kye

and a longer focal length will mean less propellers in shots when flying upwind!

I agree too. Any difference between 4K and high-quality 1080 capture pales when you're watching something and the quality of the content takes over. To a certain extent IQ is one of those things that just has to be 'good enough' not to get in the way of the story. I went 4K because I was interested in pulling out still frames and printing them as pictures - treating video like a 8MP 25fps continuous burst mode if you like. If I was just shooting video I would be outputting in 1080 and wouldn't be ruling out 1080 cameras that have high quality outputs.

I agree with the OP. In terms of my own camera decisions, if there was an equivalent to the Sigma 18-35 that was on m43 it would really change my attitude to the whole m43 system. I bought a GF3 back in the day and it came with the 14mm f2.5 lens as well as the (ergonomically terrible) kit zoom, so when the GH5 came out I was very interested in the IBIS for my handheld guerrilla film-making work because I already owned one half-decent lens in the m43 system, but the lack of a killer lens and the complication of adapting the 18-35 was too off-putting for me, with the focus issues killing off the option well and truly. At the moment I'm kind of split between looking at desirable cameras and then looking for lenses that suit and also looking at desirable lenses (like the 18-35) and looking for a camera body to suit. Right now I've got the XC10 which has a good 'body' but the lens is a bit limiting, and the 18-35 which is wonderful but my 700D / ML body is a bit limiting. A killer lens option for m43 would make me reconsider that system again.

I thought that Resolve didn't understand the RAW files from ML? I've been using MLV App to debayer into CinemaDNGs, but if I can skip this step and go straight into Resolve that would be much easier. If you could clarify / confirm this then that would be really useful. IIRC you're shooting at 3.5K on the 5DIII? I'd imagine that it would upscale pretty easily - I wish I could shoot at that resolution on my 700D!!! One day maybe

Cool! I'm not sure if you've mentioned these things in another video, but I would be curious to hear more about what equipment you are using, perhaps some specifics around settings (eg, gain at various stages). The advice of doing trial runs and what issues and treatments you encountered is useful. I personally judge a Vlog by how useful / entertaining it is, and considering this is a bit more of a practical video, the more information that you can provide that people can use in their own work the better. As with all creative endeavours, actually publishing things is an achievement and definitely one I respect - keep up the good work!

Really? Apart from trying to get shallow depth-of-field (which of course a smaller sensor can't do as well) what else would you say would give it away? This is an important question because if there isn't anything then potentially every small aperture shot on every show shot on Canon could be the XC10 and you would never know. If your set is shooting on Canon, using CFast cards with C-Log, and you had an external car shot (which is going to have deep depth of field) then why would you put a C300 on there instead of a camera that is wildly cheaper? or if you're shooting in 4K on a C200 then what would you put in there that can shoot 4K? You only have to look at a series of posts like this from the Hurlblog (http://www.thehurlblog.com/cinematography-turning-your-gopro-hero-3/) to see what kind of lengths they had to go to to make a GoPro cinematic. Would it have been cheaper just to buy an XC10, ND filter, and put it in Shutter Priority mode? You can bet your ass it would have been. If you're doing anything where there is a danger of wrecking a camera then chances are the cost of the shot far exceeds the damage done to the camera, and the savings in post of not having to grade some other completely unsuitable camera would be huge. Besides, the fact they used a GoPro in a movie all the more reinforces that an XC10 would be able to be hidden amongst footage from better cameras. Unless you think that an XC10 isn't better quality footage than a GoPro? In which case, well done for reading this far, or owning a computer, or being able to ... you know.. type and stuff. And unless there's something else that a smaller sensor can't do except shallow DOF, "go and get me B-Roll of the sunset and people playing on the beach" would be a pretty reasonable gig for a camera like this. I was never making the argument that the XC10 is the best camera in the world (it's definitely not) - it was this forum that concluded that it could never be used professionally for any paid work (to paraphrase one of the posts in the original thread). I still maintain that the people on this forum got it horribly wrong the first time around, and so far you haven't provided any counter-arguments or evidence to the contrary. I definitely agree with this. I would imagine that there is also an element of 'fishing' going on, in case you accidentally patent something that ends up being useful that you can get some royalties on down the track?

