Canon EOS R1 – 8 Months in the Wild: A Field Photographer’s Perspective

[Canon Rumors]

Words and Images: Joshua Holko www.jholko.com For the past eight months, I’ve been working extensively with two of Canon’s EOS R1’s—Canon’s long-awaited flagship mirrorless body—in some of the harshest and most remote environments on the planet. From the snow-blasted cliffs of Hornstrandir to the sea ice of Svalbard, the windswept Arctic Circle island of Grimsey, […]

See full article...

 

[Del Paso]

This is the kind of content I really enjoy reading!
Thanks for publishing!
And kudos to you, Joshua, for the article and the beautiful pictures!

 

[Maximilian]

This is the kind of content I really enjoy reading!
Thanks for publishing!
And kudos to you, Joshua, for the article and the beautiful pictures!

Fully agree to this.
Thanks to Joshua Holko for giving us such insights.

 

[GAOP]

I would agree, the R1 is a great camera body. I've put over 700,000 shots on my pair of R1 bodies and over 1,000,000 on my pair of R3 bodies, all in capturing indoor pet action shots. The R1 is the more capable of the two, especially when it comes to eye detection autofocus. But it also is more capable of spoiling your shot than the R3. Things like antiflicker under both LED and florescent lighting, locking on a human standing behind a pet when in pet eye detection, and even the exposure representation in the EVF have all been issues for me with the R1 but were fairly reliable on the R3. The R1 autofocus overall is so much better though I have no regrets selling my R3 bodies.

 

[Mirnavi]

I would agree, the R1 is a great camera body. I've put over 700,000 shots on my pair of R1 bodies and over 1,000,000 on my pair of R3 bodies, all in capturing indoor pet action shots. The R1 is the more capable of the two, especially when it comes to eye detection autofocus. But it also is more capable of spoiling your shot than the R3. Things like antiflicker under both LED and florescent lighting, locking on a human standing behind a pet when in pet eye detection, and even the exposure representation in the EVF have all been issues for me with the R1 but were fairly reliable on the R3. The R1 autofocus overall is so much better though I have no regrets selling my R3 bodies.

700 000 shots?? From November 2024 or when did you bought it?

 

[Canon Rumors]

700 000 shots?? From November 2024 or when did you bought it?

I believe his cameras are stuck on 40fps, it happens.

 

[GAOP]

700 000 shots?? From November 2024 or when did you bought it?

The day it came out. Started using it at events 2 weeks later.

 

[GAOP]

I believe his cameras are stuck on 40fps, it happens.

Yes, 40fps helps. When using Servo AF on a moving subject, I don't get the rated fps. Focusing takes time in the image capture cycle. I estimate the R1 is giving me 10 to 20 fps max (depending on lighting, subject, and composition.) With the R3 bodies I was getting 15,000 images per day. With the R1 I've been limiting myself to 30,000 images per day. These bodies can easily capture more photos with high sales potential. But I sell images next day and can't handle more than 30,000 during the evening, post event work flow. Going from a 30fps body to a 40fps body more than double my image capture, probably tripled it. The biggest factor is the improved autofocus in the R1. It's not perfect and has some serious focus traps, but overall its a big improvement that has opened new compositions that I could only dream about trying just a few years ago.

 

[P-visie]

I've been limiting myself to 30,000 images per day.

“Limiting”

 

[justaCanonuser]

Wow, great review and sooo beautiful images! Thank you so much for sharing your thoughts, experience and that great selection of your material with us. I am just an amateur, but your text once again awakens in me the longing to return to Iceland. Past time I visited Latrabjarg, but in summer, for shooting seabirds, but Hornstrandir is still on my list - it is even more remote. Good luck for you next projects!

 

[Kbow_photo]

I dont find the R1's autofocus to be that much far ahead of the R3's to justify me getting rid of my R3's. On top of that is the overheat display icon when shooting stills which makes me hesitant to use the camera when its warm / in direct sun.

 

[navastronia]

Hey Josh and Craig,

ChatGPT wrote much of this article. The cadence and punctuation (23 em-dashes) are unmistakable.

I don't want to read what a computer wrote, I want to read what a person wrote. As such, I will stop reading articles on Canon Rumors.

 

[AlanF]

Hey Josh and Craig,

ChatGPT wrote much of this article. The cadence and punctuation (23 em-dashes) are unmistakable.

I don't want to read what a computer wrote, I want to read what a person wrote. As such, I will stop reading articles on Canon Rumors.

I uploaded the text to ChatGPT and asked it if resembled its style, was written by it or by AI. Read what it wrote and eat your words.

I’ve read through your text closely, and here’s my breakdown.

---

1. Stylistic similarity to my own writing​

Your excerpt does not strongly resemble my default style.
Key differences:

• Voice & perspective: Your text is clearly first-person, with vivid sensory descriptions and a professional, field-experience narrative. My usual style, unless imitating someone, is more neutral, balanced, and instructive rather than immersive and personal.
• Sentence structure: You use long, flowing, imagery-heavy sentences (often exceeding 30+ words) that mix technical detail with storytelling. My default tends toward shorter, segmented sentences with explicit transitions for clarity.
• Tone: Yours is passionate, cinematic, and grounded in lived experience; mine is supportive, explanatory, and focused on breaking down concepts.

