Canon RF 10-20mm F4L IS STM Preview/Review Summary

[Canon Rumors Guy]

Yesterday when the news broke about the Canon RF 10-20mm, I immediately did our comparison of the Canon RF 10-20mm and the Canon EF 11-24mm. Let’s just say the preliminary look of this lens makes it look like a home run for Canon if you shoot super-ultra-wide frequently. As I mentioned in that article, if

See full article...

 

[amorse]

Honestly, it looks like a great lens and ticks a lot of boxes for me - I'm sure this is going to be used to create some excellent images. I just wish we had a faster option wider than 15mm in RF.

 

[saudkazi]

I pre-ordered it the minute I found out about this lens. I was looking for an ultra-wide angle lens at least 11mm wide. When I saw this was 10mm, I went wild. Looking forward to getting it.

 

[Del Paso]

"Interesting" comments on DPR:
- Should have been an f2,8
-Too expensive
-Why not a 10mm prime which could be cropped to 20mm while offering as good or a better optical quality instead of a zoom?
Highly qualified...
DPR whiners are so inventive!

 

[EricN]

"Interesting" comments on DPR:
- Should have been an f2,8
-Too expensive
-Why not a 10mm prime which could be cropped to 20mm while offering as good or a better optical quality instead of a zoom?
Highly qualified...
DPR whiners are so inventive!

I found that comment about cropping! Is there another dimension these people come from?

 

[Del Paso]

I found that comment about cropping! Is there another dimension these people come from?

I guess they come from a flat earth.

 

[Smiling Eddie]

A bit disappointed to see STM rather than USM; I love instant focus. Its a feature of every L-series/pro lens I've ever used since my pre-USM 100-300.
Not extending up to at least 24mm is also a bum note for me. And Canon has still priced it as a pro-level lens. What have they done to STM to justify that?
I was so looking forward to an RF replacement for my 11-24L USM but I'm not sure this is it.

 

[danfaz]

A bit disappointed to see STM rather than USM; I love instant focus. Its a feature of every L-series/pro lens I've ever used since my pre-USM 100-300.

I don't intend to get this lens, so not defending it, but Gordon Laing said in his initial review that this lens focuses faster and smoother than the 11-24.
Also, not all L lenses focus instantly, think RF 50 and 85mm 1.2. The EF 85mm 1.2 was generally considered extremely slow to focus.

 

[Smiling Eddie]

I don't intend to get this lens, so not defending it, but Gordon Laing said in his initial review that this lens focuses faster and smoother than the 11-24.
Also, not all L lenses focus instantly, think RF 50 and 85mm 1.2. The EF 85mm 1.2 was generally considered extremely slow to focus.

The EF 85 1.2 is a shocker, completely unsuited to action such as candid shots at weddings, in my view. And I have to admit the focus speed of my 11-24 hasn't proved to be a frustration generally but if STM was a step up from USM, surely Canon would have been using it with L-series lenses more widely. It might just be that I have to overcome a prejudice that Canon has, possibly unwittingly, fostered.

 

[neuroanatomist]

It might just be that I have to overcome a prejudice that Canon has, possibly unwittingly, fostered.

STM is cheaper to produce, so it’s usually used in consumer lenses. No doubt it’s fast enough to move a light focusing group the short distances needed with such a short focal length, but I am also disappointed that Canon opted for STM instead of Nano USM.

It is the lead screw type of STM, which is quieter and faster than the gear-driven type (the latter is smaller and typically used in consumer lenses).

 

[YuengLinger]

STM is cheaper to produce, so it’s usually used in consumer lenses. No doubt it’s fast enough to move a light focusing group the short distances needed with such a short focal length, but I am also disappointed that Canon opted for STM instead of Nano USM.

It is the lead screw type of STM, which is quieter and faster than the gear-driven type (the latter is smaller and typically used in consumer lenses).

I have absolutely no idea which is more durable, STM or USM, but on the Canon Europe website is an article describing different AF drive systems. What seems most relevant is this paragraph:

"When compact size is paramount, Canon uses gear type STM technology. This uses helical gears to drive the focus without taking up much space. Larger lenses use a lead-screw type STM system. This is bigger than gear type STM units but it's faster and quieter."

The initial marketing spin seems to emphasize how amazing the small size of the 10-20mm is compared to the 11-24mm. I wonder how much of the size reduction is due to the motor. How much might be due to giving up a few millimeters of focal length at the long end?

And has Canon developed a new and improved STM to be worth of L-series prestige?

Or could it be, as you've suggested, that the UWA and relatively short distance elements need to be moved means a USM adds bulk but no benefit?

I've been conditioned by Canon to associate USM with L-series lenses, and for nearly 20 years I've had no AF problems with lenses. But that's as far as my knowledge of AF motors goes.

If Canon used STM primarily to save money, it would be a startling tact for them after how aggressive they've been with pricing RF lenses. I wish they'd have been a little more forthcoming with an explanation at time of release, but maybe they believe it's the right marketing move to just not make a big deal of it.

