How PsF video from the V1 is different than "p" or "F" video

[Steve Mullen]

Originally Posted by Piotr Wozniacki

Canon XH-series have a separate "H DTL FREQ" and "DTL HV BAL" for jus that. What do you think?

I don't expect a firmware update. Sony has often released an A version -- not to fix things, but to add features/capabilities that may not have been ready.

It really comes down to how many units they want to sell. Given the near absence of HD in R50 -- sorry but its true -- the V1 is primarily aimed at Japan, Korea, and the USA. Since they'll sell all they can build in these 3 countries -- how many R50 sales do they need?

When you look at R50 there are 3 segments: PAL DV/DVCAM, 50i HD, and 25p HD. Despite the VOCAL interest in 25p -- it's a minority market, anywhere. News, sports, corporate are 50i. And, if you look at the quality of the V1 imaging -- there are tons of PAL sales to be made. So if Sony sells all the PAL DV/DVCAM units they expect, plus all the 50i HDV units they expect -- they can afford to punt the 25p market. And, the fact is despite some complaints about 25p HDV, they'll still sell a bunch of them.

Sony makes business decisions -- and they fact is that if they had never offered progressive, they would sell a huge number. Remember, the FX7 will still likely sell more units than the V1.

Moreover everyone needs to remember HDV is a joint project with JVC. It sure seems to me that they have nicely split the market between them. Sony wins interlace sales and JVC wins progressive sales. This is how the Japanese do business.

Neither company worries about Canon -- although they should -- because it doesn't have the worldwide sales channels that Sony and JVC do. There is an upper limit to how many they can sell. Plus they both need Canon to keep HDV the majority seller.

But the real issue is 3 useable? Well its 2-clicks softer than 5 which is 2-clicks softer than 7. Since I like 9 better than 7 -- 3 is a huge drop. But, other people love turning the U models waaay down when shooting 24p. Soft is part of their film look.

So, I'll bet many will choose 3 or 4 in the USA simply because they want the softness. In fact, as Adam Wilt wrote, "One telling factor is that sharpness can be dialed down on the V1 without the picture going to mush, as it will on the Z1."

So its really a matter of taste. I don't see the need to fight about taste. People can choose the tool that will give them the look they want.

PS: the minimum iris for so tiny a chip is f/4.0. Your finding is fully valid, and may explain user error as the source of the "mush"claims. :)

[Tony Tremble]

Originally Posted by Piotr Wozniacki

Tony,

Have you also downloaded the clip showing clearly how opening aperture up can sharpen the picture, even at the sharpness setting at 3? For you convenience, here is the link: http://rapidshare.com/files/18138889...rpness3_ND.m2v

All I want so say is that by carefully tuning the parameters, it is posiible to obtain a very nice picture with sharpness=3, completely free of the dancing dots (the horizontal, contrasty line twitter have already been established beyond doubt as a viewing device-induced artefact).

As to the Canon - yes, unless you crank its sharpness up (to +3, or even higer), it looks much softer than the V1E at the default 7 setting, which leaves the room to decrease the V1's sharpness while still being on par with the A1, at least with the V rez of its progressive mode (unfortunately, de-sharpening affetcs also the H rez - hence my "wish" to Sony in the previous message). Of course, the best compromise could be setting sharpness at around 5, which is completely viable in certain conditions, where the contents doesn't have too many contrasty lines and thus the aliasing artefacts are not so strong.

I am still not quite sure whether we are talking about the same thing i.e a soft blurred image or an image without edge enhancement artefacts.

The Xh-A1 has much less EE as a default than the V1. I can only assume from what you have said regarding a wide open iris that you are not aware of the softening of an image with a small aperture due to lens diffraction. I wouldn't advise going below f5.6 if you can help it.

I simply cannot agree that the a well shot XH-A1 image is softer or lacks resolution than a well shot V1 image at sharpness setting 3.

