Video Hummus

They are already hitting limits with the gyroscope sensitivity (Olympus has said the electronics are needing to be so sensitive that the earths magnetic field becomes a problem) on the latest IBIS tech in the EM1X. Adding this “smart” EIS in addition to the IBIS seems like a logical improvement. I would say it’s pretty high on the list, if not at the top. People I know bought the GH5 for the unlimited 4K recording in ALL-I and the IBIS. People buy Olympus for the supreme ability to hand hold a 800mm lens and nail shots of wildlife. It allows something like the tiny EM5 mark 3 to take hand held 50MPx high resolution shots with slower moving subjects. I would say the super power of MFT is the IBIS. It was designed for it from the birth of the mount. The smaller sensor makes it much easier to reach the 7.5 stops of stabilization in the EM1X, for example. I doubt it’s as easy to do that with a larger sensor. In the “bigger is better” days of camera sensors we live in today it would be wise for MFT manufactures to play the strengths of the system. Stabilization IS the big one. I could imagine a day where a GH6 or a EM1.3 is capable of shooting 6K video with near Steadycam levels of stabilization with near zero levels of fuss—or taking 50MPx or 80MPx handheld shots for portraits, street, wildlife, or landscape from a compact “small”, “low” MPx sensor that has amazing readout speeds with zero or very subtle motion artifacts. A smaller sensor is a trade-off. Sony at least has mitigated the lowlight sensitivity / high ISO performance of MFT sized sensors. What is left is now trading shallow depth of field for all of the things I stated above. I’ll take that trade. /rant Sorry for the thought dump.

I just watched cinema5D video they did with the SteadXP+. Why the hell is panasonic not doing this with their cameras? They have an accelerometer, gyroscope, and rolling shutter data already in the cameras for their IBIS systems. Seems like it’s not too far of a leap to record this data inline with time stamps for external software to use. The SteadXP looks great for non IBIS cameras like my GH5S. Requiring a 1/100 shutter for handheld isn’t that terrible. You could also fake a bit of motion blur in post for the critical shots. This sure beats the hassle of a gimbal. Anybody have experience with this thing? I see mixed reviews/experience with it.

I have these as well. But you have to be careful on wide angle lenses as they will cause vignetting when used with a step up ring. At least that is my experience using a 77-82mm step up ring and a XUME adapter with my 10-25mm. Slight vignette at 10mm but is gone at 11-12mm. Id also like to add, you could always rent some cameras to try the setup before committing to dropping several thousand dollars. Don't discount Olympus with something like the EM1 mark 2. It’s the only MFT camera with good PDAF, excellent IBIS (without the warp when EIS is on), and can get decent Shallow DoF with autofocus with one of their f/1.2 primes.

Ah, yes. The perfect system that doesn't exist... Maybe, take a look at Canon EOS R. X-H1 stabilizer is as good as A7III, i.e, not great. A7III codec is lacking. X-T3 doesn't have IBIS and it's H.265 files are, currently, heavy on editing rig. GH5 lacks good reliable autofocus and extreme shallow DoF if thats important to you. EM1.2 has good autofocus, decent color, good lens selection but has the same DoF compromise as GH5 as well as only offering 8-bit codec and weak LOG recording. This is why Canon needs to release a C100 Mark 3.

I just bought one. The more I use it the more I want a camera that pretty much eliminates the need for it 95% of the time. Here is a quick test I did 4 months ago with GH5S and 50-200 f/2.8-4 and Paul Leemings HLG corrective LUTs. It's a straight HLG -> Rec.709 conversion. No additional color grading. All handheld. Mix of 24fps, 60fps, and 120fps (the 120fps is Cine-D). Posting it here for anyone interested in the lens or camera combination.

For those using GHa on the GH5, how hard is it to shoot run and gun footage with this capture flow? For example, not always possible to white balance with a grey card. Using zebras primarily to judge exposure. Natural light etc... Ive been using Paul Leemings LUT with HLG and it’s pretty good but the highlight roll off of GHA is attractive. My workflow is to capture log or HLG use a transformation LUT to get to a Rec.709 and then I grade from there with a finishing LUT. I don’t like to fuss so much with crazy grades. I just want a flexible capture with great highlight roll off that gives me a great base.

I kinda agree. Panasonic has sold or shifted away from consumer products. The camera division is very much an outlier in the company. I love Panasonic’s higher end mirrorless cameras. I agree, I can’t imagine GH5 numbers over the past two years were bad. I think their cameras balance ergos, features, and build quality the best. I do expect some camera companies to fold. I hope Panasonic isn’t one of them.

Pretty sure those are refurbished open boxes but it’s a great deal for landscape photographers. Look at the EOS R. Dropped $500 since release last year.

Introducing The DJI Wanker

Well, with the general trends in cameras anyway the camera division at Panasonic will doubtfully become profitable in 2022 anyway, unless they continue to hide its unprofitability in umbrella divisions to keep it going. This sounds like a bad sign for Lumix consumer cameras. Makes me wonder why they spent the money and marketing to promote this new organic sensor and suggest it may be coming to Lumix brand if it wasn’t even in the short term strategy, unless I guess to seal the deal of this sale. Every camera company is bleeding at the moment. Canon signaled earlier they don’t see the profits in the business anymore. Nikon is in trouble with plummeting quarterly reports. Olympus as struggled with reinvestment and restructuring. Fujifilm seems to be doing...good. Looks like the Chinese will gobble up this market as well. Shame, because Panasonic Lumix has been doing some very interesting things of late.

