Report: Canon to launch 4 fast L prime lenses in the first half of 2024

[sanj]

True. But that also applies to geometric corrections accomplished optically, with lens elements. That’s why the center is sharper than the corners.

I’d argue —and I’ve shown the images to back it up— that algorithmic correction of geometric distortion can be as effective as optical correction. Thus, the RF 14-35/4 at 14mm after the lens profile is applied is just as sharp in the corners as the optically-corrected EF 11-24/4 at 14mm

I wish for the best possible IQ from the lens itself first. Then corrections.

 

[David - Sydney]

Blue hour means enough light for a 30-60 sec exposure and a much narrower aperture.
f1.4 seems to be too shallow DoF for close foreground shots, but if the foreground isn't too close, f1.4 may also provide for foreground at night.

the challenge to have a panoramic foreground (and sometimes multiple compositions) covering the same field of view of the milky way within the limits of the blue "period" (no shadows). It isn't normally a long time and of course decreasing light during that period. Using f1.4 for speed but any close foreground elements would need focus stacking of course.
Using f1.4 for night foregrounds (maybe stacked for noise reduction) is another option. Need to pick the hyperfocal distance in that case.

 

[roby17269]

I think , "coordinated" placed in the wrong part of the sentience. I often do it and have to proof read what I wrote.

no worries! English is not my mother tongue and I often write turns of phrase that make sense in my original language but are not really good English

 

[EOS 4 Life]

True, but then even less for (ultra) wides
BTW, how can IS be "coordinated" for the 50 / 85 1.2? those lenses do not have ILIS

That is the point.
They do not need it.

Camera + lens stabilisation | which Canon lenses work best?

With the EOS R5 and R6 cameras the first to offer in-body image stabilisation (IBIS), this can work in conjunction with Canon’s optical IS systems to offer up to 8 stops. But it’s not additive.

www.eos-magazine.com

 

[Longtermer]

I wish for the best possible IQ from the lens itself first. Then corrections.

Of course that's true - but isn't the most important thing the final image (taking into account both optical and software corrections)?

 

[roby17269]

Over a year ago, there was a rummer about a 12mm f/1.2 and I was very excited, but not to long later, more info came that is was for a phone, point and shoot or something. I guess someone will say, "Sigma already had a 14mm F/1.4,"no mater how canon's version turns out finding whatever differences there are as a reason that Canon is D-worded.

Waiting for the TS lenses to be announced seems like a good idea to avoid losing money from buying and selling the 28mm f/1.4

12mm 1.2 would be sweet but probably big and expensive. Not that that has stopped me in the past
I like TS lenses, they are great fun! but I'd wager we're looking at 2+ years before anything materialize...

My other camera system is based on the Hasselblad H system. That one is truly d-worded. Guess what? I can still take photos I love with it

 

[Richard CR]

RF 50mm 1.4 delayed for 2033?

Too soon.

 

[neuroanatomist]

I wish for the best possible IQ from the lens itself first. Then corrections.

Wishes are nice.

Canon must consider factors other than ‘the best possible IQ’.

 

[EOS 4 Life]

I think , "coordinated" placed in the wrong part of the sentience. I often do it and have to proof read what I wrote.

Coordinated IS decides how much IBIS to use and how much lens IS to use even if that is 100% or zero.
Obviously, a lens with no IS would be zero.

 

[DhlcadR6]

RF 50mm 1.4 delayed for 2033?

Delay to 2077 or even 3202

Why not a 20mm 1.4? I really love 20mm for landscape.

+1, For some reason Canon never seems to like this focal length....I guess wait for Sigma RF to bring the 20mm f1.4 Art

I'm totally ignorant on these matters so could people enlighten me: why a 28 as well as 24 and 35? Is that a gap that needs to be plugged?

Smaller but still wider than 35mm? If Canon can make the 28L in smaller than the new 10-20L/14-35L, that'll be an attractive option.

I will do my best to ignore this rumor (specifically the part about the 35mm) until we get a more solid one.
I dunno. Why IS in these lenses? They did not put it in the 50 and 85 1.2 where it would have made more sense. I use those 2 and I am not missing ILIS, so I am sure that IBIS would be fine for wider primes.