TT

[Tony Tremble]

Originally Posted by Steve Mullen

PS: the minimum iris for so tiny a chip is f/4.0. You finding is fully valid, and may explain user error as the source of the "mush"claims. :)

If that was directed at me as I use the mush claims and stand by it then no. I am a competent camera operator.

I'll post some of the mush recorded using Sony's recommended workaround for a poorly performing progressive mode if you like?

Hope that helps

TT

[Piotr Wozniacki]

Originally Posted by Tony Tremble

I simply cannot agree that the a well shot XH-A1 image is softer than a well shot V1 image at sharpness setting 3.

TT

And you are right, but a "well shot XH-A1" is only possible with its sharpness setting not less than the default. Whereas the V1 at the default sharpness is much sharper (thus having an edge in the I mode), and allows to reduce shapness in the P mode to avoid aliasing (OK - 3 may be too soft for some tastes, but 5 is just fine IMHO).

Disclaimer: all this is not to say I have already accepted the V1E as is (details of my signature have not changed yet:); I probably would if it was possible to decrease H Rez just enough to avoid aliasing, but leave the V rez untouched. And I still think it could be achieved by a new firmware; whether Sony is listening or not is another matter.

Also, the tests I'm running now helped me realize how important the viewing device is. I sort of accepted my LCD can display HDV very well through component, because I first bought and tested it with the Sony HC1, which was my first HDV camera and of course, just an interlaced one. Now I can see clearly the monitor has a not-so-good deinterlacer, and is silly enough to deinterlace also the V1E's 25PsF steam. Therefore, I'm planning to run my final tests with my dealer next week, by connecting both the V1E and the Canon A1 to the same, newest generation, full resulution Bravia HDTV.

[Steve Mullen]

Originally Posted by Piotr Wozniacki

OK - 3 may be too soft for some tastes, but 5 is just fine IMHO.

The nice thing about the V1 is it adds almost no EE when viewed on an HDTV that has its Sharpness correctly set. At 15, the pix simply gets more clear while at 0 it just goes soft. The control very nicely alters DETAIL -- as it should.

Which raises the question of using 4 as a compromise. The question is how much aliasing do you see on a movie exported back to HDV? If an edited movie looks good at 4 -- then you've got a good compromise. I worry that 3 might not be good at the second generation of HDV. Equally, likely is that it will be fine.

Only a test will determine how much filtering is done during the decode and recode.

[Bob Grant]

Don't know how relevant this is now but I've just found out from the Sony Vegas team that Vegas cannot create a m2t file the same as what you'd get capturing a tape from the V1 recording progressive i.e. you cannot create a file with the "p" flag set.

Further comment indicates that playout from a deck via component or HDMI will yield the exact same thing (no flags) regardless of any flags on the tape.

I offer this without comment as I have no knowledge of what happens over a HD component or HDMI connection.

[Steve Mullen]

Originally Posted by Bob Grant

Don't know how relevant this is now but I've just found out from the Sony Vegas team that Vegas cannot create a m2t file the same as what you'd get capturing a tape from the V1 recording progressive i.e. you cannot create a file with the "p" flag set.

Further comment indicates that playout from a deck via component or HDMI will yield the exact same thing (no flags) regardless of any flags on the tape.

There are 2 MPEG-2 flags. One is the PROGRESSIVE FLAG that indicates, naturally, that video should be decoded -- and most importantly -- converted from 4:2:0 to 4:2:2. Without this flag EVERY decoder, including the in a V1, can't properly decode P-mode. This is a serious issue playing back your production.

Which makes one wonder about the MPEG-2 generated by Vegas for DVDs or HD DVDs.

The other flag is UPPER, LOWER, and FRAME. I have no idea if the V1 sets this flag. It can be set with P video to indicate the encoding was done by frame. Of course, that assumes the V1 encodes P as frames. According to some, it is not, which means the flag would not be set.

MPEG-2 flags only exist inside encoded data.

