BTM_Pix

I thought they did plenty well enough without my ham fisted blundering about ! Its portable to the NX though from what I can see of their protocol. If I had one. But prices seemed to take a big jump in the last few weeks of March for some reason

So, as you may recall, I have an interest in tweaking the combinations of the basic profiles in the camera along with the contrast/sharpness/saturation/NR parameters. The purpose of that is basically a mining operation to see if there are some combinations that perhaps may not be intuitive but whose interaction may yield pleasing results. Outside of the usual "select cinelike d, reduce everything to -5" etc stuff that we all do, who really knows if Natural with +4 contrast/-5 sharpness/+3 saturation/+2 noise reduction might actually be the golden ticket? I did a hardware controller last year based on a midi control surface that let you tweak the parameters in real time which, whilst offering a way of 'discovery through play', wasn't exactly ideal from a structured testing point of view. It also wasn't particularly practical (or cheap) for you to go out and replicate yourselves if you fancied a go ! Anyway, in pursuit of something a bit more structured, I created a very simple automated testing app which I thought I might as well hand out in case anyone else is interested in this sort of thing. Its Android only so you elitists with that other phone type are out of luck for now As its also a bit of a utilitarian thing that was only really designed for my own meddling, its not exactly pretty either but it works. Mainly. OK, so here is how it looks, how to use it and what it does. First things first, obviously, is to enable the Wifi on your Panasonic camera as per if you were using their app and then go into your phone/tablet and connect to it. In the ProfileStepper app, you then press the "Connect" button and it will take control of the camera. It might complain the first time about approving the connection but if it does just repeat and it will be fine. Once connected, you press the individual parameter buttons and select the range that you are interested in so in this example we have set it to be using the base Natural profile and how it looks with all permutations of Contrast from -5 to +3, Sharpness from -2 to 0, Saturation from -5 to +5 and NR from -5 to +1. Next up is to set if you want to use Stills or Video, which we toggle between by pressing the Stills button and then to create a delay (of between 1 and 10 seconds) between shots you press the button (currently set to 2 seconds) and select it. When you are happy with the settings range, press the Start button and the app will begin compiling the list of commands it needs to send to the camera. As this is a variable amount of time (in our example there are 2079 commands ) you may see a progress pop-up before it proceeds to the playout screen. Be patient. The progress screen will then appear and display the value of each parameter in the set it is sending to the camera and the current progress (in this case we are on set 4 of 2079). After it sends each set to the camera, if you have the mode set to Stills it will then fire the shutter on the camera. Obviously, make sure you have a decent memory card capacity if you are testing a lot of permutations! If you are set to Video mode, a recording will begin on the camera as soon as you pressed the Start button and the changes will all be recorded into this one recording. This is where the Delay function becomes useful as it determines how long each setting is held on the video as it progresses. As with the Stills mode, be careful regarding capacity as if you have 1000 permutations with a Delay time of 5 seconds then, well.... In either mode, pressing the Stop button will abort and return you to the main screen and if you are in video mode it will stop the recording. The purpose of the progress screen is not just to provide you with a countdown to whats going on but also meant to act as a slate so that you can put your phone/tablet in your test scene and the settings will then be embedded in the images/video when you review them. This is an example frame from a GX80 to illustrate. There is a 3 second delay built in to the beginning of the command playout process to allow you to do this but obviously as the camera is being controlled over Wifi you can place the phone/tablet in the scene and press Start from there anyway. This just gives you a chance to get your hand out of the way ! So there you go. Its a bit of a niche app to say the least but might be useful for some. Bear in mind, its not really meant to be about analysing every single permutation in one go (though it can do that if you want) its more about helping you to look at how more targeted ranges of settings interact with each other (such as Saturation and Contrast in a specific Profile for example). And what exactly is the Noise Reduction bringing to the party in terms of colour shifts and sharpening etc. Download link is here https://mega.nz/#!Vux3UDIK!TPJNYc8gJ6YrQY5T8xkZRAQ_IpBac16xIIGe3ET6R0Y You'll need to set your device to install apps from non-Play store sources etc. Its completely free, no ads or any bollocks like that. As I say, its my own test tool and wasn't designed for general release but thought it might be useful to share. If nothing else, its a quick way to get the Cinelike D hack on to your camera! Because its just a quick and dirty test tool, there is no particularly elegant error handling going on (other than telling you that you haven't connected the camera) so if it throws one up just quit and restart it. If you set the To range to be lower than the From range (i.e. the From is -3 and the To is -4) then it won't execute the commands so don't do that. NB - The eagle eyed amongst you may have spotted the Bracket function at the bottom of the screen. This does exactly what you might expect but its unrelated to the main function but I needed it to test something so you get it as a bonus! You set the shutter speed and then the 1/3rd steps above and below and press Bracket and it takes the required frames with suitably altered shutter speeds. Its a bit flaky on some Panasonic cameras because they don't all allow the 1/3rd shutter changes so your mileage may vary and, as I say, you can largely ignore it.