Here's the link to the Canon Europe article:

Canon Lens Focusing Motor Technology

Discover USM and STM lens technology. Learn about the history of Ultra Sonic Motor and how the technologies enable you to achieve sharp and in-focus results.

www.canon-europe.com

 

[neuroanatomist]

I have absolutely no idea which is more durable, STM or USM, but on the Canon Europe website is an article describing different AF drive systems. What seems most relevant is this paragraph:

"When compact size is paramount, Canon uses gear type STM technology. This uses helical gears to drive the focus without taking up much space. Larger lenses use a lead-screw type STM system. This is bigger than gear type STM units but it's faster and quieter."

The initial marketing spin seems to emphasize how amazing the small size of the 10-20mm is compared to the 11-24mm. I wonder how much of the size reduction is due to the motor. How much might be due to giving up a few millimeters of focal length at the long end?

And has Canon developed a new and improved STM to be worth of L-series prestige?

Or could it be, as you've suggested, that the UWA and relatively short distance elements need to be moved means a USM adds bulk but no benefit?

I've been conditioned by Canon to associate USM with L-series lenses, and for nearly 20 years I've had no AF problems with lenses. But that's as far as my knowledge of AF motors goes.

If Canon used STM primarily to save money, it would be a startling tact for them after how aggressive they've been with pricing RF lenses. I wish they'd have been a little more forthcoming with an explanation at time of release, but maybe they believe it's the right marketing move to just not make a big deal of it.

Here's the link to the Canon Europe article:

Canon Lens Focusing Motor Technology

Discover USM and STM lens technology. Learn about the history of Ultra Sonic Motor and how the technologies enable you to achieve sharp and in-focus results.

www.canon-europe.com

The RF 10-20/4 used a lead screw-type STM motor, so if your argument is based on use of gear-type for smaller size (seems to be, based on the sentence you highlighted), that’s wrong.

 

[YuengLinger]

The RF 10-20/4 used a lead screw-type STM motor, so if your argument is based on use of gear-type for smaller size (seems to be, based on the sentence you highlighted), that’s wrong.

Yes, I was careless underlining that sentence without adding that I thought, from looking at the photos of the various motors, that lead-screw was larger than standard STM but smaller than Nano USM. This might not be the case.

Beyond speculation, we can't know why Canon chose STM, or how the factors of bulk, engineering efficiency, and cost savings influenced the decision.

Early adopters are taking a leap of faith (and succumbing to GAS!).

Does lensrentals still do tear downs?

 

[Del Paso]

Yes, I was careless underlining that sentence without adding that I thought, from looking at the photos of the various motors, that lead-screw was larger than standard STM but smaller than Nano USM. This might not be the case.

Beyond speculation, we can't know why Canon chose STM, or how the factors, bulk, engineering efficiency, and cost savings influenced the decision.

Early adopters are taking a leap of faith (and succumbing to GAS!).

Does lensrentals still do tear downs?

I asked Lensrentals a few weeks ago. If I remember well, they explained the missing tear-downs with having moved or transformed their existant building. But they also added that they hope to resume this activity as soon as possible!

 

[EricN]

I asked Lensrentals a few weeks ago. If I remember well, they explained the missing tear-downs with having moved or transformed their existant building. But they also added that they hope to resume this activity as soon as possible!

Great! I love seeing those articles!

 

[neuroanatomist]

Yes, I was careless underlining that sentence without adding that I thought, from looking at the photos of the various motors, that lead-screw was larger than standard STM but smaller than Nano USM. This might not be the case.

Beyond speculation, we can't know why Canon chose STM, or how the factors, bulk, engineering efficiency, and cost savings influenced the decision.

Early adopters are taking a leap of faith (and succumbing to GAS!).

Does lensrentals still do tear downs?

Nano USM and lead-screw STM actually use a similar mechanism to hold the focusing group and allow it to move. In both cases, the focusing group is mounted on a pair of guide bars. Nano USM uses a piezoelectric linear motor to slide the focusing group back and forth on the guide bars, lead-screw STM uses rotating motor to turn a screw that moves the focusing element back and forth on the guide bars.

Stepper motors are cheaper than piezoelectric motors, so STM is cheaper than Nano USM. Nano USM is faster, especially when the focusing group has to move a long distance. That, for example, is why the RF 100-400 and RF 24-240 use Nano USM even though they're consumer lenses. When STM is used in lenses with longer distances to move anyway, it results in 'slow' AF such as noted for a lens like the RF 85/2. Both Nano USM and STM work with relatively small/light focusing groups. When more torque is needed to move a larger focusing group, Ring USM is used. Even then, if the focusing group is very large/heavy (such as the EF 85/1.2L II where the massive 'focusing group' comprises 7 of the lens' 8 elements), focusing with Ring USM can be 'slow'.