[Mikko Lopponen]

Originally Posted by Steve Mullen

The nice thing about the V1 is it adds almost no EE when viewed on an HDTV that has its Sharpness correctly set.

You mean with a sharpness so low it starts to smooth? The problem with most HDTV's is that we don't know what sharpness setting is 0. And if your source is a dvd/v1/other camera it most likely has ee already.

It would be easy to find though just by plugging it into a computer and looking at black text on a white background. LCD sharpening usually looks pretty horrid.

[Bob Grant]

Originally Posted by Steve Mullen

There are 2 MPEG-2 flags. One is the PROGRESSIVE FLAG that indicates, naturally, that video should be decoded -- and most importantly -- converted from 4:2:0 to 4:2:2. Without this flag EVERY decoder, including the in a V1, can't properly decode P-mode. This is a serious issue playing back your production.

Which makes one wonder about the MPEG-2 generated by Vegas for DVDs or HD DVDs.

The other flag is UPPER, LOWER, and FRAME. I have no idea if the V1 sets this flag. It can be set with P video to indicate the encoding was done by frame. Of course, that assumes the V1 encodes P as frames. According to some, it is not, which means the flag would not be set.

MPEG-2 flags only exist inside encoded data.

The V1P at least certainly records the flag.
Shoot 25PsF and 50i. Capture clips using V7.0d and drop them on the T/L and RClick the events. Under Media Properties the PsF footage shows as Progressive and the Interlaced footage shows as Interlaced. This is old news, I posted links to some mixed clips weeks ago asking if anyone could guess which was which but it was a no brainer for anyone with Vegas.

I can take those same clips with no FXs and PTT using my M15U and capture that tape and again Vegas correctly reports the footage as P or I.

What I cannot do is encode a file with the flags set, ready to PTT. I can encode to 25p however the deck will not accept the stream, it comes up with an Invalid Format error.

I'm told that I can encode as 50i and PTT and then dub via component to HDCAM and all will be well, I guess that makes sense as the stream now contains 50i not 25PsF so the flags shouldn't indicate "P" anyway.

[Mikko Lopponen]

Originally Posted by Steve Mullen

There are 2 MPEG-2 flags. One is the PROGRESSIVE FLAG that indicates, naturally, that video should be decoded -- and most importantly -- converted from 4:2:0 to 4:2:2. Without this flag EVERY decoder, including the in a V1, can't properly decode P-mode.

That would be true for the xh-a1, but doesn't V1 encode everything into an interlaced file? That way the flags will be actually set for interlace.

The other flag is UPPER, LOWER, and FRAME. I have no idea if the V1 sets this flag. It can be set with P video to indicate the encoding was done by frame.

Isn't that the progressive flag?

[Steve Mullen]

Originally Posted by Mikko Lopponen

You mean with a sharpness so low it starts to smooth? The problem with most HDTV's is that we don't know what sharpness setting is 0.

I use test discs. My set goes to smooth (zero EE) at 8, so I set it at 16. One wants no visible outlines. Turns out that before I had the disks -- I had already chosen 20. Fact is that anything from 15 to 25 is fine.

I have to wonder how many reports of CAMERA EE come from folks who leave monitor EE at 50 which they assume is correct.

[Steve Mullen]

Originally Posted by Bob Grant

I'm told that I can encode as 50i and PTT and then dub via component to HDCAM and all will be well, I guess that makes sense as the stream now contains 50i not 25PsF so the flags shouldn't indicate "P" anyway.

There won't be any MPEG-2 flags in the component video going to HDCAM. The deck simply takes fields. More interestingly would be how/if you would get a DVCPRO HD or D-5 deck to record 25p.


"What I cannot do is encode a file with the flags set, ready to PTT. I can encode to 25p however the deck will not accept the stream, it comes up with an Invalid Format error."

... ready to PTT?

If you can't record 25p -- how do you do the following:

"I can take those same clips with no FXs and PTT using my M15U and capture that tape and again Vegas correctly reports the footage as P or I."

