Video Priorities

After lots of research, far too many videos and articles, I am starting to get a handle on this video thing.

So, after all that what is actually important?

Sensor size. Contrary to popular belief, sensor size is not the be-all of video. The generally assumed larger sensor advantage in low light is partly off-set by the depth of field advantage of smaller sensors, especially for manual focus (f1.8 acting like f2.8), but individual sensors and cameras are important by comparison.

The thing that bigger sensors do have an advantage with until tech catches up is dynamic range. This again can often be controlled or worked around and even the best cameras need some help in some light, but when things are not under your control, higher DR does help. Rolling shutter, colour, lens quality, stabilising and handling are also effected by sensor size one way or the other, so choose based on your needs, not other’s opinions or baked in pre-conceptions.

Resolution is also not the king of all considerations. 1080 is the current requirement for most clients in the regular real world. Shooting higher is fine, especially if the camera feature or your post processing vision needs it, like cropping in, shifting across the frame etc, but few people want more than 1080 for online or even TV use and up-scaling is a thing.

The trick is, it needs to be good 1080. I have a 2k TV and I can tell the difference between good quality productions and lower ones. The Rookie is one that I regulalry reference, becasue it looks super sharp and clear on my TV and is recorded in RAW 720! Theoretically it is not even enough to fill my screen, but it does and it looks great doing it. The best quality you can do on any format is going to be better than the next level up done at a stretch.

For example, I shoot the EM1x in 4k, simply because it is better base quality, pixel for pixel than it does with 1080 (don’t know why). I also make sure not to use that camera for some jobs. This not because 4k is fundamentally better than 1080, but because their 4k down scaled from an EM1x/Mk2 is just better than their so-so 1080. I then down scale it to 1080p on export. With Panasonic cameras, 1080 is fine and unlocks a huge array of features. Most cameras have a sweet spot, which is where you need them to perform, so buy the sweet spot, not the stretched outer limit.

Dynamic Range. This one is interesting. It is totally possible to use a low (10-12) DR range camera for serious work, especially if the footage is taken in a LOG or RAW format (see below), you just need to take a stand on what is worth keeping and what can be lost. Moody and dark themed projects can actually benefit from inky blacks and low contrast scenes will not produce horror stories. More DR however opens up lots of creative choices and allows you to shoot with fewer on-the-spot concerns.

The sort of high DR image I am aware of looming in the future.

Bit depth is important, but only to a point. 10-bit is the basic requirement these days. It is not that 8-bit cannot do the job, but more depth makes everything easier down the track and avoids little issues like banding in solidly coloured transitions like the sky. It is not like the difference between RAW or jpegs in stills photography (see colour codecs below for that), but it does help sometimes.

It is also a relatively small added footprint compared to other quality increases, especially for the return. Higher than 10-bit does not seem to be of much benefit to an intermediate grade shooter only cinematic work, so like resolution, 10-bit is the benchmark and some programmes cannot even handle more than 8-bit.

Colour depth. 4:2:0 seems to be enough here. 4:2:2 is better apparently for green screening and heavy grades, although modern software seems to be able to handle nearly anything, so 4:2:0 is a good bench mark. Drastically lower is an issue, especially if any grading is needed. In the last two 10-bit/4:2:0 seems to be a good balance.

End Format. MOV is better than MP4, but either will do and the difference can depend more on other factors. MP4 has more acceptance post processing, so if you are unsure of the end use, use it, but otherwise MOV is better.

Colour codecs are like the difference between RAW and jpegs in stills photography.

It goes something like this;

  1. RAW. Great if you can get it, but also have the tools and need maximum grading quality. Apple ProRes and Black Magic B-Raw are often at loggerheads, so be sure what your processing stream, end use or user will need.

  2. LOG. The blue ribbon for most pro-am users, true LOG is enough to make movies with. Every brand has their own, but if true LOG is available, it guarantees the best DR and allows serious grading and a measure of processing safety.

  3. LOG-light. An intermediate step, likely soon to be a thing of the past. LOG-L offers the flattest profile for grading outside of true LOG, but lacks the smoothness and dynamic range of the real thing. Some LOG profiles are probably closer to this in reality, like OM-LOG 400.

  4. Cine-D/HLG/Flat etc. Best in show for most recent generation cameras, these supply a semi-LOG flatness by fair means or foul. Never without their quirks, these were often the favourites of the last generation of camera users and were sometimes even seen as preferable options to light LOG profiles. For example, I like Olympus FLAT over their OM-400 LOG profile., because it behaves more predictably, even though it is not theoretically as powerful.

  5. Standard colour profiles, often “flattened” in camera. The jpegs of the video world, these are fine if you get the rest right. In the pre-easily access to LOG era (just the other day) the biggest issues are baked in colour and limited dynamic range, but often the OOC results had good noise control and as with stills, they all end up jpegs in the end! Choosing or controlling light and contrast are important, but if you can get close to where you want to end up, these save a lot of hassle. Panasonic Natural (often at -5 sharpness, contrast and saturation) was the staple of most GH5 and earlier shooters, often sticking with it in preference to other options until true LOG became available.

The lure of true LOG processing and the safety it offers is too hard to resist these days. These are well supported by most processing software (DaVinci has a LUT* for VLOG to Rec 709). Down from that, I feel that just shooting OOC Standard or Natural profiles are plenty for quick drop work.

*

So, we have 1080, MOV (output), from input of 25 to 150 fps, 4k or 1080, 422 or 420, 10-bit (8-bit if necessary), LOG or as close as you can get resulting in at least a DR of 12 stops. It is surprising how many cameras offer this level of input/output. The “ancient” hybrid G9 does all but the highest frame rates in 4k, but does not have LOG. The OSMO pocket gets close with Cine-D and the S5 sits very comfortably in this space, unless long recording is needed where it drops to 8-bit (which I would use for this type of work anyway). Between all my cameras I have all my needs covered one way or the other and usually all at once.

Other factors.

Light matters. Good light fixes much of the above, bad light makes the simple impossible. If you go artificial lights, work out first what you actually need. A couple of decent and powerful lights are all that is needed rather than a clump of weaker ones. In a studio, a couple of 60W COB’s are fine, but outside a reflector or diffuser may be more useful as the sun will win that fight.

Filtering. ND filters are a must. If you intend to obey the base rule of cinematography (the 180 degree rule), then ND filters will give you control in an otherwise restricted space. Depending on the look you are after, some type of soft filtering will help reduce dynamic range, soften (bloom) highlights and add that currently in fashion Netflix vibe. Anything that helps remove the super sharp digital look will work.

Handling. Get this bit right for you. Tripod, cage, gimbal, shoulder rig. All useful tools, but not all needed for everyone especially not all at once. With IBIS, often the simplest path is the best. Everyone has an opinion here and few are wrong, just wrong for you maybe. The best way seems to look at rigs made for your camera and work style specifically and to be honest you will likely find one. Most of them are modular so little is ever wasted. I just cancelled a cage order for the S5, because I actually prefer a half cage and already have one in my Camvate universal. The handle on the camera is usually the most comfortable.

Think before you jump. Try to “virtually” step through your needs and working preferences before committing. I have toyed with follow focus rigs, but think they are needlessly complicated unless you have a shoulder rig. Shoulder rigs also appeal, but again, they come with catches. Screens are a help, unless they are not. The touch AF on the Oly and Pana cameras is such a useful tool, I find a bigger screen with out that facility just gets in the way, so I avoid them unless I am working statically. If you only need a mic on the camera, then you maybe need no accessories.

My method (just me, maybe not you), is to focus with the lens as normal or the touch screen, view with same, unless I have a need to share. For sound I use a Zoom interface with either a shotgun or other mics piped in or my Sennheiser Mke 400 straight. A basic cage helps here, until, again it does not and I need to mount the H5 on it’s own platform (hence the Zoom F1).

Top handles I am undecided on. If I am literally running-and-gunning with the rig, they are the best with soft hands, but otherwise I find they just get in the way. Best thing I did was get a cold shoe removable one.

Focus. Focussing is of course important for any camera, but video has differnt needs to stills. With stills a single shot needs perfect focus, then the next is aquired and so on. With video you need both more control of focus movements and the smoothness these are made with. Vloggers etc may have a need for AF, and at this level, AF is often a replacement for years of focus pulling practice. Manual focus is so much more important in video than stills, because the process is on-going. It is totally possible to use AF for Vlogging, interviews or even action where consistent acquisition of a target is all that is needed, but for film making or story telling, it is a constraint. I have had some success with touch AF and it is generally smooth and consistent, but again, it works at the speed it chooses, not necessarily what you need.

The simple answer though is manual focus. The simple answer unfortunately, is often not that simple. You need manual focus with linear focus, which is to say, you need the focus to be mechanical, predictable and smooth. Non-linear focus means you actually change the focus throw with the speed you turn the ring. Go fast and you over shoot, go slow and it takes forever. Linear focus travels the same distance regardless, which is one of the benefits of cinema specific or legacy lenses.

Lenses. Cinema lenses are not sharper or custodians of a special sauce for cinematic looks, they are however often easier to use for video. Linear manual focus, often with a ridged focus ring to accomodate a focus pulling device, “T” stops rather than ”F” stops and attention to things like focus breathing and size consistency, all help, but they are not needed to produce good video. Many companies are paying special attention to their lens designs with video in mind.

An example relevant to me is the Lumix S range, which helped me to decide on the S5. Four prime and a few zoom lenses all with matching filter threads, programmable linear focus throw and nearly identical handling make life easier for the videographer.

Battery life. Not much point in having unlimited recording if the battery will not last. An C-type power connection is ideal. It allows for portable power packs or even direct to wall charging.

Sound. This is the big one! Sound makes or breaks video. Nothing that sound adds will make your footage look better, but poor sound will ruin it. How far you go is up to you, but at least invest in a decent shotgun or LAV mic for interviews, maybe a sound processor or portable mixer if you need or even something that can plug into a sound mixer. I have taken a belt and bracers approach using some Zoom products and a variety of mics (brand and type), so I can handle anything from standard on the fly interviews to band recordings direct from their own sound mixer. Again like all the above, opinions are many, so read and listen until patterns emerge.

*A LUT or Look Up Table, is a preset for processing RAW or LOG formats. Because consistency is important, applying LUTs can make all the difference for projects and save a lot of micro adjusting.

Summer Glow

“Can you work late?” usually ends with something like a politician is announcing something bland and unlikely, but sometimes, just sometimes it results in a free pass into a low key music event, likely the best place to be on a gentle summers evening.