I largely think we've managed to get things arse backwards when it comes to emulating a film look. Because everyone now has the ability to change everything about an image in post there is barely any emphasis on getting it right - or at least consistent - in camera first. If a camera manufacturer said : "OK, we are putting all our eggs in one basket here but we have absolutely perfected an in camera look that is identical to Kodak Vision 250D (for example). Our entire sensor and processing design is based on purely just replicating that but we guarantee that if you shoot with this it will be indivisible from if you'd used that stock. Oh and you'll only be able to use it to shoot up to ISO800 though". Would you buy it? I think I would. Personally speaking, taking away all those variables that offer endless rabbit holes to explore but often just disappear down and making something work within a constant framework would be challenging but ultimately more creative and productive. This is a really interesting piece about preparation in camera to match film and the follow up piece is here : http://yedlin.net/OnColorScience/ Its quite timely that this has come up because I've been interested in tweaking combinations of base colour profiles and constrast/saturation/sharpness etc parameters in Panasonic cameras for a while with a view to mining some in camera combinations that may not be obvious ones but actually work well visually. The timely part is I've just finished an app today that does automated stepping through all combinations (from selectable ranges) whilst either capturing a still per combo or as one continuous video with the changes embedded. I'll give it its own thread tomorrow if anyone is interested.

OK, I've made the app and it does + and - 18 step bracket in 1/3rd stop steps (so, 6 full stops in both directions).

I think where we may end up with this is concluding that the only way to truly differentiate the motion cadence between two cameras is by shooting unicorns in full gallop. So it doesn't matter whether you go there for fuel or chicken you're going to end up with gas either way. Smart marketing.

Well, that's dinner sorted tonight then... https://www.tripadvisor.com/ShowUserReviews-g293925-d4492725-r189425765-JJ_s_Fish_Chips-Ho_Chi_Minh_City.html

The workaround in this case as far as I can see thus far is to switch it off Honestly, this isn't about unrealistic expectations or wives or anything else - it just isn't working very well.

Its a bit of a how long is a piece of string question but if you based it purely on this year's Oscars you'd say 'how many does the Alexa have?' which is 14. Purely arbitrary though as, judging from that clip, I'd say you're doing just fine without the other 3

No, its not necessarily about the AF performance per se just about conflicts with lens corrections combined with the ones it is making with IBIS. Even trying it in MF might be worth a go.

I couldn't agree more with this. I got accused of having 'real problems' a few pages back for actually trying to look at exactly what is going on with it. From everything I've seen, the IBIS on this camera - even taking into account the sensor size - is a long way even from from being 'isn't that great'. Its not about expecting miracles.

Its better.....but the right hand side of the cabinet is going berserk at 0:37. Have you got a non-native lens you can put on it to rule out additional lens and AF corrections it might be making at the same time?

I'm not suggesting he did anything to mislead but the screen shots of him adjusting WB etc were of a completely different scene to the actual comparison footage with the RED. Can't see an ND filter on it (as they're usually clip on with smartphones so would be visible on the stuff where he's showing it on the gimbal with the RED). However, perhaps the most salient point here is that neither can you see any ND on the lens on the RED, so that would suggest they would both be on 1/800 hence why they aren't showing any difference.

I think the BMPCC/BMMCC/BMCC etc are fantastic sensor and storage solutions to base your own custom cinema setup on. Whereas I think the GH5 is a fantastic camera. With the BM stuff, they give you the insides but you have to provide the camera that goes around it, if that makes sense. Both very worthy approaches but both very different and one doesn't negate the other. I look at the BM stuff versus a GH5 like the Sigma Foveon cameras versus something like a Fuji X100F. Both APS-C, both fixed lens but the Fuji canes the Sigma at every turn for low light, speed and functionality but then you look at the RAW images from the Sigma. Thats not to say the GH5 can't make fantastic images because it demonstrably can but just that - like the Sigmas - within a smaller operating window of and with a lot more effort in post the BM stuff can make images that truly shine.

If you are OK with this being done from an Android app then its not a massive deal. You'd have to set the base exposure in manual mode from the app but as long as the calculated brackets don't exceed the particular camera's preset shutter speeds (possible with those sorts of ranges) then it will be OK.

It depends on the size of the power bank but we are measuring in hours rather than minutes. I ran that test the other week with a 4000mAh one and got bored after two hours of powering one and it had run down one out of 4 bars on the indicator. They are literally credit card sized but a bit thicker so even carrying a spare isn't going to be a hardship. Something like this one will work as is and even has the acrylic case which you could store the power bank in (in the space where the raspberry pi would have gone) I think @Kisaha has just ordered this one or one near identical to it https://www.ebay.com/itm/3-5-HDMI-LCD-1920x1080-Touch-Screen-IPS-Display-Kit-for-Raspberry-Pi-1-2-3-EB/332253259667?epid=24003437661&hash=item4d5bd66793:g:CLQAAOSwH09ZNoG6 The ones I'm sourcing are designed for VR headsets so they're OLED and better native resolution but the price is a fair bit higher. I've got a 5.5 inch 2K 450Nit one just come in from China to evaluate from the same supplier so I'll see how that goes first. With regard to the remote control, this is a readily available one that covers most functionality. https://www.cinegearpro.co.uk/products/cgpro-bmmcc-bmmsc-focus-zoom-record-remote-controller There is scope to do something like that integrated in a handle for example with a display as well and even an Android remote interface to it. Basically, these things are all Arduino based boards that talk to the camera so there is no end of things you could attach to it in terms of thumbsticks and switches to have dedicated direct control of all the key functions. It all depends on what someone would want. That one for £99 is not bad though.

