In development...

[Mark L. Pederson]

The primary advantage to RED is REDCODE. Compressed RAW at over 23.976 fps. This is a core invention of RED that's full effects have not been seen yet.

I think lots of folks overlook the relevance of this.

When I was first told the target "specs" of the RED ONE, I struggled a bit to wrap my head around it - personally, I assumed RED would only deliver half of what they said - and as far as I was concerned, HALF of what they promised would be enough for me to get excited about.

I certainly did not expect REDCODE to work the way it does, with the quality of image it has, with the compression ratio it has.

I was lucky enough to be standing inside the Red booth at IBC 2006, the day BEFORE the exhibition with the Red Team setting up the booth and watched as Graeme and Rob got one of the first alpha's of REDCODE to play back 4K on a LCD monitor - it wasn't in real time, but it was a full 4K debayer - I looked at the monitor for a long moment, then turned to Jim, who was smiling with his signature cigar, and I said, very calmly ... Holy Shit.

And I knew at that very moment, they had lightning in a bottle.

They had the "secret sauce" ... and it was not just ketchup and mayonnaise mixed together.

You see, you can get all the armchair experts (and even actual experts) to post here and debate compression and algorithms, jpeg2000, variable bit wavelets, who "invented" RAW and all that jazz until the cows come home - but it's really about the quality of the image.

High resolution, high dynamic range images in a "hard to comprehend" file size.

A friend of mine sat down with a key Panavision executive recently in Los Angeles. As anyone who is paying attention knows, Panavision has motion picture cameras on MANY episodic television shows and has much to loose if these shows move to RED camera acquisition.

My friend asked "what are you saying when people ask about the RED?"

And his response was "I tell them that they should not shoot on RED because the compression ration is too high - there is too much compression".

The funny thing about that statement is, that is exactly WHY they should shoot on RED.

As we all know - the proof is in the pudding.

And the pudding is going to be on more screens than anyone can imagine so very, very soon.

REDCODE is the real key.

If you really think about it for a while, you can start to see what doors it opens and games it changes.

 

[idyll23]

DSLR talk

DSLR talk

I'm amazed by the thought of a RED SLR but not because of the video potential (though I admit it's a fantastic feature) I'm more captivated by the sensor's fantastic Dynamic Range.
If RED partners with good lens makers that can indeed compete with Nikon or Canon it would be fantastic.
Hell I'd even love to see a partnership with Olympus to use their Zuiko glass!

Keep making these dreams real!

 

[Clint Johnson]

You know what I always wondered about current DSLRs? If the read/reset time is good enough to take video, who needs a mechanical shutter? I know that the mirror is nice for focusing and all, but wouldn't the camera be much more durable, and take far more fps, if you didn't need the mechanical shutter? What if there was a super high resolution live-view viewfinder instead of a mirror reflex system? And wouldn't that be safer anyway because it'd let you see if your sensor had dust on it?

Just thinkin' out loud I guess.

[Edit] By the by, Monstro is due out near the end of 2010, right? And the DSLRish thing is supposed to be using that too? So... DSLR near the end of 2010?

Yup, the average life span of a mechanical SLR shutter is only good for about 200 minutes of 24 fps footage even when using one of the most durable and highly rated cameras - the Canon 1D Mark III at 300,000 actuations.

Single Lens Reflex is not up to cinematic work... if it is a mechanical shutter, it needs to be something really simple and robust... like a spinning disk with a wedge cut out of it.

Mechanical is in its twilight years.

Unless it is millions of nanotech robots- one at each photocell with a selection of filters that it can flip out tens of thousands of times a second. It would have completely opaque, IR-Cut, IR-Allow, UV-Cut, red, green, blue and ND .1 through 3.0. This way, each and every photocell would have its own exposure time and filter... including the ability to take multiple exposures per frame as long as the cumulative time falls within the overall frame rate.

Cell 11847049 might take three exposures using; [48th-second/IR Cut/ND.3/Red], [48th-second/ND.6/Green] and [96th-second/IR-CUT/UV-Cut/Blue]... while cell 06019128 might take a single [24th-second/Blue exposure].

How much "fun" would it be writing the software to decipher that? The metadata would be almost as big as the image data. Red would have to develop a supercomputer A.I. to do it... let's call it Graeme 2.0.

 

[Lauri Kettunen]

I certainly did not expect REDCODE to work the way it does, with the quality of image it has, with the compression ratio it has.