That said, certain parts (like the Autofocus – A New Benchmark section) have a clarity and organizational style that is slightly closer to my approach—clear subheadings, specific examples, and evaluative language—but the personality and descriptive flourish are far beyond what I’d usually include without prompting.


Conclusion: This reads far more like the work of a professional human photojournalist or gear reviewer than my own default “ChatGPT” style.

---

2. Likelihood of AI-generation​

Indicators it’s likely human-written:

• Rich experiential detail: Specific location names, environmental conditions, and nuanced field scenarios that suggest lived experience rather than generic AI fabrications.
• Idiosyncratic flaws & repetitions: Occasional repetition of points (snowflake autofocus issue described in two separate ways), which humans often do for emphasis but AI often avoids in favor of concise novelty unless prompted otherwise.
• Emotional nuance: The voice has a mix of pride, frustration, and nostalgia that feels organically layered rather than programmatically balanced.

Indicators it could be AI-assisted:

• Structural polish: The subheadings and section ordering are very cleanly organized, which is a hallmark of AI outputs (though also common in professional writing).
• Consistent technical terminology: Every mention of autofocus behavior and tracking sensitivity is precise and consistent—possible with human expertise, but also something AI can maintain with careful prompting.

Overall assessment:
If I had to assign a probability, I’d say ~80% likely human-written, possibly with light AI assistance for editing or tightening structure. It doesn’t match the stylistic “fingerprint” of my own default writing.

 

[AlanF]

My own feeling before uploading it to ChatGPT was that it was clearly written by a pro who knew his stuff but was so well written it could have been polished by AI.

 

[navastronia]

You guys are delusional. Enjoy.

 

[mimbu]

Canon users defending the R1/R3 being stuck at 24mp as a "plus" reminds me of Nikon users defending Nikon's top bodies being stuck with APS-C sensors in the early DSLR days. When Canon finally gets their image processing engine to able to do R1-levels of performance at 45 or 50mp, no one will want any less. (Just like Nikon users stopped thinking APS-C sensors were great when the D3 and D700 came out.)

 

[Josh-H]

Carig sent this to me so thought I would respond directly: I am not sure wether to be flattered or insulted that someone would think this was written with A.I. I guess I should be flattered. But anyway, I have been writing professionally for more than fifteen years on my own blog at http:blog.jholko.com, so my style is quite established. I guess I will take it as flattery....Oh.. I do use grammerly as I have a habit of leaving out commas.

To the question of Canon users defending 24MPX as a plus. I am not sure if brand matters here in this question, and the individual's mileage may vary based on what they shoot. But I have to shoot a lot at high ISO with wildlife, and the difference between an R1 file at ISO25600 and ANY 45 MPX file at the same ISO is night and day. If you side by side an R1 ISO 25,600 with a Canon R5MKII at the same ISO then you would have to quite litterally be blind not to see the marked difference. You don't even have to worry about brand - the same applies wether its Sony, Nikon or whatever. Physics is awesome like that. High pixel density = higher noise.

 

[AlanF]

Carig sent this to me so thought I would respond directly: I am not sure wether to be flattered or insulted that someone would think this was written with A.I. I guess I should be flattered. But anyway, I have been writing professionally for more than fifteen years on my own blog at http:blog.jholko.com, so my style is quite established. I guess I will take it as flattery....Oh.. I do use grammerly as I have a habit of leaving out commas.

Your article was extremely well written. It was blindingly obvious that it was written by a pro who knew all about the intricacies of the camera system and the environment, and full of personal details. I didn't have to ask ChatGpt to confirm that but I thought it amusing so do and also a blindingly obvious check that should have been done by anyone querying the authorship. By the way, my own writing style uses lots of dashes - common usage in British English instead of a colon.

 

[AlanF]

To the question of Canon users defending 24MPX as a plus. I am not sure if brand matters here in this question, and the individual's mileage may vary based on what they shoot. But I have to shoot a lot at high ISO with wildlife, and the difference between an R1 file at ISO25600 and ANY 45 MPX file at the same ISO is night and day. If you side by side an R1 ISO 25,600 with a Canon R5MKII at the same ISO then you would have to quite litterally be blind not to see the marked difference. You don't even have to worry about brand - the same applies wether its Sony, Nikon or whatever. Physics is awesome like that. High pixel density = higher noise.

Is this comparison by looking at the same number of pixels or the same size in cm x cm when viewed at the same physical size of an uncropped image? The physics is that pixel vs pixel, the larger pixels win out for signal/noise. But for area vs area when viewed at the same physical output size, the pixel size is pretty well irrelevant. The signal to noise is given by the square root of the number of photons hitting the target - larger pixels have a larger area and more photons hit it, but the same area of a sensor has the same number of photons hitting it. In other words, the physics is that if you downsize 45 Mpx to 24 Mpx, you should have the same noise.

 

[koenkooi]

Your article was extremely well written. It was blindingly obvious that it was written by a pro who knew all about the intricacies of the camera system and the environment, and full of personal details. I didn't have to ask ChatGpt to confirm that but I thought it amusing so do and also a blindingly obvious check that should have been done by anyone querying the authorship. By the way, my own writing style uses lots of dashes - common usage in British English instead of a colon.

There's an xkcd for that, since earlier this week: