AG-DVX200 4K Large-Sensor Handheld Camcorder

[Benjobe]

I don't understand why people claim that 10-bit 4:2:2 is necessary with this camera. Have they seen this video?

Skip ahead to 4:20 where he demonstrates that there's virtually no visible difference between 8-bit 4:2:0 and 10-bit 4:2:2 even when he pulls a hard secondary color in post.

I've been shooting 8-bit 4:2:0 on my GH4 for more than a year now without any difficulties color correcting in post. This DVX200 4K camera looks like an great companion camera to my GH4.

 

[Barry_Green]

but where 10 bit undeniably has a serious edge is when you start to use log type curves, where the expectation is for a lot of processing in post.

10-bit definitely does have an advantage there, yes. But, then again, Canon introduced their C300 with C-Log at 8-bit, and apparently to a thunderous reception. I'd prefer 10-bit, obviously, but I doubt 8-bit is going to prove to be some unmanageable problem. The VariCam's lofty Film Rec has been being graded in 8-bit DVCPRO-HD for well over a decade. 10's better, but 8's not some manner of disaster. And if you absolutely need 10-bit for some purpose, the DVX200 can do it through an external recorder.

I've been thinking about this a lot, and ... I dunno, let me put it like this -- what circumstances are you going to be shooting in, where you absolutely HAVE to have 10-bit 4:2:2, but where you don't need a monitor? Are there any? And if you have an external monitor... then you also have an external recorder, if you use something like the Odyssey 7Q+ (which would need an upgrade to HDMI 2.0 to take full advantage of what the DVX200 can do, 4K/60P).

When I'm on a serious shoot, I have a monitor. If I'm shooting a corporate film or something like that, frankly, I have never needed 10-bit 4:2:2 (as evidenced by that I've done hundreds of those jobs in 8-bit 4:1:1 SD and 8-bit 4:2:0 HD over the years). Keep in mind we're talking about an all-in-one camcorder here, not some Alexa replacement, that is probably going to have a street price of around $4500 or so.

So where are we going to be needing this 10-bit 4:2:2 as such a crucial feature? Maybe when the camera is used as a B-cam or C-cam for major Hollywood motion pictures? Well, considering that they've been using 8-bit 4:2:0 DSLRs to do that for years, I don't really see how all of a sudden the DVX is crippled for that purpose...

I'm not saying "you don't want 10-bit 4:2:2" ... obviously we all do. Better is better, we all want better. But I am asking -- what job have you been turned down for, or what job will you be turned down for, because it doesn't have onboard 10-bit 4:2:2? And -- if you really were actually going to lose a job because it doesn't have it, then -- it provides the capability to output that 10-bit, yes, if you'll supply the monitor/recorder for it. Just like people have used EX1s for years in BBC production by attaching a Nano. It's not ideal, but it satisfies the requirements and you still get the job.


I wish my car had 600 horsepower, but it doesn't. And it still gets me to work and where I'm going, just fine. I probably wouldn't take it to drag-race against Tesla P85's though, and that's a compromise I have to live with. In the same vein, I wish the DVX had 10-bit 4:2:2 on-board, but it doesn't. However, I suspect that for the type of jobs that it's aimed at, it'll do them extremely well. You may not be using it to do special-effects shots on the next Avengers film, but then again, that's not what it's aimed at or made for.

If I was a Panasonic marketing man, what would currently be giving me sleepness nights at the moment would be the worry of such as Sony bringing out a camera with something like the front end and lens of the new RX10, with better zoom speed and manual controls etc, with the back end and codecs of the Z100. If they were to announce something like that at IBC, I'd expect projected DVX200 sales to drop substantially overnight. And conversely, if Panasonic announced better codecs for the DVX200 at IBC, I'd expect it's projected sales to rise!

How about you let the product actually, you know, at least finish development, and maybe actually come to market, before you proclaim its sales to drop overnight? Seems a silly and slightly inflammatory thing to say, doesn't it? Sheesh. People said the same thing about the DVX100, "if Sony would only put 24p on the PD170"... but they didn't, and the rest is history. Will it repeat? Maybe. Maybe not, maybe the day after the DVX200 comes out we'll find out that Lego is now producing professional video cameras and selling them for $19.95. Who knows? Why even make such a statement?

 

[Emanuel]

(...) you never hear of Panasonics being two, three, four, six, or 10 months late to market. Back in April they said it'd ship in October, and it'll ship in October. They know what they're doing.

True. Panasonic, Sony, Canon, etc are from another league indeed. But, the newcomers are the reason why the majors have lowered some of their prices and we can find some broadcast features in consumer products.

