This might amuse Graeme...

[Graeme Nattress]

I personally don't see this as competition. But I do want to see an end to calling Bayer a "4:2:0" sensor, which is plain daft. There's some awfully good science and understanding from John and Larry in the whole thing - very good stuff, but there's also some bits which either make me laugh, and it's mostly in the section on Bayer, which for full (or as full as you can achieve) understanding, I do think you need to have designed the algorithms and played with patterns, pixels, OLPF etc. etc. to actually visually see what the compromises you choose look like. Having a higher resolution like 4k gives you a very different overall picture than trying to do Bayer on an SD camera, for instance. Graphs don't tell you that, but images do!

Graeme

 

[David Mullen ASC]

I agree that it's not a great argument to lean on, this whole "4:2:0" thing, and many people on the CML have also poked holes in that argument, which I'm not sure even John Galt takes too seriously. I only think he's getting a bit tired of the pixel numbers marketing game because he seriously wants people to start thinking about resolution in terms that go beyond "4K" versus "HD" that ignore the finer points about how that information (both in terms of detail and color) is being generated. But to some extent, it's a losing game because people, especially producers, like simple concepts that are math-free, and don't want to hear arguments between engineers regarding MTF versus aliasing artifacts, etc. Ultimately I suspect Panavision will just give in an make a 4K Bayer camera someday, if only for marketing reasons.

However, from Panavision's perspective, I think think the Genesis was a brilliant idea for its time, because it fit into the HDCAM-SR post-production world that has emerged in the last few years and avoided cutting-edge (and thus somewhat experimental) data recording, management, and image processing issues that make studios nervous. It fit into the 35mm camera production model that Hollywood crews are used to as well, which is why in some ways, it is the most successful digital cinema camera on the market in terms of being used on major productions. This is different than how the Viper and Arri-D20 were received, sitting around barely used for years until recording to SRW1 decks became an option. The Genesis may not hold that number one spot, but it was the right digital camera to come out two years ago.

Even now, it's not the RED camera that scares some producers and companies, it's the post workflow. We just have to admit that data recording for entire features is still rather rare and a new concept just now coming into existence. It will take a few years for it to become commonplace and as painless as a tape-based video workflow.

 

[Jason Ing]

Graeme, reading your posts as a non-tech like me can give me brain cramps at times, haha, but I enjoy trying to understand them.

But I also get a sense of practical, down-to-earth desire that at the end of the day of number wizardry, you just want to see a great image.

So, I'd like to know what are your favorite movies (visually). David Mullen has mentioned his favs in his thread. What's yours?

 

[Jeff Coatney]

Fascinating and informative. I think its interesting that the Genesis camera is based on 1991 camera technology. Seventeen years worth of potential development is spent solely finding efficiencies in established, long-entrenched technologies, regardless of their relevance to the particular application. Please don't interpret my comments as criticism, its an observation that sometimes it is necessary to take certain technologies to the absolute limit of their operational envelope to ensure the next wave of innovation. However, that kind of thinking is counter-intuitive to developing a product with any degree of longevity. You really have to believe that there will be zero innovation, zero increases in image resolution and the deliverable architecture (HDTV) will remain fixed and unchanged for the next quarter century at minimum. Which is a belief you could hold if you control the market or think you do. I think it was the natural and necessary next step for Panavision to take, looking at the future from the 1990 vantage point. It's also very clear that the Genesis imager, for reasons that the Panavision presenters themselves point out, will need to evolve to be a truly relevant acquisition system in a 4K market.

 

[Jeff Coatney]

Genesis is 33%:33%:33%, and we'd have to reduce that number a bit for the OLPF, so I'm being generous here, and we've still not accounted for, that spatial offset in the R,G, and B, and and in Bayer, there's a maximum gap of 1 pixel before you hit one of the same colour. In RGB strip, horizontally, you have a gap of 2 pixels before you hit the same one. That's like having a much lower fill factor horizontally than vertically. At least Bayer is nicely symmetrical!

3 chip of equal total photosite count is therefore also 33%:33%:33%, and again, I'm being generous. We do have to ensure the chips are perfectly aligned though, and we can suffer from prism artifacts.

Graeme

This fact has repercussions in manufacturing as well, since it is cheaper to produce one chip than three. And aligning those chips to produce a resolved image free of artifacts would require tolerances and a QC regimen that would negate any competitive price advantage.

You're right, Graeme, it isn't a competition. Not in the least.

