Sure the sensor is smaller but that opens the door to the use of things like this...

Philip Bloom Blog Archive 7d with a BIG cinema lens on it (http://philipbloom.co.uk/2009/09/01/7d-with-a-big-cinema-lens-on-it/)

Holy &%#$!

I Know, not very constructive

Holy &%#$!


Exactly what I said! - it may not be constructive but it's entirely appropriate under the circumstances

Who makes that adapter?

That's why you're excited, Phil. I'll pass.

I mean it's cool, but only for maybe .001% of people who would actually buy this camera.

Me? I'm just hoping the HDMI-out isn't crippled like the 5D.

That's why you're excited, Phil. I'll pass.

I mean it's cool, but only for maybe .001% of people who would actually buy this camera.

Me? I'm just hoping the HDMI-out isn't crippled like the 5D.

I assume you meant buy that lens? cause a-lot more than .001% will buy this cam ;)

Rodney is just compensating.

Dan

Wow, that is some setup! Thanks for sharing, Philip! Now if I can only afford one of those lenses... :)

Andrew
Andrew Oh on Vimeo (http://www.vimeo.com/andrewoh)

I guess the point of this that virtually no-one owns Panavision lenses, but for the right production you could rent one and use it with the 7D. Potentially changes the game of budget HD filmaking for TV dramas etc...

Also apart from Cine lenses there are no ultrafast wide angle primes available for the aps-c sensor size. The Sigma 20mm f1.8 is about it, and thats not too good or too wide.

Dan

Where this starts to get interesting is with productions that already have a full complement of lenses and larger bodies. In this instance, a show that shoots with Panaflexes or on Genesis could utilize 7D's as crash cameras, point of view cameras, you name it--all with the existing lens package. Likewise, for Arri shows (or other PL mount cameras, including--hah!--RED), this should also be a great companion. I haven't heard yet whether there are any restrictions on the depth of the mount for either camera (I know the Panny mount for the 5D only worked for lenses longer than 35mm).

You may have the seen the pic I posted here a while back with the 5D and Panavision Primo lens on my lightweight Steadicam (http://www.dvinfo.net/conf/canon-eos-5d-mk-ii-hd/239682-5d-bigtime.html). It was a fantastically light setup for such "big" images!

Stephen's point is probably valid, although I'm not exactly sure about the numbers. Fortunately for me, I'm in that tiny percentage as I work with these lens packages all the time so I'm looking forward to the possibilities with this camera (except for the bloody form factor)!

Yeah, I like the 5DMKII and I like this one...but I keep wondering why, if they can put a bigger chip in a still camera and sell it for under $2K, why can't they put the same damn chip in a real video camera, even if they sell it for double or triple that?

On another topic that's related...I have a friend here who got the Panasonic micro 4/3 Lumix G1 and says he has an adapter so he can use C-mount lenses. But I've read you can't do that with an APC size camera because the lens would have to stick back too far into the camera, and also the lens wouldn't cover the whole field of view of the chip. Does the C-mount work on the micro 4/3 camera because the chip is so small?

if they can put a bigger chip in a still camera and sell it for under $2K, why can't they put the same damn chip in a real video camera, even if they sell it for double or triple that?Could be that such a thing is already in progress... we've been speculating about it quite a bit:

http://www.dvinfo.net/conf/area-51/236874-canon-xl-h2-when.html

C-mount adapters work on M43 because the sensor is smaller, and there's no mirror box, so the sensor is closer to the lens mount.

...if they can put a bigger chip in a still camera and sell it for under $2K, why can't they put the same damn chip in a real video camera, even if they sell it for double or triple that?

Because soon, there won't be any such thing as a 'real video camera', or for that matter, a 'real still camera'. Because we are on the cusp of a major paradigm shift, and the manufacturers are still tryng to figure this out.

The distinction between still cameras and video cameras has been an artificial one whose days are numbered ever since the imagers were first put on integrated circuits, and the control functions handled by microprocessors.

Soon we will just buy a 'camera'. With whatever mix of features and form factor we like, which in the better models may include stills, video, 3D, even projection out of the lens. This will happen for the same reason that we no longer use typewriters, or dedicated word processors.

