p.28 #2 · Sigma 14-24mm f/2.8 DG DN first impressions
SDTrojan wrote:
Fred, any chance you've tested the 6 stop gel yet? I think if that works as well as the Lee 4 stop, I'm fully sold on using/keeping this lens. Just ordered the lens yesterday.
I found that the Kodak Wratten (version 1) had higher color cast than the Lee. I also saw very noticeable contrast loss with the Kodak that is not present with the Lee ND gel. I have not tried the Wratten version 2 yet.
So far, I only recommend the Lee ND and Polarizer gels.
p.28 #3 · Sigma 14-24mm f/2.8 DG DN first impressions
JeyB wrote:
Hello
I have just received my new Sigma 14-24 f2,8 DG DN. Please, would someone tell me how to check corner consistencies?
Do I have to focus on the same object on each extreme corner? Or should I focus at middle point and then recompose, without focusing again, to the same object on every extreme corner?
I've been making tests this evening. Compared to my Samyang 14 ED AS IF UMC, the Sigma FL is noticeably shorter at 14 mm. Or looking from the other way, the Samyang is wider at 14 mm. Has anyone observed this issue? Is the 14 mm FL really 14 mm? I don't think this is related to the Metabones V I am using on my Alpha 7 III.
But the Sigma's sharpness is clearly superior in every f-stop than the Samyang, in any part of the frame. I even think the Sigma is clearly sharper at 2,8 than the Samyang at its sweet spot of f5,6.
Very happy til now. Tomorrow I will check the extreme corners, but after reviewing the photos I took today, all the corners seem to be very even, so maybe I was lucky and received a well centered copy.
Another question flying around my head. Maybe I should upgrade to an A7 RIII in order to extract the maximum sharpness of this lens. Do you think than the A7 III sensor is subpar regarding this lens resolution or the 135 GM I also own?
p.28 #5 · Sigma 14-24mm f/2.8 DG DN first impressions
JeyB wrote:
Thanks :-)
I've been making tests this evening. Compared to my Samyang 14 ED AS IF UMC, the Sigma FL is noticeably shorter at 14 mm. Or looking from the other way, the Samyang is wider at 14 mm. Has anyone observed this issue? Is the 14 mm FL really 14 mm? I don't think this is related to the Metabones V I am using on my Alpha 7 III.
But the Sigma's sharpness is clearly superior in every f-stop than the Samyang, in any part of the frame. I even think the Sigma is clearly sharper at 2,8 than the Samyang at its sweet spot of f5,6.
Very happy til now. Tomorrow I will check the extreme corners, but after reviewing the photos I took today, all the corners seem to be very even, so maybe I was lucky and received a well centered copy.
Another question flying around my head. Maybe I should upgrade to an A7 RIII in order to extract the maximum sharpness of this lens. Do you think than the A7 III sensor is subpar regarding this lens resolution or the 135 GM I also own?
There can be a small FOV variation with every lens, especially zooms. So, I don't doubt some copies may look more narrower than others at 14mm. There is also the possibility the Samyang is a little wider than 14mm.
Sigma's specifications mention: 114.2-84.1° at infinity distance.
114.2° Diagonal for the widest FL on the range is exactly 14mm.
Regarding sensor resolution: The two lenses you mentioned (135/1.8 GM and Sigma 14-24/2.8) are both outstanding even on the A7R4 (60MP). So, yes, they will perform great on the A7R3 as well.
On the A7R4, the Sigma 14-24 @24mm (which is its weakest FL), performs slightly better than the FE 24/1.4 GM prime at f/5.6. https://www.fredmiranda.com/forum/topic/1609212/20#14997501
p.28 #6 · Sigma 14-24mm f/2.8 DG DN first impressions
There can be a small FOV variation with every lens, especially zooms. So, I don't doubt some copies may look more narrower than others at 14mm. There is also the possibility the Samyang is a little wider than 14mm.
Sigma's specifications mention: 114.2-84.1° at infinity distance.
114.2° Diagonal for the widest FL on the range is exactly 14mm.
Regarding sensor resolution: The two lenses you mentioned (135/1.8 GM and Sigma 14-24/2.8) are both outstanding even on the A7R4 (60MP). So, yes, they will perform great on the A7R3 as well.
On the A7R4, the Sigma 14-24 @24mm (which is its weakest FL), performs slightly better than the FE 24/1.4 GM prime at f/5.6. https://www.fredmiranda.com/forum/topic/1609212/20#14997501...Show more →
Thanks again. Yes, it must be that the Samyang is really wider than 14 mm.
Wow, the comparison between Sony 24 GM and Sigma zoom @24mm is impresive. Maybe it is coming a time where high resolution and corrected zooms are really challenging the best primes at coincident f-stops.
By the way, I use LR to develop my photographs, is there any way to find the profile for the Sigma 14-24 DG DN? I've only found the DG HSM A018 lens profile version. But this one does not correct adequately the big geometric deformations of the DG DN version, ie lots of barrel distortion and straight lines not really straight. I've found that the Canon 17-40 L profile is a good recipe at 14-18mm, but not above 20mm range.
