HD For Indies

Wowsers. Can't wait to hear more Mike. Red's gonna rock Hollywood !
 

Wowsers. Can't wait to hear more Mike. Red's gonna rock Hollywood !
 

oh my god
 

Thanks for the report Mike. I'm just picturing clipped highlights shooting a white car with a video camera... now with Red I'm picturing something else.
 

PICTURES??
 

Awesome.
 

Sounds great, but as always, I say let's not drink the coolaid so fast that we choke on it.

Good to hear this test footage was @$&#@@$%@ awesome, but as described, this test was uncompromised output from the full frame of a chip, not the output of a camera.

When the circuitry that can control that wonderous chip finally fits inside their itty-bitty techno-groovy aluminum camera body -- and the beautiful (but reduced) image completes it first marathon run through the thousand electronic miles of still-being-designed wiring, processing and codec magic -- then we might have a preview of the imaging capability of this camera.

Hope that's soon. I totally love this flavor of coolaid, but it makes me pee too much.
 

Being the owner of the 89th cup of Kool Aid, I'm obviously very excited to hear about it looking so good.

I'm also aware, however the difference between the DVX and the DVX with andromeda which taps you in closer to the sensor. So - I'm keeping a perspective until we see some footage from the camera. Actually, I am hoping I'll be able to shoot a little dramatic material for RED as a test during development.

But all the reports sound exciting of course.
 

There's no reason why the footage from the camera won't look a lot better than the footage you saw today. What you saw was the raw workflow, in software and hardware prototype. The Quvis media server compression is probably less efficient (i'm only guessing, but it's an educated one) than REDCODE RAW.

Remember, RED is not a video camera, it's a digital cinema camera, and all that that implies.

And yes, the circuitry can fit inside....
 

Mark - you are correct that those are two different challenges, but considering that the image comes off the CMOS digital, and then it is either RAW passthrough over a serial bus or into the compression engine, that bodes well for maintaining quality. Next challenge - how good is the compression from the codec? Pretty darn good. More soon, stand by, typing....

-mike
 

Mike, any hints on the cost and resolution of the viewfinder?
 

Viewfinder is full HD, and a lot cheaper than you'd think. It's pricing fits in with the whole RED philosophy.

Graeme
 

Wow, that was fast. None other than Mr Nattress is taking the time to answer my inane viewfinder question! I feel good now.

Bruce
 

mike,

ha ha you supposed to be first time in Europe and sounds like American newbie tourist ha ha i could understand but Europe is a mother of the most things to touch us everywhere so do not be surprised so much because of the differences etc...

i would like to see a pictures of 4k red show up in theater etc

regards

s
 

Thanks to :- Chris Seivard,Alan Thatcher, Timothy Sassoon, Chris Rasmussen & Jacques Haitkin for a contribution to the running costs of CML __ [steve shaw] Still not sure what setup was actually used (camera kit or post-production), but they were real moving images.

[lucas] Steve - if you noticed the credit page... all the grading for the IBC presentation was done on SCRATCH. I've been working with the RED team for a little while now on those images and on preparing them for IBC. All the images shown at IBC went through SCRATCH at 2K, 4K, and (yikes) 5K.

The post path to get those images was:

1) Record off the camera and onto disk.

James Masters - part of the Oakley/RED production team - was the DP for the shots. I actually saw most of them being done, so can attest to the the fact that they were recorded directly off the camera and onto disk. There was no "funny business" with any of the shots. They are all pure RED. Everybody was a little nervous lighting around one of Jannard's 959s (the white porsche at the end of the sequence) but other than that, it was all good. I know that most of the shots were through a Cooke S4, but I'm pretty sure the girls with cigars were the RED lens.

2) Debayer / color matrix images

The original images off disk were greyscale 16-bit TIFF at 4980x2560. The sensor is Super35mm -- 1:9 at 5K. After processing the images, we had 5K RGB 16-bit TIFF.

3) Downrez to 4K

Although we could have worked at 5K, we knew this was destined for the Sony4K. So we downrezzed to two sets of files - 4096 x 2160, and 2048 x 1080. The IBC Theatre is 4K. The SCRATCH system currently in the RED booth at IBC is showing 2K - so we needed both sets.

4) Grade, grade, grade.

We graded all the images at Hollywood DI on the old Warner Hollywood Lot through SCRATCH on two different systems - a Barco 2K (Neil - pipe in here with the exact specs,) and a Sony4K. All the initial grading was done in 2K - because it was just easier that way - less storage, less transfer time, less hassle, etc, etc.

Once initial grades were completed in 2K, we then transferred those grades to the 4K images. We looked at the 4K through the Barco, but also loaded them up on a Keisoku 4K DDR (courtesy of Asia Media Products - thanks Grady!!) that was playing through the octal-SDI links to the Sony 4K (provided by Keycode Media - thanks Mike!)

Jim (Jannard), Ted, Jarred, and James from the RED team all supervised the color sessions and made final decisions. The colorist was Bob Gill from FilmworksFX in Santa Monica.

I must say - I'm very impressed with the RED color team. From the time we started this process until the time we did the final images - their debayer and color matrix process advanced very fast. Smart guys. The last images they gave me were worlds better than the first.