Mostly local talent, of which there is plenty, with an out of town headline act, the event promised nothing if not a relaxing way to spend a Friday night.

SPKEZY, a veteran ensemble and really good guys.

The Over The Shoulder Portrait Turns 90

I had a short job today talking to (well photographing), a ninety year old local menswear store proprietor, Don Pitt. Managing the store for 74 years, mr Pitt is still as sharp as the pins he uses when altering a suit.

While Duncan our journalist interviewed him, I did my usual “over the shoulder” portraits, which often end up being the ones we use. Quite the story teller, mr Pitt was fully aware I was snapping silently away (can you “snap” silently?).

Happy 90th mr Pitt!

Tech stuff; EM1 mk2 or G9 with 25 and 45mm Oly lenses wide open at f1.8. I will say it again, what I love about this system is you can use them wide open and get just enough smooth blur for a decent portrait without loosing the context. I thought the silent shutters had some banding issues under these lights, but in processing they seemed a lot less obvious.

CIA Refined And Consolidated.

Something that has been working for me, when I remember it at least, is the C.I.A. mantra I have been using for my recent jobs. Pics from two small jobs today.

C for Control and Composition. Get the elements you want defined, exclude what you do not want and explain your idea if necessary. Share what your goal is. Often explaining it will identify other options or subject objections and hearing it out loud often helps settle things.

How do you photograph 51 interns without it turning into a simple mob shot? You don’t. Use the space, the larger group as secondary subjects and focus (literally) on the main subject, the administrator holding the class. I could have worked the interaction better, but I am getting there. The exclusion element is shallow depth of field, hard but not impossible to achieve with a 9mm lens!

This is often a combination of a pre-conceived idea and then an adaption to reality. I try to sum up the image idea in three words. For the images below it was girl-sails-red.

I is for Interaction and a point of Interest. Once the subjects are arranged and the “shape” of your image is settled, it is time to make sure nobody has empty hands, limp arms or a look of being lost and uninvolved. As simple as a piece of paper, something to settle the viewer that this is a photo of an active person, not just a set-up shot (which of course it is).

A is for Action or Angle. Either or both, which allow you to capture any movement and/or work the shooting angle, which to a certain extend also sets the mood and power of the shot. Shoot from above, below, lengthways, through, but work these angles. Avoid if you can a flat wall of people in front of a flat wall. Movement also helps the subject to feel like a natural fit in their own photo.

Using a foreground element that matched the shape and colour of the sails on the t-shirt (as well as adding relevance, because the Mirror class of boats always use red sails) helped add a pleasingly settled feel to this otherwise straight portrait image.

What Ever Is Needed In, 1080 Out

I have been looking at a LOT of videos on quality, format, bit depth, etc.

I am still sorely tempted to get the GH5 Mk2, simply because it is such an easy and cheap fit for my kit, keeps the M43 dynamic alive and is enough to fix my immediate needs, but I have been caught up in the full frame thing.

For stills I know that for all my needs, M43 is not only adequate, but actually preferable to full frame. Massive expensive lenses on equally large and even more relatively expensive cameras is just not a good or logical road for me to take. A full frame 3:2 ratio camera for landscapes would be interesting, but I know that it is not needed.

For video, there are a few considerations.

For M43 I am pretty much sorted. The G9 is a match for a GH5.2 except for a few areas like VLOG-L (not the full thing and upgradeable), All-i (very big files), continues recording (the deal breaker) and a slightly contrasted look, but otherwise plenty and for my needs, really enough. What it lacks, the OSMO or EM1x provide to an extent, so I have to remember to hero these very capable cameras, because they all are. Indeed any of them can make professional content and regularly do. I have bought well so far.

So why an S5?

The S5 adds only what my M43 kit cannot and does it with minimum redundancy.

Very clean high ISO capture, 2 stop wider dynamic range, unlimited (to card/power) recording, the elusive and possibly irrelevant full frame look.

It seems that the S5 can produce very nice quality even in 1080, 420, 8-bit, LOG or Natural footage. Footage that has a certain something that (maybe) even M43, 10 bit, 422, 4k can only just match? Hard to be sure.

Richard Wong and Ryan Harris and many others can blow my mind with their M43 4k from even the older GH5, G85 or even G7’s, but the footage from the S5 just seems to be pro-grade with minimum effort. The S5 does not ever break 200 mbs, even in 10-bit and only has 8-bit continuous recording, but nobody I have come across has ever complained about the video (or stills) quality, even with the kit lens. My analogy of the quietly humming V8 compared to the screaming, turbo charged 4 cylinder holds true.

It seems to force limits that make practical sense to me. I would not shoot extended footage in 4k, 422, 10-bit, because, well, no-one would reward me for the massive files I would dump on them. It even crops to “super 35” with its 4k/50, but the ease with which it produces such beautiful footage is very, very appealing and the crop can even be useful. It has 10-bit 422 4k, as does the G9, so when I feel the best only will do, but again massive files, rare need.

So, if excellent 1080 is my ideal, then when I achieve that, 4k can even be up-scaled successfully from that.

The question of 1080 brings up the inevitable “future proofing” question, but I look at it this way;

If anyone is interested in my footage in 20 years time, the look, the fashions, the format and realistically the subject will all be 20+ years older, which like now, will look right for the time, but also, technology will likely allow for smart upsizing, 3D rendering or any other current fashion anyway so why make life unrealistically difficult now. We live in a “new only please” world. Archiving is for museums.

It is a bit like vacuum sealing all of your food, just in case, when you can buy fresh later anyway when your mood changes. I have seen digitally coloured black and white footage from the 1940’s that looked ok, so anything will be possible, if anyone cares. Technology will (or has) make most arguments against 1080 (or pixels, bit depth, formats etc) irrelevant quite soon. Recently Nasa enhanced an old medium format negative to show a whole face, hidden in shadow until now. What Nasa did this week, we will all be able to do in a few years.

The key is the quality of the captured footage, not necessarily the quantity. As I have written before, there are countless examples of very high quality video and movie footage taken in lower resolution formats, then upscaled (or not) without anyone ever noticing a difference. “The Rookie” for example is recorded in a 720 format, a very high quality 720, but 720 and Game of Thrones was filmed 1080 and up-scaled to 4k.

In my world, none of my few video clients have ever asked for 4k or if they have, it has often been based on misguided assumptions of required professionalism, something that they do not question on receipt of the footage. It is a bit like clients asking for high megapixel RAW files for magazine or online work, not realising, they have little clue what to do with them. My temptation is to shoot 1080 only, but the reality is, if 1080 is your intended output, 4k is still a useful recording format, because it allows you to crop to 1080 within the frame.

The G9 as a great example, has some very impressive options in 1080. It does 180 frame high speed (slo-mo), 10-bit, 422, 60p and even has focus shift (using 4k) and time lapse options. It also focusses faster and requires relatively little card speed or space. Going into 4k, it, like most cameras offers a much smaller range of options, topping out at a respectable 420, 10-bit, 60, but that is it. In 1080, you decide what you want, in 4k you chose from what you can.

The other advantage of course is, almost all slo-mo, time-lapse, long recording and other special effects from pro-am cameras, are limited to 1080/FHD. Even the GH6 is a powerhouse of slo-mo and other effects, but most are available only in 1080.

This file is a near impossible combination of M43 math (hand held 2000mm eq), software (for heavy De-hazing) and wet feet. In years to come, processing will make it even better.

The reality is, if you stick to a 1080 work flow with most modern video cameras, all of their best options are often unlimited and unrestricted. The rest of the world is not ready for your 6k RAW footage!

My hope with the S5, and my research seems to bare this out, is that the camera will produce superior 1080, which offers all of it’s sensor size benefits unconditionally, and from there I can do what I want.


Final Thoughts On The S5 (Before I Jump)

Why am I probably adding an S5 into a happy M43 users kit?

Needing a camera that would break the 30 minute recording limit, which is a major pain, but has to be addressed, I responded to an add for a cheap GH5.2.

Could have bought it and moved on. The same site had an add for the S5, GH6 and others all equally cheap in a combination of Boxing Day sale, new releases and pending end of model run-outs.

Light breaking through at last? Getting a lot of mileage from my crane shots :).

The S5 floated to the top for several reasons.

  1. It gives me a better video camera than the G9/EM1x/OSMO setup I have now, by simply having a larger, dual gain sensor and full LOG. The many reviewers* all agree that with very little effort and no special considerations, it can give me better base image quality**, with better processing and longer recording times. It has decent battery performance of over 2hrs in 4k, delivers 14 stops of dynamic range without tricks and great results in ordinary recording modes, uses two regular SD cards (V30’s will do me) and provides clean ISO 8000 plus.

  2. The G9’s and EM1x’s actually cover my M43 needs as long as my other concerns are met. High quality 4k 422 10bit, slo-mo, great stabilising etc are all done here. Sure the GH6 may better these and add a ton of codecs and colour profiles, but still it is just more M43 for a 1080, 10 bit, 420 shoot and drop user.

  3. It is great value. The body is basically the same price as the GH5.2. It needs the excellent kit lens, but for only a little more it still under cuts the GH6 body alone (I can even add the 50mm for less than the GH6 and a CFX card). It can also adapt legacy glass on a 1:1 basis. I can even use my existing cards, when a GH5.2 would need something faster for it's 400mbs All-i.

  4. The GH6 is overkill in many regards just to get the other elements the S5 can match (DR/ISO) or the EM1x (stabilising/AF). To get the most out of it, I would need to go to a place that I do not want to go.

  5. All of it is additive. The GH5.2 is a G9 with video features, and only a few for me are genuinely better. The GH6 is the same, but even harder to justify as it is the equal of my EM1x/G9 combo for stabilising and AF, but again, not genuinely better for my meagre needs. If I got another M43 camera, my excitement would be tempered by the knowledge I have too many already. The S5 adds a new format, a new lens path, a better video interface and even better stills if needed.

  6. It fits my philosophy of “best bang for the buck”. The G9’s are the best value M43 hybrids, the EM1x’s are the best value pro sports/adventure camera system and the S5 is the best value full-frame semi-pro hybrid.

If the GH5.2 had full LOG, was a little cheaper (or the S5 dearer) it would likely have won out simply because it is the simplest path and has access to the best/only lens range I have, but only VLOG-L with a small saving are not balanced out by the seldom used 4k All-i recording, dual UHD-II slots and slightly better stabilising (especially considering the G9’s actually provide all of these except All-i and unlimited recording).

