Article on November in new videography

Re: Article on November in new videography

<< yeah, its kind of annoying hearing you guys complain about the movie being soft at points or blocky at points. its a $3000 video camera! thats the point.>>

lyons den,

I say in my post that it is stunning and fantastic, but I also point out the warts. You're right it's a $3K camera and it does make some great images.

Scott
 
Re: Article on November in new videography

I thought Phantom Menace was quite disappointing-looking, given the hype surrounding it. I saw it off a film print (where it was muddy and flat) and from digital projection (where it looked very digital and blocky).

However, I thought "Once Upon A Time In Mexico" looked perfectly acceptable. And "28 Days Later" was completely unacceptable (yes, I know they were "going for a look", but the thing looked completely out of focus at all times on the big screen).

Haven't seen "November" but I sure would like to. The DVX footage should look substantially sharper than "28 Days Later", although obviously it's still SD-res.
 
Re: Article on November in new videography

I'd love to see "November" as well, and I guess I will...but I have yet to see any blowup of a 3 chip dv camcorder movie that I liked. And I've see them all, save maybe "Personal Velocity"...

You can shoot super16 with a 80k budget.

Now if you're shooting porn, gotta run with the dvx100 :-)
 
Re: Article on November in new videography

I was at the screening and agree with Scott Spears: I too thought it looked stunning. Maybe I just have not watched enough SD video movies projected on the big screen (only 2 instances I can think of - an indie shot on 16mm and finished on Betacam SP, and the 28 days later trailers at the theaters), but I spent most of the time thinking to myself "no way this was shot on a DV camera". This compared fine with low budget HD work I have seen projected on the big screen (Step into Liquid), and was way way way better looking than the trailer for 28 days later in terms of apparent resolution. I was expecting to see that 28 days later like full-on hardcore big grain video resolution that almost makes you cross-eyed :-X but when the movie actually started I probably looked pretty stupid with my jaw hanging. :o Maybe it had something to do with the upres conversion they did to 1080 HDCam, maybe it was because the screen wasn't a full-size commercial movie theater screen (it was probably about 10' high by 20' wide - wild guess), I don't know.

Yes, there were problems in some scenes: about 3 or 4 shots had glaringly obvious big time blown-out highlights, probably 3 had noticeable low-resolution stairstepping, and the dark-room red-lit scene had the stairstepping and a very weird artifacting, almost like looking through privacy glass with vertical lines. But think about it, that was less than 5% of the whole movie. I think it could have been avoided, cut out in editing, or reshot. The remaining 95%+ had barely barely noticeable resolution stairstepping/aliasing. I seriously don't think the general public would notice it. This screening really was a revelation to me. It showed me some promise that it finally may be possible to make a DV movie that can be projected on the big screen and won't have a completely horrible resolution that would probably put people off within the first few seconds. I hope this makes it into theatrical release because I want all of you to see it. :)
 
Re: Article on November in new videography

the screen was probably double that size , probably 20x40
 
Re: Article on November in new videography

ok, well phantom menace was maybe a bad choice, my point was it aint all in the image, i can live with a harsh indie look if the story and acting are top notch.

and also i think "once upon a time in mexico" looked awesome! i love watching that dvd and rodriguez always has some awesome dvd extras...

do you guys think november will get a good release? since i live in the middle of nowhere (as far as smaller releases go) i will have to wait till video. :(
 
Re: Article on November in new videography

Why in the hell are you guys talking like "Phantom Menace" was shot digitaly. "Attack of the Clones" was the one shot with the Sony HDCAM. I just had to clear that up :).

I saw "Novermber" at the meeting also. It looked okay to me. There was some really good stuff in it but I'm not sure it should have won the cinematography award at Sundance. From what I've seen of "Art History", that looks better to me. I have to see the DVD first before really making a decision. I did like "November", it just seems like they could have done a little better job with it.
 
Re: Article on November in new videography

DOH! Good catch. Let me just wipe the egg off my face... yes, it was AOTC, not TPM (although there is a digital shot in TPM, when Qui-Gonn extracts Anakin's blood...)
 
