Sigma's new HSM 12-24mm DG II
Forum rules
No more than three images or three external links allowed in any post or reply. Please trim quotations and do not include images in quotes unless essential.
No more than three images or three external links allowed in any post or reply. Please trim quotations and do not include images in quotes unless essential.
-
- Site Admin
- Posts: 5985
- Joined: Sat May 19, 2007 1:14 pm
- Location: Kelso, Scotland
- Contact:
Sigma's new HSM 12-24mm DG II
It's a very, very good lens indeed if my test sample is typical. Much sharper (without any unsharp zones or corners etc) than the original, and better than the 8-16mm which I already think is a stunning lens.
See original Alpha 900 shot, taken at f/9 and 12mm - camera was set to ISO 320 but this lens tends to produce a stop of underexposure (grid screen interacts badly with ultrawides!) on my A900, so it's actually been boost by a stop in ACR making the real ISO 640.
Link: http://www.pbase.com/davidkilpatrick/image/143072179
Taken many more test images tonight walking round, checking its performance on loads of subects from 28cm close-up ground level stuff to cropped shots of architecture corrected tightly. This lens is making me reconsider what to take for landscape shooting in the summer - A900 just works far too well with it.
David
See original Alpha 900 shot, taken at f/9 and 12mm - camera was set to ISO 320 but this lens tends to produce a stop of underexposure (grid screen interacts badly with ultrawides!) on my A900, so it's actually been boost by a stop in ACR making the real ISO 640.
Link: http://www.pbase.com/davidkilpatrick/image/143072179
Taken many more test images tonight walking round, checking its performance on loads of subects from 28cm close-up ground level stuff to cropped shots of architecture corrected tightly. This lens is making me reconsider what to take for landscape shooting in the summer - A900 just works far too well with it.
David
-
- Viceroy
- Posts: 1198
- Joined: Wed Oct 22, 2008 7:41 pm
Re: Sigma's new HSM 12-24mm DG II
David,
Somehow the right side of the image appears sharper than the left one. Sorry to poke...
Somehow the right side of the image appears sharper than the left one. Sorry to poke...
-
- Site Admin
- Posts: 5985
- Joined: Sat May 19, 2007 1:14 pm
- Location: Kelso, Scotland
- Contact:
Re: Sigma's new HSM 12-24mm DG II
I think that's just the difference in distances. As with all wide angles I've ever used on my A900, sometimes with a HORIZONTAL pic one side will appear sharper than the other, especially with SSS enabled. But I have not noticed a top-to-bottom sharpness difference even with the old Sigma 12-24mm (except on vertical compositions, where I have used the misalignment of the lens and sensor to give me a sort of 'tilt' substitute with the original 12-24mm).
David
David
Re: Sigma's new HSM 12-24mm DG II
Makes me wonder if this is similar to what I saw on some A580 shots with meduim tele focal length when I thought I might have a misaligned sensor.David Kilpatrick wrote: As with all wide angles I've ever used on my A900, sometimes with a HORIZONTAL pic one side will appear sharper than the other, especially with SSS enabled.
David
Will try a selection this weekend with and without SSS.
Amazing what you learn on this forum.
David, are you sure you don't have a USB socket so we can download you.......
Mike
All my Sony SLT gear gone. Still got my RX100 though.
- Dr. Harout
- Subsuming Vortex of Brilliance
- Posts: 5662
- Joined: Wed May 30, 2007 7:38 pm
- Location: Yerevan, Armenia
- Contact:
Re: Sigma's new HSM 12-24mm DG II
I like that... I mean to download him too...mikeriach wrote:David, are you sure you don't have a USB socket so we can download you.......
Mike
After all, it's because of him that i own a Distagon T* 2/24 SSM Zeiss lens...
-
- Viceroy
- Posts: 1198
- Joined: Wed Oct 22, 2008 7:41 pm
Re: Sigma's new HSM 12-24mm DG II
David,
So far, I think that Nex7 sensor does a much better job at rendering tight meshes, as it does it with virtually no artifacts like colour substitution and/or of any other kind of sensor/toppings induced aliasing. That actually makes it a better body for landscapes.
But definitely, having a proper lens for the task is more important...
