John Matthews

I will say Panasonic does do some magic here. I've had cameras that produce a very digital line around anything blown out. Panasonic seems to render with a little more gradient (and grace) so that it's not as noticeable. The goal was to do "run & gun," not think too much about the shot. To bring her face out more, I would have installed a reflector had I thought a little more. But even then, it's all a matter of preference I think. Some people really like a darker image than others.

The GX80 is meeting my expectations for run & gun. I think this shoot exemplifies expectations one can have of this camera for your quick, set-it-up and shoot style videos… family, street, etc. Here’s my setup: GX80 Panasonic 25mm F1.7 Zoom H1 Sony ECMCS3 Clip style Omnidirectional Stereo Microphone Settings: GX80: Natural profile (Contrast = -5, Sharpness = -5, Noise Reduction = -5, Saturation = -5), White balance (Sunny, A3, G3), ISO (200), Sutter Speed (1/50), Aperture (F2.8), IBIS (on, but no electronic IS) Zoom: WAV (24-bit, 96K Hz sampling), Level (70) Description of shoot: This took me about 5 minutes to shoot. I saw an interesting opportunity, went to get the GX80 and Zoom H1 from my office. For the settings above, I only need to put the camera in “C” mode to start. The Zoom is always put back with the level at 70, I only need to place it near the subject and start recording. Post Processing: Imported sound from Zoom H1 (just connect, don’t take out card) - 30 seconds Imported video ~3 minute video file - about 3 minutes Synced files - 10 seconds Threw it on a 4k timeline - 1 second Expanded audio - 1 second Select camera audio and push “v” key - 1 second Added 1db to Zoom H1 audio - 1 second Collapsed audio - 1 second Make cuts - 10 minutes Color corrected (highlights = + 15%, midtones = +7%, shadows = +1%, Global color 110° = -3%) - 2 minutes Added FilmConvertPro2 effect (KD P400 Ptra, Grain = 0) - 10 seconds Added a simple title - 10 seconds Rendered video - ~20 minutes Total Post Processing time = ~40 minutes (with rendering) Re-rendered video at constant bitrate of 35000 kbps in 4k with Handbrake, resulting file 690 MB- ~10 minutes Uploaded to Youtube - 90 minutes Youtube processing - 10 minutes Final Product: Picture Quality: Hindsight is always 20/20. I think the scene was a little dark as her face falls into shadow from time to time. I probably could have bumped up exposure and just avoided the window blown out in the background, but it gives you an idea of the limitations in terms of dynamic range. For me, the colors are quite nice and pleasing after applying Filmconvert… before that, not so much as white balance was off due to keeping blue channel from going too low- I guess many cameras need to be tweaked. Notes: This is a video of my daughter trying to set-up a Canon Powershot A410 that I gave her. She practices from time to time and enjoys shooting. Capturing moments like this is one of the main reasons I bought the GX80 for family. Please excuse her coughing as she’s getting over the flu. Also, note that she’s constantly switching from English to French. Good luck trying to understand, but you still have an idea of the audio quality. Of the video I cut out was a portion that was probably unusable as I wasn’t holding the camera steady enough as I moved to another position. Yes, it’s prone to jitters if the IBIS goes past its limitations, but I could easily have avoided this hadn’t it been to user error (movements were simply too fast). Feel free to make comments on any of the above as I’m trying to improve everything! By the way, here’s a photo she took of me looking serious during this shoot.

I'm going to make a few more comments on settings. I did some testing and confirmed that Natural (-5,-5,-5,-5) gave best results in terms of noise in the blue and red channels. Also, I did a test for WB(A3,G3)- this makes sense to pull away from blue because the blue channel seemed to fall below 0%, green did pop up a little too much though and I might try A3,G2 instead. Finally, I'll be shooting with zebras set at 100% rather than 105% due to banding in the blue channel when clear sky is present. I think the top 5% has very little latitude and that's why banding was present. I've been slapping filmconvert (KD P400 Ptra, grain at 0) and it seems like a stellar image. For audio, I have my Zoom H1 with a cheap omnidirectional lapel mic with a small cord that I place next to the subject. So far, so good.

I wrote Gordon and here's his answer (thank you for the quick reply):

GX80 manual says the same... I don't know why Panasonic is doing that to us- then they don't tell us if it's 8-bit 4:2:2 or 4:2:0. I believe the GX80 one-ups the GX8 on that point because it will record internally at the same time. I would say both cameras are very capable. However, I believe this person is after value based on what he said... I'm going to say the GX80 beats the GX8 because it's less expensive. If he wants weather-sealing though, the GX80 is NOT the way to go- choose GX8 of GH4. I just pixel-peeped it and still couldn't tell a difference. Staying with the original file.

First, let's get the naming right! It's GX80... I mean GX85... I mean GX7 mark ii. If value or bang for the buck is what you're after, this is it. The other cameras you mention probably all have significantly better EVFs... probably due to form-factor. The IBIS in the camera is stellar. The 4k crop will be less than the GX8 and GH4. At the end of the day (Brit expression), it's most about which feels best in your hand and how important portability is for you. For me, it was THE reason for me to consider MFT. The other cameras are all bigger than the GX80.

When I first got a RX100 a few years back, someone recommended using 5dtoRGB. Does anyone use this software? It's supposed to retrieve a maximum amount of information out a 8-bit file. I tried it for fun on a GX80 file, but I really couldn't see much difference.

I'm really happy with the GX80 that I purchased over a month and a half ago. The community for this camera and the activity around have been great on EOSHD. Like all systems and cameras, they're going to have faults, but this one is great for photography and video. I'm sure the GH5 will continue the tradition.

