kye

I'm aware you shoot with a range of equipment as you've mentioned it in other threads, but you did literally just say "Give us S35/FF shooters a major incentive to revert to a smaller sensor" only 5 hours ago! The way that I think about the GH5 is that it fits a niche. That niche is where: you shoot handheld in rough enough conditions to need top-shelf stabilisation, or you shoot manual lenses handheld, or you want it for really long focal lengths you want 10-bit internal with >100Mbps codecs you care about camera size and weight Put simply, it's anyone that shoots in difficult uncontrolled conditions. Adventure film-makers, more wild events, serious travel, etc. 1) The stabilisation criteria eliminates most FF cameras as they don't have the IBIS travel. I've been reading carefully and almost all FF camera discussions include a few reputable people saying something like "the IBIS is good but isn't GH5-level". That leaves a number of MFT cameras and a few S35/FF cameras. 2) The 10-bit internal criteria eliminates most of the other MFT cameras, and the >100Mbps codec criteria eliminates most S35/FF cameras up until the most recent batch of RAW-capable ones (eg R5) or the high-bitrate ones (eg A7S3). Most of those are pretty expensive too. 3) The size and weight criteria eliminates almost all the rest, like the S1H, and eliminates all the ones that rely on external RAW and have poor internal codecs. I'm sure there's a few other candidates that meet the above, but it's not a huge number. There's a strange mindset in the camera industry that says you either: want a small camera and therefore you're an amateur who doesn't care about image quality want image quality and therefore you're happy to have a huge camera and rig it out This doesn't really cater to people who want a great image but want their setup to be super-portable and inconspicuous. Being inconspicuous is often misunderstood to mean doing something wrong, but in reality it means not having a massive impact on the things you're shooting. Having a large / complex camera (or a tripod) when shooting in public just means that instead of getting shots of real life you get shots of people all staring at you and then you get asked to leave by security. There's a reason that people who shoot in public a lot (eg Philip Bloom) use really freaking long lenses - not much use if you'd prefer a wide. The niche of small camera with great image quality is a very strange place, with things like the GH5, Sigma FP, BMPCC and precious few others, and almost all of those don't have stabilisation. I agree that those that don't fit this specific niche may be well catered for in other systems, or even the OM-1, but I still think there's a real niche where this is by far the best option, and depending on your situation, perhaps the only option.

Is this because it will impact the weather resistance? If that's what you mean, it makes sense. I've always been too afraid to test that kind of thing!

I agree. AF testing is one of the tests that is done the least realistically and also the least consistently. I'd prefer someone like DXO or similar to setup an automated test that simulated the various common scenarios can be repeated reliably between various cameras, lenses, firmware updates and various modes.

That's what I'm waiting for 🙂 I'm having devious thoughts about a fourth setup, but I'm not sure if I'd find the motivation to shoot and edit 4 submissions!

@Attila Bakos - great presentation! It's nothing to do with resolution @tupp, otherwise the RAW files would be impacted too, it's processing. I'm not convinced it's chroma NR either, as normally NR is just blurring. Could it just be extreme compression? If it gave the Y channel most of the bitrate maybe that's what happens to the colour channels?

It's a rare day when a camera ticks all boxes for anyone.. I'd suggest buying a lottery ticket! It will be interesting to see people really put the OM-1 (and GH6) through their paces, both technically and also with the image. Unfortunately cameras tend to be mine-fields of incompatible features, like not supporting feature X while feature Y is on and it's in resolution Z, etc. It needs the Gerald Undone treatment essentially. Hopefully someone will post some SOOC footage soon and we can see how robust the image is, etc..

There's obviously a pretty significant difference between you and I. It's probably safe to say that we don't understand each other. That's fine, and I'm very aware that people come from different backgrounds and have different perspectives and values. But here we are in a Panasonic thread talking about the GH6... ...me posting here as a long-term MFT shooter and GH5 owner who doesn't use AF, who has posted a lot about wanting Prores and better noise performance and various other things that the GH5 struggled with and the GH6 seems to have, who owns lots of nice fully-manual MFT glass, and who will probably buy a GH6 in the next couple of years, and I'm saying that AF isn't necessarily a critical factor. ...and you're posting here as a non-MFT user, who seems to have heavy interest in huge resolutions and in PDAF, disagreeing with me about what the average GH6 user would want. I guess I'm just wondering, what makes you think you understand the target audience for this camera? Like, at all?

