Panavision Genesis

[Jason Rodriguez]

Are they using an Accuscene viewfinder (that's what it looks like from the pictures)? If so, then yes, it's very nice, it's an OLED screen, and I loved it at NAB, especially it's "color zebras".

[Tenolian Bell]

I just got home from Cinegear and watching test footage of the genesis.

[Tenolian Bell]

I tried to ask as many technical details as I could but the Panavision guys were pretty mum on the details of how it works.

 

It doesn't use Bayer. It does have RGB pixels. But they wouldn't go into any more detail than that.

 

It uses a custom made Panavision Accuscene viewfinder. It is very bright very sharp, better than any HD viewfinder I've seen to date.

 

The dynamic range was great. I pointed the camera out into a crowd that was partly in bright California sun and partially in shade and it held all of it. Highlights rolled nicely and did not clip sharply. In the Accuscene viewfinder the clipped highlights still had an electronic look to them.

 

The deck can connect to the camera on top and the back, and can be easily taken off (even though no one ever took it off to demonstrate). I asked about the type of connections from the deck to the camera but never really got a straight answer.

 

It was larger than I thought it was going to be though. The XL2 was demure and petite compared to the Genesis. The 1000' magazine on the XL2 was far larger than the camera body.

[Tenolian Bell]

They showed a demonstarion of the a scene where film and Genesis where intercut. Allen Daviau shot the test scene. Panavision rates the Genesis at 400 ASA. Allan shot 5218 rated it at 400 ASA.

 

It was an interesting test and I would have to say finally a digital camera that looks just like film.

[Tenolian Bell]

There were no tell tale signs of digital in the shots. No aliasing, banding, or even highlight or shadow clipping. No hot spots in the highlights. No signs of artifical sharpening. Very good detail and color reproduction.

 

In some scenes there was more apparent sharpness than the film, and in some scenes the color was more saturated.

 

But looking with the critical eye. I did see some things that were still better with the film. We know 5218 is a rather latitude flat and color neutral film. In that context it was not made to produce overly vivid colors, more neutral colors, which could be percieved as dull next to a saturated color.

 

I know no one will believe this but the '18 was less contrasty and saw into the shadows better than the Genesis.

 

While looking at the test I could see in some shots more saturated color and in some shots where the shadows were deeper. I know '18 is a neutral film so i knew it didn't produce the vivid colors, but I also thought the Genesis was seeing into the shadow's better. But it turns out that was the '18.

 

My personal taste would be for the neutral color. I know from shooting thousands of feet of '18 that you can always saturate the color well beyond the edges of the vectorscope. I'd rather have the choice. Plus some of the colors in the Genesis were somewhat too colorful almost cartoonish. Especially a scene where they rack focus to some violet flowers. They were very very purple, but purple in an unnatrual way.

 

I can certainly see this camera taking over television, it looked really good.

[Tenolian Bell]

I joked with the Sony guy if we were going to see the test with the new 4K projector. He said "no do you know what Sony means?" I said "no". He responded "soon, only not yet".

 

I would be curious to see Genesis material in 4K projection. Even though the sensor is 4K, ultimately its still down sampled to 1080.

 

With film you still can shoot 5212 (100 ASA) scan that at 6K 16bit (Arriscanner) scale that down to 4K 10bit and project it 4K.

[Michael Nash]

I was at the screenings as well today (we were probably in the room at the same time). What can I say but, UN-F*&^ING-BELIEVABLE. The Genesis footage was a litle different from film -- it appeared sharper! Some of this may have been a function of contrast (the Genesis footage was slightly more contrasty than the flatter 5218), but certainly not because of any visible edge enhancement.

 

I should point out that both the film and Genesis footage were color corrected digitally, including the use of power windows to some extent. The purpose of the demo was not to showcase all of the camera's features, but rather to demontsrate how seamlessly it can be integrated with film footage. Honestly, you could scarcely tell the difference between the two formats, except that the film started to look soft in comparison after a while.

 

Regarding things like shadow detail, the Genesis has fundamentally the same menus as the F-900. So things like black gamma and color saturation can be tweaked to taste, in camera or in post. It's no different than any other HD or DI color correction that way. So they could have dialed the contrast and chroma to EXACTLY match the 5218 if they wanted to (with this footage). Panavision claims an extended highlight latitude that's better than any knee functions in previous Sony cameras.

 

Someone pointed out the difference in hue rendering on magenta tones (the flowers in the FG of the swimming pool; the girl's hat with the couple stacking dominoes). Like all video cameras, magenta comes out a little more red and not a true purple, but this difference is akin to the hue differences you'd expect between film stocks.

 

Also, HDCAM SR (the tape format) is HD resolution (1080 x 1920). The output to film we were seeing was from that. By virtue of oversampling (more pixels in the chip), the picture appears sharper than what you get from a 1080x 1920 chip. The film was scanned at 4K, and if I'm not mistaken then downresed to HD so all the footage could be color corrected in the same pass.

