kye

I heard this recently and think it's pretty interesting.  I'm not sure if it's the best definition I've read, but it's more practical than other ones, so is useful from that perspective.

“He who works with his hands is a laborer.
He who works with his hands and his head is a craftsman.
He who works with his hands and his head and his heart is an artist.”
- Saint Francis of Assisi

I'm 100% for not gatekeeping.  Even from a practical perspective, saying someone/something is or isn't 'art' doesn't mean anything, and people who like to be critical are really just telling us about themselves, not the thing they're talking about.

Luc Forsyth likes it, or what he saw at NAB anyway.

This is a setup they had with a broadcast servo-zoom lens on it.

His comments (link with timestamp)

• He's worked on the survival show Alone for a few seasons and they use dozens of GoPros, but the footage always looks like it came from a GoPro
• This new model with a proper lens attached looked like footage from a real camera
• The broadcast zoom setup (with a phone as a monitor) handled like a proper camera
• He doesn't use AF when rigging cameras to vehicles etc most of the time so the lack of AF doesn't bother him in that context

I think it's a loss for us, as he was doing independent testing of things that no-one else was, like DR, and commenting on various combinations of modes and features, especially which combinations of modes and features couldn't work together.  This is all information that the other people don't bother with because their 'reviews' are really just product showcases or first-looks.

As much as the camera journalism and independent review ecosystem is in a sorry state, it just got worse.

13 hours ago, fuzzynormal said:

Sure, I'll give you a deep dive.  I'll also vent a little.  You might imagine there would be some worry matching footage, but for this project, surprisingly not really much of a big deal.  We had worked with the Alexa footage for months, so I didn't fret at all that a GH5 would do the pick ups.

Why?

None of the Alexa footage was shot with a deep consideration for the lighting.  It was all very workman-like.  And the "eye" of the shooter was decent, but average.  That's really the biggest thing.

Anyway, the cinematographer and director decided to bring a rigged out Alexa to a run-and-gun-available-light shoot.  The dudes are older gents and they just felt like "the best" camera was the logical tool to use.  Not true, honestly, but you couldn't convince the cinematographer of that notion.  Which is kind of a legacy mentality with older guys, but that's what happened. 

There was a political element here too.  It's a decent budgeted doc so the "shot on ARRI" rhetoric was desired. 

Okay, so the main reason re-shoots were required: there weren't a lot of compelling shots that could juice the narrative.  The footage was decent to look at, but not dynamic.  The cinematographer really couldn't get around easy with this big 'ol rig and sticks.  Interesting things would happen situationally with the characters, but he would unfortunately deliver a single shot when dozens were needed for a good edit.  He'd just end up being burned out physically as the day went along and couldn't move into interesting places for useful footage.

Ultimately, a big powerful camera was being underutilized because of "reasons".  An Alexa camera delivers nice footage, of course, but when you're pointing it into blown out skies and shooting mid-day with it on the regular, it's not like it'll give you miraculous results. 

Here's the other rub that had me slapping my forehead, the cinematographer and the director really like the crushed blacks sort of color grade.  And they didn't mind the whites being blown, so...  That's a style that was typical a few decades ago, right?  Well, you're taking a 14 stop Alexa, throwing away a ton of information, and delivering 9 stops for the final project?   That's certainly a look.  And Michael Mann loves it as well.  But then, why the hell spend the $$ on an Alexa in the first place?

Now, in this story you're getting a bunch of bias from a guy that spent my entire career as a one-man-band.  If my background was from the more collaborative perspective of traditional filmmaking, I suppose a lot of this wouldn't even stick in my craw.  Don't discount my naivete'.

As for lenses, the cinematographer was using a very clinical variable.  Ziess cinema Zoom 28 - 80 mm.  And he liked f5.6.  Great lens, but neutral character to it with how it was used, so when we went out for more footage I slapped my Olympus 12-40 on my GH5, packed a few ND's, and went with that.

At the end of the day, it turned into an effective modest film.  Could have been better, wasn't a disaster.  imo, it was too verbose and that ends up being a slog, but all that talky stuff appeals to people that vibe on the themes of the film.  And while the film doesn't stretch to get beyond that sort of thing, the director is happy with it, so all's well that ends well.

Great write up and thanks for making the effort.

I can see that shooting the Alexa with a neutral / clean lens with deeper DOF and crushing the whites / blacks and it not having carefully sculpted lighting etc would mean it would be an easier act to follow for MFT.  As I see it, the limitations of the GH5 compared to the Alexa would be the colour science on skin tones etc, DR, and shallower DOF with character lenses..  most of which weren't significant in how it was shot.

I have no experience with an Alexa but I've heard that it's a two/three person camera and that operating it solo is difficult.  When I think about things like that, combined with the weight and form factor, I can really understand how limiting it would be to operate compared to how fast the GH5 etc are.

I do have some idea about coverage and how incredibly demanding actual "real" productions are.  When I analysed Parts Unknown and saw the quantity and quality of shots required for a 40 minute episode I was blown away.  Most shots were professional but not incredible, but there were something like 1000 of them in each finished episode.  Which they manage to get in something like 5 days on location.

I suspect the speed and flexibility difference between the Alexa and GH5 is really a microcosm of the DSLR Revolution.  Sure, some of that would be shooting style from the operators and some would be camera choice (ARRI made the Amira for being much more portable/faster) but even between an Amira and GH5, if the goal was getting as much acceptable quality coverage as quickly as possible then the lighter camera has the edge for sure.  Pair it with one of those tripods where a single mechanism releases all the joints simultaneously and you'd be able to really cover a scene very quickly.

