New! Panasonic 4K & HS 1080 Varicams

[Glen Vandermolen]

Panasonic shows us their new Varicams. The heads are swappable, so you can have either a 4K S35 or 2/3" high speed 1080P Varicam. They use the same recorder:

4K:
For Journalists

HS 1080P:
For Journalists

Plus new expressP2 cards to record the high speed or 4K video:

For Journalists

Both camera heads have 14 stops of dynamic range!

[Tom Roper]

Very cool. Did I miss what lens mounts are supported?

[Glen Vandermolen]

Originally Posted by Tom Roper

Very cool. Did I miss what lens mounts are supported?

The 2/3" is a typical B4 mount. The S35 has a standard PL mount.

[Zach Love]

Those are some amazing specs which makes me think the camera is going to cost more than my house.

14 stops of DR, 1080p 240fps. There is more, but you really don't need any more. If that doesn't make your jaw drop, then not sure what would.

[David Heath]

The 4K s35 version looks good - but I see less point to the 2/3" head. If you're going to be spending Varicam sort of money today, 4K ability seems almost essential - and that's likely to be more and more the case as the months roll on. And let's not forget the F55 will do up to 240fps in 2K - maybe the s35 Varicam will have a similar ability?

May have been different a year or so ago when 2/3" glass meant a 2/3" body to use it, but with the high quality B4-s35 optical convertors around today, I don't see much future for 2/3" bodies at the top end. (Medium budget is a different matter.)

[Steffen Sailer]

This is the best anouncement since years!

I'll go for the Varicam HS.
I have good B4 glass here (HJ18ex28, HJ14ex4.3) and they will be back to their true life again.
I can't wait to get one!
Why go for 2/3" head?
No crapy bokeh of the PL-B4 adaptors anymore (I have the HDx35, I know what I'm talking about)
No sharpness drop in the corners of the picture anymore.
No CAs in the corners thanks to a propper 3 Cmos system (lens + camera are fitting perfectly,again, no additional lenses in the adapter)
and CAC!
Pany colours
No dust problems on sensors anymore (!!!!!)
Not loosing 2 2/3 stops of light anymore, good for the 240 fps in rainforests.
Be able using my b4 mount lenses wide open again without loosing resolution.
Be able to use my HJ14 again (with the HDx35 it was too bad!)
Same picture quality in normal and high speed mode (no line scipping and other crap)
1 Pixel RGB is 1 Pixel RGB, no calculating and loosing little details while calculating bayer patterns and downres.
Propper 1000 lines of resolution with propper rollof in the 1080 area without much moiree anymore, broadcast safe, like other broadcast cameras before, even at 240 fps.
14 stops of dynamic range in a 2/3 " camera!!!
Switching to a 4K imager with 120 fps in 4K is always possible...

This is a propper 1080 HD B4 mount camera with 240 fps, built for a B4 system like in former times, remember? No compromises. Perfect for natural history filming.

My main wildlife lens is and will be the HJ18ex28. Because of reach, zoomfactor, weight...

There is no PL mount lens available (and will not be) with 1300mm on the long end, f2,8-5, 18x zoom and switchable 2x extender for 2600mm and 3 kg weight. And still superb HD quality.
So why should I make compromises here with adapters?

And a propper 4K PL Mount camera is it as well. For 4K, you should use propper 4K PL glass anyway.

I think these are enough points from my side of view to go for one...
TV will be 1080 for the next 5 years at least. In my country anyway.
And switching to 4K is easy with this camera.
One day.
When there is a PL Mount 4K Lens f2,8-5,6/75-1300mm with 2x extender, <3 kg and < 80.000 $ available out there ;o)

Cheers,

Steffen

[David Heath]

Originally Posted by Steffen Sailer

Not loosing 2 2/3 stops of light anymore, good for the 240 fps in rainforests.

The use of the adaptor means that whilst it's true the effective f stop of the lens will be about 2.6 stops smaller, the larger sensor means a corresponding higher camera sensitivity - so back to square one.

Completely wrong to think that it needs 2.6 stops equivalent more light when used with a s35 camera. With or without the adaptor, it's the same number of photons of light that are being used, and that is what determines sensitivity performance.

TV will be 1080 for the next 5 years at least. In my country anyway.

I think it's wrong to equate most common broadcast standard with what will be most used for acquisition - certainly as far as programmes of the standard of using a Varicam are concerned. Just as HD acquisition at the high end became the norm before there was much HD broadcast.

I think you are being somewhat harsh regarding quality issues with the adaptor (though some makes seem to be better than others) but the feeling seems to be that s35 is the way forward, mostly used with s35 lenses - primes and short range zooms - with B4 lenses via adaptor for shots when a long servo lens is necessary.

And switching to 4K is easy with this camera.

Is it? I assume you're thinking of changing the 2/3" head for the s35? The big unknown at the moment is pricing - not just the total, but how the cost of a system will be split between the head and recorder units. I may be wrong, but I suspect the head end(s) will make up the greater percentage of system cost - so switching to 4K could be expensive.