Hmm, A7 RIII is gaining points to be the new companion to my A7 III (which I really love). Best price I've found til now is 2.099€. Maybe I could wait a pair of months to see what happens. But I have no idea about how much stock of A7 RIII is out there... it could be too late then ;-)
p.28 #7 · Sigma 14-24mm f/2.8 DG DN first impressions
JeyB wrote:
Hello
I have just received my new Sigma 14-24 f2,8 DG DN. Please, would someone tell me how to check corner consistencies?
Do I have to focus on the same object on each extreme corner? Or should I focus at middle point and then recompose, without focusing again, to the same object on every extreme corner?
Thank you very much
Best
You will get the most accurate indication of how well centered a lens is by focusing for one corner and then checking that against the the other corners without refocusing. This gives a more accurate indication than center focusing for which field curvature can soften all your corners, making small differences harder to see. I like to use the same object placed in all four corners (without refocusing) because then you also are comparing the exact same structures.
p.28 #8 · Sigma 14-24mm f/2.8 DG DN first impressions
rvh23 wrote:
You will get the most accurate indication of how well centered a lens is by focusing for one corner and then checking that against the the other corners without refocusing. This gives a more accurate indication than center focusing for which field curvature can soften all your corners, making small differences harder to see. I like to use the same object placed in all four corners (without refocusing) because then you also are comparing the exact same structures.
Hi rvh23
Then you suggest this is a more accurate test than focusing center at infinity and then recomposing for every corner?
Supposing that your method keeps field curvature excluded from the equation, it may be indeed.
I tried some not very formal tests today and I start to suspect that left side is somewhat softer than the right side at f2,8. I suppose that this pattern will repeat itself at any aperture... Or does it get better at higher number diaphragms?
Thanks a lot for your answer. Tomorrow I'll check both methods.
p.28 #10 · Sigma 14-24mm f/2.8 DG DN first impressions
rvh23 wrote:
You will get the most accurate indication of how well centered a lens is by focusing for one corner and then checking that against the the other corners without refocusing. This gives a more accurate indication than center focusing for which field curvature can soften all your corners, making small differences harder to see. I like to use the same object placed in all four corners (without refocusing) because then you also are comparing the exact same structures.
---------------------------------------------
JeyB wrote:
Hi rvh23
Then you suggest this is a more accurate test than focusing center at infinity and then recomposing for every corner?
Supposing that your method keeps field curvature excluded from the equation, it may be indeed.
I tried some not very formal tests today and I start to suspect that left side is somewhat softer than the right side at f2,8. I suppose that this pattern will repeat itself at any aperture... Or does it get better at higher number diaphragms?
Thanks a lot for your answer. Tomorrow I'll check both methods.
Best
That's a good point but field curvature comes in different shapes.. Some are 'wavy' where the mid-zone is the area mostly affected.
I think it's a good idea to focus at center and check all 4 corners as well as the corresponding mid zone areas. As you suggested, focusing at one corner (instead of center) would also work. The key here is to find consistency and symmetry for the areas away from center.
p.28 #12 · Sigma 14-24mm f/2.8 DG DN first impressions
Very nice images, theacguy71
This is the type of shots, that i also would take with my nikon 14-24mm. Haven't used it much because, i get a lot of soft images with it (focus shift, or just bad copy) dunno, also LOTS of flare. and no lifeview focus. I'm now migrating to Sony. This is one of first lenses i will buy. The MTF just blows away zooms, and is also competitive with primes. And it's still less weight then DSLR, function button, etc.
@Fredmiranda, you mention good flare resistance. Maybe this has something to do with the newly advertised Nano Porous Coating. Sigma usually doens't overstate something, that is 'not worth it', so i believe them when they made newly improved coating.
p.28 #13 · Sigma 14-24mm f/2.8 DG DN first impressions
Fred Miranda wrote:
---------------------------------------------
That's a good point but field curvature comes in different shapes.. Some are 'wavy' where the mid-zone is the area mostly affected.
I think it's a good idea to focus at center and check all 4 corners as well as the corresponding mid zone areas. As you suggested, focusing at one corner (instead of center) would also work. The key here is to find consistency and symmetry for the areas away from center.
But no matter what the field curvature shape, for decentering, we are looking for variation in corner sharpness, so the best way to do that is make sure one corner is sharp, and see how the others compare. If they are sharp too, you can be certain the lens is well centered regardless of FC shape.
However, if you center focus, and as a result the corners are slightly soft due to field curvature, then it's harder to see differences due to centering issues. In fact, in the worst case the corners might look the same, but one side could be slightly soft because it is near focused, and the other side because it is far focused.
p.28 #14 · Sigma 14-24mm f/2.8 DG DN first impressions
JeyB wrote:
Hi rvh23
Then you suggest this is a more accurate test than focusing center at infinity and then recomposing for every corner?