I remember doing a graphic design course back in the day and they said that you can use whatever stock images you like for your projects, and as long as they don't actually clash with the theme of your project then no-one will notice.  Since hearing this I have paid attention to such things and it's definitely true - the graphics really don't have to be related at all.  I suspect b-roll is partly like this too, as long as you have someone talking and include things that are vaguely related to the subject then it'll work like forgettable eye candy to keep the viewers attention.  The Kuleshov Effect is working in your favour for sure.
I love the quote "kids love colour and motion" which I think was from a movie and used very sarcastically, but I suspect some of the purpose of b-roll is just to keep that part of our brain from getting bored while we're listening to the person say the thing.  Of course, there is an art to it for sure and talented people will be shooting and making edits that create magic by being a lot more than the sum of their parts.

Great to hear you were able to navigate the politics and that the end result was a success in the eyes of the boss.

Going back to the ARRI takeover and strategy, the fact that ARRI created the Alexa Mini as a 'special use' camera and then everyone switched to it for the whole production says (to me at least) there's a demand for smaller camera packages.  It would be amazing if the new management didn't realise this and see what they can do with smaller bodies still.  I'm sure ARRI would have a good idea about sales figures for things like the RED Komodo and Komodo-X etc, which are very small, which must further emphasise the demand for smaller packages.
I understand that cinema cameras potentially do things like heat/cool the sensor so it's at the optimum temperature and this requires size/weight for the mechanisms and also significant battery power too, so maybe making things smaller is more difficult than we'd imagine.
I like to point out to people that the GH7 has a lot more stuff in it than the smaller cameras people compare it to (IBIS, cooling, internal RAW, etc), but in this case we're comparing mirrorless cameras with cameras that literally have heaters in them, so it's not a straight comparison by a long shot.

10 hours ago, fuzzynormal said:

Here's an anecdote regarding our level: 

A-Cam was an Alexa Mini on a documentary shoot.  The cinematographer didn't really get enough variety for the storytelling the director wanted.  We tried to make it work in the edit booth.  Couldn't do it.  Late in the edit/production when the budget had been burned, the two of us went back into the field to get necessary pick-ups.  Those pick ups ended up covering close to 1/3rd of the film.  All the footage was cut together, color graded, and released on one of the major American TV networks.  Every shot looks cohesive.  

That pickup stuff was done with my used, ebay purchased, 9 year old GH5. 

And there ain't no way anyone watching that film could readily tell the difference between the two.

That said, anybody got one of those Alexa-Minis laying around they want to give me? I'll trade you my GH5.

Fascinating, and reassuring too.  This is why I concentrated on colour grading - the hardware was good enough and the gap was squarely with me.

Can you shed any light on what colour grading / image processing was done to get an acceptable match?  Was there any particular way you shot with the GH5, or lenses etc you used in order to get it to match?

I would think (if it was me) that going out shooting with a GH5 knowing it would have to be intercut with Alexa footage would trigger lots of thinking about how to best go about it so it would be good enough.

1 hour ago, Emanuel said:

https://www.blackmagicdesign.com/media/release/20260414-01

 

In Portuguese.

 

In conclusion, now in English:

The big leap in Resolve 21 is this: it is no longer just an editor and colour-grading system for video. It now also includes a new Photo page, designed for still photography, with album management, node-based grading, crop and reframe at original resolution, tethered capture with Sony and Canon cameras, support for LUTs and Resolve FX, and collaboration through Blackmagic Cloud.

On the AI side, the package has become much more aggressive. It now includes IntelliSearch to find people, objects, and even words spoken in dialogue, a voice generator from text, CineFocus to recreate depth of field, Face Age Transformer, Face Reshaper, skin-imperfection removal, automatic slate reading with AI Slate ID, UltraSharpen, and Motion Deblur.

In the Cut and Edit pages, the most visible additions are improvements to keyframes and curves, the ability to adjust Fusion effects directly inside those editors, native support for HTML and Lottie graphics, improvements to Text+ and MultiText, and more practical smart bins for filtering footage in the Media Pool.

In colour, VFX, and audio, Resolve 21 adds MultiMaster Trim Manager to generate HDR and SDR versions from the same timeline, Magic Mask render in place, list and layer views in the node editor, and group versions for grades. Fusion gains the Krokodove library, improvements to the macro editor, and an updated USD toolset. Fairlight adds track folders, 6-band clip EQ, EQ and Level Matcher, and Chain FX.

For creators and modern workflows, there are ready-made square and vertical resolutions for social media, direct upload to YouTube, TikTok, Vimeo, and X, support for IntelliScript with Final Draft, import of ATEM Mini ISO projects, and major reinforcement of immersive workflows, including VR180 and VR360, Panomap, ILPD retargeting, MainConcept H.265/MV-HEVC, and foveated rendering for Apple Immersive.

In short, Resolve 21 seems to be three things at once: more useful AI, tighter integration between areas that used to be separate, and a much broader opening toward photography, creators, and immersive content. Blackmagic itself presents this version as a major update, with the new Photo page, dozens of new features, and many usability improvements.

 

This is no small improvement. And who said they were focused only on the SMPTE crowd?! ; ) LOL

Amazing upgrade, kudos to them! : -)

It's an interesting update for sure.

While I find the upgrading process to be too much of a PITA to upgrade unless there's a killer feature in the next version I really want to use, there are a few things in there that are interesting from an AI perspective.