I also see the 4K s35 version holding it's value much better than the HD 2/3".

[David Gemmell]

Yes, I have been waiting for a Varicam HS type camera for some time.

I have some expensive B4 glass (HJ18ex28) and finally will be able to get some nice slow motion wildlife footage - without the compromises that come with using large chip cameras (with 2/3 inch lens).

I just hope the price is within my budget.

[Maurice Covington]

So what are best guess for pricing? Is this a $20k camera?

[Glen Vandermolen]

More pics of the Varicams:

Varicam Modularity

[Phil Goetz]

Looking forward to seeing these at NAB!

[David Heath]

Originally Posted by Maurice Covington

So what are best guess for pricing? Is this a $20k camera?

In terms of specification, the s35 4K head with recorder would seem to be about halfway between the Sony F5 and F55 in terms of spec. (It doesn't have the global shutter of the F55, nor (apparently) the compressed RAW abilities.) Hence it seems reasonable to value it roughly half way between the two - in round numbers, I'd say about $25k for body only.

If it ends up at $20k, I'd say it looks very good value for money, at $25k, more or less what might be expected, and at $30k (let alone upwards) then a F55 would seem to offer much better value for money.

Let's see what gets announced.

[Christopher Young]

Somebody please correct me if I’m wrong here in my assumptions and conclusions.

The question that begs to be asked is what is the size and resolution of this new Panasonic 2/3” Varicam’s sensor. If it had three chips of around 2.2MP photosites then I would accept the statement that it is full HD. If it is a single 2/3” traditional Bayer type sensor of 2.2MP please explain how it delivers true HD 1920 x 1080. The true resolution cannot be determined from a photosites count. After the debayering and demosaicing algorithm processing the output delivered may well be a full 1920 x 1080 but... DEMOSAICING algorithms have to be used to INTERPOLATE a set of complete red, green, and blue values for each pixel. Definition of Demosaicing: “Algorithms making use of the surrounding pixels of the corresponding colors to estimate the values for a particular pixel.” Interpolating is the process of creating something from nothing and is only required in a camera if it doesn’t have full resolution which would be the case given you are trying to output full HD from a Bayer pattern CMOS / MOS sensor that has insufficient photosites as would be the case with a 1920 x 1080 sensor.

It sounds like the discussions around the current crop of ‘4K cameras’ that are using Bayer pattern sensors of around 10-12MP that have an effective area of about 8.85MP (4096 x 2160 = 8,847,360). None of these cameras are delivering a ‘true’ 4K no matter how much we wish for it. 4K RAW data from a debayered sensor is going to be more in the order of 3.2MP. Following the math through to a 2K scenario a single 2 / 2.2MP CMOS / MOS type sensor is going to deliver a resolution figure of around 1.6MP and therefore a 2/3” sensor slightly less, more like 1.5MP.

The way I see it to deliver a ‘true’ HD image the CMOS / MOS sensor would have to have a much higher photosites density and deliver virtually double HD resolution, close to 4K in other words. If so constructed then there would be sufficient photosites to deliver a full bandwidth 1920 resolution from the green channel with improved resolution from the blue and red channels. That’s why the super down sampled 2K output of the of the 4K Sony cameras looks so good according to many. Better looking than shooting at 2K.

Sony with their F65 (Q67 sensor pattern) had to go to approx 20MP photosites (figure varies depending on source) to be able to deliver a full bandwidth 4K. And in doing so they have actually use a pixel offset technology as compared to the conventional Bayer pattern. Sony has used half pixel offset for many years to bump up the resolution of the output of many of their earlier CCD cameras.

http://i1119.photobucket.com/albums/...or-650x419.jpg

So unless Panasonic can demonstrate quite clearly what the photosites count and resolution of this 2/3” sensor is I will remain a little sceptical as to its true HD resolution. A case in point is the conjecture surrounding Panasonic’s AG-HPX600 camera, albeit a much cheaper camera but... Panasonic claim it to be a 2/3” 1920 x 1080 sensor with an HD output. Many have asked why Panasonic have not divulged the technical specs on this camera’s sensor. Panasonic when asked just won’t give you a definitive answer. I hope they won’t be fudging the figures when it comes to disclosure on the full tech spec of this new 2/3” Varicam.

In earlier days Panasonic were a bit more forthcoming. With their HPX-500 which could record 1920 x 1080 images they were prepared to give you the sensor specs. We all knew it was a 960 x 540 sensor that was uprezzed to 1080 for output.

Roll out the FULL specs Mr. Panasonic.

Unless I totally misunderstand the technology involved do you think it’s a fair question to ask?

APOLOGIES!! On further reading I discover the 2/3" head is a THREE sensor camera so my reasoning is totally flawed. If it were a single sensor I would still think the same way unless someone can point me in a new direction with my reasoning.

Chris Young
CYV Productions
Sydney