Re: Article on November in new videography

Could some of the differences in perceptions at the screening you all attended have to do with where you were in the theater? From the back row, I suspect you wouldn't notice as many resolution based artifacts as you would from the front.

Also, it never ceases to amaze me how average viewers' eyes and minds gloss over the level of clarity of the medium. They focus on content alone. Unless the quality of the medium is fairly bad, they're not likely to notice a difference. And, if you ask them a week later, they're not going to be able to tell you which was "clearer," a film shot on 35mm or 16mm or an up rezzed DVX100 shot. They're just not in it for that, and it doesn't get noticed or remembered. WE notice and remember such things. They don't (again, unless it is REALLY bad).

To help you appreciate this, give friends and family members who have no interest in video (like we do) and are typical viewers the following test. Ask them to see if they notice the difference in the look of particular sit-coms they watch (currently or past sit-coms) that was shot on video vs. shot on film. Say, for example, (I'm dating myself here - you select your own sit-coms for your little test) The Mary Tyler Moore Show was shot on film, and All in the Family on video. Ask them if they notice, in their mind's eye, any difference. Chances are they may not, and if they did, they have never thought about it or noticed it before. Granted, both the video and film on network sit-coms are high quality, but my point is that people don't notice the difference unless you call their attention to it.
 
Re: Article on November in new videography

Hi all,

I too was at the screening, this being the second time I have seen the film and I have been thinking about Nancy's comment that it looked soft. To set everyone up for where we were, the little theater we were in is a color correction theater at Laser Pacific. The front row is about 20 feet from the screen. The last row is maybe 40 feet. The room is not meant for Theatrical projection but for critical evaluation.

That said, when I saw November in Park City, I was a good 100 feet from the screen, and it looked much sharper. If you sit too close to your TV set, or your Plasma screen, it too looks too soft. So much of that softness that people are reacting to here is based on the fact that we were too close.

Additionally, I believe that the scene that really blew out the highlight was orchestrated that way. Certainly you must realize that. There were so many opportunities for the highlights to blow out and they didn't, but in the scene in the bathroom was driven by the story, the main character was having a melt-down.

I thought that Nancy was in control of the DVX100 all the way through the film and did a great job.

FWIW,

Jan
 
Re: Article on November in new videography

I second Jan's comments. As the digital editing consultant on the film for InDigEnt, the look and feel of the film was controlled throughout the process. We used two complete scenes for demo purposes at Sundance from the original sources. The look and feel were all there as the color correction was used in the Avid to enhance the look that Nancy created during the production. Looking at the raw footage was a great learning experience to see how great a job she did with the camera in the amount of time alotted for post production. Like any good artist, knowing the capabilities of your tools and what they bring to the "canvas" Nancy was in control all the way.

The material was recaptured at uncompressed 1:1 at 23.976p so that Greg could work in uncompressed 4:2:2 for color correction as well as the many effects created in After Effects. He did not want to go through multiple decompress and recompress stages. Once uprez'd to HD, a final balancing and tweaking were done in HD space for the final master.

Michael
 
congrats to nancy and co...

congrats to nancy and co...

I'd love to see November (I was pleasantly surprised by Drew599 who rates "Art History" above November).

The Art History DVD's are shipping tomorrow - so people will be able to see and decide for themselves.

However - I thought I'd add to Michael P's point...as there is no cheap 24p tape based format that can be easily be used in post production - the November post process sounds like it made the same choices "Art History" did. (up until the HD uprez - which we never did)

Namely - 23.976 uncompressed.

Michael I presume when you say "recaptured" at 23.976 you don't actually mean recaptured - unless there is some way to grab 23.976 uncompressed off of DV tape that I am unaware of...you mean you just stripped out the 23.976 and didn't re-encode miniDV ever again in the post process.

Interchange between Avid and Final Cut 4 (which is what we used in art history to edit everything but the closing titles/opening titles - they were in vegas/after effects 6) was in uncompressed 23.976 - color corrected in 4:2:2 uncompressed space on an Avid Symphony

So I'd love to see how they compare...

