Maybe we can get Lumix rep from NJ HQ Sean Robinson to join this forum so he can push these ideas forward to Pany. If not, we can still forward this to him.

You can post more, we can print them formal and hand them to consumer division at PhotoPlus.

In addition to interview we plan.

@driftwood may be the best connection via PV.

Well, to be connection you need to communicate first, and Nick is not really master in it.

That was quick, here is Seans answer:

Always happy to read and take the suggestions. Can't make any promises for implementation, but we do read them and pass them along to the relevant teams

@apefos I think asking for 33 mpix and better dynamic range and less noise are mutually exclusive. So your asking for a giant leap in sensor tech perhaps to high an expectation. But what the hell reach for the stars. I think you might be dead wrong on the autofocus issue as competitors give us better and better autofocus the failure that is DFD will be more apparent and people will leave for better systems.

I am old enough that it use to be said stills autofocus was undependable and not for professionals. Until it became more dependable and faster than the best professional and it is taken for granted now. In time video will go thru the same evolution I'm sure and if Panasonic doesn't get there shit together they will fade.

As to 8K I think we will have to see how many people have 4K tvs by 2018 lol. Otherwise your talking a tiny niche group. I think 8K will be 2020 maybe and they will have to find amazing sensor tech or accept it will only be on ultra expensive high end FF cameras.

33 megapixels is not that much comparing to current 20 megapixel. It would be less than a fstop difference in terms of noise. If GH5 and G9 can do good video up to iso 1600, the 33 megapixel m43 camera could do it with same amount of noise in iso 800 or 1000.

A better sensor technology could keep the 33 megapixel sensor in the same amount of noise comparing to current 20 megapixel sensor, or even less noise.

33 megapixels would be also very good for photo work.

The sensor would need to be 7680 x 5760 in 4:3 aspect ratio, it would be 44 megapixels total. This way it would be 1 or 1 1/3 fstop difference in terms of noise.

Another approach Panasonic could do is to follow the Sony philosophy and release a Sensitivity version, with only 9 megapixels, to do low light 4k video.

A good strategy for Panasonic would be:

GH6: 9 megapixel sensor, 1,9:1 aspect ratio, 4096 x 2160 pixels, sensitivity version, 4k video, with 21,6mm diagonal sensor designed for video, and 36 megapixel high resolution photo mode. (GH6s)

G10: 44 megapixel sensor, 4:3 aspect ratio, with 7680 x 5760 pixels, with 8k video hevc 16:9, and 176 megapixel high resolution photo mode.

people who wants better low light 4k video would buy the GH6s and people who wants 8k video and better photo would buy the G10.

Or it could be two versions of GH6: GH6s (sensitivity) and GH6r (resolution).

Multi aspect sensor would be nice. GH2 was the last to have it.

Dual Native ISO would also be nice if possible.

Global Shutter version would be a dream.

Dual Native ISO would also be nice if possible.

No such thing as "Dual Native ISO" exist, it is marketing bullshit. I'll later make post about thing they call such.

Global shutter leads to significant issues with noise and hence - DR, plus it'll be sensor price hike.

Make the GH6 even easier to use. Should work better in semi-auto modes (so content creators can focus more on framing and what's going on in front of camera) and ofcourse still please the manual crowd.

I second many of the good suggestions already, including them and some other in the list below:

General:

• Reliable and fast AF
• Better low-light performance

Audio:

• Support 4 channels of audio
• Support wireless monitoring of audio via bluetooth headset

Connectivity and interface:

• Better utilisation of smartphone as second screen
• Rethink how we offload and log footage. Smart tagging etc.

Experimental ideas:

• Cooperate with Aputure (or similar) to make the Vari-ND DEC Adapter (or similar) work perfect; strip away all external boxes and eliminate need for external power, make the Canon lenses Autofocus smooth with GH6 PDAF, buttons an adaptor and make custom config button on Panasonic camera to adjust variND on adaptor.
• Integrate wireless audio reciever in camera (aka Rode Wirelink type tech) - I really miss keeping having wireless audio, while keeping camera compact stealth.

Support wireless monitoring of audio via bluetooth headset

It is really bad idea. Monitoring means that issue you hear is issue with recording, not with horrible Bluetooth performance.

Imagine the GH6 with 4096x2160 9megapixel 4k sensor with dual native iso = low light/available light monster!

4 channel audio recording with adjustable ALC on 2 channels and improved sensor (lower noise, better dynamic range) for me please.

@apefos

So how do you explain the fact that iso 5000 in varicam is less noise than iso 4000 and looks like iso 800?

I'll make detailed post later and explain why and how it has slightly less noise.

@apefos

Exactly. Mitch is best Panasonic representative I know. But he refused to tell how it is really done.

So, how it all works (it is simplified version, of course)

First. Initially you have photons passing filter on microlens above sensor

In a sensors quantum efficiency is the percentage of photons hitting the device's photoreactive surface that produce charge carriers. It is measured in electrons per photon or amps per watt. Since the energy of a photon is inversely proportional to its wavelength, QE is often measured over a range of different wavelengths to characterize a device's efficiency at each photon energy level. The QE for photons with energy below the band gap is zero. Photographic film typically has a QE of much less than 10%, while CCDs can have a QE of well over 90% at some wavelengths. Individual pixel where photons cause certain charge (measured in electronvolts).

Second. After exposure individual pixel can be seen as bucket filled to certain level (charge level)

Part of this charge is not caused by photons induced charge, but from "drunk" random electrons from sensitive area itself and stuff around it. The more heated sensor is, the higher is this noise level.

So, for now it is few ways to increase sensor performance - increase quantum efficiency (very hard) and decrease temperature (used in pro equipment like astro one)

Third. Now we need to get value of this charge (it is very small charge!) in digital form, preferably as accurate as possible. And it is big amount of such pixels.

So, standard way is to amplify signal and later use very fast ADC converters.

Here amplifier adds his own noise (as well as banding, due to simple fact that it is impossible to make exactly same amplifiers).

ADC are also not ideal, least significant digits can be digitized not very accurate.

Fourth. If you have large pixels, good sensor and good margins you can use other approach.

It is not unique, used in any audio recorder able to record safe track. As well as Alexa and BM cameras.

In one option of this approach you can focus on performance of ADCs and use pair of 12bit ADCs for example to get 14bit (or 16bit) data. Digital merge stage also most of the time located on CMOS sensor as it reduce data burden.

Thing that Panasonic mention under "Dual ISO" most probably means that they use very good high bit ADCs in both channels, like 14-16 bit ones (so you can switch to such ADC alone). And also are able to tell merging part that to do, so at high ISO they can use smarter merge.