JVC 4k camera does small 4k but at 60p?

[Robert L. Mickles]

Sorry, that means nearly nothing to me. I guess the hypothetical question I'm posing is this... How did JVC do it for $4k when RED couldn't for $10,000?

Because JVC got the technology down to a "Consumer" level. A small handheld camcorder with a cheap fixed lens, small sensor, and four streams of a highly compressed codec. It's not a modular raw capture device with a S35MM, interchangeable sensor, that you're going to be able to trade in or upgrade when something new comes out. It's something that will probably be technically obsolete within 6-12months of it's release, and anyone who buys it will be stuck with it when something new comes out. I'm sure RED could have easily did the same thing, but thats not the kind of product they are in the business of making.

As far as blending the images goes. I got to see it first hand at NAB and speak to A JVC Rep about it. It was quite impressive for what it is. But the rep said it will blend the images automatically through plugins and software that will be available for most major NLEs and Mac and PC.

 

[Lee Saxon]

That's a bit harsh. JVC's implementation is just a means to an end. I remember not long ago pros considered the RED workflow absurd. If they have a fairly usable utility to stitch the 4 streams together then they'll be putting 4K footage on inexpensive media. That's pretty remarkable. Anyway this camera should be an effective stepping stone...

I'm not saying it's a bad idea because of the workflow annoyances. I get way less worked up about that than many people, probably because I'm such a computer nerd.

I'm concerned about the four streams being synchronized and actually matching each other. I'd worry about that even with RAW, but with a high compression codec being applied...

Hopefully at least they encode to H264 or whatever BEFORE they split it up.

 

[Stuart English]

...I'm concerned about the four streams being synchronized and actually matching each other. I'd worry about that even with RAW, but with a high compression codec being applied... Hopefully at least they encode to H264 or whatever BEFORE they split it up.

In the following link it implies they are splitting the image up into 4 x 1080p tiles before H.264 encoding.

http://pro.jvc.com/prof/attributes/tech_desc.jsp?model_id=MDL102132&feature_id=02

" The image is separated into 4 quadrants, each recorded onto a separate SDHC card "


Of course that's marketing material so it could be wrong ...

 

[Dan Kanes]

I could tape 4 Panasonic GH2 cameras together with 14-42 kit lenses for $3200 and get 400+ mbps "4k" if I wanted to take this stupid of an approach.

They should have at least made a 4x sd cartridge like an open face p2 card to make management easier.


Maybe they will sell 2 of these. Ever.

 

[Hector Berrebi]

this camera is actually an interesting tool.
its 5k$, so cheap.
its fixed lens and small chip, so deep focus all the way
it can live-stream (first camera with this capability) 4K via 4 HDMI to a projector or monitor (and maybe to 4 ninja-like recorders?)

its amazing for sports, news, event coverage, live projections in concerts and other large gatherings, conventions and conferences, scientific use... this list is endless
cameras like this will bring in the 4K era.

it is not a camera for cinema use, or most narrative (and even documentary work).

 

[Bruce Allen]

I could tape 4 Panasonic GH2 cameras together with 14-42 kit lenses for $3200 and get 400+ mbps "4k" if I wanted to take this stupid of an approach.

They should have at least made a 4x sd cartridge like an open face p2 card to make management easier.

Maybe they will sell 2 of these. Ever.

OK, I'll bite: you're forgetting - this isn't a RED camera where you have to swap SSDs often and large-capacity ones are really expensive. Your open face p2 card idea makes no sense.

DURING DAY:
1. Buy 4 x 128GB SD cards (total of 512GB for under $1000 - nice!).
2. Put in camera.
3. Record for EIGHT HOURS.

END OF DAY:
1. Stick those SD cards into your laptop and back up to multiple drives (use the built-in laptop SD reader, plus an external SD reader and you can do 2 at once...)
2. Lock original SD cards away in a secure place.
3. Take out fresh set of 4 x 128GB cards (SD cards are cheap) and put them in the camera for the next day.

OPTIONAL: do a mid-day extra backup if you want to.

Sorry... but how is this more difficult than a RED Scarlet shoot? Unless you have a whole DIT team or $10K worth of SSDs?

Bruce Allen
www.boacinema.com

 

[Felix K.]

OK, I'll bite: you're forgetting - this isn't a RED camera where you have to swap SSDs often and large-capacity ones are really expensive. Your open face p2 card idea makes no sense.

