RF 200-800mm vs RF 100-500mm vs RF 800mm on R7 and R5

[neuroanatomist]

Unfortunately, it seems that there is a quality discrepancy between the copies released on the global photographic market.

That's true for all lenses, not just this one. If you're familiar with The Digital Picture reviews by Bryan Carnathan, for the EF 24-70/2.8L II he ended up testing four copies because the first two he bought were optically flawed (in different ways).

Basically, quality assurance to the point where there are no bad copies sold would be exorbitantly expensive (for the manufacturer, but of course it would need to be passed through to the customer). It turns out to be more cost effective to aim for a low but non-zero failure rate then let affected customers return the defective items. If the warranty services on a product point to a design or component flaw, that's quietly fixed in subsequent batches (or if it's bad enough, there's a recall).

The other thing to keep in mind is that while some of us carefully test new lenses (guilty – I have ISO 12233-type charts that cost more than some L lenses, and I thoroughly test a lens when I buy it), most buyers probably won't notice if one corner is a little soft or the vignetting is not perfectly symmetrical.

 

[Frodo]

I found significant differences between two 16/2.8 and 50/1.8 lenses and selected the best ones. I had two decentred RF 24-105/4L lenses. This may also explain diferences between reviews.

 

[AlanF]

I found significant differences between two 16/2.8 and 50/1.8 lenses and selected the best ones. I had two decentred RF 24-105/4L lenses. This may also explain diferences between reviews.

I will never buy a lens without testing it first or being able to return it after testing. The alternative would to be to buy one of @neuroanatomist 's used ones.

 

[docsmith]

Another thought, but in my experience, refurbished lenses are more consistent than new lenses. In fact, most of my lens lineup was purchased refurbished from Canon USA. While I have returned several lenses that were "new", I do not recall returning any refurbished lenses.

The simple thought being, new lenses come off an assembly line where 1 in x are tested, but every refurbished lens is tested.

 

[Del Paso]

That's true for all lenses, not just this one. If you're familiar with The Digital Picture reviews by Bryan Carnathan, for the EF 24-70/2.8L II he ended up testing four copies because the first two he bought were optically flawed (in different ways).

Basically, quality assurance to the point where there are no bad copies sold would be exorbitantly expensive (for the manufacturer, but of course it would need to be passed through to the customer). It turns out to be more cost effective to aim for a low but non-zero failure rate then let affected customers return the defective items. If the warranty services on a product point to a design or component flaw, that's quietly fixed in subsequent batches (or if it's bad enough, there's a recall).

The other thing to keep in mind is that while some of us carefully test new lenses (guilty – I have ISO 12233-type charts that cost more than some L lenses, and I thoroughly test a lens when I buy it), most buyers probably won't notice if one corner is a little soft or the vignetting is not perfectly symmetrical.

Which (partly!) explains why Leica lenses are so expensive.
As far as I know, every single Leica M lens is factory tested before being sent to a dealership or customer.

 

[koenkooi]

Which (partly!) explains why Leica lenses are so expensive.
As far as I know, every single Leica M lens is factory tested before being sent to a dealership or customer.

Sigma claims to do the same, I guess it’s a matter of how stringent your QA targets are.

 

[EricN]

Sigma claims to do the same, I guess it’s a matter of how stringent your QA targets are.

I'm wondering what the minimum is to be able to say it was tested.

 

[AlanF]

I'm wondering what the minimum is to be able to say it was tested.

I seem to recall from some years back that Sigma said that they checked each lens was within spec at each stage of its manufacture, a continuous, automated process. Trouble is that what they would consider within spec at the end would include a range we finickety ones would consider soft, partly decentered etc.

 

[EricN]

I seem to recall from some years back that Sigma said that they checked each lens was within spec at each stage of its manufacture, a continuous, automated process. Trouble is that what they would consider within spec at the end would include a range we finickety ones would consider soft, partly decentered etc.

