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 :)

JVC DLA-RS35 for FILMOUT GRADING?

Hi Ido,

We just finished a 1 minute spot for theatres with this projector and Lustre, at the lab and they printed on celco. Worked really well. But I think it would be better to get the input from somebody who uses it at their own facility.
 
Ido Karilla said:
No DLA RS 60 experience around?

I will have to gamble and buy?

Thanks.
Click to expand...

Ido,if you can wait a few weeks then i believe there will have been a review from the Uk site and quite a few people will have installed and posted their own critiques by then,i think JVC missed getting the PJ`s out for christmas.As you can see if you look at the forums it`s reports from the one fellow who got his hands on the X7 (the mid range one for non Uk readers).Personally i`d hang on for a while if you can.Do you definately have to gamble and buy right now?..that is ,if you could actually get your hands on one of the new range..
If i learn any more i`ll post an update..
 
Last edited:
I'm awaiting delivery of an RS-60 but it'll be a couple of weeks before I've got it fully set up. Considering the previous models have successfully been used on many feature film projects I'm pretty confident this will work out fine.
 
Barend Onneweer said:
I'm awaiting delivery of an RS-60 but it'll be a couple of weeks before I've got it fully set up. Considering the previous models have successfully been used on many feature film projects I'm pretty confident this will work out fine.
Click to expand...

Hi Barend, did you successfully set up your RS-60?

Did anyone learned any thing about its behavior under 10 IRE?

Thanks.
 
Barend Onneweer said:
Hi, I'm taking delivery of the RS-60 this week, should have it fully set up with Cinespace within a couple of weeks.
Click to expand...

Thanks, which probe will you be using?


Do not forget to update. I want a new setup as well.....

Thanks.
 
Also, are you aiming to calibrate based on rec709 primaries or are you aiming to squeeze the projector to as close as posse Le to dci spec?
 
I'll be calibrating using Cinespace and the Hubble. The idea is to calibrate to P3 DCI colorspace.

I'll be sure to keep you posted once everything is installed tweaked.

Barend
 
Hi Barend,

Do you have any news for us?

Thank you.
 
Hi everyone,

I wanted to wait with the update until I had everything sorted out and tested.

Here's my setup:

I'm running SDI out of Scratch into a Cinetal Davio.

The Davio converts the SDI signal to 10-bit HDMI into the JVC RS-60.

The RS-60 projector has an optical colour filter that brings the gamut closer to DCI - if you set colour mode to "Cinema 2". You can hear the filter click into the lightpath when switching between colour modes.

I set the RS-60 to the 6500K preset, Cinema 2 mode, high-lamp mode and got around 15 ftl on a 2.4 meter wide screen (matte white).

Then I used the Hubble probe with Cinespace and the Davio to create a profile for the projector. Using the profile I created a 3D LUT to get as close as possible to P3.

With these settings the JVC doesn't achieve the full P3 gamut, but it's really close. The most saturated greens and blues are out of range, but it's really rare to see colours that saturated in live-action material.

I'm currently using a 3D LUT that keeps the colours most accurate, which drops the white point to around 12 ftl. The DCI target is 14ftl.

Cinespace also has an option to do whitepoint adaptation which will retain a brighter image, but mathematically less accurate colours. I haven't tested this option yet.

After three weeks of grading a feature film we created a 10 minute test sequence for viewing on a DCI certified Barco. I'm just back from watching the reel with the cinematographer and director and the DI supervisor. The colours matched perfectly. The Barco projected a slightly brighter image (14ftl) on a larger screen so the image looked a bit more punchy. And honestly the image projected by the Barco looks slightly more stable and sharper. But the colour grade matched perfectly and there's absolutely no need to adjust or tweak the grade. The blacks and shadow detail matched really well.

So in short, we were able to grade in P3 colour space on the JVC RS-60 and the grade carried over to the Barco extremely well. Pretty cool huh?

Barend
 
Very cool ,

Thank you so much for your answer, Barend.
What distance from the screen is your projector ? What brand is your screen?
So...the Huble performed well...I was thinking about the Klein ?
 
Barend Onneweer said:
After three weeks of grading a feature film we created a 10 minute test sequence for viewing on a DCI certified Barco. I'm just back from watching the reel with the cinematographer and director and the DI supervisor. The colours matched perfectly. The Barco projected a slightly brighter image (14ftl) on a larger screen so the image looked a bit more punchy. And honestly the image projected by the Barco looks slightly more stable and sharper. But the colour grade matched perfectly and there's absolutely no need to adjust or tweak the grade. The blacks and shadow detail matched really well.
Click to expand...