Mark, your post makes interesting points. Perhaps as I had experience of Cineform wavelet compression and had seen what they mean by "visually lossless" compression never had any doubts of REDCODE. Still remember when read the first technical explanation of what Red One was going to be thought that it was simultaneously so clever and obvious that it will have to work out. The only techniacl concern I had was the amount of power the camera will need and the other concern was whether the group will have the resources needed to R&D --at that point I didn't know anything about Jim and his background.

Besides all technical achievements I also find the business model very clever. Everything is going through the web-pages implying low marketing costs. A rather large group of people almost addicted to follow the development process etc. In the end all this results in technically first class equipment with reasonable cost and a very large group of happy people. This is just great and rather difficult to copy.

 

[Eirik Tyrihjel]

I agree with Jim, Mark and many others here that REDCODE is the real magic that makes this camera work practically! Jim & Co could have just made a 17.500US$ 4K camera, but without the genious compression and high quality of REDCODE, Most of us would be helplessly stuck with Terrabytes of data, and no practical way of using it.

I am so happy to be aboard the RED train, my clients love the images I give them, and there is so much to come!

Before RED I would always miss film when shooting digital, after I got RED I stopped looking back, I want to use RED - it´s good, it´s practical and it´s convenient.

 

[defenderoftheo]

We see the future very clearly... trust us (he said)...

Jim

That you do! I can't wait to see a RED DSLR, especially if the pricing is as competitive as the pricing on other RED products. Keep up the great work Mr. Jannard!

PS. When you start needing sales reps let me know. =)

 

[Bruce Allen]

Hi Bruce,

I beg to differ. Red Raw is for me what brings it all together. Without Red Raw, what would you record to ? What would you edit on ?

I'd be fine with them downsampling to 4:4:4 10bit 1080P and recording with a HDCAM SR-like datarate. Or if they talked to Apple or Avid and made a higher-datarate version of ProRes or DNxHD - and integrated a ProRes or DNxHD encoder like the AJA IO box, Ikegami editcam, etc do.

Anyone who doubts that'd be sufficient quality hasn't done many filmouts IMHO.

Couple this with a "RED playback deck" - eg the circuitry (without the sensor part, scaler, audio inputs, etc) for $4000. It would output decent-quality 1080p HD-SDI (PsF if needed) and also good old Standard Def.

Smart people like us would do the file-based thing. Clients who demand realtime legacy workflow could just rent the playback deck.

Imagine the Red One without Red Raw. You would need high bandwidth, ie. expensive and cumbersome RAIDs to record. Then you would need even more expensive and cumbersome hardware and software to play back that data.

On the contrary, you'd have a workflow that would plug right in to everyone's current systems the day it was released.

And you'd have a camera that records beautiful images with 35mm DOF using PL or still camera lenses for much less than any of its competitors.

And you'd still have an awesome RED company with great community, customer service, etc.

If you want, offer an upgrade to a compressed RAW recording mode. But I don't think it benefits people much now.

Anyone want to give a real-world example of a paying project where it was critical for them to have 4K instead of 4:4:4 1080P?

I can give many examples of paying projects where no transcode time, easy 1080P output, a cheaper playback deck, etc would be a plus.

As to Raw compression being revolutionary, well maybe others had thought about it, maybe even others had done it, I don't know. But Red are the first to offer a camera and workflow using it, and that I can afford, and that is revolutionary from my point of view, and particularly for my business practice. It has allowed me to step up from HDV and DVCproHD and enter a whole new world, and a beautiful place it is !

Agreed it's beautiful, but a HDCAM SR class codec would have been fine too IMHO.

Bruce Allen
www.boacinema.com

 

[Jon Howe]

Echoing many of the feelings that Mark and many others have expressed, I too have been more than impressed by the ethics, integrity and sheer technical genius and scope of vision that the RED team has exemplified. I've learned not to try an 'predict' what will happen next and to just buckle up for the ride and be ready to discover what is possible with enough imagination, passion and a generous dose of know how.

Personally I'm waiting for Scarlet as it will serve my needs the best, but it never stops me from marveling at the ambitious moves that RED is making that will disrupt (in a good way, in favor of innovation) the way we content producers will capture images. This DSLR thing? I could never have seen that coming, but as I think about it, it make complete sense. My mind instantly goes to thoughts of but wait, Canon and Nikon are so established, and wait, whose lenses are we gonna use and wait, this and wait that, but shoot if anything they've proved to my satisfaction is that Jim and the guys deliver on their promises; and thank goodness they're around.

