Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Redcine Cost?

HDCam decks only work with footage shot with a Sony camera, but you don't see Sony giving them away for free!

Just my 2p

Nick

EDIT
OK that's slightly unfair. But you also don't see Sony allowing any system except an Xpri to work with native HDCam material as far as I'm aware. Or if they do, they certainly don't give the licenses away for free.
 
Nick Shaw said:
HDCam decks only work with footage shot with a Sony camera, but you don't see Sony giving them away for free!

Just my 2p

Nick

EDIT
OK that's slightly unfair. But you also don't see Sony allowing any system except an Xpri to work with native HDCam material as far as I'm aware. Or if they do, they certainly don't give the licenses away for free.
Click to expand...


Hi Nick,

I thought Red was revoloutionery!

Stephen
 
Boothba said:
And I'm afraid that I'm going to have to bug Greame again (sorry) about including a 3-2 oulldown function in REDCINE to cine-expand 24p footage to 59.94. Maybe I will be a bastard and post a thread about this soon.

It's a great, great shame - but most editors I know prefer to receive their rushes at 59.94. Some of them don't even really know what 24p is, as they always receive their film footage converted through a telecine to 59.94, or through a dubbing house (for HDcam stuff) to 59.94. I see this as a potential roadblock to the fast and wide acceptance of RED in the film community.
Click to expand...

Pulldown isn't a bad idea if you only plan to deliver for broadcast, but not because the editor wants it.

I say editors that don't know how to work in 24p need to learn and best for them to conform to the a proper workflow instead of the workflow conform to what their whims are.

It's hard enough having to do things fast and accurate, but it's even harder when we have to conform to someone's habits.
 
Might this become somewhat moot point if the AAA editors incorporate some RedCode features themselves? RedCode could be the next DVCPro (you know what I mean)!
 
And I'm afraid that I'm going to have to bug Greame again (sorry) about including a 3-2 oulldown function in REDCINE to cine-expand 24p footage to 59.94. Maybe I will be a bastard and post a thread about this soon.

I already did. :D
http://www.reduser.net/forum/showthread.php?t=151&page=2


It's a great, great shame - but most editors I know prefer to receive their rushes at 59.94. Some of them don't even really know what 24p is, as they always receive their film footage converted through a telecine to 59.94, or through a dubbing house (for HDcam stuff) to 59.94. I see this as a potential roadblock to the fast and wide acceptance of RED in the film community.
Isn't that partially because no SD tape format supports 24p without pulldown? (Or am I misinformed/confused here?)
 
Boothba said:
And I'm afraid that I'm going to have to bug Greame again (sorry) about including a 3-2 oulldown function in REDCINE to cine-expand 24p footage to 59.94. Maybe I will be a bastard and post a thread about this soon.
Click to expand...

Thomas Mathai said:
Pulldown isn't a bad idea if you only plan to deliver for broadcast, but not because the editor wants it. I say editors that don't know how to work in 24p need to learn...
Click to expand...

I disagree about supporting any kind of pulldown in REDCINE, for a few reasons:

The two most popular editing platforms out there now are Final Cut Pro and some flavor of Avid. Both of them support native 24-frame editing, and the Kona card (FCP) and Meridien/Adrenaline (Avid) all support 3:2 insertion when laying off to tape. If someone needs an NTSC or 1080i master for some reason, it can be done on output. Inserting pulldown unnecessarily into the post path before then just causes all kinds of crazy problems. Generate 24fps Quicktimes or file sequences out of REDCINE, and let the editorial package deal with pulldown insertion.

REDCINE - to my knowledge - is just outputting files. So you want to create Quicktimes with pulldown? Why, when both FCP and Avid deal with 24fps native and can do realtime pulldown insertion on output.

The sooner that pulldown and pulldown legacies can be removed from all our lives, the better. Any attempt to extend their lives should be dealt with swiftly and severely. :)

Lucas Wilson
------------
Assimilate, Inc.
Los Angeles
 
GlennChan said:
Maybe I will be a bastard and post a thread about this soon.

I already did. :D
http://www.reduser.net/forum/showthread.php?t=151&page=2
Click to expand...

Thanks Glenn - missed that. I will henceforth post my steaming screed over there. But first....

GlennChan said:
It's a great, great shame - but most editors I know prefer to receive their rushes at 59.94.
Isn't that partially because no SD tape format supports 24p without pulldown? (Or am I misinformed/confused here?)
Click to expand...

Three big reasons:

#1.) As far as I know all SD tape formats (Dbeta, D5, dv, mini-dv, etc.) work at 29.97 or 25 (for PAL.) Not so with HD.