The logic of putting money into glass is also not lost on me. Ryan Harris and others have done much with their GH5’s (even mk1’s) and good lenses, so the 10-25 f1.7 or cinema lens would also be an option, but I am only saving $500! In balance, 4k/30p or 1080/60p, 422, 10 bit with the same stabilising and fast primes from the G9’s and better stabilising and AF from the EM1x and OSMO seem like a decent support act.

If the GH6 did not have strange card needs, it would make the logical choice. I can get around the other issues, but if I need a redundancy (second card), it is just too much for a simple job. Adding an SSD is an option, but that excludes a power pack etc, etc.

The only real negative is breaking away from my M43 only habit, but what ever works.

I am still fairly soft on this choice, sometimes wanting to stay sane and maximise compatibility (GH5.2) or go future proof and get the best of all worlds (GH6), but time is running out. The sale is over as of weeks end, so I need to get on with it.

I guess the big questions are;

Do I just want to fill the holes in my game?

Do I want to up my game easily, but at the expense of my M43 roadmap?

Do I alternatively want to go “ultimate” video in M43?

*Geeky Nerdy Techy, Sam Holland, Ryan Harris, Caleb Hoover and others.

**I want the best 1080p capture I can get. 4k is great, but 2k is my practical upper limit. The S5 can even be up-scaled to 4k. The S5 can capture “big” looking 1080, up to ISO 8000+, with a huge dynamic range for more than 2 hrs per battery charge at high speeds (or slo-mo) and 422 10 bit depth. The G9 can also do some of that (just the ISO thing), so they will be nice backups.

And The Winner Is............. The Gut, ............ Or Maybe The Heart?

I get carried away. It’s a thing. Sometimes I catch myself before something silly happens, sometimes not.

Serious choices especicially tend to snowball, stagnate, re-animate like a tired zombie that can hardly be bothered eating free brains, then I often settle quietly on the right choice or the cycle continues and a decision is never made (which means to me, take the hint and don’t make one). Often the choice, if made, was staring me in the face the whole time!

If you asked me what I would have liked two weeks ago for video, I would have answered “continuous recording and a choice of better codecs, but nothing that makes my current gear irrelevant or incompatible”. The camera that would have come to mind as I said this would have been the GH5 or GH5 II or maybe the G85, which I have always had a soft spot for.

The reality is, I am more than happy with my current video kit, with the exception of the continuous shooting issue which is rare, but sometimes unavoidable and the lack of a LOG format, even a light one, which forces careful shooting and an acceptance that some things cannot be retained.

So far, my processes on the G9’s have settled on 1080p, 10 bit, 422, Natural or Standard, occasionally Cine-D and a concerted effort to get it right in the camera. The quality is lovely, but sometimes I am forced into a situation that stretches this past my happy place. This does not however tax my system or my skills if all goes well. I am basically looking at the best I can get with the straightest and simplest path.

The little email trigger that my local B-a-M camera store sent me was for a heavily reduced GH5 mk2 (sub $1600au). It quickly turned into entertaining ideas of equally reduced GH6, S5’s, then to the BMPCC4k etc.

Remember, continuous recording, better codecs, system consistency…….. .

Of the three contenders, the one that made the most intellectual sense was the GH6. This has been pushed aside simply because it does not feel like a balanced choice and has some major hurdles to overcome unnecessarily ($300 cards needed for backup, for features I will not use). It is easy to get caught up in the “big talk” of 6k, ProRes Raw etc, but this is all well beyond my needs, or even my capabilities to use. 10 minutes of 6k 422 10 bit would see my system have a little break down and no client I have would be well disposed either. 4k, usually down sized to 1080 or just 1080, 422, 10 bit is my limit. So, no GH6 for me, even at the current price.

*

The most exciting idea is the S5, which not only fixes the main issues I have, but also adds massive increases in dynamic range, ISO performance, full VLOG and even a new lens for the same body price as the GH5.2 (lens extra). The only accessory this needs is a cage, as the basic battery performance is impressive at 2hrs+ recording in 4k. I have a universal cage at hand and maybe a lens or two down the track, maybe no. Lenses are dearer on the whole, although the 50mm is actually not and legacy glass is a 1:1 conversion.

The S5 is a G9 upsized, but also to an extent a GH series with more video-centric features, then getting the job done with minimal effort. All reviewers agree, it gives you the goods easily and quickly, which for me is important and it has the best battery life.

Never going above 200mbs (V30 cards), it delivers something special even in 1080p, Natural, 8-bit, 420, and some of its negatives are actually positives, like 50p 4k cropping (increasing the effective reach of the fast end of the kit lens). Remember I have other cameras that do their thing well, while this does its. If I need to in the future I can up-grade to off board 6k, 12-bit RAW, but I seriously doubt I would ever need this.

If it has a concern, it is that the G9’s etc may not be up to matching the footage as B cams, but I feel they will when treated well, being essentially GH5.2’s without a few bells and whistles.

*

The gut choice, the one that nagged the loudest was the GH5.2, not because it is the best value, nor the best performer, but because it is simply more than enough and had the fewest “conditions” attached (the G85 is probably enough even). The only accessory it may need is a power pack or second battery for long term recording for under $100. It sits in the middle of the pack in most important areas, but it does feel like a compromise, or more to the point a missed opportunity.

The GH5.2 is effectively the match of the G9, just the video-centric hybrid with all the above conditions met, but in real terms (for me), not a major jump up. It is of course a good partner to these, but for just a little more, I can improve and deepen my kit, even address a few small issues like adding linear focus and full frame stills.

Overall the S5 seems to offer the most and threatens the least system wise. It means Trunnigntwo systems, but is that such a bad thing if the benefits are delivered effortlessly. I do not need another M43 camera, so why not introduce a different growth path?

Hand held run-and-gun or B cam, sports and street stills? G9, EM1x, OSMO.

Better low light, high DR, best upgrade path (the G9 is often not fully compatible with Ninja V’s etc), highest DR stills, easily controlled best quality, and/or long recording, then the S5 is the one.

We all need help to see the way sometimes.

This is me being real, not “hypothetical” me. Hypothetical me may need to buy a better video camera one day, but not this year and Panasonic is about to drop a lot of goodies on us, so a bad time to spend too much unwisely. A new S5 has been announced and with it will likely come firmware updates for existing models. Maybe these will trickle down to the S5, GH5.2, even the G9, maybe not, but again, it is enough.

*

When you look at the card and storage needs of higher res formats, the reality quickly dawns on you that even increasing to 422, or 10 bit, or 4k, let alone all of these, puts you into painfully heavy storage space bracket. Massive cards get dumped into massive hard drives and computers groan under the weight of it all, then often only turning out 1080p for a client.

Shooting codecs are more important. RAW sits at the top, full LOG formats come next, then light LOG, down to HLG, Flat or Cine-like until you finally get to regular picture profiles, Natural being a Panasonic favourite.

The video I get out of my G9’s in Natural or Standard, especially with a 1/8 black net filter is impressive enough to my customers. More is simply for me to fix errors, help control the uncontrolled, fiddle or show off. Just like in stills, a perfect jpeg is the end point, not a step. The S5 effortlessly adds more safety net, without massive overheads.

The big question is, which is better, M43 All-i 1080p VLOG-L or the S5’s Long-GOP with a bigger sensor and VLOG full? The GH6 and S5 seem to be a close match, but often the S5 is not trying as hard and still wins in DR and ISO performance. S5 for me.

Compromises

I have tried to write this post three times now, but hope that this angle will work.

Compromises.

What compromises can I live with, which ones can’t I.

Balance is the key.

The S5.

The S5 is the favourite at the moment, the “heart” choice. It is simply all of what I need, nothing of what I don’t.

I can live with all of its compromises, because of what it offers in codecs, recording formats, battery life (2.5 hrs), card needs (V30), ISO range (8000+) and DR (14 stop) performance as well as the added choices full frame offers;

  • The limited format and bit rate choices. What I want is there, in the form I want it and the benefit of the full frame sensor seems to allow it to perform big tasks with a small footprint. The cropped 4k 50p makes the kit lens longer/faster in length (30mm f3.5-90mm f5.6) and the very long recording time in 4k, 8-bit (or 1080p, 10-bit) is how I would roll, simply for practicality.

  • The AF is workable, but also not my only way of operating. It has hacks, alternatives (EM1x. G9, OSMO) and anyway manual is my preference. I will stick to Pana lenses and the ones that have a good rep (20-60, 50).

  • The different format and lens mount. Fact is, this is the secret of its success and adds another arrow to my quiver for stills and video. I could logically even go for a different brand, but the Panasonic quality and menu systems work for me. The future? No harm done either way. It may be a one hit wonder or the foundation of my future direction, but for the cost, it only needs to do a handful of paid jobs and I am square. M43 is my stills format, but video is open to what is needed.

  • The kit lens, which seems decent and the format is again the saviour. The slow aperture of the lens helps focus and retaining depth and it is video centric. Only $500 gets me a very good fast prime, but I may never need it and using legacy glass is an option. Back in the film era, medium format had better quality, even with a sub-par lens. In the modern era, full frame has become our new medium format, but the lenses are not a compromise, even the kit ones.

  • The small HDMI port. I can use a dongle, or not worry. I doubt I will use an off board recorder and if I do, I will simply be careful and build a good rig with cable protectors etc.

The GH6.

This is the “head” choice. It is the biggest, most expensive and by far the highest specification. I would never need the spec (6k, ProRes), or at least doubt I will and if I do, it will not be tomorrow. The S5 can be upgraded to this with a Ninja if needed.

I would happily use the GH6 for it’s merits, but this is something I cannot get past;

  • I cannot live with the card thing. I work with a redundancy, as any professional should and will not need the speed of the CFexpress only for backup. A $300+ card is too much for my work flow and I still have to address the card reader issue. An SSD would help and be cheaper, but then no power pack, so screwed either way.

The GH5.2.

This one will not go away, being the “gut” choice.

  • The only real issue is the lack of advancement the GH5.2 offers for nearly as much as the other options. It does however come without strings. It adds video options aplenty, just no DR or ISO mitigation and for the cost of only the kit lens more, the S5 does add these. I have several other good M43 video options, so adding a true video camera into this stable needs to add more than just a few more bloated codecs, VLOG-L (when the other two add VLOG) and unlimited recording. The G85 does most of that for half as much again. Still in the mix, but doubtful.

Arguments For And Against The Front Runners.

Looking at my needs*, this is the state of play as of today;

The S5 ($2100 with lens + cage = $2200/$2700 with an optional 50mm later on) is the front runner so far, but the thinking is complicated.

