Ex1 best format for Pal SD delivery (720p or 1080i)

[Walter Biscardi]

Both will work fine, but I HIGHLY recommend you invest in an AJA Kona 3 to perform this conversion.

Basically you shoot and edit in your preferred format of High Def. Personally I like 720p because the file sizes are even smaller. For NTSC, we usually shoot 720p/60, though we have one client that loves 720p/30 and both downconvert extremely well to SD.

When your edit is completely finished, you use the Kona 3 to convert your timeline to SD on the way out to DigiBeta. You have three options, Center Punch, LetterBox, Anamorphic for the downconvert.

Using this method, you stay in HD all the way through the Post process so you can also lay off an HD master for future use.

Walter Biscardi, Jr.
Biscardi Creative Media
HD and SD Production for Broadcast and Independent Productions.

Read my Blog!

STOP STARING AND START GRADING WITH APPLE COLOR Apple Color Training DVD available now!

[Jerry Hofmann]

Probably a toss up, but I’d rather work with progressive video…

Jerry

Apple Certified Trainer

Author: “Jerry Hofmann on Final Cut Pro 4” Click here

8-Core 3.0 Intel Mac Pro, Dual 2 gig G5, AJA Kona SD, AJA Kona 2, Huge Systems Array UL3D, AJA Io HD, 17″ MBP, Matrox MXO, CD’s

[Rafael Amador]

Hi Johny,
Having the Ex-1 a 1920×1080 CMOS and going to PAl, I think that 1080i50 is the best option for broadcasting.
I think that, at least in theory, 720 should yields the best quality;because you have 35Mbps to write a much smaller number of pixels than in 1080.
But you don’t want Progressive look, so no 720p25. And if you go 720p50 there are two issues:
– You may lose quality on compression because you are duplicating the number of pictures you are writing with your 35 Mbps.
– In the end when you go to PAL that is ‘i50″ you will discharge the 50% f the “p50” picture information, and this is a big waste of resources.
It would be great if would be possible to record 720i50 (just a SONY Firmaware update).
Shoot 1080i50 and edit directly in an PAL Anamorphic ProRes sequence.
Before rendering set “render all YUV in High Precision YUV” and “Motion Effects Rendering: BEST”.
Cheers,
rafael

http://www.nagavideo.com

[Rafael Amador]

As Walter suggest, if you have an AJA or BM you can get the best downconversion and in a fraction of time than FC or Compressor.
The last SONY Transfer software lets you upload EX-1 footage back to the SxS card. You can cut in FC and send your EX-1 sequence to the SxS card and capture it as SD ProRess or 10b Unc. Really time saving.
Cheers,
rafael

http://www.nagavideo.com

[Jonathan White]

Thanks guys for all the informative answers. Just on that last one Rafael… are you saying I write my finished sequence back to SxS as XDCam, then bring it back in using AJA or BM to downconvert or are you saying bring in downconverting in the camera??
If using a card why not just write to tape using AJA or BM downconversion?

I’d love to get a new card (currently on a decklink extreme SD) but I think I’d have to change my computer as well (currently on a pci-x G5, reckon all the good stuff is now pci-e ?? ). So maybe I’ll suffer the render time hit as long as quality isn’t compromised..

Thanks again,

Johnny

[Rafael Amador]

[Johnny White] “are you saying I write my finished sequence back to SxS as XDCam, then bring it back in using AJA or BM to downconvert or are you saying bring in downconverting in the camera?? “
You can downconvert with the camera too but i trust more my AJAioHD to do that.
I don’t know if would be possible to downconvert with the camera and record directly in a DigiBeta desk.

[Johnny White] “If using a card why not just write to tape using AJA or BM downconversion?”
You can do it too. Edit in HD and rendering in ProRess would be a good option. Then downscale with the card directly to the video desk. But rendering in HD would take always longer.
However I prefer to convert the EX-1 footage to Prores with the card not only for the perfect downscaling. As important as the downsizing is the Chroma filtering when passing from the 420 of the Ex-1 to the 422 of the ProRess.

[Johnny White] “I’d love to get a new card (currently on a decklink extreme SD) but I think I’d have to change my computer as well (currently on a pci-x G5, reckon all the good stuff is now pci-e ?? ). So maybe I’ll suffer the render time hit as long as quality isn’t compromised.. “
I think that with your system you could downconvert with the camera and capture as ProRess or even Uncompress.
Rafael

http://www.nagavideo.com

[Jonathan White]

I’ll experiment with the various options. Thanks for all the advice.

Johnny

[Jeremy Garchow]

I am with Jerry, I always like to shoot and edit in progressive. If you don’t like the 720p25 look, then shoot 720p50 which will translate very nicely to 525i (each p frame gets mapped to an i field). By the way, this is probably how most cameras sample images nowadays. They take a progressive image and make interlaced fields out of it. I know the HVX200, the chip is basically scanning 1080p60 (which is not recordable) and all resulting formats get scaled and then interlaced (or not) from that size. Making interlaced images from progressive captures is pretty much the norm. If I were you, I’d shoot 720p50.

Okay, why progressive? It’s much easier to edit (in my opinion) and deal with in post. All graphics and photos and everything else can be handled without fields. Making interlaced images is a snap and with 720p you have twice the temporal resolution as 1080i (but you have smaller pixel count, since you are mastering to SD, this really doesn’t matter TOO much). Making Web movies or DVD movies for review on computers will yield much higher quality results as those displays are not capable of displaying interlaced fields properly, so you will have to deinterlace. Why deinterlace when you can just shoot progressive? If I were you, I’d do my down convert in hardware with a Kona (or send out your HD file to a post house that has any number of methods of hardware down convert). At the very very least, use Compressor as it will handle all the fields correctly and FCP won’t. You have to be careful with your text and graphics when downscaling. Things that look good in HD might not look good in SD especially when it comes to font size and weight. Thin fonts will sometimes not transfer properly so it’d be best if you got yourself a Kona, edited in HD and double checked everything on an SD PAL monitor using the real time down convert option on the Kona.

Hope this helps.

Jeremy

[Jonathan White]

It does help, I like 720p50, shot my last job with it. I am tempted to invest in a Kona but I think it means buying a whole new G5 as well ( one pci-x G5 with Decklink extreme sd for sale!! )….
Or maybe I find a post house as you mentioned, I hate to compromise on broadcast quality….
Johnny

[Jeremy Garchow]

You can get a Kona 3X but you can’t CAPTURE ProRes with it, you MIGHT be able to lay it off, don’t know.