Canon RF 10-20mm f/4L USM announcement tonight

[neuroanatomist]

IS on a 10mm lens, it’s just a gimmick.
The IS is utterly pointless,

To you, perhaps. On an R/RP/R8, 5-stop IS means a 3 s exposure at the wide end. On a body with IBIS, it’s a 6 s exposure. That means, for example, silky waterfalls without a tripod.

As someone who has shot waterfalls in winter standing on a few meters of powder with a tripod set up on the snowshoes that I was wearing…it doesn’t sound like a gimmick.

 

[AlanF]

To you, perhaps. On an R/RP/R8, 5-stop IS means a 3 s exposure at the wide end. On a body with IBIS, it’s a 6 s exposure. That means, for example, silky waterfalls without a tripod.

As someone who has shot waterfalls in winter standing on a few meters of powder with a tripod set up on the snowshoes that I was wearing…it doesn’t sound like a gimmick.

You have 3 feet?

 

[Dragon]

You have 3 feet?

Easy to spot your native land. As a Brit salesman I once employed was wont to say "British humor, mean but dry".

 

[dirtyvu]

this is the first lens that I preordered on announcement day! so excited. B&H says November 2 it'll ship

 

[AlanF]

Easy to spot your native land. As a Brit salesman I once employed was wont to say "British humor, mean but dry".

He would have said "humour".

 

[shadowsports]

this is the first lens that I preordered on announcement day! so excited. B&H says November 2 it'll ship

This is always exciting. I've had the opportunity to do this a few times with lenses and a body. I think the 10-20 will be a winner, for stills and video. Looking forward to seeing some of the images.

 

[Smiling Eddie]

Also IS versus none.

Agree - one small step for a man... I wish I had that on my EF 16-35 II.

 

[Dragon]

Agree - one small step for a man... I wish I had that on my EF 16-35 II.

IS works well on the RF 14-35.

 

[David_B]

Is this the patent post for the RF 10-20 f4/L?

Canon files a patent for a wide-angle RF zoom

Correction: These optical formulas are for full-frame sensors. We know that an RF-S wide angle zoom lens is coming later in 2023 to take the place of the EF-

www.canonrumors.com

At the wide end, the image circle will not cover the full frame sensor. The patent doesn't even go all the way out to 20mm, so what's going on after 18mm? Just pretending? Or is the "zoom" end only 18mm and the "20mm" is just marketting?

To get a full frame picture from the patent'd lens, Canon will need to use the camera's firmware to stretch the image out.

Is this what passes for "pro" level lenses these days?

Are there any photography industry journalists that go back to manufacturers with pointed questions on topics like this?

 

[neuroanatomist]

Is this the patent post for the RF 10-20 f4/L?

At the wide end, the image circle will not cover the full frame sensor. The patent doesn't even go all the way out to 20mm, so what's going on after 18mm? Just pretending? Or is the "zoom" end only 18mm and the "20mm" is just marketting?

To get a full frame picture from the patent'd lens, Canon will need to use the camera's firmware to stretch the image out.

Is this what passes for "pro" level lenses these days?

Are there any photography industry journalists that go back to manufacturers with pointed questions on topics like this?

No, that’s not the patent. But in the patent where this lens is as an example, the optical formula is similar in that it doesn’t cover the full frame image circle at the wide end (which is 10.3 mm). Correction of the lens’ geometric distortion fills the corners.

This is old news – the RF 14-35/4L has the same characteristic, and it launched 2 years ago.

 

[David_B]

This is old news – the RF 14-35/4L has the same characteristic, and it launched 2 years ago.

Waiting for Sigma/Tamron to make proper RF wide angle lens.

 

[Dragon]

No, that’s not the patent. But in the patent where this lens is as an example, the optical formula is similar in that it doesn’t cover the full frame image circle at the wide end (which is 10.3 mm). Correction of the lens’ geometric distortion fills the corners.

This is old news – the RF 14-35/4L has the same characteristic, and it launched 2 years ago.

And once you start using it, the 14-35 is very easy to love. Digital correction is just a new tool in the lens design toolbox that some folks aren't ready to accept. Adding a half dozen more elements to stretch the corners would not add resolution to said corners. In fact, the result would likely be worse and the lens would be heavier and have more flare. Given how well my 14-35 works, I am tempted to pick up the 10-20, as the only lens I have that is nearly that wide is a very old Sigma EF 12-24 that is geometrically almost perfect, but the resolution stinks. The later Sigmas were less well corrected, but had much better resolution. The other new tool in the box is the highly aspherical molded plastic lens as used in the new RF 28mm f/2.8, and that technology also produces excellent results, but does not appear to have been used in the 10-20, albeit it has some less aggressive aspherical elements.

 

[David_B]

Digital correction is just a new tool in the lens design toolbox that some folks aren't ready to accept.