@sam PM sent to take this topic offline. In future it would be better if someone has a problem with something I've said to reply to that thing, instead of just rolling it all up into negative remarks about me personally. Thanks.

I've noticed that the speaker on my iPhone can be muted almost completely if you can get a complete seal over the speaker, so maybe. If it has separated speakers then maybe try a few layers of gaffer tape and see if it's enough. As you say, it may not be!

At over 1500 posts I'd be hoping they'd have covered a lot of ground... but ouch!! I swear.. ML would be a great candidate for someone studying PR to get lots of real-world experience. And help a huge number of people at the same time!!

I assume you're referencing "90a58a5 lossless: experimental resolution overriding that should cover all DIGIC 5 models (to be tested)" ? I am yet to add modules manually, so that's another learning curve I'm yet to ascend. Considering that the crop_rec_4k is called "bleeding edge", and then we're talking about adding modules, and especially modules that say "to be tested" in the description... I feel like we've run out of words for how many disclaimers should be applied!!

I think mine won't let me select above the ~1700 pixel wide, even in crop mode. I'm using the "magiclantern-crop_rec_4k.2018Mar10.700D115.zip" - is this the right version to get the extra resolution? TBH the different builds and lack of documentation is really confusing.. I've been watching the SD controller hack thread as people get good results and occasionally someone kills a card and waiting for it to be stable enough to start using. I'm not sure how long it will be until it gets added to the crop_rec branch? I mentioned the 80D as I noticed that Alex (who I believe is developing the 700D crop_rec port) appears to have paused posting in the 700D thread but is active in the 80D thread. I did also note that they're not even up to the point of ML loading, but the thread certainly indicates that people are looking at it. Considering it's got DPAF, is reasonably priced, and (if the comment in the thread is correct) it has 80MB/s write speeds, it would be a great camera body to have ML on. I realise that the A7III can't do RAW (which isn't that important to me - I'm happy with either RAW 1080 or high-bitrate 4K) and that it's expensive - especially with the Sony lenses!, but I'd still look at it as it gives a huge low light performance boost from both my XC10 and the 700D. In a way it should also "just work" which I get the impression that ML is still a fair way away from.

One thing that you might watch out for is the resolution if you're planning on switching in and out of crop mode. With my 700D I found it has different resolutions available in crop mode vs normal mode, and when I set the resolution in one mode it was within the writing speed of my card, but then when I swapped crop modes it was over what I could write continuously. It seems to be that you can "choose" a resolution that's greater than what the camera can do and ML then uses the closest possible mode, which I found was different in crop vs non-crop modes. If you're taking lots of time between shots or if you're not maxing out your cards write speed then you're fine, although I'm not sure what people do when they can't shoot in the resolution they're outputting in. With the 700D I'm shooting in something like ~1700 pixels wide and outputting in 1920 wide so perhaps slight changes in resolution don't matter as much because I'll be scaling up anyway, but your situation might be different. Downscaling is probably fine too. In terms of crop mode being a Killer Feature it might sway me into buying a 5DIII in order to have it. My camera options are currently: The first option is the XC10 which I already own, which has high bitrate 4K video, a long zoom, great ergonomics and battery life, good image stabilisation, but doesn't do shallow depth of field. The second option is the 700D with Sigma 18-35 f1.8 that I also already own, which has RAW in ~1700 pixels wide (possibly more with the SD controller hack), with crop mode it is reasonably long zoom range, does shallow depth of field spectacularly well, but doesn't have as good ergonomics and has no image stabilisation. My alternatives to the above are that I could sell them and get a BMPCC 4K, 5DIII or 80D with ML, A7III, or whatever else has been released by the end of the year. I'm in Italy for a month later in the year and was planning on taking the XC10 and 700D so will be able to put them both through their guerrilla film-making paces and really get to know them both. I've shot a few trips with the XC10 (maybe 20 days filming total) and it feels familiar and comfortable in the hand, but haven't used the 700D / ML in the wild yet. Considering that I'd basically have to sell everything and start again with a new camera & lens system for most of the above it's not a decision I'll be making quickly or without proper testing and evaluation.

