RED #8 testing team & some details

[Bruce Allen]

I don't think it make that huge of a difference since you need all three colors to key material.

No you don't. If you were keying, say, a red bucket against a blue screen versus against a green screen, you'd be getting keying info from only your red and blue or red and green channels.

In which case you'd have much higher resolution on your Bayer sensor if you shot against green.

A green color has an absence of red and blue... you need red and blue information to determine that absence.

In practice, I'd like to see greenscreen material with shadows on the greenscreen (e.g. folds in the greenscreen material, and cast shadows).

You're going to spend $18,000 on a 4k camera and not fix the folds in your greenscreen? Okay, okay, I agree, the world isn't perfect (just look at my own webpage for a folded greenscreen frenzy). And shadows are sometimes unavoidable...

Shadows tend to have more noise and therefore be difficult to key.

I think we agree there...

Bruce Allen
www.boaciema.com

 

[Ketch Rossi]

Sorry for the delayed response Ketch!

Thanks for that offer of grip and glass - much appreciated. I'm sure we could use that. I'll definitely discuss your offer with the other team members. We'll analyze the space requirements at the testing studio and also the aggregate list of tests we'll be doing, and then I'll get back to you sometime soon via private message.

Thanks again...

No reason to apologies Gibby,

what you and the guys are to do is much so appreciated, so if I can help in any way, glad to do so.

I imagine that money is not at all an issue for this planned test but in any case let me know.

Ciao,

KETCH ROSSI
www.KETCHFRAME.com

 

[Gavin Greenwalt]

I wouldn't be so hasty to say we'll only have greenscreen footage. We might only have Bluescreen footage. It'll partly come down to what our facilities have. Which is still an unknown.

One thing I would like to avoid however is shooting 'bad' greenscreen footage. (Definition of bad: rotoscoping shots.) There is a point where it's just testing the compositor not the camera.

 

[Michael Schrengohst]

You forgot about the drop test.
How well will the RED survive a
4 ft drop on to the concrete floor
of the studio from the greasy hands
of the first A/C?

 

[david farland]

Red Guy....i think you just talked your way into the test team...when does your camera arrive?

Test team....(including their newest member....Red guy)

I’m sure you’re already thinking of this, but I’d document the test setup well, as Bruce mentioned, so if there was a problem you’d have a better chance of going back to it weeks after and redoing it or maybe someone else could.

For the first few months these test results may be major source of information on the Red One camera. The results of one camera’s erroneous setting, internal fault etc, could cause some bad results and talk. I don’t know if you’re thinking of doubling up on cameras for certain tests, but if you use a second roaming ‘safety’ camera that could take a sub-sample of tests, it would provide a greater level of security, accuracy. For example for latitude tests, the second camera may take 3 stops, upper, lower and middle.

With that in mind and all respect to Gibby and the team (so much!) & Jim, if I were Jim, I’d be surprised if he doesn’t…. I’d want to get my own set of controlled test results out and release them when the camera is released…, running them over weeks in a controlled environment, showing the best of what the camera can produce. We all know with the luxury of time, repetition etc, some things will improve.

Cheers,

Dave,

 

[GlennChan]

Jeff and Gavin: Cool. I wouldn't be too bothered if you didn't shoot both green and blue to see if a particular color was better. Because even if you do shoot it, some people (like me ) could potentially complain that you didn't use the most saturated green/blue possible (unless both keyed well).

The other thing is that you don't want to over-extend yourself by doing too many tests.

2- Speaking of which, it may (or may not) be good to include other cameras for comparison. The SI-2K for example is very comparable to the Red (similar price, workflow, target market) and shouldn't introduce too many variables into the testing (low # of menu settings). If you can get it. Or the Dalsa, Viper, or F23 (or D-20?). If you want to get fancy, you'd throw in film or other video cameras (F900, Varicam, etc.). But then you run the risk of over-extending yourself. (EDIT: Sorry, missed Gibby's post on this. The plan is to test Red only.)

Putting another camera in there might help put some perspective on the results. Some tests aren't standardized (e.g. aliasing / zone plate, exposure latitude / noise) so the results by themselves may not be that informative without comparisons to other cameras.

3- For S&G, could you do a difference key test? It might be an interesting use for Red since you can shoot virtual sets with perfect spill. Or, a greenscreen where the green is fairly dark. I'm guessing with the Red you can get away with it since the images so far look fairly noise-less (even with Redcode compression).

 

[Casey Green]

I wouldn't be so hasty to say we'll only have greenscreen footage. We might only have Bluescreen footage. It'll partly come down to what our facilities have. Which is still an unknown.