CR2 is not RAW if camera does correction.

 

[koenkooi]

CR2 is not RAW if camera does correction.

The CR3 files are all uncorrected when it comes to lenses. Your RAW converter of choice does the correcting.

 

[neuroanatomist]

Waiting for Sigma/Tamron to make proper RF wide angle lens.

I suspect you'll have a very long wait. While you'e waiting for a 'proper' WA lens, here are a few things to consider:

• Do the 'proper' Sigma/Tamron WA/UWA lenses require distortion correction on Sony/Nikon mirrorless bodies? Sony, Fuji and Olympus/OM have forced distortion correction on their mirrorless lenses for years. Much has been made (inside this little forum teapot, at least) of Sony and Nikon having 'open' mounts, meaning they officially work with Sigma and Tamron. I have not looked at any Sigma/Tamron patents, but given that the corresponding Sony and Nikon OEM lenses have the same 'forced corner stretching' as the RF 10-20 (and RF 14-35L, 24-240, and 16/2.8) it's very likely that any future Sigma/Tamron WA/UWA lenses that Canon facilitates will require distortion correction just like some Canon RF lenses.
• The RF 14-35/4 at the wide end, with the distortion corrected using DxO PhotoLab, produces images with an FoV similar to ~13.5mm on the EF 11-24/4L (in-camera/DPP match the 14mm focal length exactly), and the corners are just as sharp. The EF 11-24 at 13-14mm has essentially zero distortion (that's the transition point from barrel to pincushion), yet the corrected RF lens performs just as well.
• The RF 10-20 is even wider than the EF 11-24 and based on the MTF curves (which include distortion correction, else they'd drop to zero in the corners) the new RF lens is optically better.

So...feel free to wait, perhaps forever, for a 'proper' 3rd party WA/UWA zoom for the RF mount. Meanwhile, those of us who understand the optics and the relevant output characteristics will happily continue using our small and light L-series lenses that deliver optically excellent results after digital correction.

 

[David_B]

I suspect you'll have a very long wait. While you'e waiting for a 'proper' WA lens, here are a few things to consider:

• Do the 'proper' Sigma/Tamron WA/UWA lenses require distortion correction on Sony/Nikon mirrorless bodies?

All wide angle zoom lenses require distortion correction. All. Only difference is where it happens and the level of control you have over it.

So...feel free to wait, perhaps forever, for a 'proper' 3rd party WA/UWA zoom for the RF mount. Meanwhile, those of us who understand the optics and the relevant output characteristics will happily continue using our small and light L-series lenses that deliver optically excellent results after digital correction.

Yes, I am happy to wait for a WA/UWA zoom that produces an image circle that covers the entire sensor. Meantime, there are other EF "L" lenses that with the RF-EF adapter are perfect for my needs and when I use the RF-EF adapter with the slot for a filter, it is even better than using RF "L" lenses.

You are happy with tradeoffs that I'm not and I'm happy with tradeoffs that you're not.

 

[EOS 4 Life]

Is this the patent post for the RF 10-20 f4/L?

No, obviously not but the designs might be related.

 

[EOS 4 Life]

All wide angle zoom lenses require distortion correction

Not fisheye lenses.
My only issue is how rectilinear does a lens need to be to be called "rectilinear"?

 

[David_B]

Not fisheye lenses.

Point.

My only issue is how rectilinear does a lens need to be to be called "rectilinear"?

Good question, I've never seen that defined.

 

[rbtree]

I'm sure plenty of people can find a use for this lens.

Yes, it’s fine for Astro as plenty of folks use the 11-24 with great success, and this one is wider for ultimate star trails and full Milky Way shots!
It looks stunning for vlogging!!! Like a GoPro but not puked out.
Indoor people in tight spaces, I’m using iPhone ultra wide which is ~13mm very often, but often not enough.
For architecture it rarely ever can get wide enough…
For landscape I just use stitching (doesn’t work for architecture for me as indoors needs ~28mm to stich well, which is a pain in the butt with 15+ frames)

Until I get a tracker, I won't be using an f/4 lens for astro. I prefer less wide, as I often want larger foreground objects, and then shoot rows for stitching. My astro quiver: Rokinon 8mm fisheye (seldon used) 14 2.8, 24 1.4 Sigma Art's 28 and 35, 1.4, and 85 1.8 Lots more lenses will be usable for astro once I get my first tracker.

That said, I just bought the 11-24, in pristine condition, for $1100!! Took some excellent shots with it today, paired with my R5. Yes, it's heavy. It will get used!
I also ordered a used EF-R filter adaptor from borrowlenses.com for under $100. Comes with an ND, hopefully a VND... And I found a clear filter for the adaptor for under $35. The adaptor will come in handy for filtration needs on some other EF lenses, I'm sure.