TDP Image Quality posted - 24-105L II

[ahsanford]

At long last, here it is -- the new 24-105L II.

5DS R + 24-105L II vs. 24-105L I @ 24mm f/4:

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=1072&Camera=979&Sample=0&FLI=0&API=0&LensComp=355&CameraComp=0&FLIComp=0&APIComp=0

Wow. The I version looks considerably better than the II in the center and mid-frame areas!


5DS R + 24-105L II vs. 24-105L I @ 105mm f/4:

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=1072&Camera=979&Sample=0&FLI=5&API=0&LensComp=355&CameraComp=979&SampleComp=0&FLIComp=5&APIComp=0

Looks like a fair amount of purple fringing at 105mm for the II version and not the I version.

Color me disappointed. Most Mk II versions of L lenses don't get out the door without a clear bump in performance. I know those are just two simple comparisons (and he's only got one copy of the new lens), but they should have been softball pitches to swing at for Canon.

Wow.

- A

 

[ahsanford]

24-105L II vs. the non-L 24-105 STM @ 24mm f/4

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=1072&Camera=979&Sample=0&FLI=0&API=0&LensComp=961&CameraComp=979&SampleComp=0&FLIComp=0&APIComp=1

The non-L wins in that comparison, IMHO, esp the midframe area.

24-105L II vs. the non-L 24-105 STM @ 105mm f/5.6 (aperture stopped down on the L to match the non-L):

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=1072&Camera=979&Sample=0&FLI=5&API=2&LensComp=961&CameraComp=979&SampleComp=0&FLIComp=5&APIComp=1

That's a poor result. The 24-105L II -- stopped down one stop -- is considerably softer / more fringing than the non-L shot wide open. Wow.

- A

 

[Random Orbits]

It IS designed as a kit lens. The 24-70 f/4 IS came out at more than 1500 USD and that was a ridiculous price. The 24-105 f/4 II has a lower initial price, and is probably designed to minimize cost so that the kit prices can be slashed in the future.

 

[meywd]

It IS designed as a kit lens. The 24-70 f/4 IS came out at more than 1500 USD and that was a ridiculous price. The 24-105 f/4 II has a lower initial price, and is probably designed to minimize cost so that the kit prices can be slashed in the future.

That's ok but why release a version two if its worse?

 

[ahsanford]

It IS designed as a kit lens. The 24-70 f/4 IS came out at more than 1500 USD and that was a ridiculous price. The 24-105 f/4 II has a lower initial price, and is probably designed to minimize cost so that the kit prices can be slashed in the future.

That's ok but why release a version two if its worse?

+1. I'm not condemning this lens from a few IQ samples from just one lens, but I'm eager to read Mr. Carnathan's writeup in light of this result. Knowing him, he might pursue a dialogue with Canon professionals who have sorted out similar nutty findings in the past. Who knows -- he might have gotten a bad copy.

- A

 

[ahsanford]

Hang on -- if you run my original comparison on the 7D2, it's a lot rosier for the Mk II lens, especially on the long end:

24mm f/4:

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=1072&Camera=963&Sample=0&FLI=0&API=0&LensComp=355&CameraComp=963&SampleComp=0&FLIComp=0&APIComp=0

105mm f/4 (THIS is what I was expecting to see):

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=1072&Camera=963&Sample=0&FLI=5&API=0&LensComp=355&CameraComp=963&SampleComp=0&FLIComp=4&APIComp=0


Would that imply we've got an AFMA issue going on?

Call me crazy, but I sense a retraction / retest / story from Carnathan to be forthcoming on the FF result.

- A

 

[ahsanford]

Holy cow, from Bryan Carnathan himself just now, when I asked him if he got a bad copy:

"That is a question that I have to answer. While I was not expecting a significant improvement over the version I lens (based on the MTF charts), I was hoping for a little better than this. My outdoor test images do not paint a better picture and a friend was underwhelmed with their copy, but this lens has some other nice advantages that I'll share in the review. Also keep in mind that CA correction helps in the periphery."


Oof.

For those that follow TDP, Bryan saying "I was hoping for a little better than this" simply never happens. And it would appear he's also seeing this result in his standard outside of his house aperture / lens mouseovers of crops of bushes and trees.

Stay tuned...

- A

 

[Loswr]

Would that imply we've got an AFMA issue going on?

Call me crazy, but I sense a retraction / retest / story from Carnathan to be forthcoming on the FF result.

From TDP's desciption of their ISO 12233 chart testing:

[quote author=Bryan Carnathan / TDP]
Tests are conducted using computer-aided Live View manual focusing (with center-point-only autofocus additionally tested). The best of the many re-focused shots (typically at least 10 sets and often 15-20) are used for the results for each camera/lens/focal-length/aperture combination.
[/quote]

 

[ahsanford]

Would that imply we've got an AFMA issue going on?