I wish I'd been at that socal meet...

nick
 
Re: Article on November in new videography

They shot ADVANCED for the production and the offline process was a combination of Xpress Pro for capture at native 23.976p. They chose to capture at 28:1 due to the large amount of footage (5x more drive space that DV25) and they knew they were going to recapture at 1:1 uncompressed. The did this in the Adrenaline which can either take the SDI signal from a DV deck (DSR200) and still remove frames from the ADVANCED pulldown, or capture via FireWire where a decompress happens then is captured to disk at uncompressed 1:1. In either case, it is 4:2:2 on disk with native 24 frames (23.976). Sine the Adrenaline can mix all codec and resolution types in real time, Greg could keep the offline at 28:1, while do all the effects at 1:1. Over time, the no effects elements got recaptured till the whole timeline was at 1:1 for final color correction etc.

They could have easily moved the entire project and media to a Symphony if they wanted that type of color correction.

There are other workflows available depending on the type of program. A no efefcts story dirven film could be done entirely at DV25 with similar looking results.

Michael
 
cool...

cool...

it's the SDI from a DV deck step that I'd never considered.
I assumed (hey it gets me every time) that since the files are encoded on tape as with DV codec there has to be a decoding step to get them to be uncompressed.

Even if that decoding step happens in hardware as far as I know you can't decode or extract "better" frames than are actually on the tape.

so if the capture by firewire (standard ohci 1394 - not via sdi) is effectively a straight file copy from the tape...why would capturing via SDI give you any better quality ?

obviously the next step after firewire capture has to be straight cuts or uncompressed out...but there has to be a dv codec DECODE somewhere - that step is unavoidable...

nick
 
Re: Article on November in new videography

Yes, the first step is unavoidable, it's all the following renders and being in 4:2:2 color space that helps keep up the quality from the original capture. It can never be better than the original capture, but this workflow allows it to be maintained as much as possible.

In the Avid Adrenaline worlflow, the capture can be done via SDI from a DV deck, or capture via FireWire directly with the decompress happening in the Adrenaline and a capture as uncompressed in the Avid. The Adrenaline allows capture on the fly with ADVANCED pulldown (flags or not) to any resolution supported in the application.

Michael
 
Re: Article on November in new videography

[quote author=MichaelP link=board=nws;num=1079584115;start=25#32 date=03/22/04 at 04:46:14] which can either take the SDI signal from a DV deck (DSR200) and still remove frames from the ADVANCED pulldown, or capture via FireWire where a decompress happens then is captured to disk at uncompressed 1:1. In either case, it is 4:2:2 on disk with native 24 frames (23.976).[/quote]

I reread this a bit more carefully...so the two methods are essentially the same...ok...

Thanks michael - when we were doing this I kept wondering whether anyone else was bothering to go to the length we were on the post - and now I know...

Glad to be in that level of company.

nick
 
Re: Article on November in new videography

Yes, the files on disk end up being the same, but by having Adrenaline to the decompress you avoid a more expensive deck that has SDI for DV. You can just use any DV deck with FireWire.

Michael
 
Re: Article on November in new videography

There could be one point made here, even though I'm not sure it would make any difference - But hey, that's just me... Sort of pointless at times :D.

If you do use SDI capture instead of firewire, you will be using the same physical (in-camera) hardware codac to decode that was used to encode.

If you use firewire to transfer, you will be using a different (system) codac to decode.

Now the question is, will the camera do this decode better since its using the same codac used to encode? If it is, is s-video out going to introduce more distortion or in anyway limit effective quality.

Just something to ponder.

-Rodger
 
Re: Article on November in new videography

S-video will definitely compromise quality. The video on tape is recorded as component video. S-video blends the chroma signals together into one line, separating luma from chroma but blending chroma. Once you blend that chroma you can never recover the original component signal.
 
Is there a list...?

Is there a list...?

Does anyone know if someone is keeping/posting a list of features that were shot on the DVX100, or on MiniDV in general? It could certainly be useful as a point of reference benchmark for our collective aspirations.
 
You must log in or register to reply here.
Back
Top