As a moron who loves to put full-frame RF lenses onto my tiny white R50...OIS/Lens IS is much needed. Thx Canon

Canon currently is going after Sony's design philosophy (Even though Sony's lens designers are ex-Canon personnel), lightweight+size over f1.2. And I believe Canon might try to use the same housing for the 24/28/35L primes as they have a trend of "cost control"

 

[roby17269]

As a moron who loves to put full-frame RF lenses onto my tiny white R50...OIS/Lens IS is much needed. Thx Canon

You are a bit of an outlier though ... and even if Canon will be happy to sell expensive lenses to you (and to everyone else willing to buy them), I believe that they market the slower / cheaper 1.8+ primes which have IS in many cases for buyer of cameras such as the R50.

Canon currently is going after Sony's design philosophy (Even though Sony's lens designers are ex-Canon personnel), lightweight+size over f1.2.

Maybe? I am not sure that this is true

And I believe Canon might try to use the same housing for the 24/28/35L primes as they have a trend of "cost control"

Don't really care about this, but there are other reasons for it beyond cost reduction: having lenses mechanically compatible is good for videographers

 

[davidespinosa]

why a 28 as well as 24 and 35?

That’s a religious war — some people love it, some hate it. Depends what you’re used to.

For some, it’s a good compromise (best of 24 and 35). For others, it’s bad (worst of both).

I’m excited about the 28 — best of both, half the cost, and I don’t have to change lenses. Like the pancake, but better IQ, 2 stops faster, and maybe IS.

(Yes, I know this post is obvious )

 

[Fischer]

All good primes. The 35mm being 1.4 wouldn't bother me but it would have to beat the EF version which let's face it is pretty perfect.

Then it's down to paying twice as much for the RF version than the EF version and an adaptor.

Have mine on an adapter as my very last Canon lens. After getting the 20-70 f/2.0 it does not get a lot of use, so contrary to what I originally expected when shifting to "R" I'll probably give it a pass. However, I'm sure Canon will make the new one shine.

 

[Dragon]

True. But that also applies to geometric corrections accomplished optically, with lens elements. That’s why the center is sharper than the corners.

I’d argue —and I’ve shown the images to back it up— that algorithmic correction of geometric distortion can be as effective as optical correction. Thus, the RF 14-35/4 at 14mm after the lens profile is applied is just as sharp in the corners as the optically-corrected EF 11-24/4 at 14mm.

Quite true and the previous poster is not entirely correct in his assumption that algorithmic correction is lossy. If the camera has a higher pixel density than the the resolution of the lens (true more often than not in the corners of most lenses), then the sampling is Nyquist and scaling is not intrinsically lossy. If the lens out-resolves the sensor, then scaling can create some aliasing issues in addition to some loss of resolution. Sampling theory makes a good argument for high pixel count cameras when algorithmic correction is used.

 

[Nemorino]

15mm macro!

The RF 15-30mm has a magnification of 0,5 in manual focus at 15mm!

@Nemorino has posted some wonderful shots with one int the mushroom gallery.

Thank You very much!!!

 

[Nemorino]

I wish for the best possible IQ from the lens itself first.

Big words!
Even if it is 1,5kg lens with 120mm diameter and a price of 5000$ ?
(just an example without a real lens in mind!)

 

[Quarkcharmed]

True. But that also applies to geometric corrections accomplished optically, with lens elements. That’s why the center is sharper than the corners.

I’d argue —and I’ve shown the images to back it up— that algorithmic correction of geometric distortion can be as effective as optical correction. Thus, the RF 14-35/4 at 14mm after the lens profile is applied is just as sharp in the corners as the optically-corrected EF 11-24/4 at 14mm.

I suspect a proper optical correction (or just quality glass to begin with) will be better. Note the software corrections will depend on the sensor resolution and will get worse with lower-res sensors

 

[SwissFrank]

Because you're losing image quality when doing corrections, or reveal imperfections that were there already.

First, the question isn't whether digital corrections can theoretically or even practically lose quality, but rather, why would it lose MORE image quality than doing the same corrections optically?

Second, have you ever seen an example, even one, where digital correction of distortion actually does "losing image quality when doing corrections, or reveal imperfections"? If you've seen even one example of this is it something you can share?

 

[SwissFrank]

That was me who showed those images to you.

BTW thanks for that! The image taught me something about the corner noise issue. I'm hoping you have time to teach me something about how software correction of distortion has bad effects too. I can't see how it would but again I'm happy to admit you were right about the vignetting correction increasing corner noise in more than a merely theoretical manner in some situations.

 

[SwissFrank]

I wish for the best possible IQ from the lens itself first. Then corrections.

Why does it matter? Why shouldn't we only care about the final result, and not care at all about what things may look like at mid-points in the processing?