Yeah, the Video Assist is what they would like you to use with it but, like all the 4K recorder ones for GX80 etc being 7", you end up with a ridiculous situation where its dwarfs, unbalances and generally takes away the point of a compact system. I've found some 3.5 inch 720p displays that I'm going to pimp my DIY EVF with which would be ideal as they can be powered off a gentle following breeze (well, a credit card thin USB power bank). LukiLink would be a good solution as well when it ships. I've no idea what the sales have been like for the micro but I've always expected them to drop in price and they never have as I'd probably be very tempted at a lower price. It just that little bit too much when I add up all the other bits I'd need to make it complete.

These adapters can help the battery situation so you can use long run Sony batteries. https://www.ebay.co.uk/itm/Change-Canon-LP-E6-to-SONY-F570-F770-F970-Battery-Pack-Adapter-Holder-to-Sony-/310795410767 As for a monitor, I think BMD are keen on everyone buying their Video Assist for it The remote control options via the expansion port cable set are very interesting with the BMMCC. I've seen some quite cool home made hardware remotes for it.

The theory is that its this one. Based on the specs, it would add up. https://www.sony-semicon.co.jp/products_en/new_pro/may_2017/imx294cjk_e.html It is also used in astronomy cameras as well so there does in theory exist a way to make your own very ghetto 4K raw C mount camera system by attaching it to a high end windows tablet. As the module is quite small, it wouldn't be too dissimilar conceptually to what is being shown on this thread attaching the RX0 to an Atomos. Not dissimilar prices either. https://www.firstlightoptics.com/zwo-cameras/zwo-asi-294mc-usb-30-colour-camera.html Someone has done a video here

EDIT>>>>> I'm trying to be less argumentative so I've deleted my original response to this but suffice to say that you couldn't have missed the point of my posts more if it they'd been written in invisible ink. I do hope you can see the irony in your offering the sage advice that 'not every camera is for everyone' immediately after criticising someone for explaining exactly why - in its current form - that camera isn't right for them though. <<<<< EDIT

It reminded me of this

Yes, its 3. I've had a few times with the X-T2 where it doesn't run them down in a sensible fashion where the internal one drops before the other two which means I have to take the grip off to get to it for swapping out which is a pain in the arse. It can run without the internal one and just uses the ones in the grip so thats often a better solution as you can leave the grip on permanently then. The problem I have with the battery life with my work is that I have to leave the camera on as it needs to be grabbed and be ready for work instantly like a DSLR would be and I found it a bit too docile in waking up again if I had the power management on. Added to the fact that I have to be in boost mode as well doesn't help so whilst it won't chew through all 3 in a game, it can, depending on how busy I've been with it, need one swapping. What doesn't help with the X-T2 as well is that the power switch is extremely easy to move to the On position when its being transported in a bag so I had a few times of arriving at a stadium and unpacking a very warm camera that had ate 3 batteries so now I make sure its got none in when its travelling !

Its quite informative to take frame grabs from YouTube clips and bring them into FCPX and look at them on the scopes and see how much some people often leave behind and how other people are eeking out everything thats there. You can do it in real time too if you attach your Atomos as a second screen on your computer and display the YouTube window on that

It certainly enjoys consuming batteries when its in boost mode thats for sure. The good thing about the grip though is being able to power it externally if needs be. They make the AF pretty tuneable so hopefully they might do the same with the IBIS but there is a danger with that much tweakability you can end up in a bigger mess so I could understand why they wouldn't.

Indeed. I wasn't expecting miracles from it but I suppose the rule of thumb is if you can see it working then its not really working as you need it to and its just too jarring for me. The IBIS has gone from a deal maker to a deal breaker for me when it comes to switching from my X-T2 as it was really the single most compelling reason. Still, its Fuji so it ain't over til the firmware fairy has been!

Yeah, the second one was one of the ones I saw where I thought "Have you actually got IBIS enabled on the A7iii ?" and looking at the comments, I wasn't the only one ! In the first video though, whilst it obviously canes the A7iii again, I can still see the same sort of coarse changes. If you look at between 2:07 and 2:10 just as an example, the water pipe does an abrupt move to the left from one frame to another. I've pulled three frames here and done a difference overlay to show how it is transitioning from the first two frames (where nothing much but the bush is moving) to the one where it has to do some work and makes the jump. In the brickwork, there is slewing as might be expected but also some scaling changes that all adds up to a bit of a mess that I think is what is making it so jarring. I'm not some mad pixel peeper as a rule but some of these transitions I'm seeing are very rough as are the less than subtle scaling changes even when its not being taxed too much. Its like a Scooby Doo transition at times though