Supposing that your method keeps field curvature excluded from the equation, it may be indeed.
I tried some not very formal tests today and I start to suspect that left side is somewhat softer than the right side at f2,8. I suppose that this pattern will repeat itself at any aperture... Or does it get better at higher number diaphragms?
Thanks a lot for your answer. Tomorrow I'll check both methods.
Best
As you stop down the DOF will increasingly mask mild decentering. All the copies I have tried, and most I've heard about form other FM members seem to be very good by the time you are at 5.6.
p.28 #15 · Sigma 14-24mm f/2.8 DG DN first impressions
Matti6950 wrote:
Very nice images, theacguy71
This is the type of shots, that i also would take with my nikon 14-24mm. Haven't used it much because, i get a lot of soft images with it (focus shift, or just bad copy) dunno, also LOTS of flare. and no lifeview focus. I'm now migrating to Sony. This is one of first lenses i will buy. The MTF just blows away zooms, and is also competitive with primes. And it's still less weight then DSLR, function button, etc.
@Fredmiranda@, you mention good flare resistance. Maybe this has something to do with the newly advertised Nano Porous Coating. Sigma usually doens't overstate something, that is 'not worth it', so i believe them when they made newly improved coating....Show more →
Thanks, here is a snapshot i took just to check out flare, straight towards sun, pointed up at 45 degree angle
Not too bad, i lifted the shadows to get the building details
p.28 #16 · Sigma 14-24mm f/2.8 DG DN first impressions
My point is that it does not really matter as symmetry is what we are testing for. One will be able to detect lens tilt even with FC.
I see what you are saying though and it makes sense to me.
rvh23 wrote:
But no matter what the field curvature shape, for decentering, we are looking for variation in corner sharpness, so the best way to do that is make sure one corner is sharp, and see how the others compare. If they are sharp too, you can be certain the lens is well centered regardless of FC shape.
However, if you center focus, and as a result the corners are slightly soft due to field curvature, then it's harder to see differences due to centering issues. In fact, in the worst case the corners might look the same, but one side could be slightly soft because it is near focused, and the other side because it is far focused....Show more →
p.28 #17 · Sigma 14-24mm f/2.8 DG DN first impressions
Fred Miranda wrote:
My point is that it does not really matter as symmetry is what we are testing for. One will be able to detect lens tilt even with FC.
I see what you are saying though and it makes sense to me.
I think the point rvh23 is making bit me on the bum once.
I was testing a lens with a fair bit of FC so that centre focussed all the corners looked bad and i couldn't see much difference.
But the lens had bad tilt; so the right hand side was softer. Of course the right hand corners were softer than the left ones too, but this was disguised by the overall softness of all the corners.
Focussing on the corners to overcome FC doesn't help, because then they all look good!!: tilt just changes the focus plane on one side, so if you focus on something it's still OK.
What helps to pick tilt in this situation is to compare the centre with the outer midframe on both sides of centre, and see if they are the same.
The good news is that with most modern lenses there is little enough FC that tilt will be revealed by a centering test looking at corners.
p.28 #18 · Sigma 14-24mm f/2.8 DG DN first impressions
DavidBM wrote:
The good news is that with most modern lenses there is little enough FC that tilt will be revealed by a centering test looking at corners.
My experience as well. FC is never too severe with modern lenses as long as they are designed for the Sony sensor or SLR adapted (with an adapter up to specs)
From the Loxia line, both 21/2.8 (inwards) and 35/2 (wavy) have moderate FC and focusing on the center always did the trick for me when testing for tilt. I don't think focusing on the corner would be a bad idea either...
p.28 #19 · Sigma 14-24mm f/2.8 DG DN first impressions
BTW: I just tested the Kodak Wratten version 2 (6-stop) and it's way better than the older Wratten version 1.
Colors look accurate. (only slightly warm)...However I still see some contrast loss with Kodak gels (Contrast +35 is needed in post to match the original scene). I find that undesirable.
p.28 #20 · Sigma 14-24mm f/2.8 DG DN first impressions
DavidBM wrote:
I think the point rvh23 is making bit me on the bum once.
I was testing a lens with a fair bit of FC so that centre focussed all the corners looked bad and i couldn't see much difference.
But the lens had bad tilt; so the right hand side was softer. Of course the right hand corners were softer than the left ones too, but this was disguised by the overall softness of all the corners.
Focussing on the corners to overcome FC doesn't help, because then they all look good!!: tilt just changes the focus plane on one side, so if you focus on something it's still OK.
What helps to pick tilt in this situation is to compare the centre with the outer midframe on both sides of centre, and see if they are the same.
The good news is that with most modern lenses there is little enough FC that tilt will be revealed by a centering test looking at corners....Show more →
If you focus on only one corner, and don't refocus when you look at the other three corners, you will see any tilt. That is the method I am advocating. If you are being very fussy, which I tend to be, I think the Loxia 21 is actually a good candidate for using this method over center focus.
