Canon EOS R5 - Real World Tests - Climbing Snowdon

This is the first in a series of ‘real world’ tests I’ve carried out with the Canon EOS R5. Apart from shooting with it in the studio and a handful of times for client work, I’ve not really fully pushed this camera to discover its strengths and weaknesses.

With every bit of equipment I invest in, I like to test it in a bunch of scenarios. A large portion of my work involves me capturing subjects as they happen and reacting to the environment and setting. It also happens that my girlfriend and I both love hiking, so we decided to put the R5 through its paces capturing a hike to the summit of mount Snowdon.

This allowed me to see how this camera will respond in near darkness, low light, sunrise and harsh light (midday). It also allowed me the chance to see how it will fair when being thrown around a bit, being turn on and off repeatedly, switching from photo to video mode often and how the autofocus will react throughout the day.

As I always like to mention; this is not a review - these are my thoughts having used the camera. If you want a full review of the camera then there are plenty of better deep-dives dotted around the internet.

Early Morning, a test in darkness.

The following shots were taken in complete darkness, before sunrise at around 5am, with only a small amount of torchlight for illumination. It’s clear to see that the Canon R5 is extremely capable when the ISO is pushed very high. But it has it’s limit, of course. And it is definitely not a low light monster like some of the lower megapixel Sony competition. This is to be expected and is evident in the below image taken at ISO 160,000 where the image noise destroys this picture. Lets be fair, this is ISO 160,000 which is far beyond the realms of any ISO setting I would choose to shoot at. In this situation, if I needed to shoot commercially in this low light situation I would bring some small lights with me, but it’s nonetheless interesting to see the limits of the camera. Interestingly, the camera was able to focus on Genna despite there being almost no light what so ever (I could barely see without a torch).

Noise is very evident at the ridiculously high ISO setting of 160,000.

EOS R5 with RF 28-70mm f2. ISO160,000, f2, 1/25mm

As the light started to come up I was able to bring the ISO gain down to a more sensible setting. The following images show the camera is able to capture incredible dynamic range, colours and keep image noise under control even at higher ISO settings. It’s also worth noting that the IBIS was a massive help in keeping images sharp when the shutter speed was dropped.

Canon EOS R5 w/ RF 28-70mm f2 - ISO 1000, f2, 1/80s

Canon EOS R5 w/ RF 28-70mm f2 - ISO 640, f2, 1/13s

Canon EOS R5 w/ RF 28-70mm f2 - ISO 500, f2, 1/80s

In these early morning shots its clear to see this camera is pretty happy with little or no light. But at the time it was hard to tell how these would turn out just by looking at the back of the camera and I did worry they would exhibit too much noise or the colours would wash out.

However, once I got back to process the images I was astounded. I haven’t even really pushed these too far and there is tons of highlight and shadow data left to play with. One thing worth noting however, although the camera was very impressive in being able to autofocus at all in almost no light, it did throw the occasional tantrum and just start hunting. To be expected of course, our mere human eyes could barely see at all let alone the camera’s autofocus system be able to detect critical focus. When it did start hunting, switching into manual focus was swift, however I did hate the focus by wire on the RF 28-70mm f2. It just lacks that tactile and predictable manual focus feeling. It might be that I’m not use to it yet or it may be that I need to adjust the settings on the camera for how manual focus operates, all I knew at that moment in time was I would really love a standard manual focus lens.

Canon EOS R5 w/ RF 28-70mm f2 - ISO 250, f2.5, 1/50s

It wasn’t long before the sunlight was starting to creep around the mountains just enough to bring the exposure settings into a space that didn’t give me image-noise-related-anxiety. It also meant that autofocus was really playing ball and tracking without any hiccups or issue.

It was this point in the hike where I could really see how the camera was playing to its strengths. We stopped for a quick bite to eat and to take a few images. Autofocus was remarkable to the point that I stopped needing to even thinking about it. When Canon said they used ‘deep learning’ to program the new Dual Pixel Autofocus II, I can honestly see why. No matter what happened, the autofocus system would detect Genna and track focus. Even if the wind blew her hair in front of her face or she turned to face away from me.