The first is the AI Face tools, with AI Face Reshaper & AI Face Age Transformer.  This is interesting because it shows their ability to track and understand faces is vastly improved from the previous generation of Face Refinement tool, which was obviously designed to have very soft masks because their tracking wasn't that great.
I did an excellent course in Beauty Retouching which used Resolve and basically you apply different treatments to each area of the face as each has a different tone/colour/texture and you had to mask each one manually yourself.  The ultimate would be for the AI face tools to detect the face and output a mask for each area of the face, automating the masking/tracking.

The second is the Adjust Focus with AI CineFocus, which simulates a shallower DOF, and is a combination of a blur plugin with their depth map plugin.  When the depth map plugin came out I tried it on some deep DOF shots to see how it did, and the results were worse than the iPhone 'cinematic mode' with the edges being a very obvious blurry transition, and you couldn't apply anything more than a barely perceptible blur before the edges ruined the shot.
The fact this is now an integrated plugin means it's gotten better to the point they're willing to put it forward for this application.  It's probably still a long way from blurring the background but keeping each hair on the subject in-focus, but it shows increased confidence.

I know they are also doing tonnes of little things in the background too.  I went through a phase of posting to the BM forums and suggesting features as I came upon things that annoyed me, and to my amusement I had professional colourists (including from Company3) reply and say they've been suggesting the same improvements to BM for year after year, and I notice that a number of these have gotten fixed in the last few versions.

Still, there are gaps in the things I'd really like.  One is the stabilisation, which can't handle any kind of shot that isn't perfectly rectilinear, and has no support for removing rolling shutter etc.  This is possible, and I went down a deep dive at one point some years ago looking for a solution and there was a product that did it flawlessly, but the product was in the thousands-of-dollars price range so wasn't worth it for me.
The stabilisation also lacks the ability to stabilise the tilt/pan/roll/zoom in different amounts.  If I shoot with the BMMCC and an OIS lens for example, the lens stabilises the tilt/pan quite well but has zero roll stabilisation.  I'd like to stabilise the roll almost to 100% to keep it almost perfectly fixed, while also stabilising the tilt/pan maybe 40% just to smooth off the rough edges.  This isn't possible, except if I build something in Fusion, which apart from forcing me to learn Fusion, also requires I go into the Fusion window to track the shots as well, I can't build a custom OFX and then apply it in the Edit or Colour page.
I don't know why BM didn't just make the stabilisation occur in a node, that way you could just apply it several times however you wanted, but it's a 'special' thing that happens once in the image pipeline, and once only.

My biggest wish for Resolve 22 is lens emulation.  Like the Face and CineFocus tools, the lens emulation ingredients are all there if you combine them yourself manually, but integrating them into one plugin would be pretty sweet!

CineD posted this interview.  I haven't watched it yet, as I'm not exactly in the market for an ARRI anything, but I'd be curious to know if there are any plans in there to go for smaller cameras..  With MFT bodies getting larger and larger and ARRI bodies getting smaller, maybe we're at the point where they will meet!

10 hours ago, Snowfun said:

It’s for a personal travel/ski film next winter in northern Finland (supplemented by a Sony rx0ii). I have several manual Nikon lens which I previously adapted to the Pocket 4k. The 85mm is a favourite. Plus all my M/ZM/VM (one of my favourites is the Voigtlander 50mm f3.5 Vintage Line Heliar). I want the Panasonic/Leica 100-400 (not sure why but it seems ridiculously fun to go to 800mm equivalent!). I’ve also ordered the 9mm f/1.7 Panasonic/Leica to replace a MFT Voigtlander 10.5 (far too heavy). A healthy mix of MF and AF. Shallow dof is certainly not a priority.  Interestingly, I am finding that the “fun of using” is far more important that ultimate IQ… after all, none of this really matters! 

Yeah, agree with you, and you should have fun with your lens choices - some cool stuff in there.

One thing you might have fun with is matching the 9mm PanaLeica to your less clinical lenses, I have the 9mm and it's incredibly sharp - my sharpest lens by a long shot I think.  Just remember, amateurs ask "what is the best way to sharpen your footage?" and the pros reply "actually, I soften the image on most projects I grade".

I find the GH7 to be almost invisible, which seems bizarre when you consider the size and weight of it, but it's true.  With my AF zoom lenses I find a composition by eye, raise the camera, adjust the screen if required, adjust the zoom, hit the AF-ON for it to focus, check the histogram to make sure exposure is good and adjust vND or ISO if required, then hit record.
With manual lenses I do the same thing but normally hit record then manually focus the lens.  In situations where the light isn't that variable, it's just a matter of vND -> hit record -> focus.  It's like the camera is just a screen and a record start/stop button on it, and the rest of the action happens on the lens and in front of the camera.

The experience is more like I'm operating a lens rather than operating a camera.

People talk about how complicated the menus are in this camera or that camera and TBH I don't really know what they're talking about.  You buy a camera, work out what modes you might be interested in, test them, then choose which mode you'll use and save that to a profile, and from that point on the camera is essentially a box where you only adjust something once in a blue moon.
The reviewers act like you're taking one shot in 48p 5.7K Prores HQ V-LOG and the next in 23.976p 1080p h264 HLG and the next in 30p C4K h265 in Sepia or some BS.  
Quick - the bride is about to walk down the aisle - change settings!  Why can't I assign the shadow slider in the picture profile menu to a hot button?!?  It would save me so much grief!!