 

[Barry_Green]

FS7. (sounds crazy? ... the Japanese conglomerates play by own rules ... Sony protected GH4 for a full year ... that's how it goes)

The FS7 with the kit lens is well over twice as expensive, and (based on MSRP) closer to almost 3x as expensive as the DVX200. And that's with a 5x zoom, not even remotely near the range of the 13x zoom; if you were to put a 13x zoom on the FS7, it would be vastly more expensive. So -- I know what you're saying and I do think there's plenty of collusion going on behind the scenes, but I can't see how anyone would see the DVX200 and the FS7 as being anywhere near competitive with each other.

 

[philiplipetz]

They presented the nonworking prototype to key dealers before NAB. Asked for feedback and then ignored advice for 4:2:2 and any color but red.

 

[samtheman]

Downloaded the 4K file from Vimeo. Looks amazing, just found the bokeh a little distracting, looks weird. Nonetheless, looks like a promising option. Can't wait for more footage. I started my career with the DVX 100, DVX100B, HVX200 and HVX200A. Amazing mojo.

 

[combatentropy]

Skip ahead to 4:20 where he demonstrates that there's virtually no visible difference between 8-bit 4:2:0 and 10-bit 4:2:2 even when he pulls a hard secondary color in post.

Pulling a hard secondary is not what I think would show the difference between 8 and 10 bit. Higher bits per channel lets you stretch your picture further before banding breaks in. 10 bit is quadruple the number of shades per primary color. The main time that people "stretch" the picture is when they raise the contrast of the mids, chiefly when the source material has a video look, and they want to turn it into a film look.

Of course every film look you've ever seen on DVD, Blu-ray, Vimeo, and Youtube was presented to you in 8-bit, 4:2:0. So having a great look with those specs is of course possible, especiallly if you can arrive at the look you want before it gets baked into 8-bit 4:2:0. So theoretically if this camera has settings to where you can get the look you want before it hits card, and you don't have to do any serious grading, 8-bit 4:2:0 is enough. 10-bit 4:2:2 is more like a safety net. If you have that, you don't care if the camera's paint settings are feeble, and you likely won't get criticized by your producer, editor, or whoever, for recording in too thin a codec. It's about being above reproach. Can you do without it? Of course, but knowing that you are recording 10-bit 4:2:2 frees your brain that much more, so that you can pay that much more attention to a million other things while out there in the field trying to get some great shots.

 

[Mike Harvey]

Im sure this was covered somewhere in the previous 74 pages, but what camera are they trying to protect by crippling it to 420 8-bit?

I have a theory... simplicity. The DVX200 is going to use the same general codecs as the HC-X1000 and GH4. All three are also 4k cameras. My guess is they are going to share a lot of the same internals, which should help cut costs. Realistically, I doubt they would sell that many more cameras if they included 10bit 4:2:2 codecs, given who the target market is for this camera. My guess is they figured the cost savings + sales with existing codecs made more sense than trying to reinvent the wheel and come up with a new set of internal processors and stuff that do 10bit 4:2:2 on a non-P2 camera in order to sell a few more cameras.

Still would have loved AVC-LongG or something similar, but I've sort of made peace with that.

 

[Phil vG]

Everyone put it truthfully. Is there anything you don't want the camera to have, and all in the $5000 price range, so no camera will cost more than that ever again.

 

[joe 1008]

Minimal requirements regardless of the codec: sharp, artefact free HD with 5-axis IS, C100-like DR, clean ISO 3200, slowly ramping lens. @Emanuel: you have exceeded your space for private messages. Check skype. Lisbon sounds good!

 

[Alpha Beta Omega]

It makes no sense today to exclude a high quality codec that can do 10-bit 422 in a pro produkt. It's just software written in a chip so to me this looks like Panasonic doesn't want the DVX200 to be to good. A shame.

 

[DLD]

... So -- I know what you're saying and I do think there's plenty of collusion going on behind the scenes, but I can't see how anyone would see the DVX200 and the FS7 as being anywhere near competitive with each other.

Right. They're not. Which is the point of the collusion (if there is one). For argument's sake, let's say DVX200 had a better zoom, 4:2:2 10-bit internal, 400 Mbps AVC-Ultra. Let's also assume that would have brought the price up to ~ $6,000. At that point, the two cameras would still be different but not so far apart that choosing one over the other would be farfetched. At $4,500 vs. $8000 (+ the lens, as duly noted), there are really quite a way to go. Which may be a shrewd approach from a marketing POV, regardless of whether there is a collision or not. Basically, you're marking your territory and letting your competitor mark his and it makes sense in what is essentially a nascent 4K market. In a year or two, when everything is in 4K, the market will have many models resembling each other and the preference will be made on the grounds besides purely technological.

