UncleBobsPhotography

From real works test max seems to be around 900 MB/s: https://petapixel.com/2020/09/22/cfexpress-a-real-world-performance-comparison/

Figured out the Zebras. As gt3rs mentioned earlier in the thread, I have to set second level to 95 instead of 100. I need to learn how to read more carefully.

Quick test both Zebra 1 and 2 shows up and I have it set at 60 +-5 and 95 in CLog3 (100 does not work in log)

Clog 1 is 400, not sure about clog 2 though.

It is interesting as they said in their test: https://www.slashcam.de/artikel/Test/Der-Dynamikumfang-von-Canon-EOS-R5---R6--Panasonic-S5-und-Sony-A7SIII-im-Vergleich.html that the R5 is actually on pair or even a tad better than the A7S III even before the CLog3.....  
"Im besten Fall sehen wir den Vorsprung der Panasonic S5 gegenüber dem restlichen Feld (Sony A7SIII, Canon EOS R5 und R6) bei maximal einer nutzbaren Blendenstufe. Sony nimmt den den Marketing-Mund mit 15 Blendenstufen am vollsten und landet dennoch in unserer Einschätzung auf dem letzten Platz, was auf den ungünstigen Mix aus Rauschunterdrückung und Kompressionsartefakten zurückzuführen ist. Es handelte sich bei unserer Betrachtung jedoch noch ausdrücklich um ein Vorserienmodell. Wir warten bereits sehnsüchtig auf ein Serienmodell der A7SIII, um dieses Verhalten noch mit mehr Messungen untersuchen zu können. Auch lässt sich mit dem gewonnenen Eindruck eventuell noch an den internen Parametern der Kamera einiges "drehen".

Die von Canon genannten 12 Blendenstufen scheinen uns dagegen eher etwas untertrieben. Sowohl die R5 als auch die R6 können in der Dynamik gut mit den aktuellen 14+-Spitzenreitern von Panasonic mithalten."
 
It is so hard to measure DR that is not really worth to look at numbers imo. In RAW R5 DR is more than fine if Clog3 brings it there (as I expect) then I really don't see why people would complain.... 

I had the edelkrone sliderone V1 (and headone V1). Typical Edelkrone build (and high price) but I didnt particularly like it. (Actually the first one I bought juddered but Edlekrone quickly replaced it.) It basically only works through the app and it was near impossible to set it to very long duration slides for timelapses.
A couple of months ago I bought the ifootage nano. It really is so much more intuitive and easy to use - you dont really need to use the app for standard slides - it considerably cheaper too. I wish I had had a chance to use it more but am Covid constrained. Gerald Undone has done a good youtube review of it.

I don't consider it one, it IS one.
So is the Varicam, Sony FX9/Venice, EVA-1.
Despite what people claim, the a7SIII and FX3 are NOT dual native ISO. They are dual gain. As is the RED Gemini.
Basically, dual native ISO is that there are two actual circuits underpinning each photosite and it will switch between them. With dual gain, the process is happening in a second ADC, but each photosite still has only one circuit.

There's always that contrast between mainstream and niche. Boring is subjective and relative, across culture and throughout time. As a child, you found cartoons interesting and the news boring; the inverse is (generally) true as an adult. 
The difference is that the internet has just exponentially increased the amount of media to a wider audience. Of course, the resulting popular media that emerges is quite vanilla--but many people like vanilla. If you're looking for the meaning of life in TikTok videos—that's a you problem. You're looking in the wrong place.
The number of innovative or experimental movies has increased as well. It just takes work to discover them. As it should. As it always has.
Some remember what it was like reach into the unexplored shelves of independent video stores. 
That was sometimes the best part.
 

Huh, yeah interesting, I guess people have different tolerances. I often shoot in a sound-controlled environment so the AF noise is pretty important to me. Especially with smaller crews now, it's good to have another usable audio track from the camera. 
STMs aren't uniformly quiet. USMs are also not all the same. I have a 24 f2.8 IS USM that is much quieter than the 35 f2 IS USM or RF 50 1.2.
The Nano USM models are all great, virtually undetectable. I just wish Canon would make fast primes w/ Nano USM. 
Anybody have experience with quiet, fast Sigmas?