20 hours ago, Snowfun said:

I needed a colour sensor for a specific project (my primary camera being a M11M) and just picked up a G9ii - it’s a fun little thing. Seems well built and easily customised to play nicely with individual requirements. Even with a (Smallrig) cage it’s a convenient size (identical to LUMIX S range I think - although lenses are presumably smaller). Currently just the Lumix 14-42mm but plenty of exciting options. Until I read this thread it wasn’t really an option I’d have considered. Glad I can read!

What are you shooting and what lenses are you contemplating?

I can't recall what your other normal equipment is, but MFT opens up a Pandoras Box whole world of possibilities with adapters etc too (if you don't need AF) so that can be fun too.  I've been shooting street video at night with two combos that work incredibly well..  the first is my 42.5mm F0.95 with the Sirui 1.25x anamorphic adapter on the front, making it an equivalent of 68mm F1.5 and creating beautiful rendering wide open, but the combo is 1.3kg / 46oz so I also got a Takumar 50mm F1.4 on a speed booster, which is equivalent to a 71mm F2.0, which is much more vintage but is tiny and almost a full 1kg lighter.

I'm contemplating a native 35mm F0.95 or F1.4 to replace both and have a less vintage but still lightweight for travel shooting.

I also rock the 14-140mm for day shooting while travelling, and in brighter places the 12-35mm F2.8 is pretty hard to beat.  So many great lenses.  If you don't need crazy DOF then MFT is a great option, even with the larger body sizes.

9 hours ago, Netelia said:

I'm trying to figure out whether a shorter focal length C mount lens is safe in both an OM1 MKII and an E-P5 using a Fotodiox MFT-C adapter, mainly being concerned about sensor contact with IBIS movement and shutter hits.

I have used several c-mount lenses on my GX85/GH5/GH7 cameras without incident, along with many others on these forums.  I have the right adapter that goes in a bit further and gives infinity focus, so there should be enough room.

I can't speak for other adapters or OM1 cameras, but it's definitely possible with the MFT flange distances.

9 hours ago, Emanuel said:

Slowmo modes are actually OOAK for this price range — 9600 pictures in 10 seconds burst of 1080p is something indeed: more than 6 minutes and a half going with 24p, oh well...

To this day, it’s still the fastest camera sale I’ve ever made, though. Just look at the number of subscribers on their official YouTube channel. These are numbers and facts.

Getting good affordable 960p would be cool for lots of people.  I see the science explainer channels showing bad quality 960p and the richer channels with Chronos setups.

Don't get me wrong about them not being cameras that appeal to a large number of people.  They're very good for getting the new "EVERYTHING IS AWESOME AND WIDE AND SMOOTH AND DEFINITELY SHARP SHARP SHARP!!!!" style of video that looks more like video than anything ever made before, but as soon as they say it's a cinema camera, there are 27 things they have to change from every other model ever made, and to bet they'll get every single one of them right is a very long shot indeed.

10 hours ago, Anaconda_ said:

Is there really no electronics on the lens mount? That's a bit of a shame. 

Is there any info on crop factor, as I understand it's not an MFT sensor, only MFT mount. 

Well WTF - turns out "1 inch sensor" is marketing BS and it's actually 13.2mm wide, making the crop factor 2.73x, and only just a touch larger than Super 16 which is 2.88x.

Quote

Every image sensor marketed as 1-inch sensor measures 0.52″ (13.2mm) on the long side, 0.35″ (8.8mm) on the short side, and 0.63″ (15.9mm) diagonally. This also applies to every sensor size that is derived from it. For example a 1/2″ sensor does not feature a physical diagonal measurement of 1/2″, but is in fact much smaller.

Source

Lots of MFT glass and also lots of c-mount options too as already suggested.  Ironically, with the lack of electronic contacts you can't control the aperture on most modern MFT lenses, so much of the sharpest glass will be unavailable (making the 8K sensor spec rather redundant!).  
However getting shallow DOF will probably be more difficult, especially as we have no idea what kind of focus assists it will have, so stopping down might be the best move (focus wide open then stop down to eliminate any slight errors) and that will sharpen up older / lesser lenses.

This might end up being the mythical tiny S16 cinema beast that people have wanted (or said they wanted!).

It's much smaller than even the BMMCC and that's before you realise the BMMCC doesn't have a screen so you have to rig it up to use it.

I am still skeptical though - there's lots of stuff we still don't know about and given GoPros history I have complete faith that they'll include at least one fundamental fatal flaw that will prevent this.

Looks interesting with an MFT mount and the claim the Hypersmooth works with any rectilinear prime (their stabilisation is a pretty important feature, especially with how small they are).

Of course, if they don't have animal eye-detect PDAF then the internet will skin them alive!!

25 minutes ago, Aussie Ash said:

"without drawing to much attention"-having the camera hanging on the neck strap about 1.1m above ground and therefore looking down at the fold out screen most of the public don't even know they are being filmed.It is also quite easy to do steady pans and tilts like this.

Even better than that, I have the camera on a wrist strap and shoot with it at chest height like you describe, which means that when I'm walking / standing around the camera is barely visible, unlike a shoulder-strap where the strap and camera are front-and-centre all the time.

Lots of other things come to mind..

If there are people standing around in clumps, stand right next to one of them.  
This way you'll sort-of become part of the group, so people walking by will just identify there's a group of people there and 'see' all of you as one thing and walk around you, and people looking around won't be drawn to you as much as if you're on your own against a clean backdrop - this is sort of like camo clothing where you are trying to obscure your silhouette.