PS. Been checking out some G7 footage on YouTube and, for a ~ $700 camera, it's fantastic. If DVX200 has the same sensor, G7 is probably as good of a preview as any.

A graded G7 (also 8-bit 4:2:0) sample

https://www.youtube.com/watch?v=h8lu4WbzaJg

 

[Danilo Del Tufo]

DVX200 footage with no post grading, link : https://www.youtube.com/watch?v=MAAQ-yq4rWE

 

[Bassman2003]

Pulling a hard secondary is not what I think would show the difference between 8 and 10 bit. Higher bits per channel lets you stretch your picture further before banding breaks in. 10 bit is quadruple the number of shades per primary color. The main time that people "stretch" the picture is when they raise the contrast of the mids, chiefly when the source material has a video look, and they want to turn it into a film look.

Of course every film look you've ever seen on DVD, Blu-ray, Vimeo, and Youtube was presented to you in 8-bit, 4:2:0. So having a great look with those specs is of course possible, especiallly if you can arrive at the look you want before it gets baked into 8-bit 4:2:0. So theoretically if this camera has settings to where you can get the look you want before it hits card, and you don't have to do any serious grading, 8-bit 4:2:0 is enough. 10-bit 4:2:2 is more like a safety net. If you have that, you don't care if the camera's paint settings are feeble, and you likely won't get criticized by your producer, editor, or whoever, for recording in too thin a codec. It's about being above reproach. Can you do without it? Of course, but knowing that you are recording 10-bit 4:2:2 frees your brain that much more, so that you can pay that much more attention to a million other things while out there in the field trying to get some great shots.

This is it. Professionals are expected to deliver a consistent product. More information recorded helps you deliver a consistent product by giving you a deeper well to draw upon. If you are shooting interviews in a windowless room with controlled lighting then yes, 8bit 4:2:0 is going to get you to the promised land. But if you are shooting in more "uncontrolled" situations (what this camera is being touted as...) then the odds of needed to bend the footage go up.

Which leads me to what has bothered me from the start - is this a pro camera or an in-between? Seems to be a pro-ish camera but then again DSLRs are used in pro situations a lot too.

Also, to state again - Us 10bit 4:2:2 people only are looking for this in 1080p/720p.

As far as price vs pro. Well take a look at the PX270. You can't get any more pro with that camera and its price is not that far away. Panasonic is playing a little game of chicken here imho. They are putting together a lot of things that people have been asking for for quite some time, but leaving out the huge thing they have developed over the years - P2. Now the Micro P2 is part of the ecosystem and is more affordable, why would you not lead with this to sell your new design? Instead they are leaving a hole where the recording specs are concerned which only adds some doubt. You can still offer the 4:2:0 usual suspects and have P2 in the same camera. The PX270 allows you to use regular SD cards. I just don't understand building a system then not utilizing it.

 

[DLD]

There's no need for P2 on a 100 Mbps camera and, frankly, with the HEVC on the way in some form or another and the ever increasing speed and size of the SD cards and SSD's, P2's time in the spotlight may already be over.

 

[KINOKS]

There's no need for P2 on a 100 Mbps camera and, frankly, with the HEVC on the way in some form or another and the ever increasing speed and size of the SD cards and SSD's, P2's time in the spotlight may already be over.

P2 is realy not about speed but about security.

 

[Gary W]

I believe Panasonic has made some important mistakes with a camera they could have dominated a segment with. I also believe that Black Magic will either dominate the market in the sub 10k camera or completely turn it on its head in the next 3-4 years. Though they are not a large company like Sony or Panasonic, I predict they will grow much bigger in the future because they listen to the industry's concerns, and try to accomadate at a reasonable price point. They will leave both company's scratching their heads asking what happened.

 

[combatentropy]

I can't see how anyone would see the DVX200 and the FS7 as being anywhere near competitive with each other.

I see all motion-picture cameras in competition with each other. The Rebel and the Varicam can be used for: a movie, commercial, documentary, event, deposition, conference --- you name it. What differs is the sticker price and the contortions they put your arms through to work them. But make no mistake, they are all in competition with one another.

How many people in the world have compared the footage from an expensive, shoulder-mount cam, and a DSLR, and said, "Forget it. I'll save the $59,000 and go with the DSLR." Sadly many of them end up spending the $59,000 later anyway, on reconstructive wrist surgery But that doesn't mean that the same people aren't keeping an eye on any and all cameras. Anyone remember when they shot an episode of House on a Canon 5D, because the Director of Photography wanted shallower focus?