DURING DAY:
1. Buy 4 x 128GB SD cards (total of 512GB for under $1000 - nice!).
2. Put in camera.
3. Record for EIGHT HOURS.

END OF DAY:
1. Stick those SD cards into your laptop and back up to multiple drives (use the built-in laptop SD reader, plus an external SD reader and you can do 2 at once...)
2. Lock original SD cards away in a secure place.
3. Take out fresh set of 4 x 128GB cards (SD cards are cheap) and put them in the camera for the next day.

OPTIONAL: do a mid-day extra backup if you want to.

Sorry... but how is this more difficult than a RED Scarlet shoot? Unless you have a whole DIT team or $10K worth of SSDs?

Bruce Allen
www.boacinema.com

I hear you. New options = good. There is a market and some companies start to see it and deliver.

 

[Steven Veldkamp]

OK, I'll bite: you're forgetting - this isn't a RED camera where you have to swap SSDs often and large-capacity ones are really expensive. Your open face p2 card idea makes no sense.

DURING DAY:
1. Buy 4 x 128GB SD cards (total of 512GB for under $1000 - nice!).
2. Put in camera.
3. Record for EIGHT HOURS.

END OF DAY:
1. Stick those SD cards into your laptop and back up to multiple drives (use the built-in laptop SD reader, plus an external SD reader and you can do 2 at once...)
2. Lock original SD cards away in a secure place.
3. Take out fresh set of 4 x 128GB cards (SD cards are cheap) and put them in the camera for the next day.

OPTIONAL: do a mid-day extra backup if you want to.

Sorry... but how is this more difficult than a RED Scarlet shoot? Unless you have a whole DIT team or $10K worth of SSDs?

Bruce Allen
www.boacinema.com

One word; fanboys.

Can we all please hold judgement until we see this cam in action?
It looks like it's going to be a very affordable 4k cam that doesn't require 4-5k in additional modules to make it boot up. Sort of like a Scarlet fixed.
I say GO JVC!

 

[Dan Kanes]

Yes, clearly a "fanboy" opinion because muxing four avchd streams is a really good workflow.

Sorry but that's a stupid idea and I'm just illustrating how "hacky" it is.

If they had muxed the streams onboard and used xqd or sxs cards they might have had something.

 

[Björn Benckert]

Takes to a whole new level DIT skills. 4 times the upload, 4 times the checking, 4 times the transferring to another HD. 4 times the chance of getting those little bitty cards muddled or lost & 4 times the stress of knowing if your "got it in the can". Plus a fixed lens how much better can it get!

I think you actuall just put your computer cable in the back of if and suck out the move files. Since its so heavily compressed you only need to do so once a day or so....

So yes, it's almost like gaffing 4 gopro cameras next to each other and say you shot 4k...

When is the data per frame or second, and true resolution going to be what defines resolution? I think the pixel count is just one parameter that is pretty far from telling the resolution of the image.

CMOS is not so good you need to downsamle at least 1/5 to get a some what got image. I think even 1/2 is not a overdue. Then there are those sensors that have RGB sensors for each pixel wich far exeeds the CMOS. and 3CCD and so on so. Pixels is one thing. This JVC camera might gather 4k but it will not be of any quality that you would even want to use.

It's a bit like comparing ALEXA with the first camcorders that had HD written on them... They sucked, just as this JVC thing will suck.

 

[CK Olsen]

Just imagine if one quadrant/card goes corrupt... nothing like trying to explain why you have only 3/4 of yout talents face in a take...

 

[Hrvoje Simic]

Easily the dumbest idea anyone has ever come up with EVER

H.264 quad HD 60p on cheap media.

Maybe you can offer a smarter solution to JVC engineers.

 

[Andrew Walker]

I'm sure there's a saying for cutting corners but lets just say that comparing this camera to a Scarlet is ridiculous. I'm sure people will be able to get a nice image out of it. I thought when the DVX100 came out it captured a nice image given the right conditions. But I have a feeling that this is going to be more trouble than its worth.