 

Again, this demo was limited and didn't fully put the Genesis through its paces. I'm sure film still has much more highlight range than the Genesis, but in reasonable conditions this camera more than passes. From this footage at least, Panavison hasn't blurred the line between film and HD; they've erased it.

[Michael Nash]

I should point out the test footage was shot with identical lenses, t-stops and filtration. Panavision Primo zooms and primes, no lens diffusion except for one shot (both cameras used the same filter). Mr. Daviau had selected an electronic "silver diffusion" filter for one sequence, and the softening on both formats essentially made the sharpness identical. Only the slight difference in contrast remained.

 

BTW, the color viewfinder is simply stunning.

[Jason Rodriguez]

Did Panvision mention how many stops over before clipping?

 

Tenobell, you said you pointed the camaera at a crowd and got both the highlights and shadows, no clipping, right? If so, would you say that the scene you were shooting had at least 10-11 stops? Would a similar situation on a video camera produce highlight clipping in your opinion?

 

Also I'm assuming that since this is a digital format, even if film has more highlight dynamic range (say by a stop or two), you can underexpose the digital a little bit to get that highlight detail back and not suffer any horrible consequences, especially with post color-correction.

[Max Jacoby]

I should point out that both the film and Genesis footage were color corrected digitally, including the use of power windows to some extent.

What's the point of color correcting film digitally? You lose sharpness, color depth and your blacks...

 

Just do a contact print off the original neg, for Christ's sake!

[Max Jacoby]

Then the CCD (or some other imaging device) size will start increase and quite soon 100mm will be reached for example.

I don't think that's going to happen, because you will need completely new lenses (not even 65mm lenses would work on this) and your depth of field would drop so much that it would become very impractical to shoot.

[Filip Plesha]

I agree with audiris. This screening showed how film compares in

HD DI to the genesis camera at this time.

 

A fair comparison of what the genesis footage looks like compared to original film footage would be if they printed the film optically from the original camera negative and uprezzed the HD tape recording to 4K and laser-printed it to film.

(well a 100% fair comparison would actually be if HD was laser-printed at 4K directly to print stock, and i don't think anyone does this)

 

What they have done is limited the film image to the qualitty of the genesis image (in both color deph and resolution) , and then compared those two. This comparison is only fair if you are considering films that go through DI. But not every film is processed in DI. This is not a comparison of the qualitty of the original data from the tape vs. the original grain structure of film negative.

[Mike Brennan]

Because the Genesis does not output a non downsampled full res image now does not mean it can't. Sony have never developed a camera without a complimentry recorder or vice verca.

So although it was (rumoured) that it is possible to get 444 out of a f900 or 950 as there was no Sony product to record it it was never done.

 

 

Having watched how sony r and d work over the years I would suggest that this isn't meant to be the killer camera for large screen but phase one. Phase two would be a Sony recording system capable of 4k then phase three a better sensor. 5-7 years maybe?

 

 

In respect of slomo the srw1 runs at double speed and can record 60fps 1920x1080.

I doubt very much that its speed can be changed on the fly for ramping.

That the Genesis camera tops out at 50fps at the moment is an indication that they are in a hurry to get it out and they are pushing the envelope.

 

In the past they have tripped up on bugs in the sytem when they have done this with their bespoke cameras like f900, but that was developed in a hurry by engineers with blindfolds.

Panavision state that Genesis has been in development for 4 years, with the background of the f900 and Panavised f900 behind them one can hope they will make a better job of the Genesis.

 

An earlier mention of digital correction to aid bluescreen is more likely to help correct fringing caused by the stip ccd than any issue with lens.

 

So already the marketing depart are calling the correction of a artifact a feature!

 

Those on a budget will still choose f950 and srw1.

I would not be surprised if the srw 1 is further delayed to the "open" market to give the Genesis a head start.

Comparisons between Dalsa and Genesis will be very interesting certainly on a 4k projector.

That topic will keep emulsion lovers occupied for 12 months, in the meantime anyone who knows how to use a f900 menu can pickup the Genesis and begin shooting...

 

Focus pullers will be in demand more than ever.

 

Mike Brennan

[painfulcrash]

after all facts and figures;

 

i think genesis will be a symbol for the digital revolution.

[Mike Brennan]

"An earlier mention of digital correction to aid bluescreen is more likely to help correct fringing caused by the stip ccd than any issue with lens.

 

So already the marketing depart are calling the correction of a artifact a feature!"

 

 

 

To clarify the above statement, press release says

 

"and digital lateral chromatic aberration compensation for improved visual effects cinematography".

 

The lateral chromatic aberration could be caused by the strip ccd.

It is very usefull to be able to correct but I wouldn't call it a feature!

