fx1 50fps SD ? possible

[Thomas English]

has anybody thried this yet.

seperating the HD fields to make 50fps at SD type resolution

the chip claims its 1440 x 1100 which gives a vertical resolution (split fields) of

1440 x 550 (on the chip).

once SD ed .... onto 720 x 576 this would be progressive scan at 50fps?

i

[Martin Doppelbauer]

I tried it already several times with Vegas Video and Virtual Dub and it works great !
For NTSC you will have to rescale from 540 lines to 480 and loose some quality in doing so.
For PAL there is an even better solution. Just add 18 blank lines on top and bottom of the 540 lines of the camera to get a 1 to 1 match with every PAL line (576 in total).
The result is brilliant, true progressive scan SD.
Almost too detailed for some TVs as I found some interlace flicker when I watched it on a cheap one. That "problem" however can easily be solved by softening the image.
Martin

[Thomas English]

cool,

how do you find the image compares to the pd150 using this process. ignoring the progressive scan non progressive bussines

i hear theres a lot less lattitude of exposure to the pd150.

thomas

[Martin Doppelbauer]

Sorry, I don't have the PD150 to compare.
I own Canon's XM2 (aka GL2) and there is really no comparison to the Sony HDV-FX1 whatsoever.
It's like going from VHS to DVD.
If you haven't made up your mind yet you should look for sample footage of the Sony.

Please be aware that some people report the color rendering and lattitude of exposure of the new HDV Sony to be inferior to the "top-of-the-line" DV-competition.
However this is usually based on lack of proper color space conversion. Like all HD cameras the HDR-FX1 uses ITU BT.709 and computerRGB color space. Most DV-based NLE-Software however will still be optimized for ITU BT.601 and StudioRGB setup. If you mix this up you will come to the wrong conclusions (and one much discussed test-report already has).

Martin

[Thomas English]

i really really appreciate your help!

yeah, they say the latitute of exposure is a full 1.5 stops less than the say DVX100.

so, correctly colour spaced- how does a chunk of video compare

XM2

to

FX1 - using the "50fps progressive scan tech" weve discussed.

in definition, and lattitude.

you see my application is a load of lower budget fashion spots. so its between getting this setup with a mini35 or hiring in varicams or dsr570 s with pro35 s. does it compete?! (with grading and messing about)

[Martin Doppelbauer]

Actually I have never compared the two cameras with respect to latitute of exposure before.
But I am also curious to find out if the "test" results on DVXuser.com are reasonable. So I made a comparison just now ;-)

Here's what I did:
I shot a front page of a magazin at very low illumination (almost completely dark) with both the Canon XM2 and the Sony HDR-FX1E.
Both cameras were set up to 1/50 shutter and 18dB gain.
I shot twice with each camera: once in full telephoto and once in full wide angle - both times at maximum aperture, of course.
It just happenes to be that the maximum aperture is the same for both cameras (tele is f2.8, wide is f1.6).

Then I imported the images to Vegas Video 5 timeline and did the proper color corrections for the Sony HDR-FX1.
Finally I took screenshoots of the four setups and compared them both visually and by looking at their histogram.

And this is what I found:
The two cameras deliver absolutely the same brightness - comparing visually as well as looking at the histogram. Actually it is amazing to see how close they really are. Looking like twins ;-)

However the Sony shows practically no noise while the Canon is very noisy.
If you wanted to you could easily blow up the Sony's picture by applying a gamma correction in postproduction.

Since the Canon XM2 has 1/4" CCDs versus the 1/3" of the DVX-100A I assume the latter has some reserves.
However that is not likely to be a serious issue as people already reported the XM2 one of the most light sensitive 1/4" cameras.
Furthermore - as I said above - the Sony shows so little noise that you can easily add more brightness in post (which you apparently can't do with the DVX-100a as it is reported to be much noisier).
Martin

P.s.: I can send you the four pictures if you are interested.
Just drop me a personal e-mail.

[Thomas English]

it seems like with some careful tweaking and curve manipulation some good results could be achieved...

i always like to expose DV cams up to a stop under, so as not to loose the un recoverable highlights, then careful "pushing" of the underexposed parts, to give a nicer "overal" image.

it looks like im going to have to just buy one... when the Z1 appears

if u could/can be bothered to email me some stills i would be very grateful! (presently holidaying in thailand)

studio*at*dustfx.com

but not to worry otherwise...

used carefully this could be a really great camera.. ..

vegas does also seem the way to go with this... bezier time remapping and all that...

i officialy owe u 3 of the finest pints!

thomas