#2). As you pointed out in the other thread, editors are often receiving additional shots from stock footage houses, production libraries, agency archives, etc, at different playback rates. This is particularly true of documentaries, but is also prevalent in film and tv title sequences and commercials. I just finished online editing a Bud Light spot where the director chose to shoot some scenes at 30p to give it a slightly more "now" feeling - this was intentional. One single shot a 30p, 29.97 or 59.94i forces your entire edit into a 29.97 timeline.

#3). I personally find great benefit in finishing in the EXACT broadcast delivery format. The alternative is to let a dubbing house 'fix' it for you, and that can have an ugly outcome. This is particularly important when dealing with post vari-speeds. A post speed-ramp in a 24p timeline (with 3/2 pulldown added at the final step by a dubbing house) can look truly awful.

Frankly I totally agree with Greame's opinion that interlacing is an antiquated pain in the ass. But it is a battle we've already lost as far as North American broadcasting is concerned. Broadcasters are continuing to demand their master tapes at 59.94i, whether that is shot and posted @ 30p, or shot @ 24p with 3/2 pulldown added, or shot and posted in the heinous 60i 'video' format. The smartest move IMHO is (and has always been) to shoot at 24p and add 3/2 pulldown during telecine. This is what all telecine suites do - and so it should be with REDCINE.

For RED, I think this is a case of picking your battles. It is one thing to try to create a revolutionary digital camera. It is another to try to get the entire broadcast world to rethink their SD & HD delivery formats - especially given that HD is in the middle of implementation at considerable expense to Networks, broadcasters, home viewers, etc. If RED or someone else can change the broadcasters minds, I will be the first to stand up and scream BRAVO, but I wouldn't wish that at the expense of the successful integration of their product across this entire frustrating industry.
 
luki said:
I disagree about supporting any kind of pulldown in REDCINE, for a few reasons:

The two most popular editing platforms out there now are Final Cut Pro and some flavor of Avid. Both of them support native 24-frame editing...

The sooner that pulldown and pulldown legacies can be removed from all our lives, the better. Any attempt to extend their lives should be dealt with swiftly and severely. :)

Lucas Wilson
------------
Assimilate, Inc.
Los Angeles
Click to expand...

Hi Lucas

I didn't see your post before replying to Glenn. You make a good argument but I think that my points #2 & #3 above still stand. The last offline editor I know has now (finally) switched over to a 24p compatible AVID, so that is no longer a problem really. I feel like a fool for arguing FOR interlacing, but I am just trying to reflect the market as I see it.

For clarity, all of my arguments relate to broadcast only, and not any other presentation format (film, web, dvd, etc.) It is indeed stunning that SYMPTE (or whoever) stuck with 59.94i as a broadcast format - I wonder if they are still restricted by the 60 Hz power frequency to ensure synch. Regardless, at the end of the day I am just pushing for a pulldown function in REDCINE as an option only, one that will hopefully rarely be used.

I should be skinned alive if anyone thinks I am actually advocating for interlaced presentation. But alas that is exactly what the broadcasters mandate. And I'm not really sure how to deal with them 'swiftly and severely'. :)
 
Frankly I totally agree with Greame's opinion that interlacing is an antiquated pain in the ass.
Definitely agreed here. Unfortunately, we have to live with it. :(

2- 60.00hz was adopted for black&white broadcasting so that ground loops and such would stay stationary (visually, to the viewer) and not look so bad. When color television came along, the color subcarrier would interfere with the signal (um, something like that!). Running color television at 60 X 1000 / 1001 Hz fixed that, and it was backwards compatible with B&W receivers.

My guess about why HD is at 59.97 is so that you don't get complicated timing issues. If you need to simulcast a live event in both SD and HD, either of the feeds will need to discard a frame or hold a frame longer. You'll get a visual glitches every ~33 seconds.
 
I will henceforth post my steaming screed over there.

It might make sense to continue in this thread and let the moderators split our posts into a new thread?

2- A somewhat academic question: What if Redcine can't add 3:2 pulldown?

One workaround would be to layout to HDCAM (SR) tape... although that really defeats the point of a data workflow. So let's look at a data, or mostly data workflow.

Suppose you have a documentary with both 24p Red and 60i sources. You'll want to make offline copies for editing- the convenient DV format won't work since it isn't 24p; if you have the storage, a bigger format like DVCPRO HD may not be a bad idea (or whatever else works well with your editing platform, like DNxHD). Suppose you edit in an NLE that can add pulldown on the fly. Ok.

Rough cut is done, the editor exports a 60i EDL. We assume that the NLE does the correct thing, matching the source 24p timecodes to the project's 60i timecode. We also assume that your NLE's Media Mangler won't have bugs regarding 24p.

(not sure) In the conform for online, I believe the 24p clips will get interpreted as varispeeds/speed changes. This might hurt performance? Ideally, you'd somehow set the online system to recognize the Red clips as 24p and add pulldown on the fly (not sure if all online systems can do this).