Although I'm not surprised that the colors seemed similar, I would also point out that unless you were looking at both projectors side by side with a split between them (commonly referred to as butterfly projection), there's no real way to say that the color matched perfectly. Human color memory is not accurate enough to make that assumption. That's why butterfly projection is often done by major DI companies to ensure that at least the first print is as close a match to the digital version as possible. Now, having said that, saying that it looked similar enough that everyone was satisfied that the creative intent was being maintained is a fair statement.
 
Mike: that's fair :-) I've got no way to measure the absolute difference in colour rendition between my setup and the Barco reference setup and the only definitive test would be a butterfly projection like you describe, but that wasn't an option for now. So we saw a selection of shots from each scene in the film, and together with the cinematographer, director, myself and another colorist we subjectively judged that what we saw matched (if not perfectly then at least very closely...) what we had intended to do :-).

And honestly, the JVC is no match for the Barco in terms of picture quality, detail and stability. There's something about the Barco that is really smooth and subtle. But everything I did on the JVC carried over really well, and that's enough for now :-)

As for my screen: it's a Projecta Tab-tensioned matte white gain 1.0. I think Projecta uses Da-Lite screen material.

I'm using the Hubble for profiling. AFAIK it's as accurate as the Klein K-10, but the Klein is a lot faster. When setting up the projector you can do quick ramps that take two minutes. But a full 3D cube takes between 3 and 4 hours. I think the Klein K-10 does it in less than an hour. But it costs a lot more. Incidentally I saw that the Barco at the high-end DI facility is also calibrated using the Hubble.

I'll check black level tomorrow but I think it was well below 0.05 cd/m2. The black level and black detail of the RS-60 is really impressive. There were a lot of night scenes in this film, shot with available streetlights. So judging subtle shadow detail was very important. And those decisions carried over to the Barco setup extremely well.

Projector is about 4 meters away from the screen.

I haven't calibrated to a custom film profile yet since this project will probably have a fully digital release. P3 is a lot wider gamut than most film stocks so I'm less worried about that. I can do some tests with the generic film profiles in Cinespace. If it doesn't need the DCI optical filter in the projector then the light output would be higher.

Barend
 
Lightspace is much faster at profiling with the Hubble than what Cinespace does.
Cinespace will take 2.5 hours while Lightspace does it in 50 minutes
 
Cinespace does a Quick cube, too - which takes 50 minutes. A quick cube is 1000 colour patches and a full cube does 4000 colour patches, which results in a more accurate 3D LUT.

Barend
 
Barend, can you verify the 0.05Cd of blacks?

That's 1000:1... for the 14ft/L

Also I do urge my clients to grade for a real film target and then use a conversion LUT to get to P3 DCI... its much better the result...

Also guys, thing about the chances to go to film for long term storage...

An LTO tape can't be storeed and ignored for decades, as a negative can...

I have a strong feeling that in 30 years most of our movies originals will be lost if they stay in digital domain...

A good 4K film recording is a very cost effective alternative... but this is not an easy option when the film is graded for DCI only...

If you try today to store a 2Tera bytes 4K data compressed files in a certified cloud storage provider for just ten years you have to pay more than 60K$...

Try to thing in 50 years will it be available a tape streamer? Film will always be possible to scanned...
 
A 10 point cube is seriously more than enough - and the 1000 patches used with LightSpace is quicker than with other systems, depending on the probe in use, especially when using Display Independent Profiling.

LightSpace can actually use any size cube from a 3 point cube (3x3x3 - just for a rough check) to a 17 point cube at 17x17x17, so 4913 patches.

However, the variation in accuracy from 10 points to 17 points is rarely worth the extra time. I have only only used a 17 point cube once, and when I then did a 10 point profile of the same display the difference was impossible to see.

Also, P3 is NOT a larger colour space than film. It is larger than Rec709, but it is still smaller than film. More importantly the 3 main colour spaces occupy different areas of colour space, so it the areas of overlap that are important. Using gamut as the main definition of colour space Rec709 has an approx 50% overlap area with film, while P3 is approx 75%.
 
Last edited:
You must log in or register to reply here.
Back
Top