EVA1: EVA1 8/3/17 Press Release and Specifications

" I might use AF to get initial focus sometimes, then deactivate it, zoom out and compose my shot." Guess you have parfocal lenses.
 
" I might use AF to get initial focus sometimes, then deactivate it, zoom out and compose my shot." Guess you have parfocal lenses.

Yep. Just one: a Fujinon. You really only need one fast parfocal servo zoom lens.

20 years in the broadcast industry. Used lots of parfocal servo zoom lenses attached to some kind of video camera and recorded to all sorts of media, analog or digital. Going from that to DSLRs and stills lenses was always a :/ for me, especially changing focus during a shot or attempting a zoom. Plus the 8bit 4:2:0 and h.264.... egh.

Anyway, stills lenses are for taking stills. If you're going to use them for motion, there's caveats. But if I had to donit, i would, but I'd just reverse my setup: frame and then focus. Easy enough to remember.

I prefer motion picture primes, but if I have to have a zoom, I'd prefer it be a proper motion picture one. They just behave the way that I'm used to. And I like my tools to help me work. I don't want to work around my tools...not in operability and ergonomics. Lightweight fly-by-wire stills lenses for stills? Absolutely. I do it all the time. I have some nice Canon L lenses just for that. They most likely will find themselves on EVA1 from time to time, if I buy one. Parfocal zooms for motion pictures? Well, yeah. Pity I can't mount my Fujinon B4 to EF. Oh, well.
 
I believe it is a data throughput limitation. 5.7K is a LOT of resolution, so getting that out at 30p over 6G is pretty much saturated at 10-bit. Same goes for 4K at 60p.
So for 4K at 60p we will be able to get out raw - and for 4K at 30p too? Well, not bad - given the fact that Sony uses 3G-SDI what is limited even more.

But is it valid to state that the camera has a 10bit internal signal processing?
 
Don't trust a machine to be creative for you. There are times AF has its place. It's nice to have for that 1:10000 time I ever needed it whilst recording, but the rest of the time I focus manually, as do all the other pros I know. I might use AF to get initial focus sometimes, then deactivate it, zoom out and compose my shot. Then rack as needed, if needed. I would never use AF for creative control at anytime...not on a gimbal, or otherwise flying, not on a jib, not as a crash cam. Never.

Unfortunately, from my experience with still cameras, they will do away with manual focus as soon as they get AF to work about 1/2 the time. Anyone who has played with toy video cameras knows that the is nothing more annoying than AF that keeps changing on you willy-nilly.
 
The main problem is the that there are allways things that the AF needs to operate contrast, faces etc so instead of getting a specific shot you are constantly trying to appease the AF system.
 
Just to jump into the Autofocus debate (for no real reason, TBH) I agree that I've never used AF professionally, although if I'm stuck with a gimbal or steadicam and no remote focus - I would love to have autofocus I could rely on, even if it was limited.

Having said that, I can also foresee a point when autofocus of a sort is something I would use all the time - and that's when we get autofocus that corrects my errors. In other words: I pull to a subject, but I'm a little off, so the AF snaps it into sharp. Apparently you get this on some stills cameras already - and I'm sure it will come to film at some point too.
 
Last edited:
I own a set of voigtlanders as my M43 workhorse and will be going with a set of Milvus with the EVA-1. Suffice to say I am a manual focus guy. But saying that you don't use AF and pros don't use AF is irrelevant to the point people are making. Today AF largely sucks (with the exception of DPAF, which is still in infant phases) and manual focus is superior. Arguing about the future use case of AF based on the present day, or the historical implementation of AF, is silly at best.

Maybe there's a bit of fear of focus pullers and manual skill becoming irrelevant. I get that. But it's not the point, and I don't think that will happen anytime soon, even as AF develops.

The future of AF is incredible and will offer DPs and Filmmakers extraordinary creative freedom. The ability, as a OMB, to lock onto a small object, hit f/1.4 on a telephoto, on a gimbal, and run full speed tracking it and never loose focus. The ability to set razor thin focus and push/pull into objects, rotate around them, at high speed in slow motion and then slow it back down with focus bang on the entire time. The ability for a OMB to capture, handheld with IBIS, moving action sports at shallow DOF with precision focus.

