 2 file. Now the most interesting thing is when you have your DVD
> ready. You play your DVD and you convert the digital signal *BACK* to analog to display
> on TV. So basically what you have is Analog->Digital->Compression->Compression->Analog.
> And this is just as some claim to preserve a footage 'cause VHS deteriorates. But VHS only
> deteriorates when you use it. When you just keep them in cool storage, nothing will happen
> to them. So I found a good solution. Just copy VHS to another VHS or S-VHS. You have the
> master copy which you store in your storage. To playback just use the 1st generation copy. Yes
> there's a quality loss, but it is a tiny one. You don't go analog->digital->analog. You just go
> analog->analog. The results are much better. I don't know about you, but I'm off DVDR market.
>
> --Leonid

All good points Leonid,

There's an old saying, "Garbage in, Garbage out". However, there's some
trouble with your theories.

1) Yes you digitize it, but you'll get an exact copy of how it plays
right this minute. And yes that will be about as good the original
copy.. However, I personally have a Canopus-300 which actually cleans up
some of the shortcomings of the video signal, plus once you digitize
your footage, and can run a few filters and a little tweaking on color
corrections, and you can actually end up with a better copy of that
footage. I capture with a very high data rate (not DV) in a near
uncompressed format. The losses from this are indistinguishable to the
human eye, if any.

2) I don't know if you've been to the local Best Buy lately, but it's
getting more and more difficult to find a single VHS tape player these
days.. At the current rate the VHS player will go the rate of the
turntable by next year. So while yes, you might have that perfect copy
of the Master tape, you'll not find a way to every play it again in a
few short years.

3) In another theory, A final step of outputting to an analog TV will
also be a thing of the past, as we look forward to HDTV. So your theory
here is also a moot point too.

But, yes.. True that the original would have been much better off going
straight to digital DV, but I think the whole point is not to try to fix
what was done in some kind of remaster, but more like to create
something that will allow you be able to play it at all years down the line.

-Richard

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

Interesting discussion. I am beginning to think about converting some of my
most precious VHS tapes. As a start should I copy them to MiniDV maybe via
a bit of editing on the PC and keep a copy on DV tape? I don't have a DVD
writer yet. That's another consideration - do I buy a burner for the PC or
an under TV recorder? All these things to addle the brain!!

Cheers

Margaret

Remove giggling if replying by email
"Richard Ragon" <bsema04NOSPAM@hanaho.com> wrote in message
news:QP7bc.12134809$Id.2029878@news.easynews.com...
> Leonid Makarovsky wrote:
> > Going from VHS to DVD is a waste of time! I just recently realized it.
Just like many of you,
> > I'd spend hours and days trying to convert some VHS footage to DVD. But
the end result
> > was always worse than original, no matter what you'd do. The idea is to
have the end result
> > identical to source. It will never happen. Why? Just think about it:
first you digitize the video.
> > When you digitize the video, there's a certain degree of degradation.
Then whatever codec
> > you use Huffyuv or DV would compress your footage. And then you further
compress it
> > to DVD compliant MPEG 2 file. Now the most interesting thing is when you
have your DVD
> > ready. You play your DVD and you convert the digital signal *BACK* to
analog to display
> > on TV. So basically what you have is
Analog->Digital->Compression->Compression->Analog.
> > And this is just as some claim to preserve a footage 'cause VHS
deteriorates. But VHS only
> > deteriorates when you use it. When you just keep them in cool storage,
nothing will happen
> > to them. So I found a good solution. Just copy VHS to another VHS or
S-VHS. You have the
> > master copy which you store in your storage. To playback just use the
1st generation copy. Yes
> > there's a quality loss, but it is a tiny one. You don't go
analog->digital->analog. You just go
> > analog->analog. The results are much better. I don't know about you, but
I'm off DVDR market.
> >
> > --Leonid
>
> All good points Leonid,
>
> There's an old saying, "Garbage in, Garbage out". However, there's some
> trouble with your theories.
>
> 1) Yes you digitize it, but you'll get an exact copy of how it plays
> right this minute. And yes that will be about as good the original
> copy.. However, I personally have a Canopus-300 which actually cleans up
> some of the shortcomings of the video signal, plus once you digitize
> your footage, and can run a few filters and a little tweaking on color
> corrections, and you can actually end up with a better copy of that
> footage. I capture with a very high data rate (not DV) in a near
> uncompressed format. The losses from this are indistinguishable to the
> human eye, if any.
>
> 2) I don't know if you've been to the local Best Buy lately, but it's
> getting more and more difficult to find a single VHS tape player these
> days.. At the current rate the VHS player will go the rate of the
> turntable by next year. So while yes, you might have that perfect copy
> of the Master tape, you'll not find a way to every play it again in a
> few short years.
>
> 3) In another theory, A final step of outputting to an analog TV will
> also be a thing of the past, as we look forward to HDTV. So your theory
> here is also a moot point too.
>
> But, yes.. True that the original would have been much better off going
> straight to digital DV, but I think the whole point is not to try to fix
> what was done in some kind of remaster, but more like to create
> something that will allow you be able to play it at all years down the
line.
>
> -Richard
>

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

On 2 Apr 2004 03:27:36 GMT, Leonid Makarovsky <venom@cs.bu.edu> wrote:

>Going from VHS to DVD is a waste of time! I just recently realized it. Just like many of you,
>I'd spend hours and days trying to convert some VHS footage to DVD. But the end result
>was always worse than original, no matter what you'd do.

The end result of ANY analog copy is always worse than the original.
It can be a big difference, or a very small one, but that IS
guaranteed.

>Why? Just think about it: first you digitize the video.
>When you digitize the video, there's a certain degree of degradation.