Yes, but there is kinda of a catch-22 with the hardware performance these days and performances increases from generation to generation seem to be rapidly shrinking. Also most all of the performance increases in VR have been software based. So like I said it’s: Pixel size (because viewing device is so close to the eye) Latency (because of the real-time physical immersion effect) Resolution (for more visual vitality) The big problem is bigger resolution means higher latency at the moment. Hence why VR screens are trading resolution for latency and denser pixels because it is more important in this application of VR. Now VR video I think is not a cinema thing. It’s not really a good viewing experience either. But VR video is really cool for action cameras where moving composition to post is amazing for the creator. Asking the viewer to control composition just doesn’t work. It’s a neat party trick but not immersive for cinema or filmmaking. There will be 8K TVs. There will be a big push from TV brands to sell 8K to the mass public. Will the mass public see much of a difference? Eh, not really. It’s quickly diminishing returns.

Even in VR, especially in real-time video games refresh rate and pixel size is still way more important than 8K. 8K won’t mean shit if you get motion sick. They are already optimizing VR rendering, for games, that takes into account that humans don’t see detail on the periphery. So why waste time rendering detail? 8K+ VR action cameras are the future though and I’m pretty excited for that. The perfect action camera is the one that captures everything.

If it doesn’t have any required proprietary bullshit I think this camera will be a big deal for a lot of indie/small budget guys and a gateway drug to more expensive RED cameras. Especially in the light of the new metabones EF-RF speedbooster.

I’ll try it. Definitely some kind of DfD frame sampling mumbo jumbo.

Face detect AF at 30p is surprising usable on GH5S on my first inspection. However, 24p is definitely worst than 30p. I see no difference at 24p after update. So weird 6 frames more per second makes a difference I guess. 60p with face detect works really well and even better with 1-area for the classic “hold thing in front of camera”, especially if hold the object up first and then bring it towards the camera versus just flicking it into frame.

I think they have an incredible high bar to meet so they are taking their time. The GH series cameras from the beginning are all about pushing the limits. I don’t need HFR 4K. I want practical features. Improved IBIS, recording to SSD via USB-C, USB-C power delivery, better DR, internal ND, improved weather sealing. A GH6 with these features and the 10-25 f/1.7 or a speedbooster with a vintage lens would be an absolute beast. An internal ND + speedbooster would be very compelling for even FF zealots. Removing the shutter, and replacing it with some kind of ND unit would be the single best thing they could do! Sensor readout speeds on MFT are really good. Just have electronic shutter for GH6 and the future photo-centric G9/G10/G11 or whatever can keep the shutter. AF, eh, I’m fairly happy now. Would I like PDAF? Yes, of course but I’ll take next gen DfD too if I get an internal Nd solution. Please don’t repeat another GH6S debacle. I would say they would do better to minaturise the EVA1 to fill the role of any potential GH6S.

Glad to see AMD back in the performance game with the same low price strategy. I still think the future (at least in the mac world) will be fast but extremely efficient ARM cpus with dedicated hardware acceleration chips (like Apples T2, Afterburner, Neural chips) that will accelerate things like H.265 decoding/encoding, ProRes encoding/decoding, and Neural net algorithms they are starting to use for things like object removal, stabilization, and 3D tracking. A MBP with the T2 chip just chews through 5K 200Mbit/s HEVC footage from a GH5. Future cameras, I think, will be replacing H.264 with HEVC for a multitude of reasons. You almost need hardware accelerating encoding/decoding to play back that footage and edit it in real time.

Hopefully we will see a Gen.4 for GH5S. Great work these look incredibly similar.

I’m up in the air wether I want to do a custom AMD PC build with windows (I hate windows) or pay the tax for a decently spec’d iMac or go the MBP eGPU route. I really think all I need and would want to carry traveling is my iPad Pro, but now Kye has me thinking about eGPU again.

I wonder if they are using a Sony Semi. Sensor for the Komodo and future “budget” friendly red cameras and selling the name and perhaps color science and ecosystem. If the dynamic range is good and it has REDCODE and I’m not locked into buying silly overpriced red branded junk accessories I would be very interested. And I’m a hobbyist enthusiast.

Love how you think you have it all figured out but self inflict ignorance upon yourself. Winning strategy you have there. ? It’s not a judicial setting and why I think impeachment, as laid out in the constitution, is fundamentally broken. Impeachment will never work when facts and honorable behavior are not required in the process. Trumpers are a cult and they would rather drink the koolaid to the end then look at facts.

If it has better sound than the current Videomic Pro and the Deity D3 Pro it’s a win in my book and is worth the $230 they are charging for it. Criticizing the product for not being pro when as a audio pro you KNOW exactly what quality you need, is a bit weird. Cheap mics are anything below $1000 basically. I have a feeling Curtis’ Review will be positive and a highly recommended buy for people with a $300 budget.

Looks like a great value mic that would find a home doing something just about anywhere. Putting it in my list. I’m sure the sound performance is good.

Hope someone makes a MFT speedbooster with a eVND element in it that is powered off the pins with AF pass through. Have no clue if technically possible but I know I would pay good money for the convenience.

Thank you Andrew for calling bullshit for what it is.