I also gave the camera to Genna to see how she would get on with it, bearing in mind Genna is not a professional photographer. The result? Take a look below for yourself, yes - unfortunately the image where this time I am the subject. Perfectly in focus, with zero user interaction. Remarkable.

Canon EOS R5 w/ RF 28-70mm f2 - ISO 125, f2.5, 1/60s

Canon EOS R5 w/ RF 28-70mm f2 - ISO 125, f2.5, 1/60s

Rare image of yours truely. Captured by Genna D. Canon EOS R5 w/ RF 28-70mm f2 - ISO 125, f2.5, 1/60s

Canon EOS R5 w/ RF 28-70mm f2 - ISO 125, f2.5, 1/60s

Canon Colours, oh Canon Colours

I’m not about to get scientific on you. That’s not my style nor my interest. And as I’ve mentioned before I’m not going to bash another camera manufactures ‘colour science’. However, I do want to take a moment to harp on about the colours this camera can capture. There is something extremely smooth and beautiful about the transition between colours which just create these wonderful gradients which appear gloriously rich. It really is a massive step up from the 5D MkIV and canon have done an impeccable job here.

Skin tones are honest and real feeling. There is tons and tons of data in the raw files that it makes processing the images very satisfying should you want to push the images to the extreme.

Please keep in mind that the images you’ll see below are compressed for web and therefore the lush smoothness and rich colours of the camera will not be fairly represented. But, you won’t have to just take my word for it. I’ve uploaded a few samples to my dropbox so you can download them and check them out for yourself. Link at the bottom of this post.

I also should mention the RF 28-70mm f2

A camera is only as good as the glass you attach to it. So it is only fair that I do just send some praise and criticism in the direction of the Canon RF 28-70mm f2.

Before buying, I read a number of articles about this lens, saying that this ground breaking, world first f2 constant zoom is prime quality but with the versatility of a zoom.

Personally, I wouldn’t call it prime quality. There is always a trade-off somewhere. However, it is pretty damn close to being prime lens quality with just a few drawbacks. I want to keep this blog post on topic of ‘real world’ use case of the R5 and not focus too much on the lens, so I’ll wrap this up quick.

Pros: Colours are stunning. Versatile as hell with 28-70mm being an excellent all round focal range. Super fast, constant f2 aperture for a zoom. Bokeh is lovely and appealing, but not quite as smooth as a high quality prime. Autofocus is near silent and extremely quick, quite possibly the fastest autofocus I’ve ever witnessed. Weather sealing. 9-blade rounded aperture. Min aperture of f22.

Cons: VERY heavy. At 1430 grams this lens on the R5 will really drag you down and give you a neckache after a few hours of being on location, specially if you’re hiking or traveling large distance with the camera out of the bag. It’s also not subtle in the slightest, in fact I had a number of people actually comment on the size of my camera (I knew they actually meant the lens, as the R5 is actually fairly small. But I wasn’t going to be pedantic and correct them). As I was heading down the mountain one lady even said to me “Wow, fair play to hiking to the top with THAT camera! Look at the size of it!”. She wasn’t wrong, the damn thing definitely made things a little tough going. However, for the image quality and versatility it was an even sacrifice. It’s not prime quality. Don’t buy this lens if you want it to replace your set of high quality primes, buy it as well as your primes for it’s versatility. I did a small test vs my Sigma Art f1.4 35mm and 50mm and the main difference being the out focus areas are just SO much smoother with the primes and lens flares much less fussy. I also really miss 24mm, I know it’s only 4mm difference between the 24-70mm f2.8, but there were a few occasions that I really wanted this lens to just be that tiny bit wider.

