1080p > SD could be better than DVCPRO50

S

someday

Well-known member
I think that despite DVCPRO50 has less compression, the CONTOUR is more noticeable than shooting 1080p (or 720p) and downsizing to SD.
The horrible contour generated by the sharpness setting makes the difference between video look and film look, the contour size is crucial for the look difference. But if you shoot at DVCPRO50 without sharpness you will get a defocused image.
Film has no contour, but has huge definition. So if you Shoot at 1080p with sharpness to +2 or more, and you downsize, the contour will be lost in the interpolation process but not the perceived definition.
So i suppose that 1080p (or 720p) looks "nicer" than DVCPRO50 anyway.
But that's al theory... my HVX comes next week, then i will know.
 
As soon as you get it shoot some tests - same shot, same lighting, then post some side by sides.

My guess is, you won't see much difference between the DVCPro50 and a downrezed 1080P shot.

Just my guess though - I haven't done any testing on it, but I'd be curious to see it.
 
I believe it's true that everything the HVX does is internally downrezzed from native 1080i, so in theory, nothing is gained by doing so after the fact. I know Barry has weighed in on this a number of times here, so I'll let him have the definitive word on this. I imagine he'll pop up on this thread quite soon.
 
Hum, yes i forgot that the HVX do that already internally. So the crucial question that i have is: the sharpness is applied after the conversion to SD or before? And what does it means, does it matter or not?
Yes i will do comparison shots, here what i want to do (please suggest me more if necessary)

SITUATION 1) very complex image (indoor, artificial light to avoid changes)
A: DVCPRO50
B: 1080p resized to SD with many sharpness settings
C: 720p resized to SD with many sharpness settings
C: 1080p resized in order to fill the 4:3 area (why not this could be interesting)


SITUATION 2) LOW LIGHT GRAIN
Low light situation at all different gamma settings

SITUATION 3) SHOT FOR CC
all knee and gamma combinations, and strong CC, comparison to similar light situation on ARRI footage
 
Popping up...

The edge enhancement needs to be controlled, certainly. Too much edge enhancement ruins everything.

But the point of downrezzing from HD really depends on where you're going with the footage. If you're keeping it in an uncompressed domain then downrezzed 1080 will give you 4:4:4 color sampling in SD, which is very nice indeed. There will have been more compression artifacts in DVCPRO-HD if shot at 720, but the compression is quite mild in 1080. Downconverted 1080 can look very good.

If you're shooting for DVD release or something I still say there's no observable benefit. But if you're doing heavy compositing or something, then it may actually be beneficial to record first to 1080 and then downconvert in post to an uncompressed SD 4:4:4 master.
 
i find the DVCpro 50 yields a better looking picture than downconverted 1080/720. the reason is this, the downconvert incamera is done in uncompressed space. the down convert in post is done AFTER DVCprohd compression. just dial in the amount of edge enhancement you like, with a monitor to judge, i tend to like -1 in SD. and its easier to get the proper picture that way. trust me, down converting macroblocked shadows doest do anyone any good.
 
That's my point as well, and why I've always advocated that process. I'm just loosening up a bit here since 1080 is more mildly compressed than 720, so it's *possible* that it might deliver a better 720x480 image than straight DVCPRO-50. But DVCPRO50 is much more mildly compressed, so ... yeah, I would guess DVCPRO50 is probably the more solid way to go, but I bet going from 1080 is going to look very competitive. Should look a lot better than going from 720 would.
 
I agree with all the above, we are talking about video quality and compression ratios and this is a crucial matter to obtain a professional master. But let's talk too about look. WHY THE HELL 35mm film, without edge enhancing retains a great definition in SD while a native SD footage looks as crap*

)* to me crap=video
 
because they balance the edgenhancement and carefully monitor the compression on motion picture DVD's. They tweak it for peak performance.

35mm film is pushing out 4-6k's worth of resolution. everysingle pixel of luma in SD can have its very own value with that much information being down ressed. They start with the HIGHEST quality masters, the highest quality Digital masters, then they down res into a VERY high quality SD master.... etc. start with an amazing point you will end up with great footage. start with video, end up with video.
 
Because 35mm film is transferred to video using a multimillion-dollar telecine film scanner. Compare that to a $3,000 SD video camcorder and, yeah, film's going to look infinitely better.
 
hehehe, i should have omitted that question....
And there is one more thing:
A 1920 x 1080 video frame is out of focus compared to the same size coming from a digital camera.
Examples:

a landscape video grab:
here

a landscape photo of about the same size:
here

I wonder what's the perceived res of a 1080p compressed image...

In other words HD is still far from ... HD
 
well no. HD is HD. Digital photography is something else. that 1080 Digital photograph image was probably origionally a 7-10 megapixel image down ressed to 1080 like size. But look at 1 megapixel digital cameras. they take images about as sharp as HD cameras. roughly the same number of pixels per frame. starting with a higher quality master makes for better end product is all im saying.

The 1080 of the hvx is great, but its 720p, which is internally Downressed looks amazing! it intercuts with a varicam reasonable well. ( reasonably well i reiterate)
 
It's frustrating. SD national commercials look sharper than HD native footage because they came from 35mm.
Isn't there a workaround to maximize all those HD pixels ?
 
Start with Kodak Vision2 film stock.

but instead of getting frustrated. think about this, until 2000, when the DVX100 came out, there was NO way short of shooting S16mm film, or using a F900 Cine Alta, to get 24p. in 2007, theres a 1080p 24p HD consumer camera avaialble right now for 900$.

That would have been considered INSANE 10 years ago. so relax, in 10 years imagine what will be avialable...... or even next year for that matter.....
 
someday said:
It's frustrating. SD national commercials look sharper than HD native footage because they came from 35mm.
Isn't there a workaround to maximize all those HD pixels ?
Click to expand...
Except for the national ads that were shot on the HVX, including REMAX and Subway spots.
 
TimurCivan said:
. . . think about this, until 2000, when the DVX100 came out, there was NO way short of shooting S16mm film, or using a F900 Cine Alta, to get 24p. in 2007, theres a 1080p 24p HD consumer camera avaialble right now for 900$.

That would have been considered INSANE 10 years ago. so relax, in 10 years imagine what will be avialable...... or even next year for that matter.....
Click to expand...

Indeed, Timur. It totally blows my mind to think about how far we've come in about a decade. Maybe in another 10 years "RED-like" cameras will be used for home movies. :)
 
I love the quality results from DVCpro50 when going to SD DVD. The colors and resolution were just fantastic on some musical performance auditions I shot last week.

480/24p
cine V (default settings)
stage spot and footlights

My thought is that I can get a little bit more P2 recording time using 720/24pn than DVCpro50 at 480/24p.

The assumption is that FCP does not care which codec and nor does Compressor when I convert it for DVD Studio. So if I shoot in 720/24pn, then I have an HD copy, use a lillte less P2 space, and still get great color and res on an SD DVD.

Now, if there really is better quality in DVCpro50, I'll scrap that idea and stay with it for SD shooting.

Just how different are the two...once you get to DVD?
 
You must log in or register to reply here.
Back
Top