For;

The S5 is enough to fix the main concerns I have for video (long recording, high ISO performance, better dynamic range), add a growth path and it does it effortlesly like a stock standard V8 engine just humming along. It has the added advantages of the full frame cinematic look, potentially made even better with a prime lens.

What it avoids is overkill.

No monster bit rate formats, no tricky tech, no steep learning curve with processing power or storage concerns. It tops out at 200mbs (internal) and is actually beaten by the G9 in some specs (4k/422/10 bit/60p vs 4k/422/10 bit/30p or 420/60p), but it can go external with some big increases (6k/422/10-bit!).

The S5 being a full frame gives me (according to several reviewers), better 4k/8 bit/420 than most smaller sensor 10 bit/422 cameras (and still has 10 bit/422 up its sleeve). This means a reasonable balance between it and the backup G9’s used in 10 bit, with long 4k recording times of over 2 hours on just 1 battery and a relatively light load on my laptops. I am not yet a pro videographer, but I do want quality and versatility, which means keeping things within a realistic envelope now, but being able to up my game considerably if needed. The S5 has many of the same specs out of the box as the Netflix approved S1H.

It is balanced with my already capable kit and directly addresses my primary shortfalls, but it can also go further.

I would likely get a 50mm or similar to have super-duper low light and shallow depth shooting, but the reality is, the dual ISO sensor with full frame sensitivity will still make the kit zoom better than my best M43 lens in low light and the M43 gear can do some heavy lifting here also.

Looking at the well respected kit lens in M43 depth of field terms it is equivalent to a f1.7-2.4 lens, has excellent close focus, a 50% crop factor in 4k/50 and the camera offers ISO 640 and 4000 (8000 is clean) as its secondary, meaning it gains more than two stops over the G9 as a base line at 1600 max (some noise), so it is effectively faster by comparison as well. With a $500 prime it gains 3-4 stops!

Put simply this camera, like a big V8 engine, does the hard stuff easily. No exotic extras, no hassle, no catches and the rest of my kit stays relevant. It also adds in a 3:2 ratio full frame stills camera (96mp HR), external recording and power and anamorphic support.

If I want to use it for a long period recordings (my main need), say with a drama production, 4k/8-bit/420/25p or 1080/422/10-bit/25p LOG is the most likely maximum format for the “A” cam, simply for practical storage, battery and processing. The OSMO or G9’s as “B” camera options can then be used in 4k or 1080p/10-bit/422/25p with Cine-D (or upgraded to VLOG-L) with an eye to balancing them out in post.

It is small, has no special card or power needs and handles like a GH%-G9 lite/G85 lovechild for video. Upgrade paths, should they ever be needed, include off board recording to 12-bit/6k/422/RAW external, prime lenses and external powerpacks. Potentially a fully enabled full frame GH6 for about another $8-1200!

Perfect.

Against;

The negatives are few. The AF is fine, just not totally reliable, which ironically helps make the decision not to use it easier. The EM1x, OSMO and G9 cameras, especially with their more than fast enough prime lenses and the added depth of field of M43 will give to those will be my run-n-gun cameras. The stabilising is best in it’s class, but again bettered by the M43 cameras and OSMO. Plenty of ways to get the job done.

The little issues of mini HDMI connection and only 8-bit/4k or 10-bit/FHD unlimited recording are mitigated by my realistic needs (i.e. I would not use them outside of these parameters anyway). The fact is, 4-6k/10-12 bit/422/LOG-RAW unlimited would eat up massive amounts of card/SSD space, battery and require speed and no-one has asked for it.

The crop is also a choice for added benefit.

From the recent production of Elf, which needed at least 1hr of continuous video.

*

The GH6 ($2300au + cage and card = $2700au) was the front runner for a while, but I have misgivings. This is clealry the better pure video camera and stays within my M43 landscape, but may just be too much of some things and not enough of others.

For;

Future proof video specs (can answer any of my needs at this point and for a few more years), best video AF and good stabiliser (probably no better than the EM1x and only marginally better than the G9, so not perfect), audio display, ProRes internal, best slo-mo. There are lots more but these are the only ones relevant to me. The stabilising is class leading, but the Oly cams beat it still and the AF is still not utterly reliable.

Against;

Half the cameras capabilities may never be used. The 6k, ProRes 10 bit 422 etc are all simply too much for my needs, as well as my computer’s processing power and my storage capacity. I would struggle to justify buying a single CFexpress card just to backup my footage and would need other extras. One reviewer worked out that to shoot a full days wedding shoot at full image power, they would need close to $10,000us in cards and storage!

It is big, expensive, dear to run, over complicated and out of balance with my other kit, or looking at it another way, most of it would be wasted on my needs. Most reviewers agree that the image quality, even if you limit yourself to these “lesser” settings it is still worth it, but I feel it is just too much to fix a few small problems (this all started with a Boxing day sale add and quickly got out of hand). I would be getting it for the features the S5 already offers, but at a premium and with strings attached. The S5 can be heavily upgraded if the need arises.

*

The GH5.2 ($1600 all up) is still in the mix, but seems the least likely at the moment. It does much of what the G9’s do, but little more, just to add what they do not.

For;

The GH5.2 is from the same generation as the G9 and the cheapest by a small margin. With better video specs than the S5 in some areas and better stab/AF, balanced against poor DR, ISO performance and VLOG-Light. It offers streaming, All-i at 400mbs (needs a V60 card minimum), G9 handling, menus and performance, so it is the closest in work flow.

It leans heavily towards video, but is otherwise a G9. I guess this means it is also the most sensible in this space.

Against;

No real improvement over the G9 in DR or ISO performance except when high bit rates are used and that comes with the storage and processing issues of the GH6. It begs the question of older ISO limits (1600) and fast glass being enough. For the money, it offers the least, even though it is the cheapest (by a small margin), but needs no added extras. Even the cage I have fits. In a lot of ways the G85 makes even more sense here, but then the S5 trumps that.

*

*Maximum 4k 10 bit 422 in short bursts for short film making, FHD for the rest, some hand held, most not, with some AF, more manual focus, in mixed light, with occassional long recordings, slo-mo etc, with 1-3 camera cross-compatibility.


Video. Where To Next?

My video growth path still vexes me.

There is a super special on at the moment with the GH5 Mk2 at $1600au ($1000 off) which is very tempting.

So the cycle starts again.

What is needed by me, now and in the immediate future?

  • Flexible, high quality 1080p with options (10 bit, high speed, slo-mo, time lapse, good enough codecs and compression). Have that.

  • Good 4k with some options (10 bit, 60p, slo-mo etc and again some codec options). This is client based, not me as 1080p is fine for my own needs. Have that.

  • Something stabilised. The OSMO is my gimbal camera, the others do “hand held” look well. Have that.

  • Long period recording. The OSMO is plenty for a few uses, but then it cannot do other jobs, so another option. Very Limited.

  • Colour compatability across cameras. The OSMO and Pana Cine-D and standard profiles look similar enough, the Oly FLAT profile can be tweaked and I have multiples of the two bigger cameras, so colour matching can be achieved specifically or generally. Different angles look different anyway, so strict colour matching is not hugely important. Pretty good.

  • A flexible codec for post pocessing tough jobs. The Natural look on Panas is lovely and well accepted, but not super flexible, nor even the best they offer. Their HLG and Cine-D are better. The FLAT on the Oly’s is nice and Cine-D like and the OSMO offers Cine-D or good standard. A RAW format would be nice I guess, but to be honest is more than I need at the moment. ProRes comes with issues for me with DaVinci, B-Raw is harder to source, DNxHD is an unknown, but looks good (needs more research) and VLog-L is decent. I could also just buy the upgrade key for the G9’s VLog-L. Weak.

  • Better low light would be nice. M43 does struggle above ISO 1600 in video and that can only be fixed by (1) lighting (2) a dual ISO camera or (3) fast glass. I have two but not the third.

I ask myself firstly if I need more (only a lack of continuous recording comes to mind here and the OSMO and dual cameras actually fix that to a certain extent).

I have;

  • OSMO 1 with 4k/60 Cine-D and 2k long recording (2 hrs). This can be my gimbal, continuous shooter and “funny places” camera. Not used enough now, it has the potential to produce pro grade footage and solve many common problems.

  • 2x G9’s with time limited 4k, 10 bit, 4:2:2 Long-Gop in Cine-D/HLG/Natural (upgrade to VLog-L optional). These are so good, they really defy any need for an upgrade (for me). Continuous shooting is the only hassle and it can be avoided by using mutli camera angles, I just need to get good at synching in post.

  • 2x EM1x with good C4k/24, 4K/30, ok 1080 All-i, the best stabe (although they are all good), OM-Log and Flat built in. I like the 4k out of these. the 1080 is iffy, but the 4k is not and compressing down works. I also like the FLAT profile. The Oly FLAT 4k is nice and organic looking.

  • What’s missing?

Option 1

Thinking like this always bring up, with some issues, adding an off-board recorder.

If I add a basic one (BM Assist 3g or Ninja V) to a G9, I can have continuous recording and ProRes Raw or DNxHD from the Black Magic on the G9’s or ProRes Raw on the G9’s and EM1x’s with a Ninja V. The BM Assist 3g is the better option for similar money, but is limited to G9’s and 2k ProRes. ProRes is not an ideal fit for DaVinci. This will not make my cature better, but increases my post processing options and to a smaller extent gives me a better hand-on codec for clients.

Annoyingly there just does not seem to be an ideal fit for my two cameras. The Ninja-V likes the EM1x specifically, but not DaVinci. the BM Assist 3g likes both, but with limitations and the 12g is unfriendly to both. Ironically, if I upgrade my camera, the off board options open right up, but two birds……… .

The BMA 3g with a G9/EM1x would give me unlimted recording and with USB power on a G9 or 2 batts on the EM1x for under 1k, but I would probably stick to the same formats, so really just a glorified 1080 DNx recorder that upgrades any of these, but not all (1080 from 4k on the EM1x?).

Option 2

This one is to upgrade to another (!) camera.

For 1500au I can get (for the moment) the GH5 Mk2, which is basically the video-centric equivalent of the G9, with similar handling, colours and stills performance, while adding continuous recording, 400 bit All-i compession, Cine-D2, USB power and live streaming. It even fits into my G9 cage. Unlike the GH6, this one has no special memory card needs and the easily sourced battery life is good at 1hr/4k, especially when compared to the 20 min BMPCC4k, but it is a different type to the G9’s. The main thing it offers me over the G9 is continuous recording and streaming.