So far, I think that Nex7 sensor does a much better job at rendering tight meshes, as it does it with virtually no artifacts like colour substitution and/or of any other kind of sensor/toppings induced aliasing. That actually makes it a better body for landscapes.
But definitely, having a proper lens for the task is more important...
-
- Site Admin
- Posts: 5985
- Joined: Sat May 19, 2007 1:14 pm
- Location: Kelso, Scotland
- Contact:
Re: Sigma's new HSM 12-24mm DG II
I haven't tried the 8-16mm Sigma on the NEX, mainly because my latest 16mm pancake seems better or equal at 16mm and the adaptor for 12mm is more convenient. I've got problems with the last set of shots taken on the A77 and 8-16mm. In January, I was getting good sharp images across the frame but did not use the lens much. In April I used the lens exclusively for three different days, with the 18-200mm Tamron on the NEX-7 and 8-16mm on A77.
Too many of the A77/8-16mm shots have very soft edges/corners - far too soft to be usable, much softer than say the 16mm pancake on NEX. This is disappointing as it's always been excellent on the A55. I suspect it is a focus related issue and I should have done a micro AF calibration.
David
Too many of the A77/8-16mm shots have very soft edges/corners - far too soft to be usable, much softer than say the 16mm pancake on NEX. This is disappointing as it's always been excellent on the A55. I suspect it is a focus related issue and I should have done a micro AF calibration.
David
Re: Sigma's new HSM 12-24mm DG II
I've finally just gotten the 16mm and the ultra-wide adapter this past week based on your recommendations and I have to say I'm quite pleased. I've just sold my Konica Minolta 11-18 as the 16+UW seems much better and is much more compact and likely to go more places. The transition continues...David Kilpatrick wrote:I haven't tried the 8-16mm Sigma on the NEX, mainly because my latest 16mm pancake seems better or equal at 16mm and the adaptor for 12mm is more convenient. I've got problems with the last set of shots taken on the A77 and 8-16mm. In January, I was getting good sharp images across the frame but did not use the lens much. In April I used the lens exclusively for three different days, with the 18-200mm Tamron on the NEX-7 and 8-16mm on A77.
Ed
-
- Viceroy
- Posts: 1198
- Joined: Wed Oct 22, 2008 7:41 pm
Re: Sigma's new HSM 12-24mm DG II
The problem with SEL16 on Nex7 is that there's a considerable purple shift in the corners, much like what you get with rangefinder WA's.
Even if you correct it with some 'cornerfix', you lose a great deal of DR and colour fidelity in the corrected corners.
But I didn't check if the 12mm UWA adapter makes it any better/worse.
And I don't think that a77 and Nex7 are any similar in their performance with WA lenses. That's because the sensor toppings on the 7 must be very different due to shorter registration distance of the system. Means that the 7 should be generally much better with the DSLR lenses. And so far I see it really is.
Even if you correct it with some 'cornerfix', you lose a great deal of DR and colour fidelity in the corrected corners.
But I didn't check if the 12mm UWA adapter makes it any better/worse.
And I don't think that a77 and Nex7 are any similar in their performance with WA lenses. That's because the sensor toppings on the 7 must be very different due to shorter registration distance of the system. Means that the 7 should be generally much better with the DSLR lenses. And so far I see it really is.
-
- Viceroy
- Posts: 1196
- Joined: Mon Aug 18, 2008 11:20 pm
Re: Sigma's new HSM 12-24mm DG II
Agorabasta, are you implying that fitting an Alpha lens to the NEX-7 through one of the Sony adapters (including my LA-EA2) yields better results than using current E-mount lenses such as the SEL16mm pancake on the NEX-7?And I don't think that a77 and Nex7 are any similar in their performance with WA lenses. That's because the sensor toppings on the 7 must be very different due to shorter registration distance of the system. Means that the 7 should be generally much better with the DSLR lenses. And so far I see it really is.
I only ran one test run with my Sony 18-250 zoom on my NEX-5 and looking at 100% I rated the results as good at those achieved with this lens on my A550/580 DSLRs. I concluded that the adapter did not cause a deteriation of image quality, but then this was not a scientific test. I really bought the adapter to use that Sony 18-250 on a still to be bought NEX-7 basically for video. Waiting for NEX-7 prices to drop.