I don't know what happens to the audio/OMF.

To get material into the online system, you can either go the data route or a tape route.
Via the data route, you will be limited by hard drive speed and network speed getting material onto the online system's storage. You'll also be limited by the file formats supported by the online system, and the networking protocols and file system formats available (which might be a problem for old SGI kit?). Chances are, the online system will want to see 4:4:4 RGB or 4:2:2 uncompressed Y'CbCr. These are really big bandwidths at HD resolution.

Via a tape/SDI route, you'll need to efficiently lay the material onto tape. If you have a mix of 24p and 60i sources, the 60i stuff needs to go onto 60i tapes and 24p stuff onto 24p tapes. A kludge-y solution would be to checkerboard the project, so that all the Red footage is on two different layers. Cross-dissolves occur where footage overlaps. Output all the 24p Red stuff onto 2 24p tapes. Make an altered EDL that reflects this. So to get material into the online system, you use the altered EDL and capture away.

If you want to avoid the varispeed problem, you can set the deck to add pulldown. However, Redcine needs to reflect this in the altered EDL, making sure to perform the right timecode conversion. It also needs to figure out what reels+clips are 24p Red footage, and what reels aren't 24p Red footage.

Now that I think about it, this workflow is extremely prone to complications (some of which won't be in Red's control).
 
Editing 59.94

Editing 59.94

There are also two very different threads here - we should be careful not to confuse ourselves.

One is you have a 24fps production but the off-line editor wants to operate at 59.94 fps interlaced, and then the on-line conform is at 24fps again - even if the final product is an NTSC show created from a 24fps master tape dub. This is a case where the sooner you can move off-line to the same native 24fps timeline the better.

The second is the RED footage is shot at 24 fps but it is intended for use in a 59.94 fps finished show .. in that case off-line as well as conform editing at 59.94 make perfect sense, but the source timecode from the RED camera needs replacing (because its 24 fps timecode).

Fortunately in Europe they had the good sense to use the same 25fps timebase for film capture and tv timecode, so these issues are moot.
 
I'm SO glad I live in a land without pull-down! Everything at 25fps is just so much easier.

Nick
 
in that case off-line as well as conform editing at 59.94 make perfect sense, but the source timecode from the RED camera needs replacing (because its 24 fps timecode).

I don't think that will work?? Converting the timecode from 23.98 to 59.97 only makes sense if you are to add pulldown. Rob indicated in another thread that Redcine wasn't going to be able to add pulldown.

2- You don't really want to replace the original timecode (it's just a bad idea). What you want to do is to show two timecodes in window burns:
A- The original 23.98 timecode.
B- The original 23.98 timecode converted to 60i timecode.

If you have a 24p tape in a VTR, you can control the deck either via 24p timecode or 60i timecode. i.e. Which set of timecodes do you want to use to control the deck. This behaviour is controlled via the 24p sense setting on Sony decks.

2b- You also need to indicate where the timecodes 'begin', or where their origin is. Because the 23.98 timecode is NDF and 59.97 timecode is DF, they don't stay in sync. DF timecode moves faster than the other. i.e. 01:23:45:00NDF is not very close in time to 01:23:45.00DF.


One is you have a 24fps production but the off-line editor wants to operate at 59.94 fps interlaced, and then the on-line conform is at 24fps again - even if the final product is an NTSC show created from a 24fps master tape dub. This is a case where the sooner you can move off-line to the same native 24fps timeline the better.
This should be the easier workflow if you are shooting Red. However, in a tape/SDI-centric environment, this method is only faster (than the other) if your show is effects/roto-heavy. In a 23.98p project, you don't need to sit there and remove 23.98 cadence for every single clip you need to do specific effects on. However, you generally can't package your 60i broadcast master in a 23.98 project.
 
Editing 59.94

Editing 59.94

Lets start this over.

a) The final product is 24fps (or 23.98) That's common for film outs and also for TV episodic programming that will master to 24p (HDCAM or D-5), and then dub from that to NTSC / 720p / 1080i HD / PAL. In this case my suggestion is to shoot 24fps, off-line edit 24fps and conform at 24fps.

If anyone wants to do off-line or audio working at 59.94 in the middle of this process, good luck. Its not something we can "solve" for anyone, there are processes and fixups - as Glenn describes with stiping dual timecodes on tape - that can and will have to be used if you choose to do this.


b) The final product is 24fps (or 23.98) but you have some archive 59.94 footage. In this case my suggestion is to convert the archive footage to 24fps then off-line edit 24fps and conform at 24fps.


c) The final product is 59.94 NTSC or 720p or 1080i for television or there is a lot more archive footage than new 24fps footage then convert the 24 fps footage to 59.94 fps and edit the whole thing 59.94fps.
 