Bags of Dynamic Range

As the morning went on and the daylight became brighter (as it does), it was a good chance to check out the dynamic range. The brighter daylight between around midday is a really hard test when it comes to retaining both shadow and highlight information. The R5 did surprisingly well, but with some clear limitations, nothing that isn’t to be expected however. The thing I have come to notice though is at sunrise/sunset, the camera has so much more DR than the 5D MkIV that it is not longer necessary to capture multiple exposures and composite as a HDR in post, which I’m extremely grateful for. Highlight roll off is very smooth even in the most extreme examples and shadow detail is retained beautifully without any noticeable noise.

I did notice on a few occasions during this harsher light that the autofocus was starting to get a little confused. It seemed to pickup on the odd random object in the background rather than track my subject. This was a very minor issue and would only happen for a few shots before it quickly figured out the subject and began tracking again.

The Summit.

As we reach the summit at around 11am it was getting to that point of the day that most photographers avoid. The light was harsh as hell with only a little bit of defence from a small amount of atmospheric haze. Still, it’s a great way to show off the camera’s ability during these conditions. Processing these images it took out some off the harshness in the highlights, this camera does tend to have very nice roll off in the bright areas so after some playing around in capture one I actually quite liked the shots, despite the light being the worst!

Wrap up and final thoughts

For me and what I will be using this camera for (mostly), the Canon EOS R5 has been (so far) an excellent investment. On this test shoot it was able to capture images that I wouldn’t have even attempted with the Canon 5D MkIV or I would have had to contrive and re-shoot a number of times, then review on back of camera to double, double check I nailed the shot. The biggest draw for me is the image quality and autofocus. It’s just glorious.

I genuinely have very little to say about this camera that isn’t good. And I know that may make this seem slightly unbalanced, but after using it in this way on location, it was just remarkable. Canon, well done!

I want to note also that I used one battery on this whole shoot. I captured 1286 images in total and around 286 video clips in 4K Fine 24p. Honestly, I think that’s amazing. When I did come to charge the camera when we got back to the hotel, instead of getting the Canon charger out of my camera bag, I simply plugged the camera into my Pixel 3 XL USB C fast charger while we went for much needed dinner and a pint. By the time we got back to the room it was fully charged and ready for another adventure. That might be a small thing, but it does mean that you can either charge on the go (on location) with a portable battery bank or it means you can go on location and not have the charger taking up valuable space in your camera bag.

As for video, I’m going to leave that for another blog / vlog coming soon. Teaser: it didn’t overheat (Shock!) and it was in fact pretty bloody awesome. Stay tuned.

That kinda wraps up my thoughts of my camera on this ‘real world’ test on Mount Snowdon. I owe a massive thanks to my very patient and awesome girlfriend Genna for being my model (again!). Thank you!

Like to see some full resolution .jpg samples? Click below to download.

The Forest in December -Canon C200

In and amongst the craziness that is Christmas and New Year, We found a bit of time to head out to a forest 10 minutes drive from our house. We’d not yet explored the walk near Hurn Airport in Bournemouth and after all the mince pies and booze we felt we needed to get out and have a little bit of an explore.

It was also the perfect opportunity to do a few more tests with the Canon C200. The weather was beautiful and we managed to catch the one evening where it wasn’t pouring it down with rain.

In order to test the camera in this environment I just wanted to film as we walked as much as possible and just shoot what we saw. For these types of shoots, recording in XAVC 4.2.0 8bit is perfect. It’s lightweight, small file sizes mean over-shooting is never really a problem. I was also really keen to test Canon log3 gamma to see how it held up in scenes where a ton of dynamic range was required.

From this little test my conclusions were that the C200 handles tricky lighting conditions very well, rolling off clipped highlights nicely and keeping things very controlled in backlit scenes. Clog 3 offers plenty (13 stops) of DR to be able to bring challenging shots like sunsets back to something pleasing and shadows can be recovered without much noise if required. Colours are honest and rich, greens seem to be very vivid and I actually found myself taking back the saturation a bit in the grade to avoid things looking unrealistic. The colour science offers a fantastic starting point however for stylising the shots exactly how you’d like. Canon has done an incredible job getting this much colour information into an image that is captured in 4.2.0 8bit!