If it is at all a help, I will be able to bring some Chromatte Kits (LED Light Ring with Retro Reflective Screen) to provide another option for evenly lit Blue/Green Screen without the need for a ton of lighting. Its a great portable tool that's quick to set up and works for specific needs. (not appropriate for all types of setups, but a nice option sometimes).

Gibby, I'll send you a list of other tools I can make accessible for the tests.

Looking forward to it.

 

[Michael Schrengohst]

Red Guy....i think you just talked your way into the test team...when does your camera arrive?

Test team....(including their newest member....Red guy)

I’m sure you’re already thinking of this, but I’d document the test setup well, as Bruce mentioned, so if there was a problem you’d have a better chance of going back to it weeks after and redoing it or maybe someone else could.

For the first few months these test results may be major source of information on the Red One camera. The results of one camera’s erroneous setting, internal fault etc, could cause some bad results and talk. I don’t know if you’re thinking of doubling up on cameras for certain tests, but if you use a second roaming ‘safety’ camera that could take a sub-sample of tests, it would provide a greater level of security, accuracy. For example for latitude tests, the second camera may take 3 stops, upper, lower and middle.

With that in mind and all respect to Gibby and the team (so much!) & Jim, if I were Jim, I’d be surprised if he doesn’t…. I’d want to get my own set of controlled test results out and release them when the camera is released…, running them over weeks in a controlled environment, showing the best of what the camera can produce. We all know with the luxury of time, repetition etc, some things will improve.

Cheers,

Dave,

Darn, I would if I could - Dec. delivery for me.

 

[Jeff Kilgroe]

You forgot about the drop test.
How well will the RED survive a
4 ft drop on to the concrete floor
of the studio from the greasy hands
of the first A/C?

I've already got this one covered. The 4 camera suppliers will draw straws to see who qualifies for the 3ft, 5ft, and 9ft drop tests... One camera will be reserved for shooting the other three tests. :matrix:

 

[Don Woods]

I've already got this one covered. The 4 camera suppliers will draw straws to see who qualifies for the 3ft, 5ft, and 9ft drop tests... One camera will be reserved for shooting the other three tests. :matrix:

Count me in. :innocent:

 

[Häakon]

Anyone able to lend their Canon 10-22 for the LART?

I've got a good selection of Canon glass (including the 10-22) and would be more than happy to bring it to the LART for testing, however because that particular lens is an EF-S lens, I am not sure it will be compatible with the new mount (EF-S lenses are made specifically for 1.6x FOVCF dSLRs). What I expect is that the mount will only work with standard EF lenses, in which case the widest zoom lens available is the newer 16-35mm f/2.8L II USM. However, on a full-frame dSLR, this lens provides the same field of view as the 10-22 EF-S does. There are also 14mm f/2.8L USM and 15mm f/2.8 (fisheye) primes available for full-frame cameras.

Either way, I will make my entire inventory of glass (and anything else I own that the team may find useful) available for testing when we get underway.

 

[Mark Crabtree]

EF-s

EF-s

My understanding is that the Birger mount is EF-s and so will accept both the EF and the EF-s lenses.

 

[Steve Gibby]

Thanks for all the additional suggestions on this thread over the past few weeks! I’ve been busy and mobile, but I’ve been lurking here when I could.

If you really think about it, LART will simply be a diverse group of guys in the motion media industry getting together to check out some new toys – RED One cameras, with a variety of lenses and accessories. It’s not meant to be a media circus or Rocket Science 400, but simply a hands-on opportunity for some of us to get a first look at RED One. Over the months after RED One ships I would guess that RED One will be subjected to more tests by others who are interested in drilling down into additional details of its performance. That’s not LART. We only have 2 days to test. It will be fun to hang out with these other guys and share thoughts on, and impressions about RED One.

When all is said and done, some of the key bottom lines with any new motion media camera system are:

1) The aesthetic qualities of the images it generates – the real bottom line.
2) Its ease of integration into field and post workflows – or ability to generate new field and post workflows as needed or desired.
3) Its flexibility, utility, and scalability.

There are many other “bottom lines” that could be listed, but IMO these three are the “biggies”.

Hyper-cerebral, left-brained analysis of a camera system yields performance facts and figures, assuming that the tests are conducted over a period of time in controlled environments, are repeatable to check results, and reflect the pragmatism of the scientific method. LART will not have the time, facilities, crew, or mission to undertake testing like that. What LART can do, within the limits of the time, facilities, and crew we have, is to give us an overview of a “first look” at shipping RED One cameras, with various lenses and accessories. The results of the technical tests we do will be interesting no doubt, but again, to me the key “bottom line” from the LART testing will simply be the aesthetic qualities of the images we will be able to generate using multiple lenses, in natural light and artificial light.

