08-25-2015, 11:00 PM
Honestly ... not that hot
next PZ lens test report: Sigma 24mm f/1.4 HSM DG "Art" ... on APS-C
08-25-2015, 11:00 PM
Honestly ... not that hot
08-26-2015, 12:05 AM
Klaus, I'm sorry: Saying "not that hot" while the lens is performing better or equally to it's Nikkor counterpart at double the price leaves me head shaking <_< Any idea, why one want a FF fast prime on a half size sensor?
Also, this soft corners blurb I read very often. Apparently mine must be from a different planet - when focused properly with live view, it's as good in it's corners as the Nikkor was in center, just with less CA (that goes for the Sigma). In any case, there's no optical reason to prefer the genuine 24/1.4 G against that Sigma. Can't tell for the Canon side, but the 24/1.4 L is also a little bit dated.
You didn't provide a single test shot wide open focused in the corner - so basically, what you're resuming are the chart results, including field curvature? Btw. you shot a couple of times at f/10.
Another head shake: Who in this world buys a 24/1.4 to use it with f/10? At your FF samples:
4 pictures @ f/10,
2 @ f/9,
1 @ f/5.6 and
2 @ f/1.4 where no corner was involved to focus - that's a bit a poor way to prove "corner weakness".
What you should consider is, with high MP everything beyond f/8 will suffer in sharpness. Therefore it's good to have a lens which is very sharp already at f/2.8.
But we can wait for the rumored "new reference" Otus 24/1.4 (which I believe will be 25/2 or something) and see what this beast is about to do at maybe 6Ã— the price. Fast lenses are built to be used wide open - you're abusing them, didn't rate it or compare to the 24/1.4 L II
Honestly: Your test is "not hot" as well
08-26-2015, 12:51 AM
The 24/1.4A suffers a much worse field curvature situation than the 24L II. Or rather, the astigmatism of the 24L II is much better tamed.
24L II has 3rd and 5th order curvatures balancing each other, 24A has an exquisite sagittal plane and either a badly curved tangential one, or much worse lateral color.
Corner issues in MTF tests are not exclusively to do with field curvatures, there are many field-dependent aberrations that contribute.
08-26-2015, 05:20 AM
I understand, from a tester's point of view the lens hast to be "punished" for the field curvature. However, my point is, I would not use a wide open wide angle to reproduce flat charts accurately. And if I want sharpness in the corner and take care at focussing, the lens is not that bad as long as I don't need an exactly flat plane.
Also, I really need LiveView anyway because those 51 AF points are also not a masterpiece in parallelism and accuracy to the sensor plane. Meaning: even if the best and most plane, well centered lens is in front of a Nikon, it would be coincidence if each sensor point would deliver the same AF value. I checked with FoCal pro a couple of lenses and didn't see a perfect test row. I know there has to be tolerances and I also know not to use pliers if I have a hammer available for putting a nail into a wall. Long sentence short: Under normal circumstances one needs LiveView to operate a focus at the outer borders/corners. This is not always possible and I don't see much real situations I want to have the sharpness so far off the center.
When reading a picture, we start top left and reading down to low right - if there's not something special which draws our interest to somewhere else. Most of the time there's a sort of sky to start with, and corner sharpness wide open would only be interesting for starfield photogs?
Edit (or addition)
There once was a rating tag called "field quality", at least in the test of the Nikkor 24/1.4G by Markus. It might be difficult to set the parameters for this tag, but I think based on my experience with some 3000 shots with it, I sign the same paper. And the Sigma would not be far off that rating, if at all. Maybe I'm expecting not "enough", I just see the Sigma as the only AF competing wide open wide angle and sold at a price I would have to tolerate some flaws - I just don't discover those flaws in my pictures. The improvements are also not easy to discover, but like Markus said in the verdict "...
and stopped down it's sharp enough even for landscape work, although it's certainly not meant to be the primary choice for this kind of application.
" I just have to agree on that. I also agree it's not the hottest lens Sigma throws on the market. For owners of the Canon 24/1.4 L II there might be no reason to change ship. For me, there was. At no cost, because a 5 year old Nikkor 24/1.4 G still represents some value to "Nikon only" folks.
For others thinking about a fast lens it would be foolish not to take a glance at it.
Field curvature is irrelevant for the charts - the test spots are locally refocused.
There are sample images in the full format tests so I skipped these for APS-C (as usual).
The sample images reflect the real life.
