2-hour length, 148.50 Mhz, 1920 x 1080 progressive scan image, 1-bit object data, 193 bits of file s

[Radium]

Well, not, it's not. When someone is discussing image or video
compression, and they refer to "one bit per pixel", they are almost
certainly *not* talking about bilevel black&white display. Instead,
they are talking about either a greyscale or colour image that has been
sufficiently compressed that the *average* bit rate is 1 bit per
pixel. But when the image is decompressed, it returns to some wider
format; usually 8 bits for greyscale and 24 bits for colour.

Dave

I think this is what I am describing. I could be wrong though.

 

[Radium]

Why are you hung up on sample rate per se?

Because I dislike aliasing.

Question:
if I could decrease the sample rate by eliminating "dead"
time in the original video signal (for instance, by cutting out
excess blanking time in which no image information was
being transmitted), would you find that acceptable?

Yes. As long as no aliasing -- at *any* level -- occurs.

Oh, OK - in that case, I can give you an infinitely-degraded
image in no bits at all!

LOL. There needs to be at least one bit. Otherwise the data doesn't
exist.

So it is true that 1-bit movie cannot exist. What about a WMV file that
is 148.50 Mhz sample-rate, 1920 x 1080 progressive scan image, whose
object data rate is a CBR of 1 bit per second? Could this exist? In 2
hours or this video, the file size would be 7,200 bits.

And why do you think that "color depth" is something
you can compress the daylights out of, but by Gawd, don't
touch there OTHER things?

Because I hate pixelation and alaising with a passion. Pixelation and
aliasing make me make me sick. I don't mind the artifacts -- that I
think -- are associated with a WMV whose color-depth has been
compressed even to extremes while the sample-rate and pixel resolution
are left alone. It looks similar to what a WMA file with a 44.1 khz and
20 kbps sounds like -- I think.

What would be to the human eye what 44.1 Khz, 20kbps is to the eye?

The human ear needs at least 20 hz to hear the sound. The human eye
needs at least 60 hz for the light to appear solid. E.g. a hummingbird
wing flap is to high of a video-frequency for the human eye to see,
much like the sound of a dog-whistle is to high an audio-frequency for
the human ear to see.

WMA is my preferred type of perceptual encoding. Both WMAs and MP3s
will produce artifacts with a too-low bitrate. However, WMA's artifacts
are rather pleasant, while MP3's are digusting.

I have Adobe Audition 1.5. I generate a silent file. I save it as WMA
20 kbps, 44.1 KHz, mono. I convert file this to WAV and then back to
WMA several times. I make my last conversion to WMA and save it. I then
open this WMA file. Finally I increase the volume of the audio in the
WMA file and play. Intrigueing tones result. These tones are typical in
low bit-rate, high-sample rate WMA files. I believe something analogous
could be done to WMV video.

Not really. I've tried doing my Adobe Audition experiment with MP3. How
sickening MP3's audio artifacts are. Much like non-WMV video
compression of pixels are. Those pixelations are just nasty.

 

[Bob Myers]

Or rather, 0.33 bits/pixel..

Oh, damn - I hate it when I do that....I really
need to stop and recheck my head, every so
often...

Bob M.

 

[Radium]

Not really. I've tried doing my Adobe Audition experiment with MP3. How
sickening MP3's audio artifacts are. Much like non-WMV video
compression of pixels are. Those pixelations are just nasty.

Please ignore the "not really" in the beginning of the above paragraph.
It was a f--king
typo!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

 

[Radium]

the question is
how much of a hit you're willing to take (which in turn often has
to do with what can reasonably be expected to be heard/seen
by the human user of this data).

How about a video whose "object data" bit-rate is a CBR of 1 bit per
second with a sample-rate of 148.5 mhz and 1920 X 1080 progressive scan
image resolution? How would this video look like? In 2 hours of this
video, the file size would be 7,200 bits. Sounds like a good idea for
internet video streaming -- if it is possible to have a bit-rate of 1
bit per second.

 

[Lionel]

So is it possible to exist for the following to exist:

A WMV file of 2-hour length, 148.50 Mhz, 1920 x 1080 progressive scan
image, whose object data is only 1-bit in size due to massive
compression of color-depth? [this file would be 193 bits in size
because the the "object size" is 64 bits, the GUID is 128 bits, and the
object data is only 1-bit]?

No, it isn't possible. I've already explained why it isn't possible in
great detail, but you are obviously either too lazy to read it, or too
stupid to understand the explanation.

 

[Bob Myers]

LOL. There needs to be at least one bit. Otherwise the data doesn't
exist.

And hopefully you understand at this point why it's going to take
a LOT more than one bit, or even one bit per second. Again,
you REALLY need to look into some basic information theory.
How much information a given signal actually contains, and how
far you can compress that (and how) before you start to lose
some. Then you need to consider how much you can afford to
lose, and why. Once you do that, you will no longer be asking
questions like:

So it is true that 1-bit movie cannot exist. What about a WMV file that
is 148.50 Mhz sample-rate, 1920 x 1080 progressive scan image, whose
object data rate is a CBR of 1 bit per second? Could this exist? In 2
hours or this video, the file size would be 7,200 bits.