 

[Jeff Coatney]

However, from Panavision's perspective, I think think the Genesis was a brilliant idea for its time, because it fit into the HDCAM-SR post-production world that has emerged in the last few years and avoided cutting-edge (and thus somewhat experimental) data recording, management, and image processing issues that make studios nervous. It fit into the 35mm camera production model that Hollywood crews are used to as well, which is why in some ways, it is the most successful digital cinema camera on the market in terms of being used on major productions.

Without this step, it could be argued, a RED class of camera might not have found it's way to market so soon. Panavision should be commended for knocking down the barriers to digital acquisition early on. Once Panavision endorsed the concept, industry professionals had to take it seriously.

 

[Christopher Grant Harvey]

Once Panavision endorsed the concept, industry professionals had to take it seriously.

I agree.

 

[laguun]

To say that RED is more "successful" than the Panavision Genesis camera because it sells more units is not particularly accurate since the two companies don't even follow the same business model.

I understand how you look at it. I have a different point of view however.

Both companies focus their business around the same industry. Both have 35mm cameras at the core of their business, panavision mechanical and digital, red digital.

Look at the revenue. Its pretty easy to see that red will easily pass the 100.000.000 this year with over 4.000 35mm PL-Mount 4K cinematography cameras ordered.

Look at the acceptance in the market. Red is shipping more digital cinematography 4K cameras a month than panavision genesis and arris discontinued D20 have been manufactured since their introduction to the market in years.

Look at customers. A prominent example is Steven Soderbergh, who was using panavision for many of his movies (solaris, ocean 11/12, the good german etc), who now shot his recent 3 movies on red. You will see some people going from red to panavision, but crowds of people are adding a red or switching completely.

panavision business model is no threat to red. The other way around however, its pretty problematic.

You can't buy a Genesis even if you wanted to.

a) Sony is selling the F35, which is pretty much the Genesis if i understand Sony correctly.
b) you can buy a comparable camera to the genesis from red. For a price which is less than a month rental of the genesis.

And even if you compare a RED to a camera that you can buy, like an F23, it doesn't necessarily mean that a company like Sony won't make some money off of their product even if the sales figures are lower.

Correct. I never said that they are already loosing money. However, the marketshare is collapsing. The amount of 35mm PL-Mount cameras in the market is almost doubling in 2008 from what is was in 2007, and the times before.

It's like comparing the sales of a Toyota Camry to a Ferrari -- of course more Toyotas are being sold. That's not necessarily a problem for Ferrari IF they don't lose their customer base (not to say that Sony shouldn't be worried, just that it's not simply about number of units sold.)

a) Sony, Thomson GV and Panasonic are much less affected than Aaton, Arri and Panavision. Sonys most popular 1080p camera is the HDC1500. That camera sold 4000 units as well. Arri and Panavision, who so far only got several dozens digital cameras, are due to size more endangered. Red is an alternative 35mm format, which is Arri/Aaton/Panavision land so far. Sony didnt even offer a 35mm camera until end of the year.
b) While i see both strenghts and weaknesses in red and genesis - In many aspects, the red is the ferrari at the toyota price, and the genesis a toyota at ferrari price. Up to 120P on red, 60 on genesis. 4K on red, or lets use Graemes 3.2K, 1080p on Genesis. Lightweight Red, heavy Genesis. Raw and RGB post at red, RGB at Genesis. Arri PL, Nikon/Canon/Leica, Panavision mounts on red. The Genesis has advantages as well (uncompressed out, live 1080 signal, and better sensitivity especially in warmer light) but in many areas the panavision is the toyota.

It's like this strange argument that digital is more successful than film because more digital cameras and projectors are being sold than film ones. That's been true for years, so what?

So the editing, which was done on film, is now >95% on digital. (Steenbeck/moviola->Avid/Adobe)
So the news, which was done on film, is now >95% on digital. (S16->video->digital)
So the optical printer for Effects is now >95% on digital (flame etc)
So the lab is becoming the digital intermediate. (baselight, lustre etc)
So the multitrack tape became the daw (16/32 track -> nuendo/protools)
So the mechanical Nikon/canons/Hasselblad became DSLRs and digital backs.
The list could go on for sides.

David, thats is an old hat, and i think we shouldnt add this boring film/digital debate to the topic. In my humble opinion Film will stay for some years, but the mayority of 35mm cameras sold today are already digital, no mater if still or moving images.

That completely ignores the fact that a lot of movie production follows a model of rental, not ownership, and the fact that movie theaters routinely use decades-old 35mm projection equipment.