I would guess that RED is getting valuable feedback on the beta testing of RED One cameras with several high-profile production crews. That beta feedback will undoubtedly be used by RED to tweak the parameters of the camera system before it’s finalized for production.

I don’t think any of the LART team members are expecting to test a perfect camera system. Absolute perfection is an elusive butterfly that is essentially unobtainable in the electronics industry. But I think the LART team will be real happy to get together as friends and associates and play with these new toys for a few days – and in the process generate some images, impressions, and opinions that we’ll share with the rest of you. The LART team is made up of guys that I’ve had debates with on RED User and elsewhere – we haven’t always agreed on issues and details about RED. But it’s this cross-pollination of viewpoints that should give LART a balanced collective view of the testing sessions.

Again, these LART testing sessions will be exploratory - but also fun and casual. What started as me just getting a few guys around #8 to check out the camera, has grown into a bigger crew and multiple cameras. But the vibe will be the same – a few guys getting together to check out some cameras. When I produce/direct my professional productions, my set/location is always an enjoyable thing for the crew. I surround myself with a group of people who are real good at what they do, and we all then go out and work together to bang out a good product – and we have fun along the way. My approach to LART has been the same – I’ve surrounded myself with people who are good at what they do – and now we’ll jump in and have some fun!

 

[Jarred Land]

is it time for a seperate LART forum?

 

[Steve Gibby]

is it time for a seperate LART forum?

I think that's a great idea Jarred...a central area where RED User members could go to get info on LART. RED User is point central for people interested in RED to get info and exchange thoughts. A dedicated LART forum would be very cool for them as a central place on RED User to get info on LART.

 

[PaulClements]

I bought www.redwiki.org domain name and setup a Wiki on the website. If you visit it you can see that it is still quite unfinished and requires plenty of input. I simply haven't had time to put in all the information as yet.

Looking at the posts above though perhaps the information from LART could be included here. The main point of the wiki was suppose to be something that people could point their clients to in order to have all the information in one place and educate them as quickly as possible.

At the moment it's on a server that I've been having probs with the last couple of days so you'll have to bare with the speed if you see any probs - it will be improved. If anyone wants to help out putting up information about RED please feel free to signup. Or if the LART team think it could be a useful tool for including their data I'd be more than happy to help you input it in a systematic manner with tables, graphs, diagrams etc as the information from the testing comes through, otherwise you could do it yourself Gibby.

PS, this is not to take away from RedUser Jarred, I just thought maybe it'd be a better way of listing the various information that LART will create than listing it in a forum. If either you Gibby or Jarred would like Sysop control over the Wiki I'd be more than happy. I don't have the time at the moment to do any work on it. I'd also be happy to give you control of the domain name and contents Jarred if it means the data could be linked to RedUser and have a greater reach than me trying to advertise it on my lonesome.

Cheers

Paul

 

[Gavin Greenwalt]

I wouldn't say there isn't time and personnel to do some empirical evaluations of the camera...

Unless of course you mean you want those evaluations to be accurate.

 

[Steve Gibby]

Empirical and accurate would be way cool :nerd:

For those scratching their heads for a definition of empirical, here it is:

Main Entry: em·pir·i·cal
Pronunciation: -i-k&l
Variant(s): also em·pir·ic
Function: adjective
1 : originating in or based on observation or experience <empirical data>
2 : relying on experience or observation alone often without due regard for system and theory <an empirical basis for the theory>
3 : capable of being verified or disproved by observation or experiment <empirical laws>
4 : of or relating to empiricism

:wink:

 

[Steve Gibby]

Paul,

Thanks for that generous offer. If Jarred feels that is something he would like to do I would think he would contact you.

Thanks again...

 

[Michael Hastings]

Birger mount works fine with ef-s lenses 10-22

Birger mount works fine with ef-s lenses 10-22

I've got a good selection of Canon glass (including the 10-22) and would be more than happy to bring it to the LART for testing, however because that particular lens is an EF-S lens, I am not sure it will be compatible with the new mount (EF-S lenses are made specifically for 1.6x FOVCF dSLRs). What I expect is that the mount will only work with standard EF lenses, in which case the widest zoom lens available is the newer 16-35mm f/2.8L II USM.

Birger mount works fine with ef-s lenses like the 10-22, and since the sensor size the S lenses are designed to cover is almost identical to the S35 Red size it should cover it fine. We have been discussing the 10-22 over in the underwater bubble blowers group for about six months as it is the widest lens available at an economical price that has a good chance of working.

By the way, the since the sensor size is nearly identical, Digital Rebels, EOS20, EOS30 and Nikon dslrs should be excellent for getting an idea of how lenses will work with RED.