[Steve Mullen]

RETURN TO THE ORGINAL TOPIC

For interlaced [or progressive frames as interlace] video, the encoder can either keep the fields separate or combine them together into one frame before encoding, whichever is best for compression purposes. There is a flag on each image stored in the MPEG-2 stream called “picture_structure” can be either “frame” or “top field” or “bottom field.”

IT IS POSSIBLE A FRAME CAN BE RECORDED AS AN MPEG-2 BIT STREAM JUST AS EASILY AS TWO FIELDS CAN BE RECORDED AS AN MPEG-2 BIT STREAM. LIKE DV, MPEG-2 CAN SENSE INTER-FIELD MOTION TO SWITCH BETWEEN FIELD (MOTION) AND FRAME (STATIC) ENCODING. BY DEFINITION, THERE WILL BE NO MOTION BETWEEN PROGRESSIVE FRAMES -- SO FRAME ENCODING WOULD BE USED.

WHAT PROOF DO WE HAVE THAT THE V1 ENCODES AS FIELDS? MUST IT USE FIELDS TO BE COMPATIBLE WITH OTHER 1080I HDV EQUIPMENT? I ASSUME SO.

So I assume the “picture_structure” flag is either “top field” or “bottom field.”

-----------------

For 24p, there are two extra flags available for MPEG-2. These two flags are called “repeat_first_field” and “top_field_first.” A frame in the MPEG stream can have “repeat_first_field” set to “true,” and that tells the decoder to generate 3 fields from this frame, rather than 2. Because fields have to alternate between even (bottom) and odd (top), the “top_field_first” flag tells the encoder which of the two fields in the frame should be sent out first.

In Europe almost every film is encoded with 2-2 pulldown. This will be true for V1 25p and 30p.

To display a perfect progressive image from 24p, an MPEG-2 decoder outputs 1080i digital or analog video and feeds it via HDMI or analog component to the deinterlacing chip in your HDTV. The deinterlacing chip decides the video is 24p when it encounters a stream of 5-field sequences in which the 1st and 3rd fields are identical. The deinterlacer then switches to "film-mode" and combines 540-line fields 1 and 2 to make one 1080-line progressive frame—and outputs the frame three times. It then combines 540-line fields 4 and 5 to make another 1080-line progressive frame and outputs it 2 times. (Doing so converts 24p to 60p.) It repeats the process again for the next 5 fields. The ten 1920x1080 frames are sent to a progressive display panel. Because the 2-3 cadence tells the deinterlacer exactly what to do, 24P can be displayed perfectly.

When the deinterlacer doesn’t see a 3-2 cadence, it switches to "video-mode" deinterlacing. Likewise, with 25p or 30p where there are no repeated fields, the deinterlacer switches to "video-mode" deinterlacing. Unfortunately, “video mode” deinterlacing can only yield different kinds of compromised video.

IN SHORT, A DEINTERLACER MAKES THE ASSUMPTION THAT 25P AND 30P ARE TO BE TREATED AS 50I and 60I.

IN SHORT, 25P, 30P, 50I, AND 60I NEVER CAN LOOK PERFECT.

From -- http://www.hometheaterhifi.com/volum...e-10-2000.html

[Douglas Spotted Eagle]

To clarify regarding the V1:
Component output is always interlaced video (HD out would always be 1080i).

Over HDMI:

a) If a display device is capable of 1080i, the V1 outputs 1080i.

b) If a display device is capable of 480p, the V1 converts 1080i to 480p.

c) If a display device is capable of 1080p, the V1 sends it 1080i (not p even if the footage was recorded using a progressive mode).

It would seem we've gone around the mulberry bush on this one a few times.

[Steve Mullen]

Originally Posted by Douglas Spotted Eagle

It would seem we've gone around the mulberry bush on this one a few times.

The connection path has nothing to do with encoding and decoding nor deinterlacing so it's unclear why you have posted about component video. It "clarifies" nothing this group is talking about.