Movie Theaters are switching to digital at a rollout speed of >400 a month since years. Much more digital movie projectors are sold than mechanical ones.

Most of the mayor rental houses here are adding red, and its finding extreme demand. Furthermore, renting instead of buying is due to a) service and b) finance. And in finance, red simply destroys 35mm film. No footage, no lab, no scan.. if one uses a DI process the cost savings are easily in the $$$.$$$ range. And i think it pretty easy to assume that most movies will, or already are, done in DI.

I can go out and shoot a feature right now on a thirty-year old Panaflex or Arriflex and as long as I put a modern lens on it, it would look no different than a movie shot on a one-year-old Panaflex or Arriflex. So the fact that more digital equipment is sold doesn't necessarily follow that more productions will soon be shot digitally, because you can rent older film equipment for years to come -- and also, digital equipment has a faster obsolescence rate so of course the rate of sales will also be higher and faster.

I have the choice between old Arriflex, red and hdcam here. The only camera i find rather obsolete is the Arriflex.

And no one seems to notice that Panavision is doing the digital world a big favor by making digitally-shot movies so available for mainstream audiences to see and accept, get used to, and make producers and studios comfortable with digital acquisition and distribution. Right now we have in theaters, or recent to home video, several movies shot on the Genesis: "The Other Boleyn Girl", "The Forbidden Kingdom", "Deception", "21", "Walk Hard"... so more people are going to see movies shot on the Genesis in 2007 and probably 2008 than people will see movies shot on the RED, despite how many RED cameras are sold (and not all are being sold to feature filmmakers.)
The whole competition between RED and Panavision is a bit silly if you ask me. Panavision has a rental-only business model and is heavily tied to the Hollywood studio system, because they provide consistent and good service to producers that can afford them. RED is a camera for sale and while no doubt some Hollywood production will (and has already) use the RED, it will have even more impact in other areas or countries where camera ownership is a more common model for production. Sony and ARRI probably have more to worry about RED than Panavision.

I fully agree regarding Panavisions contribution and merits - After Sony, Panavision and Thomson are the both other important pioneers of digital cinematography.

Dont get me wrong, i like Arri and Panavision quite a bit, and the Genesis and the Sony F35 are some of the finest digital cameras in the market.

Therefore i was saddened to see that Panavision officially is (or was?) misrepresenting Dalsa, Red, Arri, SI, Vision Research Phantom (and btw Nikon, Hasselblad as well) by giving an rather amateurish concept of Bayer-Pattern basing cinematography. According to the marketing-math which was presented, the Arri D20 wouldnt even be a 1080p 4:4:4 camera.

Regarding the amount of released titles, i suppose it pretty easy to say that there will be shot much more on red than on D20 and genesis combined. Lots of reds DoPs however wont aim at a cinematic release, especially clip, commercial, corporate, venue etc. Anyhow, it seems that the red will have more cinematic releases in 2008 alone than the D20 since its introduction, and it might become pretty close with genesis.

As for John Galt, he's a bright man who knows more than I do, but to some extent, he's also defending a product he helped create (why shouldn't he?), and he has some salient points to make about how pixel resolution is not necessarily the way to measure true image sharpness. His comments can be more damning about film resolution in some ways than they are against other digital cameras.

I have known John Galt as a reflected, entertaining and educated man.
The last time i met him was the recent 2 IBC iirc, at Reds 4K D-Cinema presentation of crossing the line. I highly enjoyed panavisions en detail report from apocalypto. However, the D-cinema had *daily* red presentations and screenings, always all >1000 seats taken, while panavision had only one presentation which was not filling the theather to 50%.

Rather than get into the numbers game, I'd suggest people look at real world results and judge for themselves.

I agree - let me add, if someone doesnt produce excellent images with red or genesis, i wouldnt blame the cameras.

 

[Tom Lowe]

The bottom line on all of this is that RED and Genesis will deliver beautiful 1080p/2K moving images. Since 3.2K or 3.4K or whatever is not even a format, Red will finish best downsampled to 2K or 1080p, which puts it squarely in the same league as the Genesis, F35, etc. The resolution issue is getting really old. Until someone creates a 6K or 8K oversampled sensor that can finish at a TRUE 4K, everyone else is basically shooting for 2K or 1080 right now.

There is no need to worry about resolution, because we know it's going to increase and increase and increase, just given the nature of technological advancement. It's the same thing with computer processors. No one worries whether 24 months from now computer chips will be faster. We know they will.