I think we'll have to see. The OM-1 spec sheet wasn't gobsmacking, but the images looked good. If ARRI was about to release the Alexa and someone leaked the spec sheet and posted it to the forums then in only a few pages people would be dancing around each other singing "ARRIs going bankrupt!".

I have said before that one path for the GH6 would be to do what Sony did with the A7S3 - to keep the same resolution but make it a more serious camera by improving codecs and bitrates and other factors that influence the image, and this seems to be what they have done. To put it rather bluntly, the GH5 is 5K / poor DR / ok colour / ok codecs and the Alexa is 3.4K / crazy DR / spectacular colour / great codecs and the Alexa image absolutely kills the GH5 in every aspect except resolution, so the path is either to prioritise resolution and put the things that make a great image to the back burner, or to ignore resolution and try and catch up on the things that matter. Panasonic seem to have chosen to improve what actually matters... Cool - how does it look with vintage lenses? It sounds very similar to the Alexa pipeline - but that doesn't seem to be reflected in the DR numbers? Other companies who chase DR list much larger numbers in their specs. I'm not sure how that lines up. That looks like the ARRI pipeline from the ALEV sensor: (Source: https://www.arri.com/en/learn-help/technology/alev-sensors ) Any idea on how they are similar / different? Unlike @MrSMW I don't have a physics genius on the line!

Ah, I was thinking more about just image quality in high ISOs (ie, images don't look like there's a technicolour snowstorm), rather than DR in high ISOs, but DR is a reasonable part of the image so that makes sense.

I guess we'll see. Lots of people also selling their GH5 and GH5S's to go FF Panasonic or to other brands because the GH5 features have aged, and many of these express regret at having to re-buy their lenses etc. Lots of the newbies aren't buying a GH5 because of the price, but from the reputation. I've seen more than one person claim "this is my dream camera of all time" or similar. There's also a much higher diversity in those groups where lots of people are coming from countries where their greatly reduced buying power makes these things much more expensive in practice. It's always been hard to separate hype from actual sales except when manufacturers brag about sales numbers they are proud of.

Looking forward to your thoughts, I'd imagine you're scanning the threads for the questions people are curious about. Are you allowed to confirm when the release date will be?

There's one file from the Komodo on REDs sample footage page.. https://www.red.com/sample-r3d-files Download links in description of this video: I'm sure there will be others around, just google "red komodo download footage" and see what's out there.

If you think that guy with his Russian accent is hick, you need to adjust your expectations.... Compare them to the Canon L-mount glass or Sony G-Master lenses and see how that comparison goes! However, you're missing the point. You can't buy an apartment for 1/5th the price of a house just because it's 1/5th the total area - that's not how houses work. The reason that they don't scale is because both provide a place to live, and both probably have the same number of kitchens and how water heaters and air conditioners etc. A GoPro is maybe 5% of the weight of a large DSLR camera, but it's not 5% the price because both are cameras that you can use to take images, and both have a lens, sensor, SD card writer, screen, etc. MFT has the same features as a FF camera, it provides much the same value to the customer as a FF camera (takes photos, video, etc) and still takes the same kind of money to manufacture (takes the same number of chips and electronics and the same number of people in a factory to assemble it using the same number of steps in the various processes etc). FF has more options for shallow DoF - sure. If you accept that you won't match the DoF then MFT is smaller and lighter. It's not a competition.... Actually, not always. There were no cheap kit lenses for the Canon flagship DSLRs like the 5D. It came with constant f2.8 zooms, which aren't cheap at all. If you don't care about shooting a lens wide-open then actually most lenses are great. "F8 and be there" works spectacularly well on most lenses. Besides, the kit lens limitations actually tend to look a lot like the characteristics of vintage lenses....... check out this comparison I made.