This, however, can be very, very minimal, if you do it right. First
rule: capture uncompressed. Analog video is inherently noisy, and
noise is a bad thing to retain when you're using typical DCT-based
compression algorithms.

>Then whatever codec
>you use Huffyuv or DV would compress your footage.

If you use Huffyuv, though, you don't degrade the video any further.
That's a lossless compression.

>And then you further compress it to DVD compliant MPEG 2 file.

This is where you're missing what's critical in this process: kill all
the noise. You need to filter the analog noise, as best as possible.
There about 1000 such filters that run under VirtualDub. Many work
very nicely, eliminating the VHS tape noise well without causing
visible damage to the VHS image.

If you don't eliminate the noise, the MPEG-2 compression algorithms
will do their best to keep that noise visible, at the expense of some
of the actual content you want.

Another thing for VHS conversion: the bandwidth of VHS is so low,
you're really no worse off capturing to 352x480/576 vs. 720x480/576.
That effectively doubles the per-frame "bit budget" for I-Frames --
less compression means less artifacting

>Now the most interesting thing is when you have your DVD
>ready. You play your DVD and you convert the digital signal *BACK* to analog to display
>on TV. So basically what you have is Analog->Digital->Compression->Compression->Analog.

No one cares about compression in the toolchain, it makes no
difference. LOSSY compression is what you worry about. You really
don't want to capture VHS in DV, since you can't filter the VHS before
the compression is applied. Grab it directly, uncompressed. Filter it.
Only then should you encode it.

>And this is just as some claim to preserve a footage 'cause VHS deteriorates. But VHS only
>deteriorates when you use it.

No, that's not even remotely true. Tape degrades just sitting there.
And being an analog medium, the quality will degrade a bit, just
sitting there. Eventually, you'll find dropouts that muck up sync,
causing the tape to be increasingly less playable. I've done
restorations of just such tapes, only about 8 years old at the time,
but already starting to degrade.

>When you just keep them in cool storage, nothing will happen
>to them.

Also not true. Storing them properly (moderately cool, moderately
humid, stored on end, not flat) will prolog their life some, but they
won't last indefinitely. Or even remotely as long as properly stored
DVDs.

> So I found a good solution. Just copy VHS to another VHS or S-VHS.

You're going to lose quality there, as well. Worse yet, if you don't
run a TBC between the two decks, you're likely to have sync errors
too. Not that backups of any kind aren't better than no backups. But
tape copies are worse than DVD copies.

When I put my movie on DVD (restored from the original 8mm and the
SVHS master), my DVDs were actually much, much better in quality than
our original, first generation VHS copies. You lose tons of quality in
each new VHS generation, it's not a trivial amount.

>You have the
>master copy which you store in your storage. To playback just use the 1st generation copy. Yes
>there's a quality loss, but it is a tiny one. You don't go analog->digital->analog.

If you know what you're doing, making a DVD from a VHS will yield a
very satisfactory result. Like everything else with DVD, if you don't,
it won't -- there's still a bit of art and engineering to the process,
it's not a cookie-cutter thing yet.

Dave Haynie | Chief Toady, Frog Pond Media Consulting
dhaynie@jersey.net| Take Back Freedom! Bush no more in 2004!
"Deathbed Vigil" now on DVD! See http://www.frogpondmedia.com

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

> If you know what you're doing, making a DVD from a VHS will yield a
> very satisfactory result. Like everything else with DVD, if you don't,
> it won't -- there's still a bit of art and engineering to the process,
> it's not a cookie-cutter thing yet.

And that's the bottom line, right there. It won't be perfect but how
many people will _actually_ be able to tell the difference between the
DVD copy and the original VHS/Beta tape?

Reminds me of mp3s several years ago. The ng had several "golden ear"
participants who blathered on about "quality". A listening test was
devised to see if _anyone_ could tell the difference using different
encoders and bitrates. Out of several _hundred_ participants, there was
only one guy who could verifiably tell a 128 from a 320 encode. Almost
none of the "golden ear" crowd participated, not surprisingly. Unless
you were an experienced listener and knew what to listen for, you just
_couldn't_ tell the difference. I posted an original 160 encode of a
song and then re-encoded it... oh about 10 times. That's lossy
compression X 10. No one could tell which one was the original.

I've got a lot of irreplaceable family footage on 8mm and VHS tape that
I'm just getting ready to convert. Am I expecting perfection? Not at
all. But I'll feel a hell of a lot better, and will have a chance of
actually viewing the footage, in 10 years when I've got them archived on
DVD rather than a vhs tape.

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

Dave Haynie wrote:
>
> On 2 Apr 2004 03:27:36 GMT, Leonid Makarovsky <venom@cs.bu.edu> wrote:
>
> >Going from VHS to DVD is a waste of time! I just recently realized it. Just like many of you,
> >I'd spend hours and days trying to convert some VHS footage to DVD. But the end result
> >was always worse than original, no matter what you'd do.
>
> The end result of ANY analog copy is always worse than the original.
> It can be a big difference, or a very small one, but that IS
> guaranteed.

Of course, the main point is that the quality of the vhs degrades
with time. That's the main reason to story it digitally.

> >Why? Just think about it: first you digitize the video.
> >When you digitize the video, there's a certain degree of degradation.
>
> This, however, can be very, very minimal, if you do it right. First
> rule: capture uncompressed. Analog video is inherently noisy, and
> noise is a bad thing to retain when you're using typical DCT-based
> compression algorithms.

Good luck capturing uncompressed! Better use Huffyuv (like stated
below) or mjpeg (lossy).