The real issue now is not resolution, it's dynamic range. I want a camera I can aim at a treeline on a sunny day and not blow out the sky. I wish companies would concentrate more on dynamic range and less on the resolution issue, which is basically guaranteed to resolve itself over time.

 

[Cüneyt Kaya]

The bottom line on all of this is that RED and Genesis will deliver beautiful 1080p/2K moving images. Since 3.2K or 3.4K or whatever is not even a format, Red will finish best downsampled to 2K or 1080p, which puts it squarely in the same league as the Genesis, F35, etc. The resolution issue is getting really old. Until someone creates a 6K or 8K oversampled sensor that can finish at a TRUE 4K, everyone else is basically shooting for 2K or 1080 right now.

There is no need to worry about resolution, because we know it's going to increase and increase and increase, just given the nature of technological advancement. It's the same thing with computer processors. No one worries whether 24 months from now computer chips will be faster. We know they will.

The real issue now is not resolution, it's dynamic range. I want a camera I can aim at a treeline on a sunny day and not blow out the sky. I wish companies would concentrate more on dynamic range and less on the resolution issue, which is basically guaranteed to resolve itself over time.

i guess this will be Epic---maybe?

 

[Tom Lowe]

i guess this will be Epic---maybe?

Epic will bring us closer to true 4K, I'm sure.

I also understand that 3.4K might be of better use than 2K for special fx people and other niche uses, but for most cinema shooters, they will downsample Red to 2K and will have beautiful results. Once you view Red as a 2K/1080p camera, everything else falls into place rather easily and you can compare it directly with Genesis, F35, etc. Everything makes more sense when you view Red as 2K/1080p, and we won't be distracted by these tiresome back-and-forths about resolution.

Again, for me, the real breakthroughs will not be in resolution. They will be in dynamic range, which is sort of the last bastion right now where 35mm chemical film is still standing its ground.

 

[Michael "Dorkman" Scott]

I just want to say, I've keyed 4:2:0, and I've keyed RED.

RED is definitively NOT 4:2:0.

 

[Gunleik Groven]

I just want to say, I've keyed 4:2:0, and I've keyed RED.

RED is definitively NOT 4:2:0.

That's my quote of the day...

 

[Radoslav Karapetkov]

Epic will bring us closer to true 4K, I'm sure.

I bet it might even do true 40K ($)...

But... true 4K is..... more likely.

Oops. :nerd:

 

[Graeme Nattress]

The problem is, Tom, that 4k RED shot and projected as 4k shows noticibly superior resolution / detail than RED 4k downsampled to 2k. It just looks better. If we assume similar OLPF / demosaic, Epic would deliver a full 4k luma resolution, 2.5k chroma resolution, and as pointed out, what the human eye sees from this is just a superb high resolution image. Yes, RED looks superb downsampled to 2k, but don't ignore the 4k as 4k - it's beautiful and addictive to work with.

I think what Panavision have done with digital is great. But they don't do cameras how I would do a camera. If you want a camera that works like a Panavision, then Panavision is what you rent. If your camera philosophy fits more in line with RED, then that is what you buy / rent. You can like both and use them appropriately. It's all about choice. You'll also notice that both companies are big on education. That's also good. The difference is that we take a very different approach to sensors, with different design goals and different aims, but we're both aiming for a superb image. We both get a super image, but ours has a different look and feel. I'm pretty fond of our image, and rightly so.

But because we use a Bayer pattern CFA sensor does not mean we violate the laws of physics (Sampling theory and all that) and does not mean that we don't understand the compromises with regards to aliasing and moiré, which are also very important design considerations for other sensor technologies. But if I'd been sitting in Panavision's shoes, rather than Red's shoes, I'd have had the Genesis sensor as a bayer pattern, got a 4.5k luma resolution out of it with an advanced adaptive demosaic. But I'm not in their shoes, and I didn't. But I still think that would provide, in overall terms, a better looking image. But who knows? That decision was made in other ways for other reasons and other priorities.

Graeme

 

[David Mullen ASC]

As I said, I think Panavision's choices made more sense two or three years ago when HDCAM-SR was coming into common use in the post houses of Hollywood. Truth is that even now, many people want to walk away with an HD tape at the end of the day and not deal with processing and de-Bayering time, etc. or data back-up. But even more so last year or the year before. Which is why the Arri-D20 camera (or now, D-21) was first optimized for dual-link HD output to HDCAM-SR decks rather than concentrate on getting the 2.8K Bayer RAW mode to work. And also why Dalsa, the first 4K Bayer RAW camera, didn't catch on -- uncompressed data recording was unmanageable (it still is a pain) and the idea of using wavelet compression like Cineform or REDCODE is only now catching on.