 

 

 

Mike Brennan

[Jason Rodriguez]

A fair comparison of what the genesis footage looks like compared to original film footage would be if they printed the film optically from the original camera negative and uprezzed the HD tape recording to 4K and laser-printed it to film.

According to many people who've attended the 2K screenings of the StEM material, the digital projection looked better than the answer print version of the material. So I think that digital projection has reached a point where it does not inhibit the film's look, but helps it out beyond even an answer print struck from the original negative.

[Nate Downes]

While this is factinating, the proponents of digital projection still have not answered the cost issue: How do they propose to get the cost of these 4k/8k systems to the point that they can compete with film?

 

And there is another question, how does any of this compare to Maxivision48? I'd like to see how a 4k system compares to this alternative system.

[Mike Brennan]

Any sign of a mechanical shutter? This would help pinpoint the type of ccd.

 

We are assuming it samples of course rather than just output the 16x9 portion of the ccd:)

 

 

 

 

Mike Brennan

[Mike Brennan]

"4k/8k systems to the point that they can compete with film?"

 

I think the costs will continue to fall. Sony's multiplex 4k will be $60k so they say, with a big screen version for $80k.

 

According to the DLP guys, the potential market of 100,000 Digital projectors will be underwritten by the potential sales of 50 million+ home cinema theatre systems.

 

Not a nich market like film projectors but driven by mass market technology which never ceases to surprise.

 

For example according to the Market Intelligence Center (MIC) of the Taiwan government-sponsored Institute for Information Industry (III), China hit record shipments of 65.237 million mobile phones in the first quarter of this year, with year-on-year growth of 64.2%.

 

That is a 1/4 billion phones a year!

 

Nikon's camera plant is knocking out 60,000 digital pro cameras a month, Panasonic is building 2 more HD plasma display factories to cope with predicted demand from China.

 

 

 

 

Mike Brennan

[Max Jacoby]

According to many people who've attended the 2K screenings of the StEM material, the digital projection looked better than the answer print version of the material. 

I remain very cautious of such claims. The simple fact of digitizing footage reduces its quality. Just look at any DI in your local cinema.

[Balazs Rozsa]

Then the CCD (or some other imaging device) size will start increase and quite soon 100mm will be reached for example.

I don't think that's going to happen, because you will need completely new lenses (not even 65mm lenses would work on this) and your depth of field would drop so much that it would become very impractical to shoot.

Your depth of field would not drop because you would use these lenses at a higher f ratio. Actually the higher f ratio is exactly the reason why the lenses for a bigger sized CCD give you a sharper picture.

 

you will need completely new lenses

 

Pixel count will get really high, recording capacity for the data will be cheap. The only resolution limiting factor will be the lens. I think movie makers will not be able to resist building larger CCDs with sharper lenses.

[Jason Rodriguez]

I understand your skepticism, but the plain and simple truth is that DI is here to stay, and the amount of films that are undergoing this treatment are growing with every new release. So if the future of cinema is going towards an all-digital workflow after aquisition, then you can say that the best quality is a contact print, "let's see the contact prints!", etc., but the truth of the situation is that for all practical purposes, films in the near future will be turning more and more to DI. In other words, I think it's a good apples to apples comparison, because basically you're pitting the next generation of digital cameras to the next generation of film workflows, that is 4K DI on a 2K digital projector. Maybe the current method of doing film prints is better, that is the optical-chemical path, but frankly if that way is being replaced by DI, then that's like saying Technicolor was better color than Eastmancolor. Whether one was better than the other color-wise, guess which one won out? So in this sense I think comparisons of what would be the best comparison are useless, because more and more DI is being used to finish feature films, and so it just makes sense to set that as the new standard for comparing digital cameras to their film counterparts.

[Filip Plesha]

Yes it is practical to compare how would a digital camera compare to film in a DI process because the DI will be used more and more. BUT. The future of DI is

in 4K resolution. So even then, it would be more fair to compare 4K DI film vs. gensis printed to film at 4K. Using HD DI for comparing film to something else gives misleading results.

[David Mullen ASC]

A contact print from the original negative should always be available as a frame of reference since D.I.'s are so much more varied in quality. Not that comparing a digital camera to a D.I. isn't also useful info. For example, I'm sure indie people would like to see comparison of HD origination to Super-16-to-HD, both for film-outs, since that is becoming more of a common path for Super-16 post.

[Max Jacoby]

Your depth of field would not drop because you would use these lenses at a higher f ratio. Actually the higher f ratio is exactly the reason why the lenses for a bigger sized CCD give you a sharper picture.

By 'f ratio' do you mean f stop?

 

65mm has about half the depth of field that 35mm. The depth of field doubles with every 2 stops that you close down. So to get comaprable depth of field, you need to use four times as much light in 65mm as you would in 35mm

 

Your proposed 100mm chip would require even more light. That is just not feasable from a budgetary point of view.