I find this codec table useful: https://blog.frame.io/2017/02/13/compare-50-intermediate-codecs/ Prores has about the same bitrate per pixel regardless of resolution, so UHD is 4x the database of 1080p because it's got 4x the pixels - twice the width and twice the height. From this standpoint 5.7K should be about 1.6Gbps, so it's more than Prores HQ and less than Prores 4444 (which would be 2.2Gbps). Still, this is an incredible bitrate and may as well be RAW. I'd assume those would all be included. Fun times for film-makers who prefer to capture in a professional codec in the resolution they want to distribute in! If your exposure to a cameras popularity is only on forums like these then you will be missing the incredible popularity that the GH5 still has to this day. I'm in GH5 FB groups where new people are upgrading to the GH5 and posting the newbie questions regularly, like what gimbal is best? what cage? best lenses? best vintage lenses? can it shoot RAW? how can I get it to do X? etc etc.. TBH those groups are a completely different world than here, and other forums. There are people out there filming paid gigs for music videos and weddings and the like that have just bought the camera and are filming them with the kit lens and don't know that colour profiles exist. People post their (often paid) work all the time, ask for feedback, and get 50 people replying recommending they buy lights and study the framing and camera movement of movies and high-end productions. Yeah, the AF seems to sort-of jump and pulse. This might be tunable with settings perhaps. The AF issues from the GH5 were mostly it being slow or just deciding to focus on the background rather than the person in the middle of the frame, so maybe those are lesser issues. It's not like the other manufacturers have worked it out though - I still see AF fails on vlogs shot on Canon and Sony cameras. They tend to be included in the part where something funny or interesting happens and they want to include it for storytelling purposes and couldn't re-shoot it, so those moments are often little BTS on what the camera is really capturing, and sometimes the AF is seriously ugly.

Reusing parts makes sense, so potentially yeah. I dunno... if the pic didn't have a honking-great cine lens on it, made even larger with wide-angle distortion, maybe it would look more reasonable. The GH5 is very nice aesthetically and I have larger hands.

One thing that I read that apparently makes a huge difference is the inter-ocular distance (the distance between your eyes) as its critical for the angle of your eyes vs the depth of the thing you're looking at in the environment. I bought a headset for my phone that allows this distance to be adjusted, but sadly many of the VR players that websites use for streaming content offer practically zero settings. Mobile VR Station for iPhone has TONNES of adjustments, but the support to be able to stream content in it seems to be patchy and the ability to download files from many sites is blocked, presumably for copyright reasons, so you'll be stuck with whatever resolution your internet connection can provide at that exact moment. Also, the headset that I bought allows a focus adjustment (I'd assume it's similar to adjusting the focus of an EVF) and I've been able to get some great synergies between that adjustment and the distance of objects in the VR environment. It's an odd thing because the alignment isn't something you can feel directly, but when it's aligned it feels more natural and realistic. Of course, if you then look at something further away or something moves closer/further then you'd have to re-adjust that - kind of like pulling focus - so it's rather odd. BUT, it does work, even with objects that can be very close to the VR camera. The whole thing is rather sub-optimal. I believe that one of the most important aspects is any delay between you moving your head and it adjusting the image - otherwise you begin to feel drunk / drugged pretty quickly. Needless to say, that requires real graphics processing power. I suspect that this will be where dedicated hardware will come in. The latest processors seem to handle RAW 8K just fine so they've gotten much better in a hurry, so I'd imagine that we'll see decent performance from a range of products. Definitely. I've tried a range of dedicated apps, Within is one example, that allow you to download the content so you can watch it in the highest quality. Do YT or the other VR streaming platforms offer an app that does this? I can imagine we'll potentially find ourselves in a similar situation to the one we're in now for streaming, where there are multiple companies that control the content and you have to pay to get access to it. Downloading the content and then decoding it into a much easier format for playing could be a way around the processing power required - decoding to an ALL-I file or something perhaps.