....to which the answer is still "no way."

Because I hate pixelation and alaising with a passion. Pixelation and
aliasing make me make me sick. I don't mind the artifacts -- that I
think -- are associated with a WMV whose color-depth has been
compressed even to extremes while the sample-rate and pixel resolution
are left alone. It looks similar to what a WMA file with a 44.1 khz and
20 kbps sounds like -- I think.

Why do you think that, though? Remember, in the sciences
- which includes physics, and information theory, and electronics -
what you FEEL the answer ought to be doesn't count for squat
until and unless you've got some reasoning and knowledge to
back it up.

Bob M.

 

[Jim Leonard]

Because I hate pixelation and alaising with a passion. Pixelation and
aliasing make me make me sick. I don't mind the artifacts -- that I
think -- are associated with a WMV whose color-depth has been
compressed even to extremes while the sample-rate and pixel resolution
are left alone. It looks similar to what a WMA file with a 44.1 khz and
20 kbps sounds like -- I think.

I think the terms you're searching for are frequency and amplitude.
Like you, I greatly prefer video presented at a very high framerate
(frequency) and don't care so much about color quality (amplitude). In
other words, if I had a (low) certain amount of bandwidth available, I
will take a 60fps video that is 4 colors any day over one that has 256
colors but runs less than 1fps. Both theoretical videos would have
exactly the same bitrate, but one would be a slideshow and the other
would be in motion.

You are very, very confused about information theory and compression.
Here's a quick primer: Lossless compression -- where sizes are reduced
without anything being lost -- involves searching the data for
redundancy and then recoding the data to eliminate that redundancy. So
the answer to your question ("how small can a video file get without
throwing anything away?") depends on how redundant the video is. If
it's nothing but a black screen, most of it is redundant and the file
can get compressed very small. If it's complex motion with lots of
changing elements, hardly any of it is redundant and it won't compress
hardly at all.

You are comparing lossless compression with WMV, WMA, etc. which are
LOSSY techniques. Those compression methods throw information away.
That is a totally different conversation, so you might need to restate
your question a bit.

 

[Martin Heffels]

Well, not, it's not. When someone is discussing image or video
compression, and they refer to "one bit per pixel", they are almost
certainly *not* talking about bilevel black&white display. Instead,
they are talking about either a greyscale or colour image that has been
sufficiently compressed that the *average* bit rate is 1 bit per
pixel. But when the image is decompressed, it returns to some wider
format; usually 8 bits for greyscale and 24 bits for colour.

Hang on. One bit per pixel, means that it can only have one value. Of
course with a LUT you could determine which value should be displayed.
--

 

[Bob Myers]

Hang on. One bit per pixel, means that it can only have one value. Of
course with a LUT you could determine which value should be displayed.

No, again, you're missing that all-important phrase "on average."
Talking about systems that use an AVERAGE of one bit per
pixel (or even less!) over the whole data stream doesn't really
tell you anything about the quality of the video being presented.
As was already discussed, standard HDTV operates at a data
rate such that there is, ON AVERAGE, less than one bit in the
transmitted data stream per pixel of the original X x Y pixel,
N frames per second video.

How that is achieved is actually quite clever...;-)

Bob M.

 

[Martin Heffels]

No, again, you're missing that all-important phrase "on average."
Talking about systems that use an AVERAGE of one bit per
pixel (or even less!) over the whole data stream doesn't really
tell you anything about the quality of the video being presented.
As was already discussed, standard HDTV operates at a data
rate such that there is, ON AVERAGE, less than one bit in the
transmitted data stream per pixel of the original X x Y pixel,
N frames per second video.

I see. So what you mean is that of all the original bits, on average, one
is left, after all the compression is done? Interesting.
--

 

[Dave Martindale]

I see. So what you mean is that of all the original bits, on average, one
is left, after all the compression is done? Interesting.

None of the "original bits" is left. A compressor takes in a data
stream and generates a new, smaller, data stream that can be decoded to
some approximation of the original image or video sequence.

The bit rate of the compressed data stream is simply the number of bits
per unit time (frame, second, hour, whatever you prefer) divided by the
number of original pixels in that same time period.

Dave

 

[Pasi Ojala]

One bit per pixel, means that it can only have one value.

If you have some kind of predictor coding, one input bit is
capable of much more than monochrome.

And, one bit per pixel _on average_ is capable of much more
still. Ever heard of arithmetic coding?

-Pasi

 

[Martin Heffels]

If you have some kind of predictor coding, one input bit is
capable of much more than monochrome.

That's why I mentioned a LUT, where you can make that bit to display
anything you want.

And, one bit per pixel _on average_ is capable of much more
still. Ever heard of arithmetic coding?

Arithmetic has always been my weak point at school ;-)
But I'll have a look into it.

-m-
--

 

[Ken Maltby]

None of the "original bits" is left. A compressor takes in a data
stream and generates a new, smaller, data stream that can be decoded to
some approximation of the original image or video sequence.

The bit rate of the compressed data stream is simply the number of bits
per unit time (frame, second, hour, whatever you prefer) divided by the
number of original pixels in that same time period.