The 5DMKII just included a test of a feature - not much different than many other such tests that have either succeeded or failed in the past - but its wild success (and maybe Red) has made the paradigm shift crystal clear even to the most conservative of manufacturers.

It's not just the direction that a few video enthusiasts want, such as we constantly see here. Had the 5DMKII not been very successful, Canon might well have pointed to Red and said 'It's just a niche market'. It's pretty clear now that is not the case. Ultimately, not only I, but the general public, will want a 'general purpose imaging device' rather than a 'still camera' or 'video camera', just as the Wang wordprocessors died in favor of general purpose computers running software.

So what follows now will be a long period of figuring out what features and form factors work for which price points and markets, more trying one thing, then another. Business pretty much as usual in the tech world. The screams from early adopters seeing their purchases topped by new models should become loud enough that harnessing acoustic energy may well become the new alternative energy fashion. ;-)

I think you will see the form factor and features you want within the next two years. It may, or may not, be called a 'video camera'. Of course, if Apple makes one, it will likely be called the iEye, copied by Sharp as the IiEye, and cloned by the Chinese as the I-i-I-iEye. ;-)

-Mike

Am I the only person yawning at those pics? Sure it's nice and all and not something we're used to seeing but who amongst us will be able to use these cine lenses with custom made mounts from panavision? And why would I want to shoot with a massiv 135-420 (216-672)zoom? Lol maybe if I'm shooting a safari video and I had a crew with me..

Once agan yawn..

I would love to see some anamorphics being used on these DSLRs though!


In addition, don't know why everything is embracing the VDSLR form factor as if it's the way going forward. There is absolutely nothing stopping Canon from building an HV40 type of camcorder with a big APS-C to FF sized sensor. That would be the ideal form factor for me...

don't know why everything is embracing the VDSLR form factor as if it's the way going forward.They're embracing it because it's available right now, that's all.

I caveated it by only talking about the ones who proclaim VDSLR is the future of filmmaking. That makes no sense to me.

It's the immediate future of filmmaking.

A border-line oxymoron, I know.

Yep. Available now is a good thing if it's something you need now. But it doesn't matter if Canon came out with an XH A1-II that has a full size chip and interchangeable lenses and offered a 12mm-100mm f1.8 zoom and it sold for $4,000 and had wifi and an iPhone built into the side...there would still be another new camera on the horizon that would come out the day after you placed your order, or if you're lucky the day after you made your final payment on what you bought.

For me a still camera that shoots good video might be a nice backup for my XH A1. I don't know if the footage is going to look that much better. What I'd really like to see is, after the camera is in the hands of some real video guys, not still photographers, some real world clips posted in full size that I could download and drop into a timeline and compare with other footage. If the footage is comparable or better, then I could probably talk myself into this camera.

We are taught to be wary of things that don't seem right. In the case of the 7D, I am wary because it is selling for $1K less than the 5D. With its sensor size, its new frame rates, and its reported higher bit rate , there are bound to be differences. With the lower price, that leads me to extrapolate that something will have to give somewhere. Seems to me there was a lot of early praise for the Panny, but I haven't heard a lot of great things lately about it. But I haven't follow it closely.

I will be tickled if the 7D lives up to the hype. But I have only seen one purported 7 second raw ciip from the camera, and it was jelloish and ugly, and exhibited a lot of moire effect. So it is best that we reserve until this thing is put through its paces.

They're embracing it because it's available right now, that's all.

Yes. And I have noticed the common development pattern they use goes something like this:

They put a test feature in one mid range camera.

If it goes, the next step is not to put it into the high end cameras, but instead into the consumer cameras (which causes some complaints from the high end users 'why don't we get this?')

Later it appears, much refined, in the higher end cameras. They have used all the feedback they got from having the feature out to a mass market to fine tune the design before they incorprate the feature into the high end cameras - that tend to have very demanding users.

It's a very sensible way to do this.

The VDSLR form factor is not the future. It's just the test bed.

-MD

But 24p isn't rocket science. All they need to do is insert a delay between the completion of one scan and the start of the next. We will get the same quality we get today, but with the desired frame rate. The jello will be the same (25ms latency, top to bottom), so when you pan at a given rate the buildings will tilt at the same angle as today.