> >Then whatever codec
> >you use Huffyuv or DV would compress your footage.
>
> If you use Huffyuv, though, you don't degrade the video any further.
> That's a lossless compression.
>
> >And then you further compress it to DVD compliant MPEG 2 file.
>
> This is where you're missing what's critical in this process: kill all
> the noise. You need to filter the analog noise, as best as possible.

Yes, that's very important. Especially when encoding to mpeg2.

> There about 1000 such filters that run under VirtualDub. Many work
> very nicely, eliminating the VHS tape noise well without causing
> visible damage to the VHS image.

This is very false. Depends a bit on the source. If your source
is sports, music clips, etc, there's in fact no good filter which
runs under VirtualDub and can filter the noise out without
blending and smoothing too much.

You might want to look at AviSynth instead.

> If you don't eliminate the noise, the MPEG-2 compression algorithms
> will do their best to keep that noise visible, at the expense of some
> of the actual content you want.

Yes, and you will see that very clearly.

> Another thing for VHS conversion: the bandwidth of VHS is so low,
> you're really no worse off capturing to 352x480/576 vs. 720x480/576.
> That effectively doubles the per-frame "bit budget" for I-Frames --
> less compression means less artifacting

Partly false. The bandwidth is indeed very low. But, when capping
with bt8x8/cx2388x some lousy resizers kick in when capping at 384x576
(360x480 for NTSC) or lower.

Note that your capture size "is" not the used sample rate of the card.
The sample rate (using by the card) is always fixed and well above
full PAL/NTSC, after sampling the signal is resampled at the requested
capture size.

> >Now the most interesting thing is when you have your DVD
> >ready. You play your DVD and you convert the digital signal *BACK* to analog to display
> >on TV. So basically what you have is Analog->Digital->Compression->Compression->Analog.
>
> No one cares about compression in the toolchain, it makes no
> difference. LOSSY compression is what you worry about. You really
> don't want to capture VHS in DV, since you can't filter the VHS before
> the compression is applied. Grab it directly, uncompressed. Filter it.
> Only then should you encode it.

You never captured anything, did you? Or do you have a hdd of 1 TB?

> >And this is just as some claim to preserve a footage 'cause VHS deteriorates. But VHS only
> >deteriorates when you use it.
>
> No, that's not even remotely true. Tape degrades just sitting there.
> And being an analog medium, the quality will degrade a bit, just
> sitting there. Eventually, you'll find dropouts that muck up sync,
> causing the tape to be increasingly less playable. I've done
> restorations of just such tapes, only about 8 years old at the time,
> but already starting to degrade.

Yes, true.

Wilbert

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

"Leonid Makarovsky" <venom@cs.bu.edu> wrote in message
news:c4imj8$l92$1@news3.bu.edu...
> Going from VHS to DVD is a waste of time! I just recently realized it.
Just like many of you,
> I'd spend hours and days trying to convert some VHS footage to DVD. But
the end result
> was always worse than original, no matter what you'd do. The idea is to
have the end result
> identical to source. It will never happen. Why? Just think about it: first
you digitize the video.
> When you digitize the video, there's a certain degree of degradation. Then
whatever codec
> you use Huffyuv or DV would compress your footage. And then you further
compress it
> to DVD compliant MPEG 2 file. Now the most interesting thing is when you
have your DVD
> ready. You play your DVD and you convert the digital signal *BACK* to
analog to display
> on TV. So basically what you have is
Analog->Digital->Compression->Compression->Analog.
> And this is just as some claim to preserve a footage 'cause VHS
deteriorates. But VHS only
> deteriorates when you use it. When you just keep them in cool storage,
nothing will happen
> to them. So I found a good solution. Just copy VHS to another VHS or
S-VHS. You have the
> master copy which you store in your storage. To playback just use the 1st
generation copy. Yes
> there's a quality loss, but it is a tiny one. You don't go
analog->digital->analog. You just go
> analog->analog. The results are much better. I don't know about you, but
I'm off DVDR market.

Just go get a LiteOn standalone DVD recorder and you won't be able to tell
the difference between the original tape and the DVD. Then you can dup the
DVD for your family and friends on your computer. Now you can view any part
of your tape nearly instantly by jumping directly to the chapter. Random
access is probably the most important feature that DVD has over VHS, because
without it you probably won't even watch all those old videos.

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

On Fri, 02 Apr 2004 13:05:16 +0200, Wilbert Dijkhof
<w.j.dijkhof@tue.nl> wrote:

>Dave Haynie wrote:
>>
>> On 2 Apr 2004 03:27:36 GMT, Leonid Makarovsky <venom@cs.bu.edu> wrote:
>>
>> >Going from VHS to DVD is a waste of time! I just recently realized it. Just like many of you,
>> >I'd spend hours and days trying to convert some VHS footage to DVD. But the end result
>> >was always worse than original, no matter what you'd do.

>> This, however, can be very, very minimal, if you do it right. First
>> rule: capture uncompressed. Analog video is inherently noisy, and
>> noise is a bad thing to retain when you're using typical DCT-based
>> compression algorithms.

>Good luck capturing uncompressed!

No big deal for faster systems, these days. Less still if you're
capturing in 4:2:2 or 4:2:0, rather than 4:4:4.

> Better use Huffyuv (like stated below)

Yeah, HuffYUV is a fine capture CODEC.

> or mjpeg (lossy).

Partly false. MJPEG, like DV and MPEG-2, is DCT based, and is going to
have the same issues as these others formats, with regard to noise.
With a virtually noise-free analog input, MJPEG, DV, or even
I-Frame-Only MPEG-2 could be reasonable capture formats. When the
video is noisy, though, you're damaging the video before you have a
chance to work with it. This will never yield the best result.