If RED had come out two years earlier, it may have suffered the same fate as Dalsa, Viper, and the Arri-D20, which is the RAW workflow issue compared to the standardized HD tape workflow that is well-established now (I know that the Viper is not a RAW camera...) So it's all a matter of timing -- the Genesis was well-timed to come out two years ago, and the RED is well-timed to come out today, thought it is somewhat ahead of the curve. But what has been smart about RED has been to recognize the problems of data recording and RAW processing and create solutions along with the camera rather than release a camera and hope in the following years someone would develop ways of handling the footage.

Once you have blazed this trail of making a RAW workflow easy and commonplace, and most of the post houses realize they will have to provide services to deal with it... everyone is going to follow. And I suspect that Panavision will realize that a Bayer sensor providing RAW data is no longer the inconvenience that it once was.

But considering the number of people here who have asked for a 1080P output from the RED camera, I'd say that HD video workflows are not dead either.

 

[Kenn Michael]

Epic will bring us closer to true 4K, I'm sure.

I also understand that 3.4K might be of better use than 2K for special fx people and other niche uses, but for most cinema shooters, they will downsample Red to 2K and will have beautiful results. Once you view Red as a 2K/1080p camera, everything else falls into place rather easily and you can compare it directly with Genesis, F35, etc. Everything makes more sense when you view Red as 2K/1080p, and we won't be distracted by these tiresome back-and-forths about resolution.

Again, for me, the real breakthroughs will not be in resolution. They will be in dynamic range, which is sort of the last bastion right now where 35mm chemical film is still standing its ground.

Did you get a chance to hit up NAB this year Tom? Did you see Red's 4K presentation? It was breathtaking. Almost TOO clear at times. What was interesting was seeing Soderbergh's 2K footage before seeing his 4K footage. Big difference.

All this talk of resolution is moot when you see proper Red 4K projected at 4K. Incredible.

 

[Gunleik Groven]

but in the 2k example with Sotherberg, it wasn't downsampled from 4k... Because it was shot on anamorphic.

I've seen 4k projections and film-outs and tend to agree with Graeme. The difference is really obvious when the same (4k) footy is projected @ 4k and 2k, much more than what it appears to on our small screens...

The same goes for filmouts from 2k vs 4k. The 4k source "wins", hands down.

I also think 2k is not very fun to work with (after doing some 4k) but I always downsample to 2k/1080. That's my domain. Nowhere to show, print or project 4k in my neighbourhood. And 1080 from 4k looks pretty darned nice, too

But I think RED is doing all the right things with EPIC. (Less compression, higher downsampling, full sensor overcrank, 1080 monitoring and realtime downsampling). These are each important steps, and together they should create a monster.

Cheers!

Gunleik

Cheers!

 

[Dj Joofa]

Again, for me, the real breakthroughs will not be in resolution. They will be in dynamic range, which is sort of the last bastion right now where 35mm chemical film is still standing its ground.

Did you get a chance to hit up NAB this year Tom? Did you see Red's 4K presentation? It was breathtaking. Almost TOO clear at times. What was interesting was seeing Soderbergh's 2K footage before seeing his 4K footage. Big difference.

All this talk of resolution is moot when you see proper Red 4K projected at 4K. Incredible.

I think Tom has a point here. For me, myself, personally, given a choice between going from:

(1) 2K to 4K,
(2) 12-bit dynamic range to at least 16 bit,

I shall favor (2).

Red is (relatively) weak in dynamic range as reported by various users on this forum and its about time they concentrated on that.

 

[Barry Green]

Yes, I fully understand why they chose the system they do, and it's a good compromise for what they want to achieve.

That's all I was saying. And, it should also be said, they're trying to accomplish something quite different than what Red is trying to do. Panavision is trying to make 1080p, with no pretensions of going to 4K.

What I find disingenuous about the Panavision presentation is that while they will say "you can't get 4K from a 4K chip" (which is, of course, true) it's even more true that you can't get 4K or 3K or 2.4K or, well, anything better than 1920x1080, from their chip.

Graeme says 3.2K. Adam guessed 2.8K. I've heard someone test the Red as low as 2.4K. Every one of those figures is quite a bit higher than 1920x1080!

Panavision aims only at 1920x1080. In context with that, their system is a reasonable approach and can deliver 4:4:4 in 1920x1080. Red aims for much more, but Red cannot deliver 4:4:4 at 4K.