Stuart English said:
c) The final product is 59.94 NTSC or 720p or 1080i for television or there is a lot more archive footage than new 24fps footage then convert the 24 fps footage to 59.94 fps and edit the whole thing 59.94fps.
Click to expand...

Thanks Stuart. That lays it out beautifully.

For the record my request to include a cine-expand (3/2 pulldown) function in REDCINE (for rushes) relates only to your "SCENARIO C" above. And this scenario does happen - I've online edited at least a dozen projects this year which have included some 59.94i shots (a newscast, a kid's handy-cam, a cell phone cam, a cheap looking video commercial, a 30p product shot, etc.) as part of the intentional aesthetic of the final piece. This forced us into a 59.94i timeline. "Forced" is the wrong word, as there were really no compromises involved, such as working on a 24p timeline which would require vari-speeding all 59.94 footage by 125% to get it in synch, only to re-expand it to 59.94i at the end of the day for broadcast - ugly.

Personally I would NEVER mix up 24p and 59.94i editing timelines. Choose one and stick with it. Each project should be considered carefully by the offline editor, the online editor (and often the director or producer) and the finishing format chosen accordingly. Cheep software like EDL-MAX will easily convert EDLs from 24p to 30p (or 59.94) in case you need to conform your original 24p RED footage (which can then be expanded to fit into your 59.94 timeline.)

My major point is that for "North American broadcast only" projects it is misleading to assume the superiority of a 24p editing timeline, as everything is going to be converted to 59.94i at some point anyway - either in an online finishing suite, or by a dubbing house, or by the broadcaster themselves. I would much rather retain control of this myself at the online stage rather than leaving it to some under-payed midnight Tape-Op with a shitty monitor.

I think we all agree that interlacing is the enemy, which is exactly the reason I would prefer to keep it close. I keep my 24p close, but my interlacing closer! :)
 
c) The final product is 59.94 NTSC or 720p or 1080i for television or there is a lot more archive footage than new 24fps footage then convert the 24 fps footage to 59.94 fps and edit the whole thing 59.94fps.
Click to expand...
I believe both boothba and I are mainly talking about this scenario.

The dual timecode burn is for this situation; the tapes only really have 24p timecode, timecode conversion is done on the fly. You burn both timecodes into the offline copy, in case something with the 60i timecode doesn't work (i.e. you did the wrong conversion). The original 24p timecode is unambiguous and refers to an exact timecode on the camera tape. The 60i timecodes don't do this; they are ambiguous since it depends on what timecode you start dropping from (IIRC).
By 60i I really mean 59.97DF, and by 24p I mean 23.98NDF.

2- Stuart:
Sorry, I missed a nuance in your previous post.
You were talking about:
Footage in 24p.
Offline edit in 60i. (Whereas I was thinking about 24p.)
Online in 24p.

That's not really an issue if you offline edit in 24p. And not a huge issue if you use EDL Max.

Anyways, it's situation C I'm more interested in. It's SCENARIO C where 3:2 pulldown in Redcine (with dual timecode burns) would be useful.
 
Thanks Stuart. I just tried re-reading some of my posts, and I nearly fell asleep - so thanks for staying with us.

Ultimately I just want RED and REDCINE to fit into as wide a marketplace as possible - including broadcast stuff. I think a quick survey of SD & HD broadcast projects will reveal that many, many editors are still offlining @ 59.94 - via pulldown added in telecine or 'dailies down-converts'. Once RED is in the door - so to speak - perhaps we can all work on converting the unconverted to the benefits of a 24p post workflow (and dealing with quirky interlaced particulars at the online stage.)
 
I nearly fell asleep
Click to expand...
The way I see it, when it comes to reading about workflow stuff you're either confused or falling asleep. :D

2- What about in the timecode burn, there is some text to indicate the pull-down cadence. If you don't have automatic cadence detection (like Vegas does so beautifully for 24p, and 24pA), then that at least is the next best thing. Oftentimes, the cadence flags and other things (like date/time for scene detection) don't carry over particular interfaces (i.e. no cadence flags over SDI).

Once RED is in the door - so to speak - perhaps we can all work on converting the unconverted to the benefits of a 24p post workflow (and dealing with quirky interlaced particulars at the online stage.)
Click to expand...
For reasons we previously mentioned, there are still some projects that have to be finished in 60i timelines (or are finished faster in 60i timelines). For ho-hum TV series stuff like cooking shows, reality, etc. you might want to finish in a 60i project just so you can do the packaging at the same time. Commercial breaks, program start and stop times have to be exact to the frame (the broadcasters wield the power to be this picky since they are the client, unlike commercials).

If you need to make a PAL version of the program too, then unfortunately things get more complicated. Too many formats...
 
You must log in or register to reply here.
Back
Top