For about the same I can get a GH5, offering similar performance in video with some missing links like no internal VLog-L (optional upgrade like the G9), no streaming or stabiliser and now slightly dated video specs, but still very good, it is otherwise similar to the G9. This is a poor option at the moment with the GH5 Mk2 special, but normally good-ish value. It does take the same battery as the G9.

A G85. This little bugger offers very nice organic 4k and 1080p with no (battery life) recording limit for under 1k (I even get a free lens). It is pretty basic otherwise, but a working solution. Normally the price is the main temptation, but with the GH5 Mk2 on sale, that is less of a given at just over half as much (2 for the same price?). The strength of the G85 is good, natural looking 4k/24p with battery life recording limit (at least 1hr of 4k) and good stabilisation.

The reality is the G9 does many things better, but the G85 offers longer recording for about the same price as an off-board recorder and has Cine-D/Natural profiles. A neat little camera offering a basic G9-like option with long recording. This is a guilt free buy and a handy travel camera. The G9’s can do the heavy lifting for effects etc, the G85 handle the static “A” cam work.

The G95 should be a contender, but it simply is not. The sale GH5 mk2, a better camera in almost every respect is almost the same price, the G85 a lit cheaper and the G9, which does not crop it’s 4k is about the same price and again a better camera overall. All it offers is VLog-L, but that is not a deal breaker. Struggling to see the appeal of that one.

The GH6 lifts video in some areas and upgrades overall performance argueably making it the best Panasonic all-rounder, while adding complications. The card options are less straight forward, the cage is different, the battery also and the cost puts it up into the “oops, not again” catergory. A GH5 mk2 I can swing maybe as a long term layby guilt free. The other bodies and off boards are even cheaper, but the GH6 is in the last Pro camera ever catergory and I already did that with my second EM1x. It is the top dog though, developing a reputation as a solid pro choice. A bargain right now but overkill with bells on and for me misguided?

A late comer, the S5. I had no real interest in fullframe, but when I think on it (prompted by a similar price to the GH5.2), it gives me the dynamic range of the BMPCC4k, betters the ISO performance of the BMPCC4k and GH5S with a full frame and dual ISO sensor, good size, with a decent stabiliser, workable AF (the G9’s and EM1x’s stay relevant due to slightly better performance here), better codecs with VLOG full (not light-real LOG) and the continuous recording issue is addressed in 8 bit 4k, even with a single battery, which apparently lasts well over 2 hours, meaning all I need is a second battery for 4 hrs+. It actually seems like the love child of a BMPCC4k, GH5S, A73 and GH6, offering bits of each. Unlike the GH6, it has no special needs or considerations. It just works and ironically, it is smaller by a large margin.

Lenses are addressed early on with the decent kit 20-60 and maybe a fast prime later, maybe not.

The BMPCC 4k for a little under 2k is the DaVinci friendly option, lacking most of the niceties of the Panas, but just great video. Colour is different to the Panas, which is an issue that would possibly lead to more cameras, so caution needed. The camera makes more sense than the GH5s which shares the same sensor, the better BM Assist (the top 12g is not compatible with my cameras and is a similar price) or the GH6 as a static camera option. It suffers from some hardware annoyances, storage etc, and poor battery life but as a static camera, all of these are somewhat avoidable.

Random image filler.

A dedicated video camera. These are a minefield for me, but there are some good options available. Great zoom, stabiliser, sound options etc, but not my area of strength. I also have to face the reality that all of this may end up as just a stills kit with occassional video.

The OSMO Pocket 2. I hardly use the OSMO 1 that I have so why? The mk2 has a zoom, adds a second camera with a few features that can do continuous static recording, so the other can do floating work. I also have a ton of accessories for these and the quality is very good.

Option 3

This is the best likely. Stick with what I have, use it wisely.

I have good maximum quality (4k/60 10 bit 4:2:2), recording longevity (OSMO 120 min capture), multiple angles (7x 4k and 10x 2k options), rigs and accessoiries to suit and limited experience or as yet much need in this field. This also allows me to sit on this for a while, see what the future brings and adapt as needed.

Upgrading the G9’s to VLog-L or using HLG to Rec709 (various techniques available for the well known G9), synching from multiple cameras to cover stoppages, using the OSMO for continuous footage etc all help make what I have work.

*

So the short list.

If I just want to add 2 battery limited but organic and pleasant HD/4k, the G85 will do that. AF, stabilising etc are better on the G9’s, but I have 2 of those, so the idea would be to keep using the G9’s for my main cameras (my “bar”) and the G85, EM1x’s and OSMO for their respective specialist tasks. Long recording is really the only thing I lack, so this is the easy, expedient and logical fix and it sits within my current work flow envelope. For under $1000 I can also get a good no-name grip and battery and it even comes with a kit lens. It will also make a good travel option. All up $1000

If I want to up my video generally, without changing my current dynamic, the GH5 Mk2 is the one and the relatively small extra cost is soooo tempting. Basically a video-centric G9, this is a serious camera and only about twice the price of the G85 (or $2-400 more than the G9). At worst it becomes the one dedicated hybrid stills-video camera in a stills based kit, the role of the G9’s at the moment. It also needs few accessories (a cheap C-type power pack) for battery/card limited recording and it fits my Niceyrig G9/GH5 cage. I now have mains powered sound, so a mains/power pack camera would complete that loop. Maximum video improvement with minimum hassle. All up about $1700

If I want the ISO performance of a GH5S or BMPCC4K with better dynamic range, some GH5/G9 conveniences and the benefit of full frame, the S5 jumps up and says “pick me!”. It is newer and more convenient than the BM, cheaper than the GH5S or GH6, about the same price as the GH5.2 if you count the lens ($2100au) and gives me that full frame option for stills or video. It will lead to a prime or two, but all in good time. For now it is a “super G85” or even a “light GH6”. It lacks some of the video specs of the GH5.2 in some areas, bettering it in others (Log/14 stop DR), but it all feels useable! Some specs of the GH5.2 are realistically out of my league. The S5 just screams “use me” offering better base quality without stretching my processing or skill set. The kit lens is video biased, which is where I will use it (for me I have to remember that f5.6 in FF is f2.8 in M43). I also really like that the G9’s etc are still relevant. The GH5.2 does seem to risk slight overkill in a stable of G9‘s and EM1x’s, maybe just doing some things again, rather than differently or better and what it does do better may be above my pay grade. All up (to start) $2200, maybe $3000 total.

If I want all both unlimited recording and better codecs, then an off-board option is the way. The Ninja V is the best off camera option overall if 4k is wanted from all my bodies and DNxHD is acceptable. The BM Assist 3g is a better unit in many ways, but limits me to 1080p DNxHD. These are a bit of a trap. The unit at the moment is a bargain with the cast module for $700au, but you need a SSD ($250+), connections, power etc, so well over $1000 in real terms. basically too complicated. (A S5/G9/G9 combo with the module is appealing, but I do not have $4000+). All up about $1000.

840mm Of Saviour

So, I went to shoot a boat race.

I had a plan (he says). Based on a previous trip to see the same race (Launceston to Hobart), I knew that there was a reasonable chance that I could get some good, tight action pics as they moved through a tight part of the Tamar river, a spot called Garden Island. The race starts about 5km further down the river.

I remember the last time there was a chaotic tactical bun-fight at the river choke point as the yachts contended with tides, headwinds and each other. At one point it felt like we could reach out an touch the boats and thatbthey would have to touch each other!

Hmmmmmm…..thrilling. The likely winner.

I felt a bit of a twit armed with my longest lens (300 + 1.4tc which is 840mm on a full frame), my 40-150 and a hopeful 12-40 in case they got that close, as they motored past on their way to a secondary start line, just waving and relaxing!

Resigned to coming home with “file” shots of the boats, I spied in the distance the very wide sand bar that juts off Kelso Beach a little further up river.

I felt I had missed my chance because a local informed me the race would likely go out into the ocean to start. Even if I was on the other side of the river, this would be a case of “dots on the horizon”.

A nice flotilla shot, but look, there in the background.

As I drove off the point, I looked over at George Town on the opposite side of the river and it looked like the secondary start might actually be inside the heads. I hatched a quick plan.

Driving the 2km up to Kelso I was greeted by a 1/2km of sand bar and yes, the yachts were circling ready to start directly opposite me.

Off I went, determined to trash a perfectly good pair of shoes. Take them off I hear you say! No chance. The sand is mixed with sharp rocks, shells and “locals”, so shoes on and just have to try to dodge the bigger tidal pools (not so successfully as it goes).

Literally millions of these little crabs…….

…..and plenty of birds

Pretty happy with the 840mm combo (crop from above).

So, was it worth it?

Ironically now fighting strong winds, but also thick sea mist and still pretty decent distances to the boats, I got at least a feeling of the race.

The RAW files were ok, but could be better.

C1 De-haze applied, some cropping and a little colour correction and we have a decent image of the likely winner and the lighthouse as a point of interest/scale.

Lessons? There are always lessons.

  • Take my ratty old winter “mud” boots, even in summer.

  • Never give up.

  • Go to other side of the river next year (I could have taken the shots I wanted of frantic crew highly cropped tight on deck and compressed yachts).

  • My TC and 300 work well (all the shots above are with that combo). I do not trust the AF for sports without adressing the firmware on the lens, but otherwise good quality.

The Photocross 10, A Win?

The PC 10 arrived yesterday and it is exactly as advertised.

It is bigger than the TT 10 in real terms, being squarer and deeper overall. My 40-150 f2.8 with a screw on metal hood (much less fragile than the Olympus one), sits in it nose down as does another zoom (8-18) and EM1x body or G9 and third lens. This is ideal as my sports companion.

They look similar here.

My F2 will remain my day bag though, because I have found it very useful for taking notes on with a A5 note pad. The pad sits on top, which allows me to shoot and take names as I go.

Full, this bag feels heavy, so I guess the bigger 13 would be quite uncomfortable full.

New Workflow?

I decided to duplicate my base model MacBook Air M1 to reduce hassles going back and forth from work, add a backup and depth (my iPads have both died and my desktop is getting old).

I installed Capture 1 23 and ON1 22 as will as DaVinci Resolve 18.

This thing is fast! It uploaded 60 odd RAW files today in about 5 seconds, previews included!

C1 and ON1 seem to be better together in this form, but I was reminded by ON1 to use DNG (RAW) files not TIFF’s that I use based on an early recommendation.