Of course, if Canon wants to take their time and give us better compression, 720p60, 48 kHz audio and other goodies, I'll appreciate and use them. But just give me 24p with the current quality level, and I'll be quite pleased. (And if the HDMI stayed at HD for critical live focusing, that would rock too. There are no workarounds for trying to focus at HD levels with an SD monitor. But I'd understand if that feature were in the next gen.)

The VDSLR form factor is not the future. It's just the test bed.-MD

Correction - VDSLR is the future. However, the form factor of large sensor video cameras for us serious videographers/filmmakers will not be in the form factor of VDSLR.

In the future, the image sensor part of the system will not vary much from still to motion applications. The only thing that will vary is the form factor appropriate for the intended application

But 24p isn't rocket science.

Jon, that's not being fair to Canon. I think you may be looking at how quickly Tramm has built up the Magic Lantern firmware with dozens of valuable features. But it's no wonder he can work so quickly, just look at all of the factors that are in his favor:


Little or no money.
Whatever spare time that can be found.
He has absolutely no Canon source code.
He has no documentation of the hardware or software.
He has to reverse engineer everything from scratch.


With those kinds of advantages over Canon, it's no wonder he can fix a host of bugs and add new features in a single day's work, or build out advanced new features in less than a week.

Canon, in contrast, has all sorts of difficulties and barriers working against them:


Billions of dollars.
Dozens of full time paid software engineers.
Full access to all source code, SCM, and project tools.
Full documentation of all hardware/software.
Direct access to the engineers who designed the hardware.


With those kinds of disadvantages it's no wonder it took Canon half a year just to add Manual Control. Imagine how much slower Tramm would be if he had full documentation. Or (heaven forbid), the actual source code. Or was paid to work on it full time. His progress would grind to a halt! The only reason he has been so quick until now is because he had to work blindfolded and reverse engineer everything from scratch.

Jon, that's not being fair to Canon. I think you may be looking at how quickly Tramm has built up the Magic Lantern firmware with dozens of valuable features. But it's no wonder he can work so quickly, just look at all of the factors that are in his favor:


Little or no money.
Whatever spare time that can be found.
He has absolutely no Canon source code.
He has no documentation of the hardware or software.
He has to reverse engineer everything from scratch.



That logic doesn't make sense. Yes, Tramm's work is incredible and it is more than anything, indicative of his brilliance with reverse engineering and understanding the firmware code. Regarding the 24P feature, no one outside Canon knows for sure what technical hurdles exist for that to be implemented. Just because Tramm was able to enable so many other features doesn't mean given enough knowledge of the firmware or ample resources, 24P can be implemented via software.

Correction - VDSLR is the future. However, the form factor of large sensor video cameras for us serious videographers/filmmakers will not be in the form factor of VDSLR.


That is what I meant, even if not so clearly stated. ;-)


In the future, the image sensor part of the system will not vary much from still to motion applications. The only thing that will vary is the form factor appropriate for the intended application

And I suspect that form factor will converge a bit more between still cameras and video cameras. I don't know what the final result will be, but the 35 mm camera shape came as a very efficient way to cram a lens mount, two film spools, a viewfinder etc into a space, and the common sort of 'long rectangular box' shape of most video cameras was needed because of long vidicon tubes and complex electronics.

Neither shape makes much sense for a modern computer and electronic sensor, and in fact increases manufacturing costs quite a bit. Both were needed so that - with this huge change to electronic imaging - the manufacturers could put something in the hands of a photographer where they could feel some sense of familiarity - 'Looks like a camera, feels like a camera, works like a camera, yes, maybe I can work with this'.

It's interesting watching the shapes slowly evolve into something more suitable.

For professionals, I just hope that shape includes space for decent sized XLRs and audio level controls. ;-)

-Mike

That logic doesn't make sense. Yes, Tramm's work is incredible and it is more than anything, indicative of his brilliance with reverse engineering and understanding the firmware code. Regarding the 24P feature, no one outside Canon knows for sure what technical hurdles exist for that to be implemented. Just because Tramm was able to enable so many other features doesn't mean given enough knowledge of the firmware or ample resources, 24P can be implemented via software.

You may very well be right. Hyperbole of that magnitude applies better to features that are guaranteed to be software-only, such as Auto ISO, focus stacking, etc.

Jon, that's not being fair to Canon...

Awesome. :)

The reason this is exciting for many filmmakers is for the exact reason this custom rig was created: for the images it can produce.

How many times have you seen images that were obviously from using a long telephoto and created that wonderful "compression" effect in-camera between foreground and background? Lot's. And typically the only way you could do this is on a film-rig that can accept film lenses.

OR, you could create a hybrid such as this. A few years ago I had a Hollywood master-lens setup guru create a special mount for me so I could use the P+S Pro/Mini-35 with my HVX200 and later HPX500 and my 600mm F/4 Nikkor (see pic used for sizing purposes). It was a monster rig that required a "real" film-head on the massive tripod to hold it all, but it worked and I got the shots I need to fulfill the creative vision.

To me this rig isn't about just grabbing the subject that is far off in the distance, but also being able to create that in-camera compression, something you can't do well in post without it looking fake.

Personally the ability to use long lenses - especially those in the EOS "L" lineup such as the 200mm f/2, 400mm f/4 and if you've got the balls to rent one the 1200mm f/5.6 can produce images that are just... stunning.

This Panavision tele-zoom to a 7D? As the Brits say: Brilliant!

Daniel, your sense of irony is even better honed than mine. I stand in awe.

who amongst us will be able to use these cine lenses with custom made mounts from panavision?

Well, me, for instance, as I detailed earlier in this thread. I also described why this has interesting ramifications for those of us who work with these lenses to begin with. OK, I'm certainly in the minority on this board, however I think that the point is not specifically about Panavision lenses but that 35mm cine lenses in general can now be used on a DSLR without cropping issues. For anyone who has a RED or 35mm adaptor and PL mount lenses, a demographic that is well-represented on this board, this should mean that with appropriate mount they can use their lenses on this camera as an adjunct to their primary camera body, if desired. What I haven't heard yet is if there will be any restrictions on certain lenses due to the placement of the mirror--we weren't able to use the Primo primes wider than 35mm on the 5DMKII, for instance.

The 11-1 (24-275) Primo zoom shown in the pictures is the standard workhorse and nearly every production I've been on carries one; for most shows where we "live" on the zooms, this will sit on the B camera while the A camera uses the 4-1 (17.5-75mm). Generally the 3-1 (135-420) is day-played as needed. It is indeed massive, its nickname is "the Hubble", but in the world of 35mm cine lenses, you needed a lot of optics to deliver the results that are expected. 420mm is actually not even that impressively telephoto; it's roughly equivalent to the long end of the XH-A1's built-in zoom!

Little or no money.
Whatever spare time that can be found.
He has absolutely no Canon source code.
He has no documentation of the hardware or software.
He has to reverse engineer everything from scratch.

Don't forget:

I have only one camera
There was a very good possibility of bricking that one camera early on during development
I can't actually change any of the firmware in the camera, so everything that Canon's software is doing must be worked-around in mine

With all of these advantages, it is amazing that I'm not developing new firmware images even faster!

The area where we have had the most success with the Magic Lantern firmware is in the audio controls. Coincidentally, that is also where we had a full datasheet (http://www.asahi-kasei.co.jp/akm/en/product/ak4646/ak4646.html), which spelled out how everything worked and what to expect when we poked at the control registers. We may even be able to enable 48 KHz mode in the AK4646 (page 20 of the datasheet, register PM2), although there may be issues with how that integrates with the DIGIC4.

I'm working on a new technique that will allow me to forcibly relocate portions of Canon's firmware into RAM, fixup any %pc relative address references and then use self-modifying code to implement the functions that I want. If this works it will remove that last advantage from the list and allow us to start overriding significant portions of their code.

Ditto. That was some funny sh**! Between that and the "Hitler hates the 7D" thing that Phil posted, I've had a nice afternoon of laughs....

john

I wonder if there'll be lens adapters so you could rent cine lenses.