I second this. I found the RF 50mm f/1.2 tempting, but after trying it I realized the focusing system was way too loud and the autofocus was quite jerky as well. I'm using an STM lens instead, and although it's not perfectly silent it's much more quiet and the focusing is smoother. It seems like the budget RF lenses are more suitable for video than the L RF glass.

The 70-200 2.8 RF is very smooth, silent and fast compared to the RF 35 1.8 so not sure you can generalize. 

STM is much quieter and faster than adapted USM lens but there is an external moving element so if you're planning on using a shotgun mic, it could still pick up movement during fast focus shifts in a quiet environment. 

I guess this could happen in theory, but I've never had any problems with it. I don't think the force will not be large enough to do any meaningful damage to your camera, at worst a small scratch in the LCD, probably not even that. The shot will be ruined, but triple tapping the trigger by mistake will ruin the shot regardless of whether the camera crash into the handle or not.

This product is cheaper than UV filters, improves image quality instead of degrading it, and protects your lens: Magic Lens Protector and Image Quality Improver
I've got a variety of filters, but don't use them all that often. The one I have used the most is variable ND filter just because of how convenience. Just got the Canon drop-in-variable ND-filter for the RF adapter, but since the sun has hardly come above the horizon I haven't gotten to use it much. The small amount I used it, it annoyed me a bit how easy it was to make adjustments by mistake. I wish it had a locking function or more resistance.

I use the screen flipped out with the Weebill Lab and the GH5 and I haven't had any problems balancing it with M43 lenses. The S5 is smaller and lighter than the GH5, so I would guess that it works, at least as long as you don't use the heavier lenses. I can't get the GH5 balanced if I use it with a speedbooster and Canon L glass.

I can see two fairly easy ways to get the effect on the humans:
-Make the people a mask by having them lit in front of a black background. Then combine the two source files in a similar way you would do to fake a double exposure shot
-Have them dress in all white in front of a background and have a projector project on them. I believe this technique is called something like "projector mapping". This is probably not done here as it's hard to avoid shadows, but a fun technique anyway.

This is just a Wishlist, not even speculation what a potential GH6 might have. But if I were Panasonic -- which I'm not! -- I'd go bold: micro 4/3s is always going to have a perceived, though not necessarily actual, image quality deficit versus 135 and APS-C formats. So play up your strengths and use the cost savings from the cheaper sensor to bundle in better tech.
For the body, pick up the refinements from the S series like that gorgeous 5.7m dot viewfinder, highres touchscreen, and 8 way joystick. Also sweat the details, for example, calibrating both the EVF and touchscreen from the factory.
Emphasize connectivity. I'd like to see the latest Wifi 6 and Bluetooth 5.1 or 5.2. Then add 2 thunderbolt 4 connectors. 
Aggressively ditch the legacy stuff. Cut the card slots and instead offer 128 or 512GBs of internal storage. Make transferring data off via thunderbolt or wifi simple, fast and dependable. 
Push the boundaries on tech. How about H.266 as an option? For stills, 10 bit HEIF and Display P3. And please, toss the proprietary raw and go with DNG. Adopt the latest 1.6 spec and use that semantic and tone mapping Apple's done on the iPhone 12 Pro (ProRAW is just a DNG file). 
For autofocus, I doubt Panasonic will go with phase detect. But maybe this'll be the update that makes DFD great. Drive that at 240 hertz with improved algorithms and processing power, especially for subject recognition, and it could be pretty good. While you're at it, profile the Olympus lenses (I suspect they may have done this but don't advertise the fact. But if that's the case, make it official. You want m4/3s to be as attractive a platform as possible). 

I am indeed! I meant to thank bloggers (including Andrew specifically) for bringing the issue of overheating up so much. The pressure about the timer nonsense forced Canon to change the way the camera monitors temperatures, which is the reason the mod works at all.