New Forest in Autumn - Fujifilm X-T3

So I actually shot this last year (2018) in autumn as a bit of a test for myself to understand a bit better what the video capabilities of the Fujifilm X-T3 were. But for one reason or another I never did anything with it.

So, instead of it lying around doing nothing I thought I’d share it on here, along with my thoughts on shooting video on the X-T3.

The X-T3 presents a new era of video for Fujifilm. They tried really hard with the X-T2 but, in so many ways it just fell short of the mark. With the new X-Trans 4 sensor its a real contender when pitted against the other mirrorless cameras from the likes of Sony and Panasonic.

The video was meant to test it’s internal recording capabilities at 4K 60fps HEVC 10bit 4.2.0.

The Good
60 frames at 4K is just awesome. And to be able to capture it at 10bit in HEVC is a treat. Colour science from Fujifilm has always been fantastic and has kept me coming back to Fuji again and again.

F-Log offers (on paper) 11.2 stops of dynamic range. But, to be honest DR is something that is banded around as marketing hype and in the real world it will always be slightly less. Regardless of what is stated, you have enough to create great looking images even in challenging lighting conditions such as this example video shot shortly before sunset.

Not to mention the Fujinon glass. Fuji have a huge wealth of experience making lenses over the years for other camera manufactures, broadcast, scientific and medical uses. If you’ve been shooting for a while, chances are at some point you’ve used or seen something shot with a Fujinon bit of glass, but you might not have realised it. As a result every lens they make is just awesome (IMHO). This example was shot using the Fujinon 35mm f1.4 and 56mm f1.2.

It’s small. Really small. So same goes for most of the Fujinon lenses. As a compete package its very discrete and perfect for travelling, hiking or if you need to move light. It’s also perfect for a ronin or other form of grip as it weighs very little indeed. However, that also comes with it issues as I cover below.

Photos and videos. In one package. Everything is there and you don’t need to travel with both bits of equipment. And for keeping things feeling and looking the same, Fuji also provide an Eterna LUT so you can restore that F-Log footage to Rec.709 with colours that feel very Fujifilm.

Slow motion at 120fps in full HD is awesome. It definitely isn’t as sharp or detailed as the other recording modes, but its a nice feature and footage still looks good.

Skin Tones are excellent in all modes. F-Log and the film ‘simulations’. But colour is what Fuji always smashes.

Auto White Balance and Exposure are reliable. Naturally it’s far better to shoot everything manual to avoid mistakes or the camera overcompensating. However, should the situation arise, you can rely on the camera to just figure things out for you.

The Bad

It’s small. This is a good point and a bad point. Bad, because as i’m sure you’ll know, a small camera can be tricky to operate and keep your shots looking nice without involving some sort of grip or elaborate shoulder mount rig and viewfinder set up. Now, this is an option of course - but once you mess around clamping this little guy into a shoulder mount and fix your gratical to just the right eye relief, you begin to wonder why you didn’t just bring your FS7/C200 etc. It’s just not very practical for run and gun or documentary work. If your in a controlled environment it can be awesome. But on location, in a field when you need to work fast with little or no grip; it can be a bit off a faff.

For example, when shooting in the highlands I had to abandon my viewfinder as the HDMI signal kept dropping out from the camera every time I moved around due to it’s pathetic little micro HDMI port - it’s just not a workhorse in the same way a dedicated video/cine camera is. Now, of course this isn’t truly a ‘negative’ point about the camera - it’s just a fact about it’s form factor. It is what it is.

That’s it really, I don’t have many negative things to say about the X-T3 when shooting video. So providing you have the right use case for it, the Fujifilm X-T3 is an awesome option for shooting video and can deliver some incredible results.