Thanks for the response @hmcindie. Cameralabs does pretty good work and I think Gordon would have gone through the right process for his shots. He could have made a mistake though... He talks about the balls in his article and the difference in size and their shape (the speculars were not round to begin with)... he doesn't give a reason though. Maybe, I'll just ask him. But are you saying it's impossible to have such a difference at F2 from one lens to another regarding bokeh balls? Is F2 always F2 regardless the lens or how wide of an aperture it starts?

Not my point... read my comments before. I was simply stating that the flange on the G-master lenses was large compared to some of the DSLR offerings. I'm hoping that this Hasselblad would not fall subject to the same problem...

So, I watched part of their livestream and they're definitely promoting these cameras for their small size. The 2 lenses they're putting out there are a 45mm F3.5 (notice they don't show the hood on the marketing material) and a 90mm F4.5. I wouldn't regard these lenses as "fast." I wouldn't consider them really pancake either. I'm beginning to wonder if, indeed, they will have the same issue as the G-masters. That said, I'm sure they'll be amazing still cameras... I doubt it for video... not unless Sony gave them a whole bunch of tech.

I'll let you judge for yourself: http://www.cameralabs.com/reviews/Panasonic_Leica_DG_Summilux_12mm_f1-4_H-X012/ Then, come back and tell me what you think as to the reason for why this is. I said front element, but I'm not 100% certain.

Please explain this then: why are the bokeh balls almost double the size at F2?

Is it possible that Sony's e-mount was originally made for APS-C, but then they realised that they could put a FF in it? Those G-master lenses are the state-of-the-art from Sony- the absolute best work they have in 2016, yet they still have that huge flange on them. Why isn't Sony producing wide, fast lenses for the mount, ones that don't have a massive flange? Correct me if I'm wrong, but I don't think they can. Also, don't they need a bigger mount for the IBIS? That sensor needs to move around in there. The rebuttal shows images of an older Fuji camera and a big lens. Let's try putting the Fuji's 16mm f1.4 on a modern Fuji body and see how it looks... just saying. Back to Hasselblad, I just thought they people should know the "pocketable" medium format camera isn't right around the corner- that's my point.

There was a rebuttal to that article. It's here: http://petapixel.com/2016/04/05/defense-sonys-pro-mirrorless-cameras/ In it, even mirrorless owners agree that mirrorless is not about size. I agree that people love the features about the A7 series (when they work). But how many of them bought these cameras thinking they would be lighter and smaller? I'm thinking there were many.

Yes, small camera, but big or slow lenses. Am I wrong? You still need to cover that sensor. People often say mirrorless is about size, but I think it's mainly about features (EVF, adaptors, etc.). Sensor size is about size. For video, I doubt this camera from Hasselbald will provide what you need (readout and feature set) for your next IMAX feature... then again, maybe it's the step in the right direction people have been waiting for and it's going to sell like crazy.

Thank you for your post and reading 22 pages worth! The GX80 crop is 2.22. Your 14mm will become a 31mm in 4k. In 1080p, it'll a be 28mm.

I just hope it doesn't run into this type of situation. Three FF cameras to scale: notice the mirrorless one has an enormous flange. I think it's just physics. Petapixel had a whole article on this here: http://petapixel.com/2016/04/04/sonys-full-frame-pro-mirrorless-fatal-mistake/

The body looks nice, but don't people think it's a bit silly to carry a bunch of lenses, each with a huge medium format flange? I don't think they're getting around the physics. Same thing happened to the Sony A7 series cameras. Upon adding a G-master lens, it's the same size as its DSLR counterpart... the only difference is that the DSLR user will only have ONE integrated flange on the body to carry around.

Ok. I get it. Just to give some of my experience with it... I've noticed that the blue channel seems rather noisy in the shadows (nothing like my old RX100 though). Red channel seems to evoke a sort of blurring/glowing effect. Green, as always, is a champion. If someone could tell me how to get the best results from these channels, it'd be very helpful. I was planning on doing some tests myself this week. I think one of the most difficult things to do with 8-bit signals is to get accurate memory colors (like skin) in shadowy areas. So far, I've been using the standard profile with reasonable results.

That was a mistake, right? Haven't tried this, but I'm certain you'd get inconsistent results doing so.

I think that's about right. 2.22 times crop from 35mm.

Thank you for responding. I'm genuinely interested in the reason for why that is because the difference seemed huge (almost double in size) between the Panasonic f1.4 and the Olympus F2! I know you can change the shape of the bokeh by placing a dark piece of paper with it cut in the shape you want in front of the lens, but I didn't know the front element size determines the size of the bokeh ball... interesting.

OMG this discussion seems pointless. FF vs. MF vs. phone vs. etc. That's not what this post was about. Let's move on... next. Guess what... "You can't convince someone against their will." Someone's grandmother once said that.

I really don't see the big deal in syncing... During the shoot, I would spend about the same amount of time setting up an external recorder versus a wired or wireless option. Throw the files on the computer. Ok... this takes the most time. Depending on the file size, it could take as long at 1 minute for a big 24-bit audio file. Syncing in FCPX: Select both files. Right-click and select: "Synchronize clips"... 15 seconds later... voilà... works like magic. In the timeline: Expand audio. Select audio from camera and press "v" to disable audio. Adjust your levels and you're done. 10 seconds. So, what we're really talking about is 2 minutes (maximum) of extra time, but you get a fantastic audio file and it costs much less. If your time is worth that much, I'll argue that you wouldn't have asked the question on this forum. If fact, the only "hard" part of this whole thing is keeping track of which audio file goes with which video file. The only real negative that I can think of to this method is during the recording. You can't adjust and monitor an external recorder (placed next to your subject on-set) if the sound becomes too loud... audio ruined. So care needs to taken when setting the levels.