Camera shake is a lot more noticeable for sure. I don't know of any other issues with it but I haven't used it much yet so YMMV. I doubt it's relevant to most people, but be aware that changing the resolution also changes the crop factor, but this is true in either crop or normal mode.

Lol, well said. We all experience life from our own perspectives. I think Total Recall was an interesting exploration of the idea of having a holiday from yourself

The recent announcement of the BMPCC 4K has prompted me to do a lot of thinking about cameras in general and what features are useful etc, and I've realised that ML RAW has a killer feature that seems to be relatively unique - the 3x Crop mode. Is anyone else finding the crop mode to be a big differentiator against other cameras? It turns my Sigma 18-35 on my 700D into a 29-56mm and 87-168mm lens, which combined with the fact it switches with only a single menu option change, and doesn't require the purchase / carrying / changing of additional lenses, makes it a hugely competitive feature IMHO. I know this is only in ~1080 and not 4K, but the fact it's RAW goes somewhat to making up for this when compared to cameras that compress their 4K output. I suppose with something like the BMPCC 4K you could just record RAW and then crop in post, but you'd be throwing away a lot of data in post. In theory, a 32MP (or above) sensor can have a 2x (or more if greater than 32MP) crop and still have enough pixels for 4K. It would be great if someone used a 50MP sensor and had a 4K pixel binning mode for the full sensor, and then a 2.5x crop mode for a 1:1 pixel 4K mode, but I suppose all the manufacturers would rather sell you more lenses than add the flexibility.

As the owner of an XC10 I find these topics fascinating. This forum (and most of the internet) completely ridiculed the first two versions of this camera, yet Canon claim they sold more than they were anticipating and the Cinematography Database YT channel seems to run into them on professional sets on a semi-regular basis. This leads me to believe that the internet doesn't understand the design brief and associated tradeoffs of this camera. My impression is that it was designed to capture footage in certain situations (eg, one-operator ENG, B-Roll, BTS, or as a professional GoPro alternative) that was indistinguishable from the footage from larger cameras - which means that we might be seeing shots from it in professional releases and just not know it. If professional film sets can include GoPro footage in feature films then they sure as hell can hide XC10/15 footage. If this is the case then how can anyone on the internet talk with any authority when according to the above they got the first two versions so wrong?

Does the GH5 have physically separated speakers? If so, would sticky tape provide enough attenuation?

Looks like I've offended you. In which case I apologise. My motivation is to share my enthusiasm for this topic, to share some of the knowledge I've picked up along the way, and help people. The fact that my advice isn't the same as everyone else on here is actually deliberate - it's useful to have multiple opinions or viewpoints when discussing a topic and I don't see any value in posting endless "me too" posts like others do here. There have been studies showing that if you ask enough lay-people about a topic then you get a better answer than asking an expert, and it comes from the range of differing inputs. In terms of my levels of knowledge, one of the topics you mention above is what I do professionally, which I've been doing for over 20 years. One thing we should all try to keep in mind is that this is the internet and there are always people posting things that are misleading or incorrect. I'll be the first to admit I sometimes unintentionally make a mistake, but we all know that there are others with vested interests, trolls, and those who enjoy flat-out lying. I suspect this forum would be just as bad as others if we fact-checked every post. I believe that people thinking for themselves when provided with differing viewpoints is better than anyone blindly following someone else's advice.