Pause a few seconds before showing the camera.
If you walk up near someone and stop, they'll probably glance at you to see who you are, what you want, etc.  If all they see is someone doing nothing (ie, not a threat or opportunity) they'll go back to what they're doing.

Shoot people who are distracted and doing things.

Most people who are distracted are just on their phones, but contrary to internet hype people do still do other things, and unless you're working on your doco series "People on their phones - Episode 27" its good to seek out these moments.

Shoot through people / things.

Be careful how you move and approach shots.

I try and be very focused on things that are just becoming visible.  As soon as you can see them, they can see you, so it's best to not get closer than you need to.  The further you are away the more likely there is to be layers to shoot through too, so that's a bonus.

People also have a sixth sense that someone is looking at them, even if you're looking "at them" on your camera screen, so although you can approach someone from the side or even from behind and they'll just turn and look right at you.  I'm not sure how to navigate this, but I'm sure there's some way to influence it that I haven't worked out yet.

This lady was facing directly away from me when I started filming and then turned suddenly a few seconds into the shot:

The guy nearest me suddenly turned around to look at me, despite none of his friends noticing me beforehand:

I know people do look around sometimes, but the timing is uncanny, so it's definitely a thing.

The old trick of finding the backdrop and waiting for someone to come into shot is a good one too, which is what this shot was.  It has the benefit that you're not coming into their environment, they're moving through yours.

Any situation where you're shooting through layers has the potential for someone to come into shot too.  

I was shooting compositions using the bikes mirrors and then a lady came and parked her bike right in front of me.  

I'm pretty sure she knew I was there, but as I was already standing there when she arrived I wouldn't have triggered that 'a new person just arrived' reaction, and also as she arrived at the situation from somewhere else she was probably quite distracted as the whole situation was new and she was trying to park her bike too, so it's possible she was completely oblivious to my presence.

Anyway, that's some further thoughts.  There's a lot online about how to stealthily take street photos (e.g. Garry Winogrand pretending to fumble with his camera, etc), but much less about street videography where you have to essentially remain motionless for many seconds while rolling, plus you can't 'drive by' people and freeze them with a short shutter speed either.  For one reason or other most of the street photography tricks don't really work.

I'd imagine that @BTM_Pix would be deep down this rabbit hole..

After digesting my trip to China, I'm now planning the next trip to Japan, where we're mostly in a remote location but we have a few days in Tokyo in the middle so I'll try and spend as much time shooting there as I can (which really means leisurely meandering around shooting and having breaks with delicious food / drinks etc).

My equipment lessons from the China trip included:

• The GH7 is a workhorse and I don't think about using it at all, just on what I'm shooting
• The 14-140mm is a great all-round day lens for home video stuff
• The Takumar 50mm F1.4 on speed booster is good, but a little soft on the sides of the frame and the rectangular insert is probably a bit much

I also learned a bunch of stuff about how to shoot in crowded situations without drawing too much attention to myself.  I suspect that this is an infinitely-deep rabbit-hole that the best shooters probably do unconsciously, but like all things practice makes improvement.
One thing I did that I think also worked well was to just shoot as fast and as much as possible.  Not only did it lead to more shots and variety for the edit, but I think it also potentially helped me be less in my head and shoot more instinctually, which I suspect will yield more creative and expressive results. 

I've been thinking a lot (and talking to friends) about what I'm learning and what equipment is appropriate:

• The 70mm FOV seems potentially universal because in crowded situations you want to focus in on something so the frame isn't just full of chaos (photography is the art of subtraction), but in situations where there are less people you get spotted at a much greater distance and so having the longer lens means you can still get closer shots of people without actually getting close to them
• If my goal is to make edits that feel more immersive, then it makes sense to shoot with a lens that's roughly "normal" so it has a perspective similar to the human eye, which is about 50mm on FF.  Having 70mm is a bit longer and would introduce a slight element of distance between myself and the subject, which is emotionally appropriate as I am an outsider in the places / cultures I visit, so this is coherent and adds to all the other decisions I'll make in what I shoot / how I shoot / how I edit / etc.
• I mentioned wanting a lens that was a bit sharper on the sides of the frame to someone and they countered by saying that having such a limitation will make my work more consistent (not only from the images themselves but also because it means I tend to compose with the subject nearer the centre of the frame) so this is a reasonable counter-point
• Despite all this, I suspect that I'll want a wider lens for when I get into the emptier narrow streets where it's more about the location rather than the people in it, and I suspect this is closer to 35mm or so

As such, I'm mostly settled on the following lens contingent:

• 9mm F1.7 - for getting that wide-angle distortion that makes buildings etc really pop
• 14-140mm F3.5-5.6 - for day-time home video shooting
• 50mm F1.4 with SB - for that 70mm "night cinema" goodness
• 12-35mm F2.8 - for the wider "night cinema" duties, and being a zoom it means that I can use whatever focal length works for this task (35mm equivalent is just a guess) but also combined with the AF I can shoot a variety of angles / compositions really quickly
  
  I'm also likely taking the following, partly as just-in-case and partly to experiment with:
• TTartisans 17mm F1.4 - if I find that the 35mm FOV is desirable then this is a fast prime I can swap to
• TTartisans 50mm F1.2 - obviously I'm a fan of the 70mm FOV and I wonder if this 100mm FOV would be useful / workable, especially as it's super fast with shallow DOF
• Risespray 35mm F1.6 c-mount - this is about a stop slower than the Takumar+SB combo but seems cleaner wide-open so is a way to challenge my assumption about needing the speed of the Tak

I'm also contemplating shooting 24p rather than 23.976p, and also 1080p instead of C4K.  Both decisions have pros and cons to them though.