 

[Barry_Green]

Which leads me to what has bothered me from the start - is this a pro camera or an in-between? Seems to be a pro-ish camera but then again DSLRs are used in pro situations a lot too.

Consider the original DVX100. It was 4:3, standard-def, 4:1:1 on consumer DV tape. Was that "pro" or pro-ish? Lots of people complained bitterly about how the 4:1:1 and 5:1 compression were just terrible, it was nowhere near BetaCam quality, no pro could use it... they complained about how amateur the 1/4" tape format was vs. BetaCam's 1/2". It was a 1/3" when professional cameras were 2/3". Fixed lens when professional cameras were interchangeable.

So, yes, it (and the PD150) kind of created the term "prosumer", but both were adopted by pros and used extensively by them.

So is the DVX200 "pro"? It's vastly more powerful than the DVX100 was, every aspect of it is fully on par with the PX270 and bigger cameras... the only compromise at all in it, spec-wise, is on the codec. Sure would have been nice if it'd had the PX270's recording module on it. But, to point out the obvious but perhaps it's not so obvious -- the PX270 is a high-def camera only. It isn't a 4K camera. Its internal processing isn't 4K. So if they'd gone with the PX270's hardware, you might have had a 1080p-only camera with 10-bit 4:2:2. That's not this camera -- this one is 4K through and through. And the processing engine it has, apparently only allows for 8-bit 4:2:0 recording.

Does that mean it's not "pro"? I guess folks are going to have to answer that for themselves. Compared to the original DVX100, it seems to me every complaint is addressed about it, except for the final recording format. For those for whom the codec is a showstopper, they'll have to go buy the other all-in-one camera that offers 4K and 10-bit 4:2:2... which is, I guess, the $10,500 FS7. To those who can't afford that, they'll have to consider whether the codec is a dealbreaker or not. I suspect that a lot of us will come to the determination that "10-bit 4:2:2 would have been nice to have" rather than "it's a dealbreaker", but -- I guess we'll see.

Also, to state again - Us 10bit 4:2:2 people only are looking for this in 1080p/720p.

If 1080/720 is your concern, then remember, you can always shoot in 4k, downconvert in post, you'll have 10-bit 4:4:4. As a practical matter, are there many likely scenarios where you're going to need to deliver 10-bit 4:2:2 footage, but that will not have had any post-processing done on it whatsoever? Doesn't seem that likely to me; if the footage needs 10-bit 4:2:2, then it probably will be footage that needs to be color-corrected and perhaps graded before delivery, right? If so, you can do the conversion in post and end up with 10-bit 4:4:4.

About the only place where I can see where this wouldn't work would be in a case of where you need to hand off 1080p footage to the client at the end of the day, and that client is expecting 4:2:2. I can honestly say I've never had such a job, but I can imagine that such may exist. And for those circumstances, however rare or common they are, the only thing I can say is that maybe in those instances you'd need to use a Samurai or Ninja or something like that. Perhaps inconvenient, but is it necessarily a showstopper? For some maybe, for others maybe not.

As far as price vs pro. Well take a look at the PX270. You can't get any more pro with that camera and its price is not that far away. Panasonic is playing a little game of chicken here imho. They are putting together a lot of things that people have been asking for for quite some time, but leaving out the huge thing they have developed over the years - P2. Now the Micro P2 is part of the ecosystem and is more affordable, why would you not lead with this to sell your new design?

I understand what you're saying -- but consider, again, the PX270 is a 1080-only system. They would have had to limit the DVX to 1080 only, to do what you're asking. If the PX270 was a 4K camera, I would be right there with you, but -- it isn't. And the only 4K P2 camera they have is the $60,000 VariCam, and I don't think it's reasonable to expect that they could have put its processing engine into the DVX200. So faced with the decision of making the new DVX 1080-only and 10-bit 4:2:2, or making it 4K even if that meant being limited to 8-bit 4:2:0, they chose 4K. Given the target audience for the camera, I think it was the right choice, but the market will speak and let them know.

 

[philiplipetz]

I believe Panasonic has made some important mistakes with a camera they could have dominated a segment with. I also believe that Black Magic will either dominate the market in the sub 10k camera or completely turn it on its head in the next 3-4 years. Though they are not a large company like Sony or Panasonic, I predict they will grow much bigger in the future because they listen to the industry's concerns, and try to accomadate at a reasonable price point. They will leave both company's scratching their heads asking what happened.

We have owned five BM cams. They are good cine cams but poor ENG event cams.