Canon RF 100-300mm f/2.8L IS USM next from Canon

[Fischer]

A question no-one's asked: is the 100-500 so good that we don't need this? With today's low-noise sensors, IS and IBIS, and the fact we don't need f/2.8 to activate all the AF sensors, does anyone actually NEEEEEED this? Can anyone point to a published 300/2.8 shot that simply would have been unsalable with the DOF or more noise of f/5.6? I've argued for a couple years now that the real trinity is the f/4 zooms and 100-500, no longer the f/2.8 zooms.

My challenge with the 100-500mm is DOF. For pure optical performance its amazing. However, I need the very shallow DOF of a 300mm f/2.8 lens. To me, its a waste of money from 100-299mm to have a glorified zoom instead of just a 300mm prime. And; we still have to see side-by-side comparisons, look at the bokeh (especially when not shot wide open) and I have still not seen anyone confirm it is actually 300mm at the long end and not "just" a 260/270/280mm. The 100-500mm for example does not go to 500mm but only around 480mm (from memory).

 

[neuroanatomist]

My challenge with the 100-500mm is DOF. For pure optical performance its amazing. However, I need the very shallow DOF of a 300mm f/2.8 lens. To me, its a waste of money from 100-299mm to have a glorified zoom instead of just a 300mm prime. And; we still have to see side-by-side comparisons, look at the bokeh (especially when not shot wide open) and I have still not seen anyone confirm it is actually 300mm at the long end and not "just" a 260/270/280mm. The 100-500mm for example does not go to 500mm but only around 480mm (from memory).

The focal lengths and apertures printed on the barrel are always rounded (up and down, respectively). The 100-500 is really 485mm at the long end, the 300/2.8 (EF) is 292mm and the RF zoom is probably similar.

It’s also important to keep in mind when people post comparisons showing how much shorter a focal length is than specified, they often ignore focus breathing. Focal length is specified at infinity focus, and with close subjects the apparent FL is shorter (by how much depends on the lens).

 

[SwissFrank]

The 100-500mm for example does not go to 500mm but only around 480mm (from memory).

Every lens I've seen exact figures for is a few percent off. I don't know if it is company policy or some industry standard but it is VERY purposeful. Even in patent applications, Canon might list three embodiments of a patented design (in other words three lenses designed according to the design point they're patenting) and each of the three will be off the nominal figure by some very exact amount. Like 24mm f/2.8 lenses will all be 25.92mm and f/2.94 or something like that.

In a way this is good news for you because if you like your current 300/2.8, the new 100-300/2.8 is probably the same focal length--288mm or something...

 

[Kit.]

Noise is one thing but there is just no way to make up for the background separation difference between those two apertures.

Technically, DPRAW could be such a way.

 

[Fischer]

Every lens I've seen exact figures for is a few percent off. I don't know if it is company policy or some industry standard but it is VERY purposeful. Even in patent applications, Canon might list three embodiments of a patented design (in other words three lenses designed according to the design point they're patenting) and each of the three will be off the nominal figure by some very exact amount. Like 24mm f/2.8 lenses will all be 25.92mm and f/2.94 or something like that.

In a way this is good news for you because if you like your current 300/2.8, the new 100-300/2.8 is probably the same focal length--288mm or something...

I have no issue with how lenses are marketed, but I would be very disappointed to see the focal length "shrink" from the prime. Time will tell what we actually can expect.

 

[SwissFrank]

I have no issue with how lenses are marketed, but I would be very disappointed to see the focal length "shrink" from the prime.

To be clear, I assume both the EF and RF 300mm primes, and the new 100-300mm RF zoom, will have the exact same focal length.

In the new "patent" lens designs shown on this site today, for instance, the wide end of ALL THREE of the zooms is nominally 15mm, but quoted to two decimal places as 15.45mm. That's exactly 3% "less wide" than advertised.

The f/stops are nominally f/2.8 and f/4 and f/5.6, but when quoted to two decimal places are actually f/2.88, f/4.12, and f/5.8: again, 3%, 3%, and 3% (within the limits of rounding) worse than the nominal number. (One figure is 2.8 not 2.88, but arguably it's a typo. I see other typos in this release as well.)

The tele ends are nominally 70mm but in fact 68.04mm (2.88% less long than nominal, call it 3%), nominally 70mm but in fact 68.94mm (3.03% off) or nominally 85mm but in fact 83.77 (this time only 1.5% off??).

OK, so there are 12 numbers, 10 of which are almost exactly 3% easier than the nominal figure. The remaining two are in fact the nominal figure (0% off, but could be a typo?) and 1.5% off (several possible explanations but cannot guess which).

I'm guessing that either Canon's internal rules, or an industry body or govt regulation, allows the actual spec to be up to 3% away from the nominal spec, at least for this range of focal lengths and apertures, and they design to that tolerance. And if they do now, they probably have for a long time. I've noticed this pattern quite a few years ago and it was the same then too.

 

[scyrene]

Focal length is specified at infinity focus, and with close subjects the apparent FL is shorter (by how much depends on the lens).

I never knew that! What about the MP-E? (Although I've always felt it even having a specified focal length was superfluous).

 

[koenkooi]

I never knew that! What about the MP-E? (Although I've always felt it even having a specified focal length was superfluous).

No idea about the MP-E, but the EF 100mm non-L and EF100L become 67-ish mm lenses at MFD and the RF100-500L is (from memory) 270-ish mm when zoomed to 500mm at MFD. I haven't looked at what my RF100L does at 1.4x, but Rudy Winston mentioned that 'focus breathing has improved compared to the EF100L'.

