A Q. from an HD newbie (tho' an SD oldie)

K

Kit_L

Active member
Hello all,

I have posted a version of this Q elsewhere, but recently used EX cameras have come on to the market, so the equation has changed.

context: I have a JVC GY-500 with a Canon wide broadcast zoom, and a Canon HV-30 with the BeachTek XLR adapter, with the wide angle adapter.

I make SD DVDs, and intend to make clips for YouTube, Vimeo and iTunes. Although an ex network director (ABC here in Australia) none of my material will ever be broadcast, nor go to film, and definitely not to Blu-Ray.

I edit in FCP studio 2, on a quad-core Mac with plenty of fast internal and external drives. I have a Sony PVM-14M2A monitor that I know won't monitor HD, but will see the MPEG-2 DVD picture.

So, I have decided that for storage and scalability reasons, 720p (probably at 30fps) will do all that I need, and my question is about total workflow as much as anything. I see from this forum that some folk use an intermediate codec for the EX footage, but I assume that this will not be necessary if I shoot 720 progressive in camera.

Am I right in thinking that a Firestore can handle the bit rates of 720p/30fps of either an EX1 or a 170 and that FCP can edit both types of footage natively? And that both cameras are similar weights and form factor (I shoot overseas twice a year and will have to carry all gear with me)?

Used EXs are now less than new 170s, which is why I am inclining in this direction. I should also say I have no brand loyalty; I am just looking for the most efficient workflow and will appreciate any assistance. TIA, KL
 
Well,

I, and I assume others, use an intermediate codec not because of speed, but because when I color grade, I prefer to have something more robust. Just my preference.

You cannot connect a firestore to the EX1 and record 720p so that negates the second part of the question. FCP will not edit native EX1 files. It currently requires a rewrap of the footage, which is reasonably fast, but certainly not instant.

I believe the Panasonic camera is lighter but I've not held it. In looking at shooting costs, the EX1 (even new) has a distinct advantage when media costs are factored in over the P2 cameras. With the new SDHC recording on the EX1, media costs are about $100/hr compared to the $1500-$1800/hr of 720p shot on P2. Using factory Sony media, the EX1 media costs are $750/hr or so. If you don't plan on dumping cards regularly, this might make a difference to you.

When Sandisk release their 32GB SDHC cards in '09, my firestore will be retired. That will give me 4 hours of recording without ever having to open the camera. That's about what my Firestore offers though only at HDV rates.
 
Perrone,

Thanks for this; a few supplementaries, if I may.

What utility does the re-wrap of the EX footage?

And, if the EX footage can't be ingested in its native state, then I might re-consider the HPX-150 too (its footage needs conforming, although I am assuming that rewrapping is faster than conforming). It is an extra step, though. When you say the rewrapping is reasonably fast, can you give an estimate of time to (say) rewrap 30 minutes of footage might be?

I agree completely re. media costs/cost of shooting for the EX1; the new SDHC + express card holder seems like an excellent way to go. Re. the Panasonic P2 cameras though: can the Firestore record the 720p/30fps that I am considering?

Thanks again for responding. cheers, kl
 
Kit_L said:
I have decided that for storage and scalability reasons, 720p (probably at 30fps) will do all that I need,
Click to expand...
1080p and 720p both use 35mb/s, so there's no difference in storage-space needed.
720p will be faster to post-process because there are less pixels, but I still would always prefer to capture 1080p, because 1080p to SD downconversion typically results in higher quality and less noise than 720p to SD.
Kit_L said:
I see from this forum that some folk use an intermediate codec for the EX footage, but I assume that this will not be necessary if I shoot 720 progressive in camera.
Click to expand...
If your computer is fast enough, then there's absolutely no use of using an intermediate codec, but there would be drawbacks:
-extra time needed for conversion
-much more harddisk-space needed
-perhaps some quality-loss

That rewrap Perrone mentioned is a process done by a free sony-app, putting the video/audio-streams of the mp4-files into quicktime-containers/files. You can use those mov-files anywhere/with all apps on a system with FCP installed, where you can use quicktime-files. Very convenient.
And it's absolutely robust. I'm wondering of what Perrone was thinking about when he said that converting to an intermediate codec would increase robustness. I never had any problems with these quicktime-files at all.
 
Last edited:
Kit_L said:
What utility does the re-wrap of the EX footage?
Click to expand...

Free Sony Clipbrowser Software. Also does metadata. Nice utility.

Kit_L said:
When you say the rewrapping is reasonably fast, can you give an estimate of time to (say) rewrap 30 minutes of footage might be?
Click to expand...

About 3 minutes on my laptop.

Kit_L said:
Re. the Panasonic P2 cameras though: can the Firestore record the 720p/30fps that I am considering?
Click to expand...

Yes. I know they can on the HVX, not sure on the HPX.
 
Last edited by a moderator:
mrjazz said:
And it's absolutely robust. I'm wondering of what Perrone was thinking about when he said that converting to an intermediate codec would increase robustness. I never had any problems with these quicktime-files at all.
Click to expand...