>> There about 1000 such filters that run under VirtualDub. Many work
>> very nicely, eliminating the VHS tape noise well without causing
>> visible damage to the VHS image.

>This is very false. Depends a bit on the source. If your source
>is sports, music clips, etc, there's in fact no good filter which
>runs under VirtualDub and can filter the noise out without
>blending and smoothing too much.

False. All [actually] 20 to 30 noise filtering plug-ins I have here
can be tweaked for tradeoffs between filtering and smoothing. I use
both spatial and temporal filters, selective blurring (either as a
filter, or via TMPGenc's block noise filter, which is basically just
an adaptive low-pass filter) to deal with this kind of noise. I've yet
to find a situation I can't fix well. It may take some effort, and of
course, knowlege of the tools. You might walk into my woodshop and
claim you can't build a proper oak cabinet with the tools there; that
doesn't mean I can't.

>You might want to look at AviSynth instead.

All of the AVISynth plug-ins are available for VirtualDub, far as I
know, and of course, AVISynth can load any VirtualDub plug-in. I
actually run VirtualDub plug-ins, most of the time, in Vegas, via a
bridge plug-in.

>> If you don't eliminate the noise, the MPEG-2 compression algorithms
>> will do their best to keep that noise visible, at the expense of some
>> of the actual content you want.
>
>Yes, and you will see that very clearly.

BUT only if you know what to look for. Beginners, folks new to digital
video, I think, are grabbing their VHS stuff, making DVDs, and seeing
quality issues. They don't know enough about getting from point A to
point B to really understand why they're not happy with the results.
Hell, there are plenty of postings here from people having trouble
getting good DVDs from clean DV sources. It is still very much a kind
of art form.

>> Another thing for VHS conversion: the bandwidth of VHS is so low,
>> you're really no worse off capturing to 352x480/576 vs. 720x480/576.
>> That effectively doubles the per-frame "bit budget" for I-Frames --
>> less compression means less artifacting

>Partly false. The bandwidth is indeed very low. But, when capping
>with bt8x8/cx2388x some lousy resizers kick in when capping at 384x576
>(360x480 for NTSC) or lower.

Sure, if you have a capture card that can't run different sample-rate
clocks (pixel clocks are always divided down from the ~28 or ~14MHz
clock you'll typically create via phase lock in a sync processor), you
don't want to make the device driver resize. The best rule for any
processing is to bring it in as dry as possible, then make it wet when
you have an "undo" button. However, IF your capture card doesn't
compromise half D1 captures, you may find raw captures are far more
practical.

>> >Now the most interesting thing is when you have your DVD
>> >ready. You play your DVD and you convert the digital signal *BACK* to analog to display
>> >on TV. So basically what you have is Analog->Digital->Compression->Compression->Analog.

>> No one cares about compression in the toolchain, it makes no
>> difference. LOSSY compression is what you worry about. You really
>> don't want to capture VHS in DV, since you can't filter the VHS before
>> the compression is applied. Grab it directly, uncompressed. Filter it.
>> Only then should you encode it.

>You never captured anything, did you?

I have captured hundreds if not thousands of hours of video, from
analog and digital sources. I've been working in computer video since
the 80's, both professionally and as a hobbiest.

> Or do you have a hdd of 1 TB?

Between the two main work systems here in my office, I have a bit over
1/2TB. Another 200GB or so between the three other office/lab
computers, a "floating" 100GB Firewire drive, a 250GB NAS box for
backups, another 100GB or so free on my kids computers, and a little
free space on the TiVo and the MAME system, all on the network (which
really does need to be upgraded to 1000-Base-T, though I run Firewire
between the two main systems).

Why do you ask?

Dave Haynie | Chief Toady, Frog Pond Media Consulting
dhaynie@jersey.net| Take Back Freedom! Bush no more in 2004!
"Deathbed Vigil" now on DVD! See http://www.frogpondmedia.com

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

On Fri, 2 Apr 2004 08:54:01 -0500, "Morrmar" <morrmar@myway.com>
wrote:

>> If you know what you're doing, making a DVD from a VHS will yield a
>> very satisfactory result. Like everything else with DVD, if you don't,
>> it won't -- there's still a bit of art and engineering to the process,
>> it's not a cookie-cutter thing yet.

>And that's the bottom line, right there. It won't be perfect but how
>many people will _actually_ be able to tell the difference between the
>DVD copy and the original VHS/Beta tape?

Most simply won't know what to look for. As well, they probably won't
care -- fact is, people were largely satisfied with VHS quality. Not
everyone -- I didn't buy pre-recorded VHS. But enough that it's long
been the acid test of consumer acceptability. Is it VHS quality? No?
Fix it. Yes? Ship it.

>Reminds me of mp3s several years ago. The ng had several "golden ear"
>participants who blathered on about "quality". A listening test was
>devised to see if _anyone_ could tell the difference using different
>encoders and bitrates. Out of several _hundred_ participants, there was
>only one guy who could verifiably tell a 128 from a 320 encode.

Funny thing was, MPEG Layer 3 was originally defined to offer
transparent audio quality at ISDN rates (64Kb/s mono, 128kb/s stereo),
for use in broadcasting. It did that job better than anything that
came before, no one complained. We had to get the geeks involved for
that.

>Almost none of the "golden ear" crowd participated, not surprisingly.

Some of those folks have issues. Some aren't really golden ears, but
got that kind of reputation, and hey, if you're not a golden ears guy
yourself, how do you know they're not wrong? Especially in the
"Audiophile" world, where most of the audio quality stuff is a little
science mized with heapin' helpin's of mythology, bad science, snake
oil salesmen, and "emperor's new clothes" purchase decisions. They
don't like too much actual reality, like objective blind tests, to be
introduced into that melee.