I once did as you suggested - thus pushing the weaknesses in the same image section. All I got was criticism that nobody would shoot images like these. At that stage I then decided that I did not want to moderate a gazillion messages about inappropriate samples.
> Any idea, why one want a FF fast prime on a half size sensor?
For better corners ?
Less vignetting ?
Less distortions ?
Better bokeh in the corners ?
Because you want to upgrade at some stage ?
Because there's no native 24mm f/1.4 for APS-C ?
For tele lenses - because you'd like more reach ?
08-26-2015, 05:55 AM
Thanks for the explanation about refocusing the corners. Wasn't sure about that.
So getting those gazillion complaints about how much people would shoot wide open with main subject in the corner does also say something about the importance of corners sharpness with this lens, doesn't it? Rate it the way you like, I'm aware of comparing Nikon and Sigma in the Canon board and grateful not to be thrown out.
And now, since I checked, I'm also aware of a price gap of 500 francs between Canon and Nikon version. And since you rated the Canon version already, the situation is different because Canon is not asking 100% more. 104% to be accurate and that goes with a bundled polarizer on the Sigma... yes, a Sigma polarizer, not a China copy.
08-26-2015, 06:28 AM
Quote:Thanks for the explanation about refocusing the corners. Wasn't sure about that.
Ratings are ratings - technical findings have a limited overlap with the real life. How limited depends on the use case.
We can't do the scoping for the reader.
If you want to have the sharpest Canon 50mm lens at f/5.6 on full format, you may choose the el cheapo 50mm f/1.8 STM. If you want to have the best bokeh, the the "bad" 50mm f/1.2 USM L may be your choice. The technical rating is irrelevant then.
In the real life I reckon I am a bit on your side. I'd use the 24mm f/1.4 at f/1.4 purely for shallow DOF thus the corners are irrelevant. I'd also use it stopped down to f/8 for infinity DoF which is where it's good. Thus within this usage pattern, the Sigma is fine. If you are taking pictures in a theatre (low light) you may prefer to have a lens with better borders just as an example.
Thus do your scoping and live happily ever after.
08-27-2015, 04:15 AM
Maybe we're just expecting each new Sigma A/C series lens to produce absolute optical miracles, every time they bring out a new breaking the barrier lens. ................................................................................;;; On the Canon FF 50 MPs sensor (which grabs all lens's resolutions by the throat) it looked to produce very good resolution on the frame edges and corners from F2.8 up and is surprisingly good even wide open where your unlikely to be cheeky enough to ask for crackling corners. The center resolution is not even in question.
One would expect less sharpness on the APSc sensor print size wise.
What you do get is a solidly made FF lens with Sigma's upgraded quality control over decentering, offering a F1.4 aperture on a wide 24mm FL. Add just for good measure no focus shift or OOF fringing. All for a price that doesn't make you wince like OEMs.
From where I'm sitting it looks a nice bit of kit, especially if you been spending the day picking bits of AF gear teeth out of an old Sigma EX lens of which you had to pack out bayonet with cardboard just to get some sort of semblance sharpness on the right hand side of the frame!
Sigma are doing very well now they've up'd the game and that's rubbing off on Tarmon!
08-27-2015, 06:08 AM
Maybe I manage not to bring in Nikon again. :unsure:
As long as we have to use a mirror box between sensor and lens, there will be no 5â˜… rating in lenses shorter than 40 mm, very especially not in the wide-angle-on-APS-C section. If one wants maximum sharpness and fine colors, sacrificing nearly every other aspect, throw away the APS-C DSLR (no matter which one) and get at the price of a lens a DP0 merrill quattro, 14/4, equiv 21 mm.
The other question is, does a lens and body have to be excellent in each aspects although it's impossible?
Klaus, I did my scoping 5 years ago and had 5 happy years with great wide-angle shots delivered by the Nikon ( :o blah, but it's fact, even if I didn't manage not to bring that in again).
If one wants
the Sigma IS an altenative and interestingly not only for those who can't / want afford a genuine one. It's performing equally or better to one of the big players and I daresay the other big player will not shine much more with their version.
That aspect runs a bit short in your test.
09-02-2015, 04:18 PM
Today, I've seen two sources comparing the Sigma 24/1.4 against it's competitors. One was DxO, the other was Lenscore.org.
It's fair to say "The best wide open 24 you can get for your Nikon or Canon" - maybe not by much difference, but much money to be saved. Lateron I will post a comparison table from Lenscore.
I know, some members here are no particular fans of both those websites. Just ignore that post then ^_^
|Users browsing this thread:|