Technology will make this possible. It's already happening. It won't be long. Take a camera with the brain of an iPhone/google image cloud software, with robust image recognition for thousands of objects. Add it to a small dual lens setup on top of a camera, which can read a scene and calculate depths based on the offset of the different focal lengths (the software calculating the iPhone blur maps is based on this, and 3rd parties plan to create depth sensing apps once the feature is unlocked to developers). Add to that laser, short high frequency sonar bursts, etc. in smaller and smaller form factors that can combine all of this together, both at a hardware and software level, with processing improvements in AF algorithms that already exist, and it's easy to see that the future of AF will be incredible.

It's coming. It's not a bad thing. It will make possible shots that would have been impossible in the past, and will make possible shots for single operators that would have been nigh impossible.

What exactly does Pros shooting manual focus have to do with this? Do any of these Pros have access to what I just described? People are talking about two different things here...

And before you jump on me... as stated, I own and use all manual glass as my main lenses. People are talking not about today, but about the future. And that future is remarkably close. Everything I just listed already exists. It's just a matter of putting it all together. Who will make the next big competitive leap toward that future and get us 25%-50% closer? Not the EVA-1, and that's ok. But in 5 years time, cameras without highly competitive AF features will have a hard time, because pros will start to utilize the incredible creative freedoms and benefits once they experience what the technology can offer artistically and practically.
 
Last edited:
My view is quite the opposite. Let's say we are shooting the guy and girl you described. Let's also assume we have the camera on a tripod and we want focus on each face at the proper moment in a narrow field of focus. First, we establish the blocking and then we set marks for the actors. Having established the focus points, they are marked and set. During the scene, the focus is pulled at the proper times and we get the shot. Except that the actor missed their mark and we didn't quite get it. Or the actor did a fabulous take that the director loves, but in that take, the actor moved in an unexpected way forward or back, out of the field of focus. The shot is lost because we didn't hold focus.

Meanwhile, the operator using auto focus taps the screen at the right time and the camera holds focus on the actor, regardless of them hitting their mark. The fabulous take is saved as the camera was able to react to the unexpected move.

Actors often don't hit marks, which is why the art of focus pulling is exactly that: an art. A great focus puller can instinctively read the body language of an actor and compensate for whatever they do.My regular focus pullers all came up in film and know how to pull the old-school ways but they all use monitors now, and at this point I feel that the percentage of accuracy is higher than it ever was. That too is a skill. interestingly, the focus puller on "Birdman" (my old college friend Gregor Tavenner) did not use a monitor and it's an astonishing piece of work in that regard.

I do however agree that the future will bring significant advances in auto focus tech and that will largely be driven by the "prosumer" market (man, we need a better name for this). I think it will be a while before the hunting issue is solved, but I can easily see a touchscreen interface in play. One of the concerns in the racking scenario above is regulating the speed of the racks between two characters, but conceivably that could be regulated by drawing a line between the two subjects which designates the speed of the rack (separate from tapping between them which would ask for an immediate rack). And with very delicate situations such as an extreme closeup with shallow depth of field where one must make a choice between the two eyes, having a focus indicator appear on each eye and using the same techniques to choose when and how the rack occurs will be imperative.

The key is that for many situations, allowing a human being to make the focus decisions is critical. These interfaces will simply be another tool just like pulling off a monitor that can be enabled or disabled as the situation presents itself.
 
So for 4K at 60p we will be able to get out raw - and for 4K at 30p too? Well, not bad - given the fact that Sony uses 3G-SDI what is limited even more.

But is it valid to state that the camera has a 10bit internal signal processing?
Published RAW output specs:

5.7K 10-bit uncompressed 23.98p, 24p, 25p, 29.97p
4K 10-bit uncompressed 23.98p, 24p, 25p, 29.97p, 50p, 59.94p

RAW output coming in a future firmware update.

Note that EVA1 RAW, like VariCam RAW, is both uncompressed and Log encoded. This is unlike Sony FS RAW, which is both compressed and Linear encoded. Linear encoding is generally agreed to be less efficient and a 12-bit Linear holds less effective information than 10-bit Log.