Experiment time :).

The RAW file processed to my liking.

Retaining natural sharpness and detail.

A TIFF in and out of ON1 with default settings.

A RAW sent directly over as a DNG, then processed when it came back to C1.

Takeaway is this (and I looked closer);

The RAW looks the most natural, but has a little noise visible (ISO 800).

The TIFF looks better all-round, but is just starting to look a little processed, i.e. a little plasticky around the bricks and some natural glow is missing (ON1 No Noise often takes out small highlights from eyes etc).

The DNG is the cleanest and sharpest, but is looking a little over done. I actually dropped sharpness back in the C1 processing.

The biggest issue for me with overly aggressive noise reduction is the loss of fine detail in textured areas like 100 year old bricks. The curious bit is the TIFF, by not being as “open” to processing I guess, is that is holds onto the RAW detail better.

Looking the other way, a church steeple.

The RAW. This is nice and looks right.

The TIFF processed as above is clearly crisper, especially on the tiles.

The DNG is up another level again. I could process the RAW sharper, but of course this does not go across and the TIFF does not take to much more post processing. The main thing is to either reduce ON1 sharpening or sharpening in C1 as the halo effect is obvious. I noticed in both the above files that the colour is less brilliant, but that is likely me processing less aggressively, probably responding to a cleaner file.

Take aways;

The RAW files look the most natural and will likely print out as well as needed.

The TIFF’s retain the look of the processed RAW, then refine. At ISO 800, there was not much to fix, so next test is to see how ISO 6400 is handled both ways.

The DNG files are more powerfully processed, but it also looks that way. The top file especially has some plasticky “over perfect” parts and some fine detail has been removed. In the lower file, it actually looks better, so the different processing flows may be best applied to certain subjects selectively.

A note on processing generally.

I have found that with C1, the whole Lightroom noise vs sharpness leading to marbling and grittiness thing has gone away. I rarely touch sharpening, never noise reduction (do not even have the tool enabled), taking what I get as it comes then apply Dehaze, Clarity, White Balance and various Exposure tools.

If noise is an issue from ISO 6400 up, the file goes to ON1, usually on default settings, then out again, a process of 20 odd seconds and I gain roughly 2 ISO settings of clarity and sharpness.

If focus or even a little noise has robbed me of some sharpness, I will often brush it in, although No Noise adds some anyway.

I have no fear of ISO 3200 or even 6400 often in C1 and not much aversion to as high as 25,600 with ON1, which tops out my needs for pretty much any job. Lightroom limited me to 3200 maximum on M43 files. By this I mean sharp and clean, not bearably awful, that may print ok small.

This is my base line, which is a C1 ISO 6400 file, no ON1.

This means also that I am fast.

Most files are fine with one or two sliders of work, many need nothing and if they need major work, I batch drop them into ON1, which now seems to be much quicker (10 files in a minute). If I could just learn to shoot straight lines straight, I would be even quicker!

Bag Retrospective, A Busy Few Months

Having bought way too many bags over the last few months, the idea of looking back and revisiting my thought processes (my spending spree), is a little daunting, but often the best lessons are learned by looking back.

It started with a perceived need for another, better bag for my work with the paper. I had a very workable Domke F802, but wanted to keep it for the school, which at the time was half of my work, and felt the layout was possibly not right for the paper. Other options were the Filson Field Camera bag, Domke F3 rugged, an ancient Domke F2, LowePro Pro Tactic 350 and the Think Tank Turnstyle 10.

Crumpler Muli (4000?). This bag prompted an article on the perils of buying sight unseen as the usual problem of the supposed capacity and the real capacity were at odds. The bag does hold exactly what some have said, but in the wrong configuration for me (read broken-down, not ready-to-go). For a single camera and a couple of lenses it is fine, but for me it excels as a getting to work bag. It is quite rigid, which makes toting a laptop and other bits reassuringly stable. Putting this down to a learning opportunity, I moved on.

Height and depth was the main issue. With my new kit (40-150 f4 and 9mm with gripless cameras) it is probably better, but at the time the f2.8 and 8-18 pushed it too far and a lack of extra pockets is a pain.

Domke 804 black. This one was a lucky find, but has proven to have the rare and unlikely issue of actually being too big. It will get, as all my Domke bags do, plenty of use in a role not yet defined, but as a day kit bag it is massive and unnessary. The main idea was to replace the F802, my workhorse with a bag that could take a body with battery grip (EM1x, EM1.2 with grip), which it turns out is total overkill. From here I went back to the F802 for the height without the depth and the added pockets I have for the F802 mean it is actually bigger in real terms.

The Domke F3x ballistic. This did not happen, but I wanted it to. In hindsight, the F2 is the better choice I guess, although my current push for a smaller kit might have been a perfect fit for it (the bag was made for a small film era kit like an F3 with drive, 20, 35, 85, 180, which is surprisingly close to a modern mirrorless kit). In other words, this bag might have saved me getting both the F2 and the Photocross 10, but more likely, each of these is a better bag at their respective jobs, but still……. . I actually have one of these in green rugged-ware (very rare BIC camera special edition), but I would prefer a lined one.

Domke F2 ballistic. This is an old friend revisited and updated. The older F2 is now 30+ years old and lacks the lining of the ballistic bags. This is a win mostly, apart from still being a shoulder bag. I will use this for full day kits as it is perfect for the kit it was bought for. The 4 compartment divider and decent main camera compartment fit my gear as well as any bag and I appreciate the pen holders, small front pockets and the way the bag sits on the floor, but the boxy shape can haep it roll off a car seat and it is big on the hip.

See a trend?

Mindshift Photocross 13. This one, like the F3x was the wanted item that turned into something else and probably for the better. I wanted a bag that could take my sports kit, but was more convenient than the Turnstyle 10 (too small) or the Pro Tactic 350 (too….backpack).

Mindshift Photocross 10. This is the (hopefully) right bag for the job above. I think that maybe the 13 would have carried too much and been too big. The 10 is more likely to do a better job of the TT10’s role.

The search for the perfect bag is as futile as much as it is fun, but the need to get something that genuinely does the job when you actually need it for work (not just a hobby) quickly sorts the junk from the winners. I have lots of bags, some not even mentioned here, but they are all useful for something.



Big NewZ

So, after being issued a mixed kit of old and new Nikon DSLR full frame gear, for me not an enticement, the powers that be at the paper have now relented and decided we need…….Z9’s!

Big news for me, this is the camera that will fix my editorial needs, but for sport I will stick with my M43 stuff. The main reason for that is my weight and reach options and the fact I have two very reliable EM1x bodies and good glass (see my recent article on full frame superiority).

What may happen though is the Lightroom processing stream will be fine for most stuff, meaning I will use my gear for some jobs, the Nikon and Lightroom for others, but the bulk of my own gear can now stay at home (to do what, I am not sure, but hopefully something).

Looking at the Z9, and being aware it sits at the top of the Nikon tree, I am happy to use it as much as possible, but the same things I have said before hold true to some extent. The weight of the camera and a standard lens are still a consideration when I can use other gear with perfectly good results (45mp FF is overkill for a newspaper, but the D6, which the other togs wanted, are nearly impossible to get).

A Z9, 24-70 f2.8 and 70-200 f2.8 will come in somewhere near 3.5-4kg. I can carry two bodies and several lenses with a wider range to that or a lot less. I will however really appreciate the quality of the Z9 for editorial work, especially its ability to shun flash, crop heavily, process easily and use the super shallow depth of field when needed using relatively average lenses*.

Not overly worried about quality, but nice to be using gear supplied, not my own.

At 149x150x91 and 1350g compared to the EM1x at 144x147x75 at 997g means it is not much bigger than an EM1x, until you put a lens on that is. Having said that, I would not be interested in an EM1x for editorial work, as I feel it would be overkill.

Bags, my other hobby, will be interesting. I am thinking the F-2, F802 or F804 for the Nikon, the PC10 for M43 kit. It may be the other way around or the whole thing may just go wrong. I have tons of bags, so I am sure something can be found even for the large Z9 body, but no more bags!



*M43 gains the benefit of about 2 stops of extra depth of field at the same taking aperture as a full frame because the lens used to match the same magnification is half as long (M43 45mm = FF 90mm etc so f1.8 looks like f2.8-ish). The Nikon Z mount uses such a wide mouth, that it actually looses a stop of depth at the same aperture as other full frame lenses (f4 = f2.8 DOF). This means that in effect the Nikon Z mount looses 3 stops of DOF when compared to M43 a bit like a small medium format camera, which for me will give me (1) lots to remember and (2) strong tools to use in tandem.

Fun In The Treehouse

Friends have a vinyard and wanted to get some nice images for their website.

Payment as promised was in kind, a dozen bottles of their excellent (and quite expensive) Pinot Grigio, but I would have done it for free.

Another reason for this post is to share the process and to reinforce the power of M43 for shallow depth of field work. I have said before, the shallow depth of field control of M43 is enough for most tasks, sometimes still too shallow. The ability to use any aperture on the lens is a real bonus.

When using full frame, I would go to anything wider than f2 very sparingly and can remember many times when I double or tripple checked focus, often using live view on the rear screen with manual focus if able. With M43 mirrorless cameras I just place my focus point and shoot.

The bulk of these images were taken wide open on my f1.8 45, 75 or f2.8 12-40 and 40-150 lenses, or close to it.

The 75 wide open. Shooting basically straight into the sun, the background Bokeh is a little busy due to the relative distances of the lens to subject and subject to background. The slightly nervous uprights are grasses about ten feet behind the bottles.

This is better with nothing directly behind the subject. The lighter coloured blurred object to the left is the vinyards treehouse.

The above two images were taken the first day just to try out some ideas.

Shooting again into the sun, this time bouncing off the water, I tried using just reflections off the house front (mostly glass).

The image above is recoverrred as far as C1 can, but failed to hold enough information for my idea.

Wanting to try to preserve the background (Schouten Island), I switched to an off camera Godox 860 flash, lens stopped down to f6.7 (f13-16 ish on a full frame) for some hint of shape, using high speed sync. Even with M43, there is little chance of holding detail when things are this far separated at any aperture, but I hope I nailed the balance.

The above image shows how useful flash can be even when daylight is strong. My wife held the flash about two feet to the left firing through a hand held 42” Godox brolly.

Next we decided to incorporate the actual tree house into the images.

First attempt is interesting, using a soft foreground for a frame using the 150mm f2.8. The 40-150 Pro has some slightly nervous background Bokeh.