-
- Viceroy
- Posts: 1198
- Joined: Wed Oct 22, 2008 7:41 pm
Re: Sigma's new HSM 12-24mm DG II
Yes, I say exactly that.alphaomega wrote:Agorabasta, are you implying that fitting an Alpha lens to the NEX-7 through one of the Sony adapters (including my LA-EA2) yields better results than using current E-mount lenses such as the SEL16mm pancake on the NEX-7?
The Nex was designed to work reasonably well with native lenses that are often too far from being telecentric, much more so than the SLR lenses are.
So an SLR lens on adapter is most likely a more telecentric setup, and you benefit from that. Means you may expect better corners at wide position.
Re: Sigma's new HSM 12-24mm DG II
I haven't done formal comparisons to see if the UW adapter actually increases purple shift in the corners (nor will I) but my impression is that it does. But with most shots none will be visible unless it's something like leaves or branches against a blown-out sky and then it's definitely there. So far I don't see that much with the 16mm and it seems more pronounced with the UW adapter, but it's still a matter of degree and the exact same thing will be done in either case by me if it is so apparent that it needs fixing. So that's really a moot issue as far as I'm concerned.agorabasta wrote:The problem with SEL16 on Nex7 is that there's a considerable purple shift in the corners, much like what you get with rangefinder WA's.
Even if you correct it with some 'cornerfix', you lose a great deal of DR and colour fidelity in the corrected corners.
But I didn't check if the 12mm UWA adapter makes it any better/worse.
I haven't used Cornerfix and if it's not fixable in ACR, I usually just locally desaturate the purple and there is no loss of dynamic range anywhere else. You can't get the detail back but that's not nearly so offensive as screaming purple.
My 11-18mm never went on extended hikes whereas the 16mm already has. An opportunity to actually use it didn't arise on that particular occasion but it was with me. I have to look at the "big picture," so to speak. A lot of people are upset that Sony didn't make a 600-1000$ lens for 250$ (or 360$ with the UW) but I'm not one of them. This one has found a permanent home.
Ed
-
- Viceroy
- Posts: 1198
- Joined: Wed Oct 22, 2008 7:41 pm
Re: Sigma's new HSM 12-24mm DG II
Ed,
The infamous 'purple shift' is the thing you may see in the exaggerated test image below. It's nothing like the purple fringing. It's the global purplish tint that mostly colours the corners of the image as the light rays hit the sensor at such an oblique angle that the Bayer colour filters don't coincide with their respective sensels in projection.
The infamous 'purple shift' is the thing you may see in the exaggerated test image below. It's nothing like the purple fringing. It's the global purplish tint that mostly colours the corners of the image as the light rays hit the sensor at such an oblique angle that the Bayer colour filters don't coincide with their respective sensels in projection.
- Attachments
-
- DSC01928.jpg
- (212.77 KiB) Downloaded 3728 times
Re: Sigma's new HSM 12-24mm DG II
I see what you mean. My misunderstanding. But that would seem even easier to fix. Does this seem an improvement?
- Attachments
-
- depurpled.jpg (224.09 KiB) Viewed 6691 times
-
- Viceroy
- Posts: 1198
- Joined: Wed Oct 22, 2008 7:41 pm
Re: Sigma's new HSM 12-24mm DG II
Ed, there no need for manual attempts.
Here's a neat little proggie that concerned types used for ages - http://sourceforge.net/projects/cornerfix/.
But I insist that it's still an awkward kludge. And that purple tint primarily means that the colour separation by the Bayer array has been severely deteriorated, so the restoration based on residual data lacks fidelity and introduces colour noise. So I find it of very little usefulness, mostly unacceptable for serious quality work.
Here's a neat little proggie that concerned types used for ages - http://sourceforge.net/projects/cornerfix/.
But I insist that it's still an awkward kludge. And that purple tint primarily means that the colour separation by the Bayer array has been severely deteriorated, so the restoration based on residual data lacks fidelity and introduces colour noise. So I find it of very little usefulness, mostly unacceptable for serious quality work.
Who is online
Users browsing this forum: No registered users and 21 guests