And no, it is not valid to make the assumption that the EVA1 has 10-bit internal signal processing. That seems to me to be an overreaching statement.
 
Mitch, any news yet on what all passes through when using a 3rd party LCD option? All menus; green focus squares; etc?

As well, any info yet on more specifics of the EIS system and what makes it unique or special, and specific recommended/not recommended use scenarios?

Thanks for the great engagement!
 
Jason R. Johnston:
Anyway, stills lenses are for taking stills. If you're going to use them for motion, there's caveats.
--
Ok..that clears things up a bit . But to those without the resources to afford such lenses shooting in situations less amenable to frame, then focus..you can imagine the benefit of AF.
 
Yep. Just one: a Fujinon. You really only need one fast parfocal servo zoom lens.

20 years in the broadcast industry. Used lots of parfocal servo zoom lenses attached to some kind of video camera and recorded to all sorts of media, analog or digital. Going from that to DSLRs and stills lenses was always a :/ for me, especially changing focus during a shot or attempting a zoom. Plus the 8bit 4:2:0 and h.264.... egh.

Anyway, stills lenses are for taking stills. If you're going to use them for motion, there's caveats. But if I had to donit, i would, but I'd just reverse my setup: frame and then focus. Easy enough to remember.

I prefer motion picture primes, but if I have to have a zoom, I'd prefer it be a proper motion picture one. They just behave the way that I'm used to. And I like my tools to help me work. I don't want to work around my tools...not in operability and ergonomics. Lightweight fly-by-wire stills lenses for stills? Absolutely. I do it all the time. I have some nice Canon L lenses just for that. They most likely will find themselves on EVA1 from time to time, if I buy one. Parfocal zooms for motion pictures? Well, yeah. Pity I can't mount my Fujinon B4 to EF. Oh, well.

Oh, but you can, my friend. I have and do, when I need to.

I have the IBE HDx35 2/3" B4 lens to s35 adapter and I can use my ENG lenses on s35 sensor cams with PL, EF and FZ(F5/55/3) mounts(they make other mounts, as well).

If you can spare a few grand and a few stops of light, you're in business.
 
Mitch, any news yet on what all passes through when using a 3rd party LCD option? All menus; green focus squares; etc?

As well, any info yet on more specifics of the EIS system and what makes it unique or special, and specific recommended/not recommended use scenarios?

Thanks for the great engagement!

The menu tree can be seen on outboard monitors, but the shortcut "Home" page cannot. The Home page is the one that looks like the Home screen on the VariCams and is similar to the layout used on many cameras (Alexa, F5/F55, Phantom Flex 4K, C700). The Focus Assist tools are not available on the monitor output.

Don't have info on EIS at this time.
 
I own a set of voigtlanders as my M43 workhorse and will be going with a set of Milvus with the EVA-1. Suffice to say I am a manual focus guy. But saying that you don't use AF and pros don't use AF is irrelevant to the point people are making. Today AF largely sucks (with the exception of DPAF, which is still in infant phases) and manual focus is superior. Arguing about the future use case of AF based on the present day, or the historical implementation of AF, is silly at best.

FWIW, Nikon recently got a patent (in Japan) for the PDAF auto focus. I assume theirs is just different enough not to violate Canon's patents. In terms of the industry, Canon currently has a monopoly on the DPAF in the camera business (others have DPAF in their smart phones) but, when two companies have it, then soon everyone will have it.

Technology will make this possible. It's already happening. It won't be long. Take a camera with the brain of an iPhone/google image cloud software, with robust image recognition for thousands of objects. Add it to a small dual lens setup on top of a camera, which can read a scene and calculate depths based on the offset of the different focal lengths (the software calculating the iPhone blur maps is based on this, and 3rd parties plan to create depth sensing apps once the feature is unlocked to developers). Add to that laser, short high frequency sonar bursts, etc. in smaller and smaller form factors that can combine all of this together, both at a hardware and software level, with processing improvements in AF algorithms that already exist, and it's easy to see that the future of AF will be incredible.
Light camera just released a bunch of photos that they claim to be equivalent to 52 MPX from their multi (16) lens/sensor camera.

https://www.cinema5d.com/light-l16-the-16-lens-camera-is-finally-shipping/

And Lytro has been undergoing trials for a few years as well.