Dave

You mean compression ratio, not bit rate.

One bit has no meaning until there is some significance applied
to it's two values. Within a computer, a bit exists as an element of
a byte. The position of a bit within a byte can be an element of data,
as well, which can change the significance of the bit's value. Bytes
are grouped/read as a word.

You don't have bytes of only one bit, even if the rest of the bits
have no significance they must be there. So in a computer you
can't have a bit stored all by itself.

Now the fact that I can assign any significance to each of a bit's
two values that I wish, doesn't mean it is carrying any more
information/data than that it is in one of two possible states.
There is no difference in the amount of information/data conveyed
between a bit where one state=0 and the other state=1, and a bit
where one state=the contents of The Library of Congress and the
other state=a picture of the moon.

The library contents nor the image data, are actually contained
within the bit. If I could somehow store that bit someplace (say
a file) and examine it outside of a program that assigns it's two
states a particular significance, it only carries the fact that it is in
one of its two possible states. The "file" with only one significant
data bit, would not contain anything more than that.

Now someone will say we are talking about a bit stream not
data stored on a computer. ( A bit stream of one bit? )
That the significance of each bit is established by the order in
which it is processed, not by a structure of the bits themselves.
( No bytes, just packets, that are only there for coordinating
the transmission of the stream.) But for this to actually mean
anything, you still need a series of bits to feed the process, one
bit won't do it.

If you could write a program to read one bit, and apply a
significance to one of the states of the bit to be; a "2-hour
length, 148.50Mhz,1920 x 1080 progressive scan image";
it would be in the program that the "image" data would have
to reside.

This is the state of affairs for any level of abstraction above
the two states inherent in the bit itself.

Luck;
Ken

 

[Bob Myers]

I see. So what you mean is that of all the original bits, on average, one
is left, after all the compression is done? Interesting.

No, let's try this another way.

Let's say you have an original video stream that consists
of - gee, I don't know - let's say 1920 x 1080 pixel
frames that come at you at a rate of 30 per second.

That would be 1920 x 1080 x 30 pixels per second, right?

Working out the above, we could say it's 62.2 million pixels
per second.

Now let's say that we compress this into a data stream which
has an average rate of - again, just to pick a number completely
at random - 19.39 Mbits/second.

On average, how many bits per pixel of the original
video stream does that compressed stream give you?

Bob M.

 

[Ken Maltby]

No, let's try this another way.

Let's say you have an original video stream that consists
of - gee, I don't know - let's say 1920 x 1080 pixel
frames that come at you at a rate of 30 per second.

That would be 1920 x 1080 x 30 pixels per second, right?

Working out the above, we could say it's 62.2 million pixels
per second.

Now let's say that we compress this into a data stream which
has an average rate of - again, just to pick a number completely
at random - 19.39 Mbits/second.

Woah - How many bpp? Is it "1bpp" monochrome? Is it
2bpp CGA? 4bpp EGA? 8bpp VGA? 16bpp XGA,High Color?
24bpp True Color? How about the 32 bit I use for my desktop?

I guess you want to use the monochrome 1 bit per pixel, right?
That would be for both the original and the compressed, right?

 

[Bob Myers]

Woah - How many bpp? Is it "1bpp" monochrome? Is it
2bpp CGA? 4bpp EGA? 8bpp VGA? 16bpp XGA,High Color?
24bpp True Color? How about the 32 bit I use for my desktop?

Doesn't matter. Again, the compressed stream is delivering data
at a rate which, relative to the original pixel rate, represents less
than one bit per pixel on average.

Now, it turns out that the numbers I picked in my example were
not (surprise, surprise!) chosen completely at random:

19.39 Mbit/sec is the actual data rate limit specified in the ATSC
(U.S. digital/HD TV) broadcast transmission standard.

1920 x 1080 at 30 frames/second, is, of course, a standard HD
format and frame rate. As it happens, such material is generally
originated as RGB data at at least 8 bits/color (24 bits/pixel);
long before getting to the compressed form of the final data
stream, it will most likely be re-encoded as YCbCr at something
less than a 4:4:4 sampling scheme - probably 4:2:2 or even 4:2:0
(which are systems that effectively reduce the spatial resolution
of the color components, while preserving the luminance component
at full resolution - this is itself a form of lossy compression vs. the
original RGB representation). The result then goes through a goodly
number of further compression processes, which results in the
cumulative data-rate reduction (compression ratio) seen in the end
result - something in the neighborhood of 80:1 vs. the original.

Bob M.

 

[Ken Maltby]

Doesn't matter. Again, the compressed stream is delivering data
at a rate which, relative to the original pixel rate, represents less
than one bit per pixel on average.

OK, so how many bits of "object data" does that put in Radium's
WMV file? More than one, I would guess. You have been reading
this thread, right?

Luck;
Ken

 

[Bob Myers]

OK, so how many bits of "object data" does that put in Radium's
WMV file? More than one, I would guess. You have been reading
this thread, right?

If you look back through the thread, you'll see that I've
been contributing rather regularly - and as has been
said several times before, by myself and others, there
have to be a LOT more than just one bit. Or even one
bit per second.

Bob M.