On 3/3/2026 at 7:03 AM, eatstoomuchjam said:

These days, my travel setup looks a bit different, but with a similar goal of traveling light.

I just bring the GFX 100 II and for lenses, the EF 35/1.4L II and the Fujinon 110/2.  I used to bring the Fujinon 32-64/4 instead, but then I needed to bring something for lower light/night shooting.  Plus a 2x zoom makes less sense when I can just crop anyway.  I was looking at the GFX 100RF, but it's really not all that small and when I realized I was willing to accept a fixed 35 with that, I decided to just do the same thing, but, y'know, fast.

For video, the GFX also makes it more like I have 4 lenses instead of 2.  I can record in 4K or 5.8K at full sensor width - so effectively a 28mm or 80mm lens in FF terms - but if I switch to 8K, the crop is close to FF so I also have a 35 and 110mm lenses - with plenty of resolution to crop in if I want to.  The fact that it's 8K is less interesting than it being a bit cropped.  😅

I don't really need to get all that wide so the 35 is usually enough, even if I crop in a little when IBIS darkens the corners a little.  For photos, there's enough room to crop in that I can get reasonable results going anywhere from around 27-28mm on full frame to...  something like 200-300mm after cropping.

For ND filters, I just bring a set of Kase clip-in filters - ND8, ND64, and ND1000.  The entire set fits in a tiny little box and weighs almost nothing.

For a tripod, I... don't.  The IBIS in the GFX is decent enough that I can get sharp photos handheld as long there's at least some light (if it's too dim for around 1/15second at f/1.4 at ISO 6400, that's rough - but it's just not a scenario I'm optimizing for these days).

Limitations?  Longer exposures for waterfalls are hit and miss.  I usually just turn on the 2s timer at about a 1 second exposure and shoot it like 5 times and if I don't get something that's sharp with smooth water, I decide that wasn't meant to be.  I have a lot of pictures of waterfalls already (I really like waterfalls).  The camera and lenses are obtrusive - the GFX 100 II is smaller than my GFX 100 was and that's nice, but it's still not small.  Rolling shutter in some modes is enough to be noticeable even with relatively small movement.
The fact that it's obtrusive can be a really big problem - in a lot of cities, it's not a camera to casually hang around the neck and stroll around.  I'm in Sao Paulo right now - there are only a few parts of the city where I'd even bring it out.

BUT the good news is that my phone (iPhone 16 Pro) works absolutely fine in all of the places where I wouldn't take out my big obtrusive camera.  The iPhone is good enough that I take out the big camera less and less, even when it would be totally safe to take it out.

Yes, lots of things to trade-off against each other.  Thus the arguments about what is best when people impose their own values and priorities onto other people then judge them for doing it "wrong" lol.  The more I refine my setup the more that other peoples approaches sound so alien to me.

I realised just recently when thinking about my UK/Europe trip that safety isn't something I think about that much when shooting as I tend to go to places that are relatively safe, mostly big cities across Asia.  I do find it ironic that shooting in London (the centre of the "civilised" world - right?) is the first time this issue has really made me question how I would address it.

On 3/31/2026 at 8:27 AM, fuzzynormal said:

Recently came into possession of this nice and functional 35mm 2C-BV ARRIFLEX.  Before I eventually sell it I'm considering shooting a 200' roll to notch that experience; just to say "I done done that"  2 whole minutes of footage!  What the hell.  Ain't cheap, kind of financially stupid to do this sort of undertaking, but maybe ultimately worth the effort, I think? Anyone out there willing and able to offer advice regarding the lens situation with these old things?

 

As this is a once-in-a-lifetime thing, and you'll likely be overwhelmed with the logistics of using a camera you've never used before + a format you've never used before + a cost per second you've never experienced, my suggestion is to keep it simple and keep it meaningful.

My suggestion for lenses....

Go to a rental house and tell them what you're doing and get their recommendations.  I'd suggest lenses that are neutral in look, easy to use, reliable, and probably not too heavy / expensive.  Perhaps something classic like a set of Zeiss Super Speeds (which were popular for a reason!) etc.

My suggestion for shooting.....

As this is a never to be repeated thing, I'd suggest shooting people and places and subjects you love.

Not only will this be a lower-stress approach, but you'll end up with lots of images that will be relevant for the rest of your life, and perhaps longer for friends and family.

Completely secondarily to this, shooting a range of different things will be fun, and it will also be great if you want to nerd out and pixel pee etc, as you'll have a range of different subjects and scenes.

23 hours ago, Aussie Ash said:

Even more amazing they were low budget B-movies with shoot schedules under five weeks !

I watched a great video talking about Christopher Doyles contributions in working with Wong Kar-Wai (I'll post below) and there's a great line in there where Doyle basically says (when comparing his films to Hollywood blockbusters)

"I think we have absolutely opposite attitudes to what's film-making.  We make the film we can, they buy the film they think they want" (around 5:00 mark)

When talking about small budgets and tight timeframes these films are often a lot more like my own travel videos than a Hollywood blockbuster.
In my videos I shoot on-location with available lighting and no control over the scene whatsoever.  In some ways I am capturing something that is more authentic, because I'm not constructing sets or rigging lighting that might deviate from the actual location, but this also means I have less flexibility to work around the camera etc (where sometimes cheating things makes them look more normal rather than less), and it requires me to capture things in a way that more authentically depicts the location rather than including/excluding things in a way that's not balanced or authentic.
Obviously these lower budget films are still working with lighting, (probably) closed sets and production design, but they're not constructing everything from scratch on a soundstage in a warehouse in Burbank.