I prefer to not use 8 bit, long GOP files when I work. But that's just me.
 
Perrone - Why? When we get our new gear, this is my biggest question.

1) Convert to ProRes before editing in FCP. Pro: Easier on CPU while editing. Con: Footage is recompressed. New file sizes are larger than original XDCAM files.
2) No conversion. Edit in XDCAM format. Pro: Footage isn't recompressed. Slightly better quality. Con: More demanding on CPU while editing. More rendering.

Have I got this right?
 
MitchLewis said:
Perrone - Why? When we get our new gear, this is my biggest question.

1) Convert to ProRes before editing in FCP. Pro: Easier on CPU while editing. Con: Footage is recompressed. New file sizes are larger than original XDCAM files.
2) No conversion. Edit in XDCAM format. Pro: Footage isn't recompressed. Slightly better quality. Con: More demanding on CPU while editing. More rendering.

Have I got this right?
Click to expand...

Mitch, I don't use a Mac or FCP so my results won't mirror yours. However, I find that for doing color correction, grading, and other work, I prefer codecs that move a bit faster on the timeline, and hold up better for re-rendering. Either that, or I work with uncompressed 480p proxies.
 
Thanks Perrone. I'll have to try it both ways and see what works best for us.

Switching from DV (Standard Def) to XDCAM (HD) has brought a LOT more options to the table. I'm looking forward to the upcoming learning curve.
 
I hear ya man. I switched over to HD this summer, but had done some HDV work over the past year. It was a fairly steep learning curve. But oddly, reading ASC Magazine and some others showed me how the "Big Boys" were doing it, and I have just adapted my work flow to something similar.

Simply moving to cutting proxies and working on those changed EVERYTHING. Basically, I render my raw footage into a full-sized lossless (or Cineform) file. Then put that away. Then I render the same timeline into a 480p Cineform or 10-bit codec format proxy. I do all edits and grading on the proxies and before the final render I simply replace my proxy with the full-res file and render out. Works like a charm, and it moves at the speed of miniDV.
 
Yes, Sony makes a solution and from what I understand it works just fine. Though it won't overcrank for the same reason the KxS solutions won't overcrank. It is also limited to 2 hours, and won't allow the door to close on the EX1 or EX3.
 
mrjazz said:
, but I still would always prefer to capture 1080p, because 1080p to SD downconversion typically results in higher quality and less noise than 720p to SD.


l.
Click to expand...

Has anyone actually done a comparison to prove this? I don't doubt that the math is there, but I'm wondering about the real world here. If so, anyone know if test results have been posted any where?
 
brianluce said:
anyone know if test results have been posted any where?
Click to expand...
I will do a test right now..., but let's talk briefly about the background of it:
Did you realize that 720p has higher sensitivity than 1080p? Let me express it briefly in this way: Manual 1080p to 720p conversion won't add gain to keep noise constant (like in-camera conversion), but would reduce noise to keep gain constant.

@Perrone
I just had to laugh when I read your signature. ;)
 
Last edited:
I now tested it. The first grab is 1080p with 0db and f3.4 downscaled in Photoshop to 720p. The second grab is in-camera 720p with 0db and f4.0.
The first grab has obviously less in noise in the darks and more sharpness (ok, not the best scene for testing sharpness ;) ).
 

Attachments

  • 1080to720.jpg
    1080to720.jpg
    110.7 KB · Views: 0
  • 720camera.jpg
    720camera.jpg
    133.1 KB · Views: 0
mrjazz said:
I now tested it. The first grab is 1080p with 0db and f3.4 downscaled in Photoshop to 720p. The second grab is in-camera 720p with 0db and f4.0.
The first grab has obviously less in noise in the darks and more sharpness (ok, not the best scene for testing sharpness ;) ).
Click to expand...

Thanks. If you get a chance, try the downconversion of both to SD. SD will be with us for a while yet. If 1080p gives results that produce a better SD image, that'd be useful to know.
 
Of course I can test it, but if 1080p gives better 720p than in-camera 720p, obviously 1080p will give better SD than (in-camera) 720p.
 
mrjazz said:
Of course I can test it, but if 1080p gives better 720p than in-camera 720p, obviously 1080p will give better SD than (in-camera) 720p.
Click to expand...
Seems reasonable, but if 720 is half a stop faster, and the ultimate downconversion is going to be comparable whether from 720 or 1080, then I'm sure people would want that half a stop. :thumbsup:
 
Barry_Green said:
I'm sure people would want that half a stop. :thumbsup:
Click to expand...
If you want that half stop, then simply use gain. That half stop of 720p itself just comes from implicitly added gain in the downconversion-process. So in terms of noise there won't be a difference.
brianluce said:
I'll be you a cup of coffee that the SD image will be indistinguishable.
Click to expand...
You're right. I tested it. left is 1080 to SD, right 720 to SD.
 

Attachments

  • 1080toSD.jpg
    1080toSD.jpg
    55 KB · Views: 0
  • 720toSD.jpg
    720toSD.jpg
    58.2 KB · Views: 0
You must log in or register to reply here.
Back
Top