I suspected it would be like that. It's hard for me to consider it as actually a test more than marketing. Still, I don't deny Sigma has fans

 

[Del Paso]

Sigma claims to do the same, I guess it’s a matter of how stringent your QA targets are.

I just can't imagine Sigma testing each single lens element, as also to have such a stringent final check. I've bought many Leica M and R lenses, never got an "Oktober Fest" one.
Yes, serious testing is very expensive.

 

[koenkooi]

I just can't imagine Sigma testing each single lens element, as also to have such a stringent final check. I've bought many Leica M and R lenses, never got an "Oktober Fest" one.
Yes, serious testing is very expensive.

During one of the factory tour videos they showed a sigma camera built into a contraption and said they test every lens on it.
So like @AlanF says, it depends on what Sigma thinks are acceptable results.
In my previous job I was responsible for automated testing, but it was management that set the acceptance criteria. We had projects where every metric showed red for months on end, but the funding for it was only for doing the testing, not for caring about the results :/ So ‘every lens has been tested’ doesn’t mean that every lens passed the tests.

 

[Del Paso]

During one of the factory tour videos they showed a sigma camera built into a contraption and said they test every lens on it.
So like @AlanF says, it depends on what Sigma thinks are acceptable results.
In my previous job I was responsible for automated testing, but it was management that set the acceptance criteria. We had projects where every metric showed red for months on end, but the funding for it was only for doing the testing, not for caring about the results :/ So ‘every lens has been tested’ doesn’t mean that every lens passed the tests.

This reminds me of cosmetics "dermatologically and clinically tested".
Doesn't mean you can't, after the tests, peel off the human guinea-pigs' skin...

 

[Del Paso]

What I can imagine is Canon testing (seriously) every single "big white".

 

[Del Paso]

I suspected it would be like that. It's hard for me to consider it as actually a test more than marketing. Still, I don't deny Sigma has fans

They have some pretty good lenses. The main deterrent for me was AF on Canon DSLRs. And also the weight of the WA, compared to the tiny and lightweight Leica M WA.
Additionally, the RF zooms are optically excellent. It also seems that Canon's automated production process has had a very positive effect on sample variation.

 

[EricN]

I'm not sure if Canon's WAs will beat Sigma's in terms of weight, but I'm confident they will be equal to or exceed them in other aspects. I have considered Sigma's 14mm 1.8 and the 180mm 2.8 macro for EF, but I'm not in a hurry when I'm very likely to buy the Canon RF answers to those lenses as soon as we expect an announcement I'll be preparing myself with coffee.

Definitely the zooms (and primes) have improved drastically over EF. Although as you mentioned, they might be too perfect depending on what we want.

 

[Del Paso]

I would definitely wait for the EF 180 macro's successor, if equipped with OIS. It really helps, even for macros, though less than for longer distance shots.

 

[Del Paso]

I'm not sure if Canon's WAs will beat Sigma's in terms of weight, but I'm confident they will be equal to or exceed them in other aspects. I have considered Sigma's 14mm 1.8 and the 180mm 2.8 macro for EF, but I'm not in a hurry when I'm very likely to buy the Canon RF answers to those lenses as soon as we expect an announcement I'll be preparing myself with coffee.

Definitely the zooms (and primes) have improved drastically over EF. Although as you mentioned, they might be too perfect depending on what we want.

I'll have to contradict you. I still cannot imagine a zoom to be too perfect...
But fact is, especially the WA zooms have become a viable alternative to many primes, with added comfort. Telezooms are a different story, they are often indistinguishable from tele-primes. In fact, my Canon zooms are visibly better than my Leica R longer tele lenses, with the exception of the Apo-Telyts. And, even there, the difference is not that huge!
I really like the EF 70-200 F4 L II and the EF 100-400 L II. Their successors must be even better!

 

[scyrene]

Just FYI, this morning I returned my copy of the RF200-800 mm lens. My copy was soft (sometimes) and other times very good. I also found the huge zoom throw a bit of a pain in the neck. Just not the lens for me and I will be keeping my 100-500 mm L lens.