In some cases, Canon has split the focusing elements into two moving groups, each of which has its own Nano USM motor. That's 'Dual Nano USM' and is found in lenses like the RF 70-200/2.8 and RF 100-300/2.8 (both of which I know from experience focus lightning fast). I suspect if those lenses had a single focusing group, they would need Ring USM and they'd be larger and heaver than they are.

The RF 10-20/4 is such a wide lens that the physical distance a focusing group needs to move to go from MFD to infinity is very short. Over such a short distance, a lead-screw STM should be just as fast as Nano USM, so Canon's first use of an STM motor in an L-series lens probably has effectively no practical consequence in terms of AF performance. While I am personally disappointed and I'm sure tech-oriented forums and YouTubers will complain about it, most buyers probably won't care. Actual reviews of the lens will say it focuses fast because it will. Buyers, for the most part, won't care.

Even though I'm disappointed in the use of STM instead of Nano USM personally, if I didn't already have the EF 11-24/4L and was looking for a UWA zoom, the use of an STM motor in this lens would not make me hesitate to buy it.

 

[Del Paso]

Great! I love seeing those articles!

They are what I've been eagerly waiting for, week after week!
I bitterly miss them, they were both entertaining and educational.

 

[YuengLinger]

Waiting for the lens, naturally I'm rewatching the first wave of marketing videos. This time around I did catch Rudy Winston, at 2:07, attribute at least some of the RF 10-20mm's compactness to the STM AF drive. He also describes STM as "proven."

I've already staked out half a dozen locations to bring the lens, but I am having second thoughts about the Kodak gels. Yes, my daughter could cut them for me, which would solve my problem with anything arts-and-crafty, but then I'd have to fiddle with them on location--unless I planned to just leave them in until coming back home.

For now I'm waiting for some other filter solution, while also considering just keeping my adequate 15-35mm for ND filter convenience. Always trade-offs with photography, and very often expensive!

 

[Del Paso]

Nano USM and lead-screw STM actually use a similar mechanism to hold the focusing group and allow it to move. In both cases, the focusing group is mounted on a pair of guide bars. Nano USM uses a piezoelectric linear motor to slide the focusing group back and forth on the guide bars, lead-screw STM uses rotating motor to turn a screw that moves the focusing element back and forth on the guide bars.

Stepper motors are cheaper than piezoelectric motors, so STM is cheaper than Nano USM. Nano USM is faster, especially when the focusing group has to move a long distance. That, for example, is why the RF 100-400 and RF 24-240 use Nano USM even though they're consumer lenses. When STM is used in lenses with longer distances to move anyway, it results in 'slow' AF such as noted for a lens like the RF 85/2. Both Nano USM and STM work with relatively small/light focusing groups. When more torque is needed to move a larger focusing group, Ring USM is used. Even then, if the focusing group is very large/heavy (such as the EF 85/1.2L II where the massive 'focusing group' comprises 7 of the lens' 8 elements), focusing with Ring USM can be 'slow'.

In some cases, Canon has split the focusing elements into two moving groups, each of which has its own Nano USM motor. That's 'Dual Nano USM' and is found in lenses like the RF 70-200/2.8 and RF 100-300/2.8 (both of which I know from experience focus lightning fast). I suspect if those lenses had a single focusing group, they would need Ring USM and they'd be larger and heaver than they are.

The RF 10-20/4 is such a wide lens that the physical distance a focusing group needs to move to go from MFD to infinity is very short. Over such a short distance, a lead-screw STM should be just as fast as Nano USM, so Canon's first use of an STM motor in an L-series lens probably has effectively no practical consequence in terms of AF performance. While I am personally disappointed and I'm sure tech-oriented forums and YouTubers will complain about it, most buyers probably won't care. Actual reviews of the lens will say it focuses fast because it will. Buyers, for the most part, won't care.

Even though I'm disappointed in the use of STM instead of Nano USM personally, if I didn't already have the EF 11-24/4L and was looking for a UWA zoom, the use of an STM motor in this lens would not make me hesitate to buy it.

I guess that if Canon are implementing an STM AF in an L lens, they sure be knowing what they do. I wouldn't even pay attention to it, focusing a 10-20 is not that critical, apart from close distances. Since I'm using such focal lenghts for landscapes, most often they are set on infinity.

 

[neuroanatomist]

For now I'm waiting for some other filter solution, while also considering just keeping my adequate 15-35mm for ND filter convenience. Always trade-offs with photography, and very often expensive!

Here's a thought... For their Wonderpana XL system (186mm filters, which is what you'll need for the RF 10-20), Fotodiox has a holder for the Sigma 12-24/2.8 Art. That lens has a specified diameter of just 1.3mm larger than the RF 10-20/4. Looking at the Sigma lens, the focus ring has a slightly larger diameter than the hood, and it's the hood that the Wonderpana filter holder slips over. The existing Sigma 12-24/2.8 holder may actually work just fine with the RF 10-20mm. Perhaps best not to spend the $230 to test it yourself (of course, you could buy and return), but maybe worth a call/email to Fotodiox to inquire.