 

[kunal2]

Whooooooooooooaaaaaaa

Whooooooooooooaaaaaaa

So there will be a RED DSLR?????..my gooooodnesss...this is good news

The primary advantage to RED is REDCODE. Compressed RAW at over 23.976 fps. This is a core invention of RED that's full effects have not been seen yet.

The secondary advantage of RED is our sensor program. Some could argue that this is number one.

Mysterium "Monstro" is a sensor program that pushes the envelope past anything on the horizon. It will go into Epic, and another camera aimed squarely at the DSLR market. Epic ships with Mysterium-X and has a free upgrade to Monstro.

Scarlet has been talked about extensively. It is a 3K, 120fps camera with a built-in 8X zoom that starts at 28mm (full frame still 35mm) wide angle. The T-stop will be "at least" as published.

We have a couple of other "tricks up our sleeves". The important message is that we are just getting started.

Jim

 

[Karl Gustav H.]

Baseless Speculation for a First Post

Baseless Speculation for a First Post

The DSLR Thing is really interesting.

Im thinking a 'barebones' Epic (just the imagery architecture "Monstro") coupled with the Scarlet backend (audio, monitoring, prehaps throughput). Recording to CF either as uncompressed RAW (if that is the correct term to describe Canon/Nikon RAW) for single frame/burst/low fps continuous, switching to Recode for video, prehaps capped at 60 fps. The Monstro spec may make a mechanical shutter obsolete, so continuous viewing wouuld be possible. The big question I guess is whether an optical viewfinder would be used, or if an all electronic monitoring solution is adoptd.

I guess this kind of growth is more manageable for RED now it is using dedicated silicon in the form of ASIC rather than FPGA's. A good decsion, and echo;s that made my the big players in the early days of DSLR's.

The body would use interchangeable mounts, ideally with an open protocol so extra lense controls could be adopted. Initially I suspect Red would offer a PL mount, as well as Nikon and Canon, both in 'smart' and 'dumb' flavours, prehaps partnering with Birger, ET et al. That would make the body a fantastic directors finder as well as intergrating with a massive installed userbase of standard glass. The open protocol would mean that all lense functions could be controlled in camera, withou having to revert to external controls.

All of the above cocnjecture is based on the idea that the RedSLR will use a 35mm sensor. Of course, the RedSLR could be an attempt to make medium format photography much more acessable using both standard lenses and Reds own "extra good and affordable" lenses using the experience gained with the 35mm glass. I guess this would be dependnat on the 'theory' that Epic is a 65mm sensor designed to take on IMAX/Vistavision and so on.
That being that, the two ideas are not mutually exclusive - a 65mm sensor could always produce a 35mm crop when using 35mm glass.

 

[Babu Kantamneni]

Gustov packing 140 miles on Cuba?!

 

[patrickortman]

In addition, before I finally step away for the night:

How is it that you guys are rocking the sensors like this? I know there are some great minds at work here, but why did no one else follow through with the tech the way you guys are, I wonder...

Regardless, you guys did it. Rock.

I was reading Scott Billup's book, and one of the things he talks about is how he visited the big Japanese companies years ago and saw all sorts of things he still could not tell his readers about. Stuff that was possibly decades ahead of the curve.

He then said something along the lines of how the big Japanese companies were carefully controlling when and how much of this tech would be released into the world at large.

My impression on reading was that the big Japanese companies were very concerned about letting "too much" out there at one time and disrupting things.

Seems to me, maybe one of the greatest things about RED is you guys don't give a damn about shaking things up. I truly hate when technology companies purposely "dumb down" their products. So, hat's off!

 

[Retired Account 000]

Hah! I can see it now.
Everyone at Nikon is jumping up and down.
"Yay! We're finally releasing a DSLR camera that shoots 720p HD! We're way ahead of the competition now!"
"Ah Fuck!"

Well Nikon and video are two terms you don't often see paired together. Got to hand it to them for trying eh?

So should I hold off on that 1DS MkIII for now?? What will the lens options be?? Another round of mounts required??

I also share these questions and what kind of market the RED DSLR would be targeted at. If it's a professional DSLR (which I wouldn't doubt, but you can never be 100% sure) with a 'professional price tag' in the thousands count me out =/ Afraid college is going to rob me of all my cash before I can even dream of a RED DSLR in my hands

 

[Ed Blythe]

While there's development talk ... RedRay, anyone?