In the video he talks about how because they filmed in real locations the actors were responding to their surroundings in an authentic way, rather than having to pretend they're somewhere that they're actually not:

"The environments that the two worked in dictated the movement, emotion, rhythm, and transformation of those locations into an active force within each film.  The physical surroundings were always used to shape psychological states. Hong Kong becomes the central site of this transformation."

This idea of filming on location and letting the day-to-day (and perhaps moment-to-moment) shooting experience influence the acting and filming reminds me of what Noam Kroll preaches, which (to me) is really the fundamental advantage of the low-budget film.   Wong Kar-Wai sometimes wrote the next days scripts the night before, which means they could adapt to how shooting was going and the weather etc.

With the technological advances (film getting faster and not needing lighting / 16mm cameras that were light enough to use without a tripod / on-location sound then sync sound / digital) that enabled Italian Neorealism / French New Wave / British New Wave / Dogme 95 it's all about it getting smaller/lighter/cheaper, so taking these advantages and then still doing a full pre-production cycle then rigidly shooting to that in prod is really just throwing away much of the new potential that technological advancement has delivered. 

Not even in 4K!

It's like they've never watched a single YT tutorial on how to make their footage cinematic.

I think your criticisms of Resolve are quite relevant and justified, and perhaps the most significant thing (apart from the overwhelming user experience when first learning it) is the workflow.  If you want a straight-forward experience then I think it's all about workflow.

Depending on how you are thinking about it, I think there's two overall philosophies you have to choose from:

1. Make it work the way you think things should be done, and don't support other ways (or even be openly hostile to them)
2. Try and make it as flexible as possible so people can choose their own workflows

I have had significant issues with the way that Resolve limits things, which are stuck in the workflows that began in the days of celluloid.  It's not that it doesn't let you do things your own way, as mostly it does, but 'their' way will involve a single shortcut key that is mapped by default, and 'my' way often involved seven functions and perhaps some of them couldn't even be assigned to a shortcut key at all.  If that's a thing that you do per-shot, or per-cut, then that's game over for that workflow - they may as well have not bothered.

If you're going for the latter, then you'll need to reach out to people with vastly different workflows and mindsets and then let them use your tool and see where the limitations and faffs are for them.  I have a lot of experience in IT and the only thing you can really count on is that some users will do things that seem completely bananas to you until they are given a chance to explain things (which often requires them explaining what their world looks like).

Even if you're going with the first one, if you are then I'd suggest be clear about it and don't get distracted with anything else.  Half-supporting a different workflow won't do either you or the people who work like that any good and is just a waste of time.

11 hours ago, eatstoomuchjam said:

If so, I'd strongly consider the 50/1.2L or the Sigma 50/1.4 Art. 

Not the CN-E 31.5-95mm T1.7 zoom? 😆😆😆

I suspect that I'd likely want to go as clean as I can afford, because the situations are amongst the most brutal possible with huge DRs from strong light sources in frame and the associated coma/smearing/etc that happens.  Almost every lens I have used looks controlled in normal high-DR situations (ie, daytime exteriors in direct midday sun) but start to look 'vintage' when out in the streets at night.

I don't generally take stills of frames with lots of issues, but this starts to hint at the territory I'm describing - the below is the 12-35mm F2.8 zoom on the BMPCC.  This has no promist filter on it, this is just the lens itself.  It's not the most clinical lens in the world wide-open, but if it performed like this on normal scenes then people would have cancelled it as being unusable, yet here the bloom extends half the height of the frame!

The images from the Voigt 42.5mm F0.95 + Sirui combo seem to be pretty good across the frame, like this one where the text seems pretty clear even on the edges of the frame:

but even in shots that don't have high DR, the Takumar 50/1.4 doesn't do a good job on the extreme edges:

Maybe the woman posing bottom left is slightly behind the focal plane, but even then the softness looks like lens aberrations and not just being out of focus, even with my crazy rectangular / moon-shaped bokeh.

Using the Takumar I found that I was composing images with the subject at the centre (or near to it) whereas I don't remember feeling like that with the Voigt+Sirui combo.  I'd certainly like to feel more free to compose how I want.

11 hours ago, eatstoomuchjam said:

If so, I'd strongly consider the 50/1.2L or the Sigma 50/1.4 Art.  The Sigmas are, generally speaking, too clinical for me, but lots of people like them - it really depends on how clean you wanna go.  The 50/1.2L has been around for a long time now.  I got mine pretty close to when they released it - it might be old enough to vote, at this point.  That's to say that it's been out long enough that you could probably find a copy with clean glass and some external dings for not too much money.  Bokeh is completely a matter of taste thing, but I hated the bokeh on the EF 50/1.4.  I remember it was ugly/busy enough that even non-photographers noticed it on some photos.  Maybe I just had a bad copy?  I liked the EF 50/1.8 more, even.

There's also Otus, of course, but used prices on them have stayed surprisingly high.  I think there was also a Zeiss ZE 50/1.4 which is probably more affordable.  I never tried it.  It's probably very good.

If considering the Sigma f/1.8 zooms, you might want to double-check how the coverage is with the 0.64x SB.  I remember the 18-35 had a pretty small image circle - the 50-100 is probably a bit better, being more telephoto.

I hear you on the character of the bokeh, I find some lenses to have quite objectionable bokeh, and in my tests with the Tak I found it highly variable actually, with the character changing depending on the focus distance and distance to the things being blurred.  