Looking at the Canon website, the MP-E has an 18 degree angle of view specified, regardless of magnification. So it might be keeping the same focal length along its "zoom".

 

[Del Paso]

No idea about the MP-E, but the EF 100mm non-L and EF100L become 67-ish mm lenses at MFD and the RF100-500L is (from memory) 270-ish mm when zoomed to 500mm at MFD. I haven't looked at what my RF100L does at 1.4x, but Rudy Winston mentioned that 'focus breathing has improved compared to the EF100L'.

Looking at the Canon website, the MP-E has an 18 degree angle of view specified, regardless of magnification. So it might be keeping the same focal length along its "zoom".

Maybe (possibly?) I'm wrong, but could focus breathing be caused by internal focusing?

 

[scyrene]

No idea about the MP-E, but the EF 100mm non-L and EF100L become 67-ish mm lenses at MFD and the RF100-500L is (from memory) 270-ish mm when zoomed to 500mm at MFD. I haven't looked at what my RF100L does at 1.4x, but Rudy Winston mentioned that 'focus breathing has improved compared to the EF100L'.

Looking at the Canon website, the MP-E has an 18 degree angle of view specified, regardless of magnification. So it might be keeping the same focal length along its "zoom".

Thanks! I only asked because it has no infinity focus, so they must have used a different method to calculate the focal length.

 

[neuroanatomist]

Thanks! I only asked because it has no infinity focus, so they must have used a different method to calculate the focal length.

The optical design has an infinity focus, and that is the basis for the 65mm focal length. The lens itself is analogous to a reversed normal lens on a bellows, and as you say it cannot be focused to infinity (or focused at all, in fact).

 

[scyrene]

The optical design has an infinity focus, and that is the basis for the 65mm focal length. The lens itself is analogous to a reversed normal lens on a bellows, and as you say it cannot be focused to infinity (or focused at all, in fact).

Thanks for the explanation!

 

[JTM-77]

How many times does it need to be said that the hood will be included, as it is with all L-series lenses? Canon sells replacement hoods for L-series lenses, because sh!t happens. If I ever need one for my 600/4 II, it will cost $700.

Not that it's your business, but my 'gear fund' bank account balance is more than sufficient to cover the $9.5K cost if accurate (even after the balance dips by $2K from my unfulfilled preorders for the EL-5 and R8). I'm fortunate that money is not much of a limiting factor for my purchases, as is evident from my gear list (which I'll need to update at some point, since I've sold the 1D X and EOS R, and bought an RF 100-400, RF 24/1.8 and a Vixia HF G60 camcorder since the last update; I also just bought an underwater photo/lighting setup for diving so perhaps I'll add that).

ok genius if the hood is included why did they mention the price of it then?

 

[neuroanatomist]

ok genius if the hood is included why did they mention the price of it then?

Because sometimes people lose or damage a lens hood. Canon also sells the E-112 Lens Cap for $25. Do you think that's because they don't include a front lens cap with the lens? It's great that we have genius brilliant smart members like you contributing here!

At least you did your research before posting something that made you look foolish.



Oh, wait…

 

[AlanF]

My challenge with the 100-500mm is DOF. For pure optical performance its amazing. However, I need the very shallow DOF of a 300mm f/2.8 lens. To me, its a waste of money from 100-299mm to have a glorified zoom instead of just a 300mm prime. And; we still have to see side-by-side comparisons, look at the bokeh (especially when not shot wide open) and I have still not seen anyone confirm it is actually 300mm at the long end and not "just" a 260/270/280mm. The 100-500mm for example does not go to 500mm but only around 480mm (from memory).

The lenses have focus breathing since they focus in part by changing their focal length rather than move the whole lens away from the the sensor as they did in the good old days with bellows. The stated focal length is for infinity focus and it gets shorter for nearer objects. I actually measured the focal length of the RF 100-500mm from the size of the image of the moon on the sensor of the R5. The diameter of 3,474.8 km when it was 400,403.56 km away gave a diameter of 990 pixels. If the pixels are 4.39µ as usually stated (TDP etc), this gives a focal length of 500.8mm. However, they calculate the pixel size based assuming the image occupies the whole of the 24x36mm sensor. If I use instead Canon's stated value of 47.1 Mpx on the sensor, instead of the 44.76 Mpx that are actually used, the pixel size is 4.28µ and the calculated focal length 488mm. I'd be grateful if someone could tell me the true pixel size.

 

[Fischer]

The lenses have focus breathing since they focus in part by changing their focal length rather than move the whole lens away from the the sensor as they did in the good old days with bellows. The stated focal length is for infinity focus and it gets shorter for nearer objects. I actually measured the focal length of the RF 100-500mm from the size of the image of the moon on the sensor of the R5. The diameter of 3,474.8 km when it was 400,403.56 km away gave a diameter of 990 pixels. If the pixels are 4.39µ as usually stated (TDP etc), this gives a focal length of 500.8mm. However, they calculate the pixel size based assuming the image occupies the whole of the 24x36mm sensor. If I use instead Canon's stated value of 47.1 Mpx on the sensor, instead of the 44.76 Mpx that are actually used, the pixel size is 4.28µ and the calculated focal length 488mm. I'd be grateful if someone could tell me the true pixel size.

Straight 10 for your deep-dive into the physics of this.