You probably have a few legit guys, too, who (like me, for example),
cannot hear as well as we could, say, back in our 20's. I still think
my ears work very well, I hear things other folks don't, but I'm not
making a living on 'em. And I'm honestly not sure I could pull 320's
from 128's. Maybe on my best sound system, on music I know very well.
But in general? Now, if you're talkin' WMAs, that's another story --
just listen for the pre-echos. That stuff gives you fatigue; lots of
people will pick those out easily.

>Unless
>you were an experienced listener and knew what to listen for, you just
>_couldn't_ tell the difference.

Officially, MPEG Layer 3 is transparent at 192Kb/s, last I saw
something speced out from Farunhofer. It's something of a judgement
call, the whole transparency thing. For most listeners, you're trained
yourself to not care about marginal audio. Thus the acceptability in
the consumer market of FM radio, Philips cassette, MTS Television, and
of course, VHS.

>I posted an original 160 encode of a
>song and then re-encoded it... oh about 10 times. That's lossy
>compression X 10. No one could tell which one was the original.

That's also a kind of best case situation. MPEG is always looking to
do the same kind of post-transform filtering. So when you decompress
back to a WAV, then recompress (especially with the same encoder),
you're not really tossing out much in the way of useful information.
If you have artifacts, those will cause this process to decay a
little, but the best compression algorithms don't (otherwise, you'd
hear them, too). Folks used to do generational testing on MD's
(ATRAC). Back in the old days, you'd hear pronounced degradation after
about 5 passes. Today, you can go 20 or 30 generations and not hear a
problem.

What you will find problematic is generational loss when transcoding.
For example, put an MP3 on an MD recorder, especially as a lower
bitrate ATRAC3 file. Change WMA into MP3. You'll hear it, pretty
easily. The different encoders all try to do the same basic job, but
they do it in different ways, and those kind sort of beat against each
other.

>I've got a lot of irreplaceable family footage on 8mm and VHS tape that
>I'm just getting ready to convert. Am I expecting perfection? Not at
>all. But I'll feel a hell of a lot better, and will have a chance of
>actually viewing the footage, in 10 years when I've got them archived on
>DVD rather than a vhs tape.

Another thing to consider -- are you doing any editing? If you're
planning to edit your home video, you'll find it easy to bring
first-generation video into digital, edit there, and produce a DVD
that's dramatically better than any basic roll-edit to VHS, in both
quality and what you're capable of doing. Especially if your original
camcorder video isn't VHS -- everything else was at least a hair
better, some formats (SVHS, Hi8) dramatically better. With Video8/Hi8
and a clean video, you can also use most Digital8 cameras to capture
your video without a trip through the video DACs of a capture board.
The quality this way is quite good, unless the video's very noisy.
Dave Haynie | Chief Toady, Frog Pond Media Consulting
dhaynie@jersey.net| Take Back Freedom! Bush no more in 2004!
"Deathbed Vigil" now on DVD! See http://www.frogpondmedia.com

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

I'd like to thank you all for your responses to my yesterday's post and wish
you a happy belated April Fools.

I had no idea that my post would start a serious discussions. I did a similar
thing on rec.audio.pro, got about 22 different *serious* responses half of
which were calling me names. While in audio what I was saying was irrelevant,
there's however some truth to what I wrote here. No matter what you do, the end
quality of resulting DVD is worse than original VHS. It, of course, doesn't
mean that you shouldn't be going from VHS to DVD. This is the only solution
these days. But if you compress the lossless Huffyuv codec 10 times, how can
you not lose the information?

The claim that LiteOn DVD recorder records without a quality loss is simply
untrue. I guarantee you that the footage that has a lots of motions will show
some small square pixels not found on original. I capture with my Philips based
TV Tuner card into Huffyuv full resolution and then post compress it with
highest settings of TMPGEnc. I guarantee you that this way is much better than
any stand alone DVD recorder.

--Leonid

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

"Leonid Makarovsky" wrote ...
> I'd like to thank you all for your responses to my yesterday's
> post and wish you a happy belated April Fools.

I was quite amused at both of your prank threads.

OTOH, if you have lurked either the audio or video newsgroups
for any length of time, you will note that your joke statements
have been essentially made in months/years past by perfectly
serious (although deluded) newbies.

It also illustrates the very wide range of people on these
newsgroups, ranging from quite knowledgable professionals
who thought the statements were too stupid to respond to, to
people just as confused (but with a strong dose of unwarranted
self-confidence) as your imaginary author of the postings.

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

Richard Ragon <bsema04NOSPAM@hanaho.com> wrote:
: copy.. However, I personally have a Canopus-300 which actually cleans up
: some of the shortcomings of the video signal, plus once you digitize
: your footage, and can run a few filters and a little tweaking on color
: corrections, and you can actually end up with a better copy of that
: footage. I capture with a very high data rate (not DV) in a near
: uncompressed format. The losses from this are indistinguishable to the
: human eye, if any.

I thought Canopus was using Canopus DV codec. Let me know otherwise. I was
looking into Canopus, but they told me the best way to buy this Canopus MBR
or smth that captures directly in MPEG2. But this way they also compress
the audio. And I want my audio to be PCM.

--Leonid

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

Wilbert Dijkhof <w.j.dijkhof@tue.nl> wrote:
:> There about 1000 such filters that run under VirtualDub. Many work
:> very nicely, eliminating the VHS tape noise well without causing
:> visible damage to the VHS image.