Likewise, sonar and laser focus assist products have been on the pro market a while as well.

I do however agree that the future will bring significant advances in auto focus tech and that will largely be driven by the "prosumer" market (man, we need a better name for this).
It's now called "enthusiast". You take a shot, then do a wave and now you're officially an enthusiast.

One of the concerns in the racking scenario above is regulating the speed of the racks between two characters, but conceivably that could be regulated by drawing a line between the two subjects which designates the speed of the rack (separate from tapping between them which would ask for an immediate rack). And with very delicate situations such as an extreme closeup with shallow depth of field where one must make a choice between the two eyes, having a focus indicator appear on each eye and using the same techniques to choose when and how the rack occurs will be imperative...
The racking speed adjustment already exists on many cameras (NorBro linked to 1 D XII earlier in the thread) and GH5 actually gave so many options on its AF system, a lot of early buyers had trouble setting it up to their liking.
 
Do the math. The limit is the 6G spigot, followed by the recording capacities on the SATA-based SSDs used by the offboard recorders. These are the pinchpoints in the pipe.

Thanks for confirming this Mitch. What about a higher quality 2k 100fps via the sdi out?
 
The smallest battery looks set to drive the camera for a bit over two hours, the largest battery will offer four hours of continuous use, and the batteries charge about as fast as they discharge. I believe Mitch said the camera will include a dual charger; I would guess it would be the same unit as included with the UX180/UX90. If so, that should mean charging should take about the same time as discharging; with one battery on the camera and two on the charger you should be set for all day use (and probably overkill; you could probably go all day with one on the camera and one on the charger).

Camera I had included the dual charger and that's all what I experienced. The 8900 batteries are killer. Bundled battery and two of those will cover just about any day without charging.
 
The menu tree can be seen on outboard monitors, but the shortcut "Home" page cannot. The Home page is the one that looks like the Home screen on the VariCams and is similar to the layout used on many cameras (Alexa, F5/F55, Phantom Flex 4K, C700). The Focus Assist tools are not available on the monitor output.

Can this be fixed on a firmware update? Before launch?

Running it by engineers, would be much appreciated!

I can live with a lower quality LCD (I assume it's an even lower res than the GH5 because there's an assumption pros will simply replace it with their own?), but it's sort of a kick in the pants to be given a crappy LCD and then be told all functionality can't be sent to your 3rd party device that's pretty much required. Luckily this seems like an easy enough fix, especially since the green focus dots are supposedly so magical, would be a bummer to lose these...
 
The flange distance is generally an issue in adapters that don't have optics in them. It seems to me that since SpeedBoosters are inherently optical adapters, that it's possible that the flange distance can be taken into account in the design of the optical element.

I also would think they'd have been all over it, so one of two things must be the case... either:
1) it's impossible and I'm a moron for thinking it's possible, or
2) they never thought about it.

I think Brian Caldwell has participated here on DVXUser before, but I don't know what his userid is or I'd send him a message to ask if it could be done.

I think the problems of trying to make adapters to suit other mount lenses to EF mount were pretty extensively covered by Mark Duclos when he was quizzed about the modifying the new Fujinon MK parfocal zooms to EF mount. A condensed selection of his comments pretty well covers the issues:

“In our efforts to adapt the Fujinon MK line to Canon EF mount, we overcame several hurdles, but simply couldn’t fit the optics into a new housing economically. We would have had to reduce the housing by a whopping 26mm in order to make the Canon mount work. In order to accommodate a Canon EF mount, we would have had to redesign and manufacture about 20-30% of the lens itself which would have put the final product into a much higher price bracket than we were comfortable with.”

It is a shame as Duclos now make an FZ mount for these Fujinons plus MTF Adapters have announced FZ and M4/3 conversion mounts for these constant aperture parfocal zoom lenses.

Admittedly the EF mount is very popular but you cannot get away from the fact that their operational drawbacks such as short focus throw, non parfocal zooming and in the case of many of them exposure ramping as you zoom in, even with lenses that claim to be constant aperture such as the 24-105mm, are far from desirable features for cine zooms.

CYV Productions
Sydney
 
Back
Top