If you're not familiar with the history then it's worth some googling - they're always first with every tech and they basically control the large technology choices. When it was VHS vs Beta they chose VHS and so VHS became the standard for consumers, when it was Bluray vs HDDVD they chose Bluray so that's what became the standard for consumers too. Essentially, whenever you get two groups of manufacturers that are about equally sized promoting different standards then the adult entertainment industry gets together and picks one and then announces their decision, and considering how much money they represent they become the deciding vote. They also have tonnes of money to spare for tech and R&D. The consensus is that they'll be the ones leading the development of humanoid robots too. Gaming is starting to become its own significant force in the market, so VR will likely get shaped by both markets. Interesting about rotation being an issue. I've experienced a number of videos where the camera moved, and even went from being stationary to moving and back to stationary and those were mostly ok, although I have gotten the sensation of losing my balance on a few of those transitions. Maybe the rotation has been implemented too jarringly? I recal Walter Murch talking about how in the early days of cinema there was a debate about if you could edit film, because people don't teleport in real-life so perhaps couldn't handle a hard cut. Similar questions were debated around the time that cars started going faster than a horse (and thus faster than any human had ever gone before) about what the limits of human capability are. It turns out that cutting film is fine, and the mechanism Walter cited was that we teleport in dreams all the time, so it's not a foreign cognitive experience. As someone who gets motion sick quite easily, I understand the source is a discrepancy between the motion perceived by the visual system and that perceived by the body (inner ear, and some other senses in the body too I think?) which is a sign that you've been poisoned and are now hallucinating, so the response is to reject whatever you ate/drank that is poisoning you. Another "things don't agree" problem in VR is when you switch from looking at something close to something further away and the angle of your eyes changes (your eyes get less crossed) but the focal distance doesn't change (you're still looking at the VR screens in your goggles). Apparently the solution is to have multiple displays (clear ones on top of others) so your eyes can change focus from one to another in the "stack". The early research I read suggested that not many layers were required, and perhaps only two were required, just so that your eyes change focus at all. I have experienced ending a VR experience and having trouble focusing afterwards as my eyes had kind-of forgotten they needed to change focus, so your brain does adapt even in a 10-20 minute experience.

Looks pretty good to me. If someone notices then they're probably not paying attention to the subject matter. I watched a YT video on colour matching different cameras and the guy said that the footage needs to look "like it comes from the same universe" which I thought made a lot of sense, as essentially as long as the colour differences aren't so obvious that they break the illusion then they won't be noticeable and so are fine. Interesting that the top one has the greener mids (the shirt) and the more magenta shadows (the shadow on the wall) compared to the bottom one, when they've been matched on skintones and the wall colour. How much colour work did you have to do for these shots?

You can do some of these, probably more than you think, but you have to do them slowly. Like all things in video - the adult entertainment industry is leading the way. There's good information around on what works and what doesn't seem to work, so the understanding of the medium is building.

Interesting lens. Panasonic released one for MFT but I've never heard anyone talk about it. In theory you could use the 5K 4:3 mode to get some reasonable resolution too. I suspect that a significant application of this could be virtual presence at events like concerts or weddings etc. In years past I have been into high-end hifi and making and recording music and the common approach to record an acoustic concert (basically anything in a concert hall) is to put a pair of microphones directly above the second row, and a pair high in the ceiling nearer the back and to mix them together so that when you listen with a high-end stereo setup it replicates the experience of attending the concert. Obviously audio is a little different to video, so for video you'd want to put the 3D camera front dead-centre for the best view. I can imagine wedding videographers offering a VR package where they put one or two of these in the good spot so that the whole event can be re-lived. Concerts might end up having a lot of them placed in various places and maybe you can swap between them as you like, or even provide an experience where every-so-often you move to a different vantage point to keep it interesting. One thing that's fascinating and we haven't really worked out yet is how you can "edit" VR footage. ie, I've seen people talk about how to transition from one place to another (basically you do it slowly through a dissolve or fade down/up so that people don't get disoriented - we're not that used to teleporting!) but we haven't really worked out more than that. It's an interesting space.

They wouldn't be the first people in history to end a relationship and get closure by bringing something new into the world, but I'd imagine it's frowned upon by their future partners....

Looks like they're taking design cues from the S1H... Let's hope that the GH6 isn't too big... I remember walking around Pompeii with my GH5/Prime/VMP+ combo in one hand and having a sore wrist just from carrying it around in one hand for 3-4 hours. Maybe there's a set of new "physical preparedness" exercises we need to be doing where we carry a brick around to build up strength. Like marathon training, 1 minute the first day, then 2, then 5, then 10... so that by the 6 week mark you can carry two bricks all day without straining any ligaments!

I'm not an expert in sensor tech, but Sony did a pretty amazing job of improving the ISO performance of their A7S line from normal FF sensors of the time, so if that can be done then maybe it is possible to have similar improvements in MFT sensors. If the GH5 is a reference point for ISO performance on MFT sensors then I'd suggest that we aren't anywhere near what is possible! I'm not saying anything definitive here, just that Sony showed that the tech can really be pushed - presumably with serious investment in the tech, and maybe that effort could significantly better low-light to MFT cameras.

That tech truly was incredible to see. It's nice to see that at least one company making money with product after product has decided to innovate - in anything other than their cripple hammers!