 

[Karl Gustav H.]

Gustov packing 140 miles on Cuba?!

Closer to 300 miles in my case. Sunny old GB.

O/T I know, but does anyone know of any RED projects coming out of Bristol (UK that is not the NASCAR track in Tenassee!) - especially from the Beeb?

 

[Kyle Hance]

Man. It's these kind of announcements that make me HATE still being in high school! I get all excited and then realize, "Oh yeah... I'm still in a crappy high school with crappy Dells.... AND I'm completely broke...."

Looks like it's time to go job searching so I can save up for a Scarlet - and this mysterious DSLR that you speak of, Jim.

 

[Brook Willard]

Well just be sure to put a Nikon mount on it...

 

[Zakaree Sandberg]

hopefully my red one will still be worth its value after all this..

 

[Alex.Mitchell]

Well just be sure to put a Nikon mount on it...

I've been tempted this last week to pick up a DSLR. The thing that kills me now is that I know this RED DSLR will be awesome, but I don't want to have an investment in lenses on one system only to have to migrate to another a few years down the line. I guess what I'm saying is that it'd be awesome if it had a replaceable mount like the RED ONE. Then I could even use my Canon lenses, which would be awesome.

 

[Damien Molineaux]

I'd be fine with them downsampling to 4:4:4 10bit 1080P and recording with a HDCAM SR-like datarate. Or if they talked to Apple or Avid and made a higher-datarate version of ProRes or DNxHD - and integrated a ProRes or DNxHD encoder like the AJA IO box, Ikegami editcam, etc do.

Yes, but for one you'd lose the advantage or RAW : the option to do and parameter the debayer yourself. Secondly, would Apple or Avid have agreed, at the time when Red was considered to be a scam ? If so, at what price ? And although, as you say the quality is sufficient for film outs, it wouldn't be up to what you can obtain from Red Raw. Thirdly, we'd probably have higher data rates than Red Raw. Lastly, Red Raw is like a digital negative. Do an HD conform today, do a 4K output two years from now for 4k theaters !

Couple this with a "RED playback deck" - eg the circuitry (without the sensor part, scaler, audio inputs, etc) for $4000. It would output decent-quality 1080p HD-SDI (PsF if needed) and also good old Standard Def.

ie. Red Ray, it's coming, for a lot les than $4000

On the contrary, you'd have a workflow that would plug right in to everyone's current systems the day it was released.

Well, if Apple or Avid had played along, and you're talking about a higher data rate than the current versions of ProRes and DNxHD, so that may require some powerful machines. That would also mean Apple and/or Avid would have to updtae their software. We could reinvent the wheel with ifs and maybes. Red have reinvented the cinema camera.

And you'd have a camera that records beautiful images with 35mm DOF using PL or still camera lenses for much less than any of its competitors.

We have that already.

And you'd still have an awesome RED company with great community, customer service, etc.

We have that to.

If you want, offer an upgrade to a compressed RAW recording mode. But I don't think it benefits people much now.

Anyone want to give a real-world example of a paying project where it was critical for them to have 4K instead of 4:4:4 1080P?

Maybe not now...

I can give many examples of paying projects where no transcode time, easy 1080P output, a cheaper playback deck, etc would be a plus.

Of course, so can I. And if that's what you want/need, the Red One can do 1080p playback now, which you can record with an AJO IO direct to ProRes, and you have the system you're talking about.

Agreed it's beautiful, but a HDCAM SR class codec would have been fine too IMHO.

Agreed, but I don't think we have a HDCAM SR codec with the same bandwidth as Red Raw.

Your points are correct from a certain point of view. But it seems to me you're not grasping the whole Raw thing. It is two things, one it makes the camera simpler, you record the Raw data, no debayer, no gamma, latitude, black stretch, white stretch, etc. settings ! This is key. Secondly, you can set all these settings yourself, try them out, and adjutst them for otpimal results, during post. If you want quick workflow, it's possible, either with the proxies, or with the 1080 playback. If you want optimal quality, you have that too, with your raw files, all that at 27 MBps !!!

Remember, this is only the begining. The Red One is blowing it's first candle tomorrow ! HAPPY BIRTHDAY

Cheers,
Damien

PS Midnight is ringing as I write.