: This is very false. Depends a bit on the source. If your source
: is sports, music clips, etc, there's in fact no good filter which

Yes my source is mostly music (live concerts) and sports (ice hockey). And I
just can't find the filter. Most of the VHS I capture from are very good
quality though so AS-IS looks just great.

: runs under VirtualDub and can filter the noise out without
: blending and smoothing too much.

: You might want to look at AviSynth instead.

And what would you recommend?

: Partly false. The bandwidth is indeed very low. But, when capping
: with bt8x8/cx2388x some lousy resizers kick in when capping at 384x576
: (360x480 for NTSC) or lower.

I capture with Philips SAA713xxx

: Note that your capture size "is" not the used sample rate of the card.
: The sample rate (using by the card) is always fixed and well above
: full PAL/NTSC, after sampling the signal is resampled at the requested
: capture size.

So would you recommend doing 640x480 rather than 704x480?

--Leonid

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

On Fri, 02 Apr 2004 10:38:25 GMT, dhaynie@jersey.net (Dave Haynie)
wrote:

>Another thing for VHS conversion: the bandwidth of VHS is so low,
>you're really no worse off capturing to 352x480/576 vs. 720x480/576.

What about capturing 704x480/576, filtering, etc., then resizing
(Avisynth's HorizontalReduceBy2, for instance) to 352x480/576? The
horizontal resolution for analog material, after all, is a subjective
thing. Thus, I prefer to rely on the, more accurate, subjectivity of
my card at capture time, then bring down the resolution to my own
subjectivity (visual limitations) when it comes to playing time. In
the meantime, filters may have a more accurate material to operate
upon.

>That effectively doubles the per-frame "bit budget" for I-Frames --
>less compression means less artifacting

With half-D1, you can encode in 100% CQ mode (D1 DVD quality is around
65% CQ) . The bitrate would then usually be 6000+, even close to 8000
at times.

>Tape degrades just sitting there.

And even worse if you leave it inside the VCR.

>Storing them properly (moderately cool, moderately
>humid, stored on end, not flat) will prolog their life some, but they
>won't last indefinitely. Or even remotely as long as properly stored
>DVDs.

Not that people should forget that discs also have a lifetime. But
then, by cloning, you can perpetuate the species.

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

"Morrmar" <morrmar@myway.com> wrote in message
news:Faebc.12$6D1.2@bignews5.bellsouth.net...
> Reminds me of mp3s several years ago. The ng had several "golden ear"
> participants who blathered on about "quality". A listening test was
> devised to see if _anyone_ could tell the difference using different
> encoders and bitrates. Out of several _hundred_ participants, there was
> only one guy who could verifiably tell a 128 from a 320 encode. Almost
> none of the "golden ear" crowd participated, not surprisingly. Unless
> you were an experienced listener and knew what to listen for, you just
> _couldn't_ tell the difference. I posted an original 160 encode of a
> song and then re-encoded it... oh about 10 times. That's lossy
> compression X 10. No one could tell which one was the original.

It's true that most people can't tell the difference, and largely it's an
issue of education. Once someone knows what MP3 does to the audio it's
really easy to spot later on (and it's annoying if you're a producer like
me. ;-) ).

It also has to do with the source material being used. If you encode a "rock
band", there's very little difference in terms of dynamic range, frequency
response, or sound staging. But play any sort of trance/electronica that has
a a very wide audio spectrum, lots of dynamic range, and seriously takes
advantage of the stereo field, it makes a HUGE difference.

So just because no one "could tell a difference" doesn't mean that it's true
across the board. It was, for THAT MATERIAL YOU USED.

-->Neil

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

In article <c4l19f$qi3$5@news3.bu.edu>, venom@cs.bu.edu
says...
> Wilbert Dijkhof <w.j.dijkhof@tue.nl> wrote:
> :> There about 1000 such filters that run under VirtualDub. Many work
> :> very nicely, eliminating the VHS tape noise well without causing
> :> visible damage to the VHS image.
>
> : This is very false. Depends a bit on the source. If your source
> : is sports, music clips, etc, there's in fact no good filter which
>
> Yes my source is mostly music (live concerts) and sports (ice hockey). And I
> just can't find the filter. Most of the VHS I capture from are very good
> quality though so AS-IS looks just great.
>
> : runs under VirtualDub and can filter the noise out without
> : blending and smoothing too much.
>
> : You might want to look at AviSynth instead.
>
> And what would you recommend?

A good quick filter chain in VDub (probably use the same
in AVISynth, 'cept you have to write the script):

1. Static Noise Reduction (strength 4-8)
http://www.shdon.com/view.php?doc=vid_snr

2. Dynamic Noise Reduction (strength 4-8)
http://www.shdon.com/view.php?doc=vid_dnr

3. Resize down to 352x480/576 if needed
- Also good to crop out tracking noise and re-center
the image during this step. Use the cropping button
with the resize filter selected, crop off the bottom of
the image, write down the resulting height (e.g. 472).
Then configure the resize dialog. Resize to the
resulting height (e.g. 472) and the desired width (352).
I use Lanczos3. Use the letterbox checkbox and set the
desired output height to 352x480/576 and it should
nicely center the cropped image vertically.

That filter chain works nicely for clean source, other
problems require different filters. Speed is around 3x
or 4x normal on my AthlonXP 2600+ with DDR333.

>
> : Partly false. The bandwidth is indeed very low. But, when capping
> : with bt8x8/cx2388x some lousy resizers kick in when capping at 384x576
> : (360x480 for NTSC) or lower.
>
> I capture with Philips SAA713xxx
>
> : Note that your capture size "is" not the used sample rate of the card.
> : The sample rate (using by the card) is always fixed and well above
> : full PAL/NTSC, after sampling the signal is resampled at the requested
> : capture size.
>
> So would you recommend doing 640x480 rather than 704x480?
>

Test your capture card, find some good clean footage and
capture at 720x, 704x, 352x, 360x, 640x and see which
comes in the cleanest. Or google around and try to find
out if anyone has posted exact data for that chip.