 

[rod bradley]

Almost all innovation in any field and in any era has been called the "dumbest idea EVER" by any number of people. As someone pointed out, the same arrows were shot at Red. JVC has some brilliant engineers and is a better
engineering than marketing company. Among other things they produce some of the best home theater projectors around and when Digi-beta was king they produced a quality equivalent at a third the price with a much
more cinematic look. They've also produced some pretty nifty prosumer cameras. I wouldn't bad-mouth JVC and their engineers with quite such extravagant delight before you actually see first hand what they've done.

 

[Maik Müller]

I have to say JVC surprised me with that camera. Its not comparable to Scarlet-X, thats clear, but it has its niche and sets a mark. All bad things aside, they're trying to push the limits and that should be honored. Did anyone else of the so called big players announce a 4k camcorder? Sure the "creative" solution they're bringing to market looks like some prototype-stuff that never should leave the labs and it made me chuckle, but in the end the only thing that matters will be the performance that you can get with it.

Being able to shoot 4k now will make a difference in the future and if this camcorder delivers good output then i see lots of happy customers. They will probably have to do some weird work now and laugh about it in the future, but they will also be happy to have shot in 4k instead of HD. And then thats the only thing that matters.

 

[Karim D. Ghantous]

Some thoughts:

1. Finally, low-budget film producers will be able to ditch those horrible 5D2s. The JVC is only twice the price AFAIK, but a bit more than twice the resolution. I think it will sell.

2. I don't have a problem with their four-card approach. But I would have preferred a different recording method: instead of one quadrant per card, have one out of four pixels per card so that each card has a complete 1080-line version of the image (makes it easy for quick previewing on cheap equipment).

3. I hope that some kind of follow-focus can be attached to the lens. Surely that has to be worth something. That is a focusing ring I see on the lens, right?

 

[Tim Nitsch]

This JVC camera might gather 4k but it will not be of any quality that you would even want to use.

It's a bit like comparing ALEXA with the first camcorders that had HD written on them... They sucked, just as this JVC thing will suck.

Hense the $5k price tag. If it had all the shit that you wanted it would be a scarlet or epic for many many times the cost. LOL what exactly were you expecting dude?

To cut costs you need to get creative, and JVC did that, very well IMO.

 

[Benoît-Joan Clariana-Roig]

Just imagine if one quadrant/card goes corrupt... nothing like trying to explain why you have only 3/4 of yout talents face in a take...

On a classic one card/SSD system, if your card/SSD is corrupted you get NOTHING.

 

[Maik Müller]

1. Finally, low-budget film producers will be able to ditch those horrible 5D2s. The JVC is only twice the price AFAIK, but a bit more than twice the resolution. I think it will sell.

Can they? Wasn't the hype around the 5D2s mainly based on the fact that you have a big sensor and can use different lenses? Oh and don't forget that super-shallow dof...

2. I don't have a problem with their four-card approach. But I would have preferred a different recording method: instead of one quadrant per card, have one out of four pixels per card so that each card has a complete 1080-line version of the image (makes it easy for quick previewing on cheap equipment).

This would work with a lossless solution, but probably not so good with h.264. Decoding such a stream doesn't give you the individual pixels back. It would also make reencodng mandatory when it comes to creating a single 4k-stream, while JVC's solution, in theory, could go without it. We have to see how JVC's plugins work, i hope they do it the smart way.

Splitting the 4k-image into 4 FullHD-images seems to be the best solution for their plan to build a cam without a new chipset.
BTW, wasn't the Falconbrid-chipsset announced last year to have 4k2k-support? And now they have to put in 4 of those processors and can't even write the output into a single stream...

3. I hope that some kind of follow-focus can be attached to the lens. Surely that has to be worth something. That is a focusing ring I see on the lens, right?

Doesn't look like a proper focus ring, i think it will just be a multi-purpose electronic ring. But do you really need a FF on such a lens and that small sensor?

 

[Benoît-Joan Clariana-Roig]

BTW, wasn't the Falconbrid-chipsset announced last year to have 4k2k-support? And now they have to put in 4 of those processors and can't even write the output into a single stream...

Doesn't look like a proper focus ring, i think it will just be a multi-purpose electronic ring. But do you really need a FF on such a lens and that small sensor?

The Falconbrid supports 4K2K but unfortunately you have to split the 4K stream because there's no single SD card today that can supports that amount of data/sec and no real 4K codec yet.

Some ppl tested the focus ring/rocker at the CES : it works very well. Focus ring seems very easy to use and effective on the prototype.

Wait and see.