A little more coherent. 75mm f3.5.

An off-cut vine used foe some interest. This vine strand did a bit of work, but after about 30 minutes, it started to look a little tired. The 45mm wide open is a little warmer than the 75, which makes processing less straight forward.

12-40 at 28mm f2.8. I used this lens often at 28mm without looking at the zoom position. I could have used my 25 or 30mm lenses if I had them with even smoother blurring.

There were a lot more images of family etc, but I will let the client use these as desired.

Back to the house and flash was tried again, used more as gentle fill than to fight the sun, but we found that the brolly alone in reflecter configuration was enough. Lesson I learned here was to clean the glass! 28mm on the 12-40 f2.8.

Going for more context, 17mm at f6.7 on the 12-40. Flash through the brolly again.

Some final detail shots while the greenery was at all useful. 45mm f1.8, the natural warmth of the lens helping here.

Gear list;

  • 1 EM1x (the new one)

  • 1 45 and 75 f1.8’s

  • 1 12-40 and 40-150 f2.8’s

  • 1 Godox 860 flash and off camera controller

  • 1 Godox 42” brolly and hand bracket

  • 2 bottles, glass (clean) and a bit of greenery

  • 1 vinyard complete with tree house

A Mindshift

The Mindshift/Thinktank Photocross 13 eluded me and at the time it was a little disappointing because I really felt it would be a problem solver, but on second thoughts, maybe it was, like a lot of things, a win in disguise.

Originally I wanted a bag that could take my sports kit, but realistically that may have been unworkable. The PC13 would have still struggled with gripped bodies (EM1x), even M43 ones, so I would have had to break down combinations, defeating the purpose of the exercise and not really achieving anything over the LP Pro Tactic 350 or other options (Domke F-804).

The Turnstyle if it was compared looks to be the next incremental drop down in size.

Looking at my day kit, trying to change up the heavy bag on the shoulder dynamic, I decided instead to use my TT Turnstyle 10L as the base, hopefully working from the sling “workbench” with ungripped cameras and a few lenses. The F-2 Domke is nearly perfect, but it is still a shoulder bag.

The Photocross 10 kept coming up a few times when I went looking for alternatives and it has a few benefits compared to both the PC 13 and TT 10.

The mouth is wider, so easier to work with gear and maybe even a two camera kit again. The TT10 is surprisingly big inside, but the two ends can get a little cramped even with the zip fully open. The PC10 is more of a three sided-flap opens forward setup rather than a single sided “gaping mouth” zip. The TT is a better travel bag, a little more secure, but the PC may be the better work bag.

It is a little bigger than the TT, squarer and deeper. It may even be big enough to effectively do what I wanted with sports, a job the TT10 is almost big enough for, but a touch more room (especially height) would make a huge difference. When I did the basketball recently, the TT10 easily took my spare lenses and accessories, even taking my second body, but not in ready-to-go form and it was a little cramped when I went to add a flash.

One helpful reviewer compared the PC13 to a Domke F-2, and they were basically the same volume if very different shapes. This means that if the PC10 is 80% of that, which it looks to be, then it should take my cut down kit, maybe even a smaller sports kit*.

The other small benefit, apart from the fact it is actually available, is that it is a fair bit cheaper. About $70-80au on average. If I love it, I can then gauge the chance of the bigger bag working and I can get some idea of the weight when loaded, so the PC13 may come later.

*For football I intend to use the 40-150 f4 instead of the f2.8, which will change things.

The Mythical Superiority Of Full Frame

This one will never go away.

There is no doubt from my persepctive* that this is just another of those phurphies told by an industry bent on selling more/bigger/newer cameras, but I know there are a few of us who disagree and can prove it if needed ;).

Shape

First, lets look at the history of “full frame”.

When the first makers of compact and portable stills cameras went looking for a film stock to make a “miniature” camera format, easily sourced 35mm film stock was the logical, the only real choice. It was first used as the movie shooters used it, but vertically being the later named “half frame” of 18x24 4:3 ratio. The later Simplexpresented a pair of options, one of which resulted in an odd shape the wider 3:2 ratio 24x36. The 3:2 ratio was never popular with print makers, magazine editors or even frame makers, but it was bigger which matterred when quality was premium. Leitz then adopted the horizontal format officially and that cemented the 35mm 3:2 format into history.

Almost all formats that came before were squarer, but this did not stop 35mm “full frame” from shoe-horning itself into the mainsteam market.

Fast forward to the dawn of digital and the term “full frame” became known as the professional format, almost promoted to a status of mystical superiority. This was because the early formats of digital, forced by technical realities of making sensors, were smaller, but also some common sense prevailed and a few deliberately went away from 3:2 full frame.

Canon and Nikon as the main players were never happy relinquishing full frame as their pro offerring**, but the reality was, the economy of sensor manufacture guaranteed full frame sensor cameras sat at the top of the tree, further reinforcing the perception they were the one true format, the format to aspire to.

Some companies looked at the whole thing from scratch, such as the 4:3 consortium, who decided to (1) go back to a better shape for print and at that time screen and (2) chose a sensor that actually made lens design easier. It was easier for them, because Olympus in particular had not much of a working legacy to support and Panasonic and Fuji effetively started their push here, choosing 4:3 and APS-C respectively as their primary formats.

While we are on it, I feel there are really only two true formats of choice, the square and 16:9 wide screen or wider. The square is convenient, equitable, expressive and logical, while the wide screen format is proven to make sense to our two eyed view of the world.

The 3:2 format is not as settled, crops poorly and forces a wide/tall choice that is often not satisfying or even convenient. The sensor is also too wide for good lens design. The designer need to cover a lot of width for a low content of height.

below are four crops (well three and an original). Which is best? Up to you, but the bottom two to me are more decisive, more poweful, the top two are much of a muchness, the 4:3 one though is more versatile.

When chosing a shape for my images I find the square is exciting and freeing, cinema wide feels “right” and a little epic looking. My standard 4:3 is convenient and less wasteful to make square, but 3:2 is a poor compromise of all of these. The reality is, 3:2 is not a shape enabler it is a shape forcer like square or wide screen, but more limiting. Interestingly, no one I have ever submit my images to has every noticed or complained about my squarer submissions.

My editors both at the school and the paper say the same thing, “shoot horizontal, so we can crop vertically if needed”. Almost all the templates we are forced to use still ignore 3:2 as the norm, often forcing us to accomodate wider or squarer.

This image has enough height at the distance shot. On a 3:2 ratio camera a little more width would have been needed. It can also go vertical with ease.

Ok, so full frame, a format that is the current pro choice is by default an odd choice and has a mixed history. Remove nostalgia and legacy and it makes little sense to stay with it in the modern era. Like the mirror, it is a hold over from the past, a convenience, a habit, supporting the legacy of the two big guns in the industry.

Quality

Next, we need to look at quality or more accurately sufficiency.

What do we actually need? The line peddled regularly is we need more, but ironically, highly ironically, the only medium that needs more resolution in real terms is print, a form of image viewing that is dying before our eyes. When we needed quality in a print based world, resolution or raw image quality was elusive, creating a desire that was hard to sate, but now we have tons of quality and the software to fake it if needed, few ever do.

Taken from a half body (+) image.

I was talking to a friend the other day who is running a 60mp Sony, but never uses his images for more than screen viewing! The proper viewing distance rule has always been the limit that addresses print size needs. Print a billboard, then stand back and look at it. Stand close enough to see the dots that make it up and you cannot see the image. How close do you sit to your TV screen? You are dot or pixel peeking and only photo nerds do that to prove a pointless point.

Enough for most people is 2mp (a 1080p screen), maybe 8mp (4k), but even then, a good screen and good base image quality will defy you seeing any real difference at proper viewing distances. We are constantly looking at high res on lower res screens or the opposite, almost never realising the inherent compromise. Some of the sharpest images you see on TV are still recorded on 720 HD. Even 4k is only 8mp.

What resolutuon did the masters of painting have or the vast majority of film photos in the 20th century? They satisfied all our visceral needs at the time and are often revered now, because they were good in all ways relevant and the technical limits did not stop that from happening. The large format film shooter had enough quality to satisfy any needs, even in the 1930’s, but most of us were happy with less.

As we move through our image making history, we are sold the idea that quality is all. This is true to an extend, but what we need to remember is quality is a combination of things all supporting each other, not a single number or value.

Ask yourself this;

If M43 or crop frame are inherently inferior, then why would any company, especially one with 100 years of innovation and excellence behind it stake it’s whole future on these formats? Why indeed would a photographer who was working in a camera shop with all it’s buying advantages buy into that system and then, with the opportunity to re-think their path, choose it all again and go even deeper?

One day a few years ago I decided to test my Fuji and Olympus cameras and lenses against each other with the intention of choosing between them in the short and longer term. All was going fine until I discovered, after a lot of normal sized viewing, that the Fuji images were actually taken as small jpegs (amazing jpegs being a Fuji thing, so that bit was no surprise).I had this set to take some web images for ebay. I had not noticed even at 8x10 print sizes, that the Fuji files were tiny. Super sharp, colourful, but tiny. One of the main reasons this went unnoticed was my determination to look at the images fairly, using the 29” screen of my Mac without peeking closer and a decent print as the gauge. Turns out the Fuji files just filled the screen with little more to offer, but were fine for that.

Contradictions are plenty. The mobile phone industry and even most full frame makers are quick to sell us on the benefits of their 1” sensor super compacts, their APS-C range or even the smaller sensors in their remaining compact cameras, but then they want you to believe that anything short of full frame is a compromise unworthy of a professional.

From 12mp on in crop frame, only printers or pixel peepers were dissatisfied. The rest of us could not tell the difference and the customer/client/employer usually does not care. My sisters favourite image I have taken for her was on a 6mp crop camera and she always comments on the “quality” of the image.

Search the internet for examples of people "fooling” clients or viewers with their lesser gear posing as top flight equipment***. M43 passing as full frame, low res prints indestinguishable from high res. The list goes on and on, all made irrelevant by application and even time itself.

If the next great thing is needed to make future images, then what about everything that came before?

In the film era, format did make some difference, because lacking computers for post we could only enlarge like to like. Physics at work. Start with a bigger negative then you can enlarge to a bigger size, but even then, some of the benefit of the larger formats was lost due to the difficulty of making better, larger format lenses. The Nikkor 55 micro blew away the standard 80mm Hasselblad lens for actual resolution measued in lines-per-mil, a lens that in itself had a near perfect reputation, but it relied very much on the bigger format for its raw quality.