I realise I'm really pushing things here to the limits, which is pretty much normal for me, but I feel like there's a lot of experimentation still ahead, once I can justify the investment required.  Plus I can always dirty things up in post if they're too clean, the Film Look Creator really changes the game in that sense.

I'm also looking at shots like this of Myongdong in Seoul and thinking that maybe this is too crowded for such a long focal length and a wider lens might also be useful:

I could even sacrifice a small amount of aperture and go with the Sigma 50-100mm F1.8 zoom (equivalent to a 64-128mm F2.3), which would have all the benefits of a zoom.  It's a pity there isn't a mid-range offering in that series so I could go both wider and longer than the 50mm mark.

There's always the Canon CN-E 31.5-95mm T1.7 zoom, which would be perfect if it wasn't 7.8lbs / 3.5kg, enormous, and USD24,000!

3 hours ago, stephen said:

I've used Metabones Ultra 0.64x speed booster with my Blackmagic Pocket 4K. Compared to 0.71x is moves your crop factor a tad closer to FF 0.64x -> 1.28x vs 0.71 -> 1.42x  and gives you a little bit more light (1/3 stops) and a little bit more DOF. You can use it not only with Canon EF but all lenses which have flange distance > than Canon EF. Like Contax Yashica, Nikon, m42, Pentax. Biggest issues for me is the weird crop factor. x2x0.64 = x1.28. Now I want an angle of view equivalent to 35mm in FF. Which lens should I use ? 28mm as 28x1.28=35.84 And so on. That's why I moved to FF. 35mm is 35mm no matter if lens is vintage or modern, plus I can use Minolta MD, and a bunch or other brands and bayonets. No need for speed boosters. Lens is used exactly as it was intended by the manufacturer. Sensor if big and has no problem with low light. For vintage lenses and in general for lenses alone FF is the way to go for me. I was tiered to calculate crop factors and deal with speed boosters.

Thanks for the info.  

They're definitely not for everyone, but I'm more than comfortable using them and all the associated math.  I'm even a fan of the in-between focal lengths.  I discovered that I absolutely adore shooting with my 42.5mm F0.95 lens paired with my Sirui 1.25x anamorphic adapter, and that's equivalent to a 68mm F1.5 lens on FF.  50mm would be too wide and 85mm too long - so the fact that FF lenses are exactly 50mm or 85mm on FF is actually a disadvantage for me.

The economics of it are also pretty straightforward, I can get one of these for half of what a Panasonic S9 or OG S5 would cost, and it means I don't have to carry around two bodies etc either.

1 hour ago, eatstoomuchjam said:

The difference is not huge.  When wide open on very fast lenses, you're more likely to see some speed booster artifacts with the 0.64x.  Whether those artifacts are acceptable is personal taste.  If the prices weren't similar, I'd just take the cheaper one.

I was also surprised to find that the 0.64x covers the slightly small S35 sensor in my Z Cam E2-S6G.

Anyway, as far as other brands, I think that Kipon's focal reducers are well-regarded.  Otherwise, most of the third-party ones are skippable/don't seem to be too good.

FWIW, you can also have your cake and eat it with a focal reducer, if you want.  Pentax K, Nikon F, Leica R, Olympus OM, and M42 lenses (including your Takumar) can all be adapted to EF mount.  I've also thought about building a recessed M39 adapter since it would be theoretically possible on any camera without a mirror assembly (M39 is 28.8mm and EF is 44mm, but I have several M39 lenses with controls near the front and that are skinny enough that they could just sit further back in the EF mount).

Anyway, for EF lenses otherwise, do you know which you're considering? 

I think focal reducers and adapters and front anamorphic adapters all provide a myriad of potentially interesting options, which I brought up in the Adapters are BACK.. and better than ever! thread.

The main candidate would be a 50mm F1.4 to be a 64mm F1.8 equivalent lens for shooting "night cinema".  This would be to replace my M42 Takumar 50mm F1.4 + SB combo (which is very vintage and has distracting bokeh and isn't great on the sides) and the 42.5mm F0.95 + Sirui 1.25x combo (which is heavy and also isn't great on the sides).

Between Canon and Sigma and Zeiss I'm sure there will be a range of lenses that are as pristine as I'm willing to pay for.

I've found that the 65-70mm range is really great for crowded street work like markets etc where you can shoot a range of compositions from wides to portraits to macros, and is also great for shooting wider shots on the other side of the street.  I'm also wondering if a wider fast prime might be useful too, so maybe a 24/1.4 (which would be a 31mm F1.8 equivalent) or a 28/1.4 (which would be a 36mm F1.8 equivalent) might also be interesting but I feel like I'm just getting started with this style of shooting.

I'm idly contemplating buying a Metabones Ultra 0.64x speed booster for my GH7.  This would take me into the world of EF for the first time.

I'm completely familiar with speed boosting and crop factors and all that jazz, with years of experience from my 0.71x M42 to MFT speed booster and (many) M42 lenses.

What's the deal with speed boosting to EF?  
Is the Ultra 0.64x worth it over the normal 0.71x adapter? (they seem to be similarly priced used).
Is there a different one I should consider (other than Metabones)?

Essentially I'd be getting it to shoot shallow DOF (like I do with my M42 Takumar 50mm F1.4, etc) but with more modern / cleaner results as M42 lenses are quite vintage and far dirtier than fast EF glass, especially when shooting wide open.  AF is of little importance to me, so I'd be expecting manual focus.