Something important for everyone to know - from the final 5K RGB images pre-color to the rendered images at IBC - with the exception of one shot (the girl blowing the bubble at the beginning of the sequence) the only color processes we did to the images were Lift (master) down, Gain (master) up, Saturation up, Lift (Blue) down. That's it. No shapes, keys, or secondaries. The RED team has done a very nice job with their debayer/matrix algorithms.

5) Approve and Render

Once final approvals happened, we rendered to a final set of 4K (4096 x 2160) DPX files. Those DPX files went to QuVis and Keisoku. QuVis went through their process to crunch the files for display. With Keisoku, it's just dragging and dropping file sequences, and it plays back uncompressed.

I believe that at IBC, playback is happening through QuVis, but I am not certain of that.

If anybody has any other questions, please feel free to ask. I've been under a pretty savage NDA, but now that the IBC press event is done, RED has blessed my talking about the workflow... If anybody at IBC wants to take a closer look at the images, they are on a SCRATCH system in the RED booth, as well as on the SCRATCH systems in the ASSIMILATE suites at the back of Hall 7.

I also have the images in LA, and once the RED team arrives back from IBC, I know that they will be setting up screenings. Probably the first screening will be for the reservation holders... :)


Lucas Wilson
 

I would like to see a video of the faces and the reactions of the public, it would be even more moving... Specially if it points the ones of the reps of other companies like, said... Sony.
 

Thanks to :- Chris Seivard,Alan Thatcher, Timothy Sassoon, Chris Rasmussen & Jacques Haitkin for a contribution to the running costs of CML __ [steve shaw] Still not sure what setup was actually used (camera kit or post-production), but they were real moving images.

[lucas] Steve - if you noticed the credit page... all the grading for the IBC presentation was done on SCRATCH. I've been working with the RED team for a little while now on those images and on preparing them for IBC. All the images shown at IBC went through SCRATCH at 2K, 4K, and (yikes) 5K.

The post path to get those images was:

1) Record off the camera and onto disk.

James Masters - part of the Oakley/RED production team - was the DP for the shots. I actually saw most of them being done, so can attest to the the fact that they were recorded directly off the camera and onto disk. There was no "funny business" with any of the shots. They are all pure RED. Everybody was a little nervous lighting around one of Jannard's 959s (the white porsche at the end of the sequence) but other than that, it was all good. I know that most of the shots were through a Cooke S4, but I'm pretty sure the girls with cigars were the RED lens.

2) Debayer / color matrix images

The original images off disk were greyscale 16-bit TIFF at 4980x2560. The sensor is Super35mm -- 1:9 at 5K. After processing the images, we had 5K RGB 16-bit TIFF.

3) Downrez to 4K

Although we could have worked at 5K, we knew this was destined for the Sony4K. So we downrezzed to two sets of files - 4096 x 2160, and 2048 x 1080. The IBC Theatre is 4K. The SCRATCH system currently in the RED booth at IBC is showing 2K - so we needed both sets.

4) Grade, grade, grade.

We graded all the images at Hollywood DI on the old Warner Hollywood Lot through SCRATCH on two different systems - a Barco 2K (Neil - pipe in here with the exact specs,) and a Sony4K. All the initial grading was done in 2K - because it was just easier that way - less storage, less transfer time, less hassle, etc, etc.

Once initial grades were completed in 2K, we then transferred those grades to the 4K images. We looked at the 4K through the Barco, but also loaded them up on a Keisoku 4K DDR (courtesy of Asia Media Products - thanks Grady!!) that was playing through the octal-SDI links to the Sony 4K (provided by Keycode Media - thanks Mike!)

Jim (Jannard), Ted, Jarred, and James from the RED team all supervised the color sessions and made final decisions. The colorist was Bob Gill from FilmworksFX in Santa Monica.

I must say - I'm very impressed with the RED color team. From the time we started this process until the time we did the final images - their debayer and color matrix process advanced very fast. Smart guys. The last images they gave me were worlds better than the first.

Something important for everyone to know - from the final 5K RGB images pre-color to the rendered images at IBC - with the exception of one shot (the girl blowing the bubble at the beginning of the sequence) the only color processes we did to the images were Lift (master) down, Gain (master) up, Saturation up, Lift (Blue) down. That's it. No shapes, keys, or secondaries. The RED team has done a very nice job with their debayer/matrix algorithms.

5) Approve and Render

Once final approvals happened, we rendered to a final set of 4K (4096 x 2160) DPX files. Those DPX files went to QuVis and Keisoku. QuVis went through their process to crunch the files for display. With Keisoku, it's just dragging and dropping file sequences, and it plays back uncompressed.

I believe that at IBC, playback is happening through QuVis, but I am not certain of that.

If anybody has any other questions, please feel free to ask. I've been under a pretty savage NDA, but now that the IBC press event is done, RED has blessed my talking about the workflow... If anybody at IBC wants to take a closer look at the images, they are on a SCRATCH system in the RED booth, as well as on the SCRATCH systems in the ASSIMILATE suites at the back of Hall 7.

I also have the images in LA, and once the RED team arrives back from IBC, I know that they will be setting up screenings. Probably the first screening will be for the reservation holders... :)


Lucas Wilson
 

And Red just got a new page - NICE LOOKING! All this news is incredibly exciting and unbelievably refreshing! Give Jim and his team a big pat on the back Mike! They deserve it. Now let's get that camera!

Yash Keough