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

In article <btrs60htmvv2m8q565bmbj036u7kkiog4f@4ax.com>,
bariloche@bariloche.com says...
> On Fri, 02 Apr 2004 10:38:25 GMT, dhaynie@jersey.net (Dave Haynie)
> wrote:
> >Storing them properly (moderately cool, moderately
> >humid, stored on end, not flat) will prolog their life some, but they
> >won't last indefinitely. Or even remotely as long as properly stored
> >DVDs.
>
> Not that people should forget that discs also have a lifetime. But
> then, by cloning, you can perpetuate the species.
>

And for personal footage, that can't be replicated,
consider using up any remaining space on the DVD with
PAR2 files (using QuickPar). That way, when the built-
in error correction can't keep up with the scratches
anymore, you have a window of opportunity to use the
recovery data to verify/fix-up the files prior to
cloning. As long as the number of damaged data blocks
on the disc are less then the number of recovery blocks
that you added to the disc, you'll probably fix all of
the damaged data blocks.

Note: This requires that your VIDEO_TS folder gets
written to disc prior to burning, and that your DVD
writing software allows you to place extra data on the
disc. Not really an option for set-top DVD recorders,
unless you're re-authoring to put different menus on the
disc.

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

> It's true that most people can't tell the difference, and largely it's
an
> issue of education. Once someone knows what MP3 does to the audio it's
> really easy to spot later on (and it's annoying if you're a producer
like
> me. ;-) ).

Can you say flange? <g>

> It also has to do with the source material being used. If you encode a
"rock
> band", there's very little difference in terms of dynamic range,
frequency
> response, or sound staging. But play any sort of trance/electronica
that has
> a a very wide audio spectrum, lots of dynamic range, and seriously
takes
> advantage of the stereo field, it makes a HUGE difference.

The test, if I remember correctly, was a tune with a huge dynamic range,
maybe classical but I'm not sure.

> So just because no one "could tell a difference" doesn't mean that
it's true
> across the board. It was, for THAT MATERIAL YOU USED.

This was more of an encoder/bit rate test. At the time, some people
swore by Xing, others mp3enc31, some APS. Others swore they could tell
the difference between a 160 and a 192. And these were people who were
quite experienced with the emerging mp3 technology and considered
themselves critical listeners. It was quite enlightening for all.
Sometimes it's not what you hear, but what you _don't_ hear.

I just threw the re-encode X 10 to deflate some egos.

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

"Leonid Makarovsky" <venom@cs.bu.edu> wrote in message
news:c4l0lk$qi3$3@news3.bu.edu...
> The claim that LiteOn DVD recorder records without a quality loss is
simply
> untrue. I guarantee you that the footage that has a lots of motions will
show
> some small square pixels not found on original.

Oh, so you have one? And how much is your guarantee worth?

> I capture with my Philips based
> TV Tuner card into Huffyuv full resolution and then post compress it with
> highest settings of TMPGEnc. I guarantee you that this way is much better
than
> any stand alone DVD recorder.

You *might* be able to get better results with your TV tuner card. Tmpgenc
definitely has an advantage, but it depends on what your tuner card is
putting out and it's a lot more work. My Pinnacle Pro TV tuner card doesn't
produce any better results than the LiteOn, and that's using HuffYUV and
Tmpgenc 2 pass vbr encoding.

Plus I was directly addressing your regarding "waste of time" issue.

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

FLY135 <FLY_135(@hot not not)notmail.com> wrote:
: You *might* be able to get better results with your TV tuner card. Tmpgenc
: definitely has an advantage, but it depends on what your tuner card is
: putting out and it's a lot more work. My Pinnacle Pro TV tuner card doesn't
: produce any better results than the LiteOn, and that's using HuffYUV and
: Tmpgenc 2 pass vbr encoding.

If you have Pinnacle PCTV, they are based on Conexant chips (the older models),
so they won't produce better results than Philips based cards. Although later
models now putting the Philips chip. And yeah, I agree with *a lot more work*,
but at the same time I have more control regarding mastering audio.

--Leonid

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

FLY135 <FLY_135(@hot not not)notmail.com> wrote:
: Oh, so you have one? And how much is your guarantee worth?

Someone gave me bunch of DVDs he recorded on high end DVD recorder from
Hi8 or betamax. These were hockey games. The quality was indeed good, but if
one pays close attention he/she can see some blockiness during fast motion.
And yeah, he chose the highest bit rate possible. I don't have his source so
I can't compare back to back. I'd like to though. Later on I recorded a game
from just a VHS with this TV Tuner card and sent it off to him. He was really
impressed with my work.

Also what do you think of Pinnacle DC1000 or DC3000, whatever is worth $1200?
My friend has it. He lives in Germany. He made a few DVD-Rs from Laser Discs
and from VHS. I couldn't compare the LD source material, but I did have oirig
VHS and I could clearly see degradation. And as a matter of fact I could do way
better quality with FlyVideo 3000FM tuner card than he did with his Pinnacle.
Maybe he didn't use a bit rate of 8000kbs like I do, but my snapshot of the
same frame looked sharper.

