Look Steve, there seems to be a disconnect between what I'm explaining
and what you're hearing. I've already said it before (but you did not
include it in your most recent message), but to reiterate in even more
excruciating detail: The reason we have drop frame time code is
because the FCC decided early in the 1950's that if we are to provide
the public with a compatible color TV system, the relationship between
the visual and aural carriers' frequencies must maintained at 4.5MHz.
A choice of color subcarrier frequency had to be made so that the
Intercarrier Frequency Offset is not changed, yet the color sidebands
can be interleaved with the sidebands of the luminance information
providing a minimum of interference with both the aural carrier and
other channels in the band. To do this, the the horizontal line
frequency was lowered to 15,736Hz from 15,750Hz, which lowered the
frame rate slightly to 29.97fps. This allowed the color subcarrier
frequency to interleave synchronously with the horizontal lines. That
decision led to a choice of color subcarrier of 3.579545MHz; a
frequency that would interleave with the new lower horizontal line
frequency.
Previously (in the years of television that preceded color TV), the
old B&W TV system used sync generators that either locked their field
rate to the AC power frequency (exactly 60Hz) or they generated their
own crystal frequency standard (not locked to AC), also running at
EXACTLY 60Hz. As everyone knows, there are 2 fields for each frame,
therefore the old B&W frame rate WAS exactly 30fps. So there actually
WAS a difference in the number of ACTUAL frames produced between the
original B&W system and the new compatible color system. That
difference was (and still is) 108 frame per hour.
But the fact is, it doesn't matter if there ever was B&W TV running at
exactly 30fps or not. The fact is, NTSC color TV runs at 29.97fps,
which is 108 ACTUAL REAL frames LESS than the number of frames that
WOULD fit into an hour if the frame rate actually WAS exactly 30fps.
So if you number those NTSC frames in sequence as if there were 30
frames in each second (which we know there are not), the time code that
results will not follow clock time. This is NDF. So to correct for
this frame-numbering error, there must be created a formula that skips
numbers and brings the resulting time code into line with actual clock
time. This is DF. Now, to further annoy and bore you all, let's go
through Steve's points as he wrote them:
On Oct 21, 2009, at 7:46 PM, Steve Hullfish wrote:
> The confusing thing here is that there's really no such thing as
30fps except on the web. In NTSC it's ALWAYS 29.97.
I answer:
It may be confusing, but there IS (or WAS) such a thing as 30fps
video. It WAS black & white video and was the ORIGINAL NTSC, NOT the
color NTSC. The color NTSC was actually the SECOND NTSC. I never
refer to the first NTSC because of the confusion. Therefore, I never
said that 30fps video (real or imagined) is NTSC (although technically,
it WAS NTSC; the FIRST NTSC. You want confusion? THAT's confusing).
Steve:
> I know there's such a thing as true 30fps, but this statement in your
post, when describing drop vs. non-drop - which is what the thread is
- about is deceiving:
> "The difference in the total number of frames between 29.97fps NTSC
> video and exact 30fps video is approximately 108 frames per hour"
I answer:
The original discussion was the question of DF, when you either start
at :28 frames or :02, what frame number you'd end up with after exactly
30 seconds. Then Phillip Hodgetts offered his blog entry with a
graphic that showed the difference between DF and NDF visually. I
mentioned in his blog and in a follow up message here that there's more
to the story. This is where the discussion went tangent. That was my
fault and I apologize for that.
Phillip has a real good understanding of DF and NDF TC. He stated in
his blog this:
"...it took three pieces of additional information before I truly
understood it: Rainer Standke's insistence that "frame is a frame is a
frame"; the realization (although known) that 29.97 is actually slower
than 30 fps with the consequence that each frame runs slightly longer
than at 30 fps; and that the correction can't be applied evenly."
Also, please note that in Steve's quote of my message shown above, I
say "EXACT 30fps", not approximate 30fps.
Steve also said:
> If, by "30fps video" you mean NDF timecoded NTSC video...
I reply:
No, I DON'T mean NDF timecoded NTSC (color) video. I meant what I
said: 30fps video. I said "EXACT 30fps". 30 ACTUAL FRAMES PER SECOND.
It's theoretical now, but existed before color NTSC TV. And it
doesn't matter if it ever existed at all because it's just a way to
divide an hour of actual clock time into a number of video frames as a
reference. It's a fixed number of frames because it's a fixed amount
of time at a specific frame rate.
Steve:
> That's why I said you were mathematically correct but semantically
incorrect. An hour (as a unit of time) has an identical number of
frames whether it's drop from or non-drop frame. It has 108 frames less
than 30 (frames) times 60 (seconds) times 60 (minutes). If you mean
real 30fps video, which is really not an acceptable form of video at
all - unless you're Canon - then there are more frames in 30fps video,
but the discussion is really about how 29.97 is named in terms of
timecode. Bringing a different frame rate into it just confuses things,
like saying "There are fewer frames in an hour of 24fps film than in an
hour of 29.97 video." It just confuses the argument by measuring apples
and oranges.
I say:
This is why I don't think you, Steve, got my point. On the above, we
don't disagree. But that wasn't what I was talking about. Perhaps
it's my fault for taking this discussion far beyond it's original
query. For that, I apologize. And for that reason, I will have
nothing more to say on it. As far as I'm concerned, it ends here.
Dennis Degan, Video Editor-Consultant-Knowledge Bank
NBC Today Show, New York
------------------------------------
Search the offical complete Avid-L archives at:
http://archives.bengrosser.com/avid/
Avid L2, Where the Answers are.Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/Avid-L2/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/Avid-L2/join
(Yahoo! ID required)
<*> To change settings via email:
mailto:Avid-L2-digest@yahoogroups.com
mailto:Avid-L2-fullfeatured@yahoogroups.com
<*> To unsubscribe from this group, send an email to:
Avid-L2-unsubscribe@yahoogroups.com
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
No comments:
Post a Comment