You may well be right - is there a clear and agreed difference between 'film' and 'video'? If there is I'm curious, but if not then I would question about what value there is in excluding one style of animated images with sound from another type of animated images with sound

Lol. It depends. I can't speak for anyone else, but in combination with not knowing a huge amount about film (in comparison to those in the industry or who have been at it as a hobby for years) I know quite a considerable amount about topics that are very near film-making that are sometimes applicable. I think it must be difficult for someone who has a built up a body of knowledge to understand that someone might not know the first thing about one thing, but know hundreds of times more about the next thing, when in their experience those two things get learned at the same time. Happy to talk further about how/why this is if you're interested. I guess my point is that you shouldn't be surprised if someone doesn't have the same distribution of knowledge as you. They might be the student AND the teacher, depending on what it is that is being discussed. I should add that I try and only talk about what I know about, but sometimes I don't know I don't know, sometimes I'm coming from such a different place that it gets all mixed up and misinterpreted, and sometimes it's impossible to get a message across - text based communication is a terrible way of communicating really.

Depends on film-making style. If you're talking traditional film-making then I agree, but those at the more guerrilla end of film-making (such as those shooting travel or filming family) size makes a huge difference.

Use the minimum equipment required to get what you need. If I'm going to carry around something that weighs a lot, costs a lot, is complicated to use, is fragile, etc. then it better be worth it. My XC10 / Rode VMP+ / Gorillapod 5K rig is way too large, but is the minimum size to get the footage I want. My upgrade path goes from small and cheap to larger and more expensive, but every 'upgrade' was only justified by the previous setup being used in the field and found lacking in an aspect that I cared about enough to spend money on. My ultimate camera would be a 16K RAW flying 360degree camera with 22 stops of DR and was invisible with infinite battery life. Anything less ergonomic than that had better bloody earn it.

I've spent a while playing the "what camera should I have" game and I realised that the process is about finding out what you already know. The two questions to answer are: What is most important to you in a camera? What have you already decided, but are now trying to convince yourself of? Maybe IQ is the most important thing for you, or maybe it's the experience using it. If it's IQ, then talk about codecs and IS and 4:2:2 vs 4:2:0 and all that good stuff. If it's the shooting experience, then work out what your perfect camera looks like and then go find it. Considering you seem reluctant to shoot with a big expensive camera, I would suggest that you start with a the smallest/cheapest setup possible and then 'upgrade' the setup based on what you need. eg, for me the logic would be: iPhone 8 No - needs to have zoom capability Pocket camera No - needs to have directional sound or mic input etc..... This would ensure your setup doesn't include anything you don't need.

In addition to my previous post, a good way to think about efficiency is that it's like a relay race where the baton is passed from person to person. The largest issue in productivity is when the baton either isn't passed properly "I thought you had it... I didn't realise you were waiting for me..." or the person who is meant to take it isn't ready yet. Obviously there are situations where it's not a straight pass between people/departments, the easiest example of that is "ready on set" where everyone has to be ready for a take, however you will find that there is only one critical path / sequence of activities that starts when the director says "cut" and takes longer than everything else that needs to be done between takes or setups. Having an AD / Director that understands the sequences of what is happening, the dependencies between them, and then continually monitors everyone on set to ensure the baton gets passed cleanly and there are no delays is how you get efficient. The "appearance" of what people are doing on set can be important too, if the client is there, but don't confuse that with efficiency. If you can, educate the client so that they can see that the baton is being passed efficiently around, but if not then manage things however you must, but be clear that managing a client who cannot see if a set is efficient or not is something that doesn't effect the final work, and what effects the work won't be visible to the client.

If you are interested in raising the efficiency of your set then I suggest two things: Look at what is lowering the efficiency of your set by actually looking at what is happening rather than what you think is happening Recruit the people you hire to also raise the efficiency of the set, which can only be accomplished through trust Most workplaces are radically inefficient, often because instead of actually looking for what is actually going on and trusting their teams to do the same, they think they already know what is happening. I remember one example of an inefficient set was because filming at a rural location the portable toilet was about a 3 minute walk away, so every time they made a change everyone had a 6 minute walk to use the facilities. When the schedule got more and more behind the response of the AD and Director was to yell at people instead of moving the portable toilet. Complete fail.