--Leonid

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

Toshi1873 <toshi1873@nowhere.com> wrote:
:> So would you recommend doing 640x480 rather than 704x480?
:>

: Test your capture card, find some good clean footage and
: capture at 720x, 704x, 352x, 360x, 640x and see which
: comes in the cleanest. Or google around and try to find

Well, all I know is my capture card definitely does *NOT* support 720x...
resolution. When I used it, it was actually doing 704x... resolution and
during playback it stretched the image to 720x... thus violating aspect
ration. 704x... resolution fit perfectly. I didn't try 640x... resolution
though. I believe you'd recommend to capture in 640x... and then during
mpeg 2 creation resize it to 704x..., right?

--Leonid

PS. There's some serious stuff here now. This thread doesn't seem to be april
fools anymore. Probably gotta change the topic.

 

[ME]

Archived from groups: rec.video.desktop (More info?)

Looks like your question did not get answered:
If you want to copy only you can just buy a standalone DVD Recorder. If you
want to do any editing or make nice menus, etc., buy a burner for your PC
and some easy to use editing/authoring software such as Ulead. You should be
able to copy from VHS to the PC by going through the Digital Camacorder in
passthrough. I do this all the time and can't tell any difference between
the original VHS and the new DVD. Maybe my eyes arn't as discreminating as
some though.
One reason for copying VHS to DVD is to preserve. DVD has a much longer life
that VHS. Something like 100 years vs. 7 years if I'm not mistaken. I also
enjoy being able to set up chapters/menu buttons and go directly to a seen.

"Margaret Willmer" <margaret@gigglingwillmer.org.uk> wrote in message
news:c4jdq0$7le$1@hercules.btinternet.com...
> Interesting discussion. I am beginning to think about converting some of
my
> most precious VHS tapes. As a start should I copy them to MiniDV maybe
via
> a bit of editing on the PC and keep a copy on DV tape? I don't have a DVD
> writer yet. That's another consideration - do I buy a burner for the PC
or
> an under TV recorder? All these things to addle the brain!!
>
> Cheers
>
> Margaret
>
> Remove giggling if replying by email
> "Richard Ragon" <bsema04NOSPAM@hanaho.com> wrote in message
> news:QP7bc.12134809$Id.2029878@news.easynews.com...
> > Leonid Makarovsky wrote:
> > > Going from VHS to DVD is a waste of time! I just recently realized it.
> Just like many of you,
> > > I'd spend hours and days trying to convert some VHS footage to DVD.
But
> the end result
> > > was always worse than original, no matter what you'd do. The idea is
to
> have the end result
> > > identical to source. It will never happen. Why? Just think about it:
> first you digitize the video.
> > > When you digitize the video, there's a certain degree of degradation.
> Then whatever codec
> > > you use Huffyuv or DV would compress your footage. And then you
further
> compress it
> > > to DVD compliant MPEG 2 file. Now the most interesting thing is when
you
> have your DVD
> > > ready. You play your DVD and you convert the digital signal *BACK* to
> analog to display
> > > on TV. So basically what you have is
> Analog->Digital->Compression->Compression->Analog.
> > > And this is just as some claim to preserve a footage 'cause VHS
> deteriorates. But VHS only
> > > deteriorates when you use it. When you just keep them in cool storage,
> nothing will happen
> > > to them. So I found a good solution. Just copy VHS to another VHS or
> S-VHS. You have the
> > > master copy which you store in your storage. To playback just use the
> 1st generation copy. Yes
> > > there's a quality loss, but it is a tiny one. You don't go
> analog->digital->analog. You just go
> > > analog->analog. The results are much better. I don't know about you,
but
> I'm off DVDR market.
> > >
> > > --Leonid
> >
> > All good points Leonid,
> >
> > There's an old saying, "Garbage in, Garbage out". However, there's some
> > trouble with your theories.
> >
> > 1) Yes you digitize it, but you'll get an exact copy of how it plays
> > right this minute. And yes that will be about as good the original
> > copy.. However, I personally have a Canopus-300 which actually cleans up
> > some of the shortcomings of the video signal, plus once you digitize
> > your footage, and can run a few filters and a little tweaking on color
> > corrections, and you can actually end up with a better copy of that
> > footage. I capture with a very high data rate (not DV) in a near
> > uncompressed format. The losses from this are indistinguishable to the
> > human eye, if any.
> >
> > 2) I don't know if you've been to the local Best Buy lately, but it's
> > getting more and more difficult to find a single VHS tape player these
> > days.. At the current rate the VHS player will go the rate of the
> > turntable by next year. So while yes, you might have that perfect copy
> > of the Master tape, you'll not find a way to every play it again in a
> > few short years.
> >
> > 3) In another theory, A final step of outputting to an analog TV will
> > also be a thing of the past, as we look forward to HDTV. So your theory
> > here is also a moot point too.
> >
> > But, yes.. True that the original would have been much better off going
> > straight to digital DV, but I think the whole point is not to try to fix
> > what was done in some kind of remaster, but more like to create
> > something that will allow you be able to play it at all years down the
> line.
> >
> > -Richard
> >
>
>

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

On 3 Apr 2004 00:31:49 GMT, Leonid Makarovsky <venom@cs.bu.edu> wrote:

> But if you compress the lossless Huffyuv codec 10 times, how can
>you not lose the information?

What part of "lossless" isn't getting through? Try it with ZIP
sometime... no loss is no loss is no loss, even if you do it 1000
times.

Dave Haynie | Chief Toady, Frog Pond Media Consulting
dhaynie@jersey.net| Take Back Freedom! Bush no more in 2004!
"Deathbed Vigil" now on DVD! See http://www.frogpondmedia.com

 

[Guest]

Archived from groups: rec.video.desktop (More info?)

Leonid Makarovsky wrote:
> Going from VHS to DVD is a waste of time! I just recently realized
> it.

Rewinding VHS tape is a waste of MY time, and I always knew that

--
www.odysea.ca