Convenience and Empowerment

I can go on forever about the massive size difference between my M43 lenses and their full frame equivalents, but lets look at this seriously.

The squarer M43 sensor is a lens designers shape of choice. A square sensor matched to a circular lens shape makes the most logical choice, so closer to square makes more sense than not. This means the sensor offers more useable area for the design and even smaller lenses.

Not a choice for me. The newer Olympus is at least a match for this well respected war horse and the rest speaks for itself.

The reality is if they went with a half sized 3:2 ratio, the sensor area would have to be even smaller or the lenses bigger. The lens mounts on M43 cameras are quite relaxed, in comparison to some full frame ones anyway (the Sony mount actually clips the sensor corners). Nikon has gone super wide with their Z series cameras, but that is again partially because of the shape of their sensor. They could actually squeeze a 4:3 medim format one in there also and may intend to.

I can realistically carry a 600 f4 around with me in my day kit and regulalry have a 300 f2.8 at hand. My 300 f4 is a toy compared to most. There are examples with every lens I own of bigger, heavier, more expensive full frame equivalents, that are often optically compromised or are even more ridiculaously oversized to avoid that. Look at the Sigma ART series for example. Packing 2-3 of their FF 1.4 primes into a bag is no laughing matter, but in M43, it is not a big deal (although they are bigger than most M43 lenses as they can fit APS-C also).

My cathartic moment came when I was showing a friend my Canon 35 f1.4L and my Panasonic 20 f1.7. The size difference was drastic, but I knew that my images at that time (5d mk2 vs EM5 mk1) were effectively the same, the Oly often winning on speed and focus accuracy and always for convenience.

Would I, knowing what I know, sacrifice my current flexibility and freedom for the assumed, but mostly un-provable advantage of full frame quality? No way.

This file is a crop from the image below. Taken on a tiny, relaitively cheap zoom acting as a ff 300 f4 equivalent, it is more than enough for any uses, even without aggressive post processing.

I work with two photographers who omly take a small part of their FF kit with them on a job, because the whole lot is back breaking. I cannot help but wonder if they miss opportunities by leaving things behind. I have a comprehensive kit including 18-300 FF equivalent focal lengths in zooms and some fast primes, two cameras, video and flash accessories, all in a bag that would only handle one of their bodies and an attached lens. I still manage to complain about weight!

Real Benefits

The most often sited benefits of full frame compared to smaller formats are better high ISO noise control and shallower depth of field when needed. These two really wrankle and are the foundation of many full frame users feeling of superiority and non full framers sense of injustice.

Depth of field is a creative tool, but who is to say one format is better than the next. If you want super creamy, super shallow depth, the difference between M43 and FF formats is not going to make a huge difference.

For a real change in depth of field try large format with movements! The rules of depth of field are based on lens magnification, aperture, distance to subject relative to their distance to the background. Basically, the same lens on any format will produce the same DOF, but will be effectively different in practical focal length. There is a lot going on here and relatively small differences in sensor size, determining actual lens magnification, is only one small element. Even a compact camera can achieve very shallow depth of field if used well.

Even at effectively 600mm at f9 on a full frame, there is not enough depth of field to get all the birds sharp. I rarely complain about having too much depth of field.

I find professionally, a little more depth of field is always a good thing. I can fake less in post, but not more. I can use my f1.8 lenses wide open more often than not, gaining the benefit of their light gathering power without worrying about stupidly shallow depth of field.

This brings us to the second point, noise at high ISO settlings. I will go on record here and say I believe software will kill this monster way before sensor size make any real diffierence. I can regulalrly use ISO 6400 in my work, which is more than enough, with little fear of dissapointing a client (they never seem to notice). There is a balance between the quality of the latest M43 sensors and good processing.

Add to this the above mentioned depth of field advantage and I can shoot two ISO settings lower than a full frmae shooter in the same situation. Sure the Full framer can buy the same focal length and speed and gain those ISO settings back, but can they?

A full frame pro body and 150-200, f1.8 to f2 lens would come in at $10-1500au+ (if available at all). The closest I have seen is the Canon 135 f2L, a great lens, but matched or actually beaten by the Oly 150 f1.8 (75 f1.8) at half the price and smaller. This on an EM1x comes in at about $3500au. The full framer can also crop if blessed with more pixels, but that often evens out the advantage as 40mp+ drop to 20mp odd, the same as the M43 camera and that resolution advantage you paid for is lost.

ISO 25,600 properly exposed with a quick trip through C1 and ON1 No Noise. More than just a rescue mission.

Relying on a full frame camera’s sensor and running the files through the same old soup is fine, but why limit yourself to using just the sensor size? If you do go the extra yard, then how much better is sensor A vs Sensor B? When you look a little deeper you see that the world has changed. We can all get enough from a lot less. I have seen amazing results from the latest phones.

Full frame has it’s advatages as does any format, but the relentless push for it to be the sole format for small cameras seems one eyed and pointless. To me it just seems to be the middle ground of the range of formats available and like many middle points, it is a compromise of ideas, master of none. Sometimes it forces a larger dynamic, but not add a decent enough jump in quality.

If the ability to shoot good enough quality for a fine art grade 16"x20” print from a well treated ISO 3200 file is enough, then M43 will do fine, maybe more than enough. For screen filling, even less is needed. Before you go and get that monster sensor camera with 30+ MP, and a lens stable to match, consider what you are actually going to use it for. It may be that less is more.

Just my take.

*20+ years working in camera shops, 35+ years using cameras of all formats and media.

**Canon never made a red ring lens for APSC, Nikon never even fully fleshed out their lens offerings for crop sensor and Sony still under sells and supports their excellent APS-C cameras. Ironically, I found many of their older lenses performed much better on crop sensors.

***Luminous Landscape have plenty, as does the Lens Rentals blog, Ming Thein, who shot a Rollex commercial with a 1” compact, etc, etc, etc.

Can It Be Done?

I am in a funk with photography at the moment. I am working through it, but still have a way to go.

One line of thought is address uncertainty with a review of processes and gear.

What do I use, what should I use and how do I use it?

What I use currently is a good and sensible coverage which forced a few new lenses in specifically to reduce weight. This is designed to handle anything, to be the swiss army knife or boy scout kit. Safe.

G9, EM1 mk2, 9, 17, 45, 12-40, 40-150 f4, flash, Led, mic.

What I should use is the gear that inspires me to shoot my way.

So the question is, if I was shooting purely for myself, what would I use?

Wanting to avoid flash, preferring single focal lengths for clarity of vision and having a depth of field advantage, I would use my fast primes for shorter work. This is less logical for longer lenses (although the 75 could do most things), so my f4 zoom usually, then the 75 for known poor light jobs.

Could I function with a single EM1 mk2, 9, 17, 30, 40-150 f4, an led panel all in a Think Tank Turnstyle 10?

The story telling 17mm just gets the job done. Why clutter that thinking with a zoom?

This is basically my travel kit, my “seeing it my way” kit.

Effectively weightless; 9, 17, mic, led.

Very light, but of substance; EM1.2, 30, 40-150 f4.

Bags.

The Turnstyle 10 is small, but can take this kit with room and the outer and inner pockets will hold the other stuff I need. It is even weather proof.

Maybe the Tokyo Porter or waxed Domke f3x?

What could go wrong?

Cameras break down, which is always a possibility, but the other two togs go with a single body and I would have spares at hand in the office (maybe put one in the car in case). I use two cards so the first level of failure is covered.

Not enough coverage? I can go 18mm FF equiv, 16 is my widest, which requires a zoom, but if 18 does not do the job, should I be taking the image anyway? Remember, my images-my way.

Not long enough? This is rarely a thing, because I know the 150 can be cropped in to crazy small frames for print media (6-800mm eq).

A 150mm lens doubled by M43 then cropped into about a 700mm FF equivalent. Fine I feel.

Too dark?

If f1.8 or f1.4 does not do it, then the LED can add some light or maybe again I need to consider the shot. We rarely shoot in total darkness and if I know that is what I am going into, I will swap the 40-150 out for a flash (same-same).

Video?

Unfortunately, there is not much happening here, but if it does, all I need to do is add the mic back in using a small bag attached to the strap of the TT. The G9’s do video better overall, but the EM1 mk2 in 4k is no slouch. The sound is better with a mic (always the case) and the 4k capture is decent, even good. The 1080p is ok, but 4k is better if I can do it.

So, relying on a single camera, 18-300 range and limiting accessories to what I actually use.

Maybe a thing?

I had the TT10 at work when I wrote this and no, it is not possible. The bag holds it all, but only just and is not convenient for fast work. The Mindshift 10 is on the way, which should do the job better.





A Little Test

A little test to see if you the viewer can see any real difference here.

No hints given what you are looking for, except to say, I have written in the recent past about the real differences between some lenses.

M43 blesses us with a large and well established lens landscape with a couple of things that you can take to the bank.

All the lenses, especialy the long ones, are relatively small compared to larger format kit.

The sensor size and shape was selected with lens design in mind.

There are a lot of giant killers in the range, whether they be pro or not.

Answer?

The 1st, 3rd and 5th images were taken with the multi thousand dollar 300 f4 Pro, which is large and expensive for a M43 lens, but a holiday in the sun compared to an equivalent full frame 600mm f4.

The rest were taken with the slightly better than kit level 75-300, usually at about 200-250mm where the extra width helps on a small ground and ironically the extra depth of field helps tell a story. This lens loves f8, especially below 250mm where it is nearly impossible to split from the 40-150 and 300 Pro lenses. At 300mm wide open, I can sometimes pick it….sometimes.

The give-away is the slightly nervous Bokeh on the Pro 300’s files with a hint of “ringlet” Bokeh (look at the roses in image 3 or the brickwork in image 1), but this is balanced with shallower depth of field for better separation .

The zoom was shot across the ground with a messier background, but still looks less bothersome (the out of focus sign on image 2 & 4 and wheel hub in image 6). The other benefit of the “sunshine” lens is it’s ability to tame strong contrast. I usually do less work to the files.

Rarely do things line up conveniently in photography, but this sunny day story teller is the perfect cricket lens (as long as the sun is out). It offers good compositional control, excellent contrast and is plenty sharp. I find on many smaller grounds, it is hard to get the keeper and batsman in the same shot or the bowler side on with a 600mm equiv and the 40-150 with tc is a little short.

For winter sports like AFL, there is no competition. It’s slow aperture and less assured AF mean the cheap zoom will go back into my home kit, but for summer sports, it is a regular option and is small and light enough to always be included.