(Water)cooled Canon EOS R5
 

Komura
Kodak/Eastman Kodak
But I am not sure how compatible all of their lenses are with your focusing system...

I got the C70 last week as part of CVP’s first batch. 
Even though I have the A7SIII, I bought it because Canon obviously read a post on this forum about what my “perfect camera” looks like. So I just had to. I sold my EVA1. 
I put it in the deep end straight away on 2 professional shoots for fitness brands. The videos aren’t finished so can’t share any footage yet! 
My impressions: 
1. Body design is fantastic. Lovely in the hand and a joy to operate. Feels cheaper than other C-series for sure but still premium. Well built and easy to use. Very light too! 
2. The screen is flimsy and should be more robust. 
3. The joystick is drunk. You want to go left, but it goes up etc. Not sure if this is a “getting used to it” kinda thing, but it’s unusually tricky to get the hang of. 
4. Record button is also a bit stubborn. You have to press that thing very well to get it to record or stop. 
5. Quick touch screen menu is amazing. There’s no need to reboot the camera to change frame rates and codec like I did on the EVA1.
6. DIS is decent! Doesn’t feel overly robotic like other IBIS systems. But there’s a slight crop. 
7. DPAF with the Sigma 18-35mm is pretty good. My Sigma 50mm hunted quite a bit though. 
8. It has no tracking like the A7S3, only for faces. Seems the face tracking is disabled in 4k100fps. 
9. Footage in all modes is crisp and clean. Not overly sharp, but detailed.  It looks a bit more organic than the A7S3, which is clinical without a Promist. Noticed the A7S3 is a bit more contrasty in SLOG3 than CLOG3. The C70 has an edge in the shadows. 
10. You can use a V60 in 4k100fps but comes up with a warning  that it won’t work. It does, but sometimes cuts off due to a full buffer. Not often though. I’d still get a V90. 
11. H265 footage plays back better than the A7S3 on my 2025 iMac. Still a sludge though.
12. Footage can be easily matched to the A7S3. Took me about a minute. Amazing how much Sony have improved it. 
In conclusion, I had a good first experience and I’m looking forward to the next shoot. I currently prefer the A7S3 but it’s too early to decide that really. Both are great and can be used together very seamlessly due to the Sony’s improved colour and codecs. 
 
 
 

I think you've understood the situation quite well. h265 is more efficient, but high bit rates in professional cameras can make up for it while using less efficient codecs. h264 is a safer codec to use if you want to minimize frustration. h265 support has improved a lot recently, and will probably improve a lot the next 1-2 years, but we might end up with a situation like ProRes vs h264, where people prefer ProRes even though it's less efficient because it's easier to work with. At higher bitrates h264 will perform good enough anyway, so why not make life easier for the editors?

I feel like there is a crowd for these videos here. Apologies for the shameless self promotion though.

Sorry my first sentence wasn't clear. ARM outsources its production and designs chips. Apple Silicon uses elements of a the ARM architecture, but the chip is designed by Apple.
For those who are quoting specs like "only 16gb of RAM" and you cannot add more, I don't think it matters quite as much because everything (CPUs, GPUs, and RAM) are located on the M1 chip itself, rendering the chip much more effective at handling processes. Also, RAM and SSDs are becoming closer and closer to similar speeds. In short, this would not stop me from buying one of these machines.
Remember what Alan Kay once said: “People who are really serious about software should make their own hardware.” This had been quoted by Steve Jobs many times. A slower spect machine can outperform the higher spect machine, given optimized software. I'm not a Apple software fanboy, as they've dumbed-down so many of their products in recent years, making them much less capable (i.e. FaceTime, iTunes), but they'll always have a serious advantage over Wintel and AMD, especially performance per watt.

I am very positive towards ARM CPUs. My biggest question mark is whether they are able to get graphic solutions that compete with nVidia. ARM's earlier graphic solutions have been power-efficient, but not really comparable with the Geforce series. I hope the new relationship between nVidia and ARM means they will end up with nVidia GPUs as well. I have only had bad experiences with Radeon/AMD gpus.