It is still a lot of lens for the price, but I decided to take my lightweight long-reach (800 mm) bird photography in a different direction; albeit at a more significant cost.

I would have taken it off your hands!

 

[EricN]

I'll have to contradict you. I still cannot imagine a zoom to be too perfect...
But fact is, especially the WA zooms have become a viable alternative to many primes, with added comfort. Telezooms are a different story, they are often indistinguishable from tele-primes. In fact, my Canon zooms are visibly better than my Leica R longer tele lenses, with the exception of the Apo-Telyts. And, even there, the difference is not that huge!
I really like the EF 70-200 F4 L II and the EF 100-400 L II. Their successors must be even better!

You're right, I was thinking of primes. Mainly the difference between EF 1.2 and RF 1.2 models. I should have written more carefully. I can't really say about tele primes because I've never seriously thought to buy one. I'm not skilled at even finding birds/wildlife let alone shooting them, but I'm getting better... From other's photos, even though the 200-800 is cheaper, it's looking good well worth the price.

 

[docsmith]

Getting back to this thread as I have been comparing my EF 100-400 II w/1.4TC vs a loaner RF 100-500 and the RF 200-800. This post is the 100-400 II vs the RF 100-500.

Previously, I knew the RF 100-500 was a bit better than the EF 100-400 II. Looking at the charts on TDP, I really did wonder how much better it was, specifically this comparison of 500 mm vs 560 mm. Already owning the 100-400 II, holding onto it was a bit of a no brainer...or so I thought.

But, listening and reading here, also having a friend that recently upgraded and singing the praises of the RF 100-500L, I decided to get a loaner form CPS.

While I wasn't able to line up species, I think the difference is obvious enough.

All taken on my R5, all of birds at the same branch at my feeder, within 15 min of each other. I measured today to confirm, the branch is 27 ft away from the window I shoot from.

EF 100-400 II w/EF1.4TC. @560 mm, 1/1250, f/10, ISO 4000 (from auto ISO, interestingly, the other lenses were ISO 3200 for the same scene...T-stop vs F-stop?)

Other than cropping, and unedited headshot of this finch:


IMO, ugly. There is detail there....ISO4000 is taking it's toll, but this is why I historically have used my EF500 II wTC at my bird feeder. I wish I could say this is a one off, but I had three different recent sittings with the 100-400 II /TC and this is pretty typical. Sure, I could have helped noise by going to f/8...that only helps so much.

Trying to clean up the image with DeNoise in LR and, I think bird tongue is funny:


So, you can get a good images. It is a great lens. But I find it interesting that 27 ft/8 m pushes it's abilities.

What a processed/full bird image might look like:

Look...for a lot of instances and a lot of people, not bad. I have walked parks/kayaked/traveled with this combination. But, I jumped on the RF 200-800 for a reason.

RF 100-500L @ 500 mm (1/1250, f/10, ISO3200)
Unprocessed headshot of a Junco...

IMO, unprocessed is still very nice, but the images clean up very well.

Processed headshot:


Processed full-body:

IMO...that's pretty.

RF 100-500 w/RF1.4TC (700 mm, 1/1250, f/10, ISO 3200)
unprocessed headshot:


Not even that above needs much work, so I'll switch birds. Processed headshot:


Processed full-body (I am loving the different hues coming out):


I do also want to note that the hit rate, while I did not calculate, was noticeably worse with the EF 100-400 II wTC. The RF 100-500 with and w/o the TC was fast and more consistent.

Overall, call me impressed by the RF 100-500L. Great lens. It definitely annoys me how you cannot fully zoom out with the TC in place, but optically, those images are up there with my EF 500 II w/TC which is one of my favorite lenses of all time.

The day after these photos were taken, I tried to compare the RF 100-500 w/TC and the RF 200-800 with and without TC. Hopefully I post that tomorrow.