Call me crazy, but I sense a retraction / retest / story from Carnathan to be forthcoming on the FF result.

From TDP's desciption of their ISO 12233 chart testing:

[quote author=Bryan Carnathan / TDP]
Tests are conducted using computer-aided Live View manual focusing (with center-point-only autofocus additionally tested). The best of the many re-focused shots (typically at least 10 sets and often 15-20) are used for the results for each camera/lens/focal-length/aperture combination.

[/quote]

I always forget -- thanks, Neuro.

- A

 

[YuengLinger]

Maybe time to slow down. Not at the CR end--we respond like lightning and damn the torpedoes, but at the review end. Don't post partial reviews or half-baked first impressions. Be methodical and certain.

 

[Random Orbits]

Maybe time to slow down. Not at the CR end--we respond like lightning and damn the torpedoes, but at the review end. Don't post partial reviews or half-baked first impressions. Be methodical and certain.

His test method is sound. What is available are IQ crops. He is planning on testing a second copy of the lens, which will take time. There are lenses which have multiple sets of results (including the 24-70 f/2.8 II, when first tested on the 1Ds III). The crop sets are available in advance of the actual writeup -- that is typical. Hopefully, the second copy will have improved IQ at least as good as version 1, but I think that expecting version 2 to be much better is out of the question (the MTFs don't imply that and Bryan's initial testing doesn't either).

It is telling (from ahsanford's post) that both Bryan's and his friend's copy exhibited similar deficiencies...

 

[hendrik-sg]

maybe it's time where economical pressure is rising. So maybe the new lens is cheaper to produce and some goodies as better IS may somehow compensate.

What was interestng for me is, that the sigma art lens is not better. this shows how good the original version is and how difficult such devices are to design.

At the moment there is much hype for the new version, most people blindly believe the new version to be "much better" and so is the price. The price will come down as for all kit lenses, i just do not know why we have the 24-70 F4 and the 24-105 STM lenses for? Both of them i never saw in the wind, where the old 24-105 is everywhere, not only on my camera

I would really like to pay for a 24-70 2.8 IS, which would improve the FF camera system by 1 stop of light gattering ability, i like to be out at night without tripod (which may be called unprofessional, who cares). Maybe i need a 35 2.0 IS

 

[ahsanford]

At the moment there is much hype for the new version, most people blindly believe the new version to be "much better" and so is the price.

I'll give you a few reasons why Canon folks were fired up for this Mk II version:

35 f/1.4L II
16-35 f/2.8L III (aside from some staggering vignetting)
100-400 f/4.5-5.6L IS II
24-70 f/2.8L II

What do all of those have in common? They were sequels to successful lenses, and every one of those lenses above absolutely mopped the floor with its predecessor -- Canon has not made a bad sequel to an L lens in recent memory. I'm hard pressed to think of even one in the last 10 years.

So... yeah. Hopes were high for this one.

- A

 

[geekpower]

i really wanted this lens to be awesome. i hate to say it, but when it was announced i did wonder what would become of the 24-70 f4/L if this was even close in sharpness but had better reach. sounds like perhaps that lens is not in danger of obsolescence after all. will reserve judgement until a few full reviews are out, but so far, not so good.

 

[Deleted member 7828]

I was REALLY excited for this lens -- and while I'll still wait for more reviews to come in, it's sinking in that I may end up keeping my version I longer than anticipated...(it seems to be one of the "good copies" so to speak, besting my 24-70 f4 in the 50-70mm range)

 

[ahsanford]

i really wanted this lens to be awesome. i hate to say it, but when it was announced i did wonder what would become of the 24-70 f4/L if this was even close in sharpness but had better reach. sounds like perhaps that lens is not in danger of obsolescence after all. will reserve judgement until a few full reviews are out, but so far, not so good.

The killer app of the 24-105 is clearly 71-105mm.

The killer app with the 24-70 f/4L IS is the macro (and the size/weight to some extent).

My 24-70 f/4L IS will not be replaced until another L standard zoom with a 0.7x max mag is released. In that light, it's place in the cabinet is all but assured for the next 10 years.

- A

 

[TeT]

did Brian do these? He may need a different copy to test or I see know reason to rush out and upgrade..

 

[AJ]

Goes to show that "L" isn't everything.

 

[TeT]

Goes to show that "L" isn't everything.

Not so, L really is everything... This is surprising. I am interested to hear what brian says in his review.

 

[unfocused]

I'm really surprised, but reserving judgment. This would be the first time in years (maybe ever) that a II version is not optically stellar. Even the non-L lenses like the 24, 28 and 35 IS versions are optically much superior to their predecessors.

I do not believe Canon would have updated this lens without giving it much improved optics. Time will tell, I guess.