SLT light show

Specifically for the discussion of the A-mount DSLR range
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.
User avatar
mikeriach
Imperial Ambassador
Posts: 647
Joined: Sat Feb 02, 2008 11:29 am
Location: Aberdeen

Re: SLT light show

Unread postby mikeriach » Mon Apr 15, 2013 12:57 pm

Can't say I've noticed significant fringing on the A77 with either 70-200 or 70-400 but I'll have a quick review of recent shots.
I don't normally shoot wide open, usually at least f4.5 on the 70-200 and f8-11 on the 70-400.
His setting differ a bit from my "norm".
Using 1/3200s, f5.6 and ISO 400 with a static subject on a tripod seems very strange to me.
Birds at 1/200s seems to be asking for "shake" issues though the effect shown doesn't look like movenent induced. Could using SSS on a tripod contribute to this?

I do occasionally get a little colour fringing but that is easlily sorted with LR.

I'll let you know what I find.

Mike
Some Sony stuff and now some Nikon stuff (shock horror). Sony cannot supply what I want so rightly or wrongly I'm branching out to someone who does.

agorabasta
Viceroy
Posts: 1198
Joined: Wed Oct 22, 2008 7:41 pm

Re: SLT light show

Unread postby agorabasta » Mon Apr 15, 2013 1:36 pm

The mirror is easy to remove. All should be clear upon doing so.

But the examples in that link show something that comes not from the mirror, I'm afraid. And the fact that it shows only above the bright parts does only tell that there's more light and ghosts are brighter and hence better visible.
I think it's the very poor design of sensor toppings; and that's far worse than some mirror-induced ghosting. To put bluntly - purple fringing comes from the reflections between sensor and IR filter, and full-colour fringing comes from reflections between sensor and the LPF.

User avatar
Greg Beetham
Tower of Babel
Posts: 6155
Joined: Sun May 27, 2007 3:25 pm
Location: Townsville, Qld. Australia
Contact:

Re: SLT light show

Unread postby Greg Beetham » Mon Apr 15, 2013 4:04 pm

I read the dpr thread (as well as the link from Barry), it was pretty much as expected, some reasonable attempts at an explanation, and also some strange explanations absolving the A77 of any blame. :roll:
I guess he was using f5.6 looking for some bokeh which is a reasonable thing to do I guess. The LPF layers in front of a sensor is complex, here is a diagram of the one in the D800 http://photographylife.com/what-is-low-pass-filter (has the D800e too)
Canon use 6 layers of filters, linear and circular polarizers plus anti-aliasing, also non reflective coatings in front of the sensor on some cameras along with two lots of band-pass block on IR no less which is probably why it’s called an LPF. http://www.sensor-film.com/filter.html
I don’t know what LPF system Sony uses, I didn’t notice anything from Sony in the search results.
He claims he didn’t have those top fringes using the same lenses on his A700.
Greg

agorabasta
Viceroy
Posts: 1198
Joined: Wed Oct 22, 2008 7:41 pm

Re: SLT light show

Unread postby agorabasta » Mon Apr 15, 2013 8:11 pm

I don't really know how it all works with a77/99. What I know is that all the Nex and a55 definitely had the topmost layer (that apparently is a hot mirror) made of plastic film that hangs about 1-2mm above the rest of filter sandwich. At the same time, the a700/300/350/900/850/550/500 all seemed to have a hard filter sandwich much like those in your Canon link.
I also know that modern Nikon bodies normally produce far less PF that Sony's do, and also less than Canon's do either; Sony is the worst at PF.
If you look into a Nikon V1/J1 cam, you may notice that the hot mirror is also a plastic film hanging about 5-7mm above the sensor; those cams produce very little PF but they also deliver very bad local contrast, especially at low-light conditions with some high intensity light sources in the frame - that's due to reflections off that remote plastic layer turning into kind of veiling.

So really there is a great variability in sensor toppings designs, that's for sure. And the designs of those are far from any perfection and they are changing rather fast.

User avatar
mikeriach
Imperial Ambassador
Posts: 647
Joined: Sat Feb 02, 2008 11:29 am
Location: Aberdeen

Re: SLT light show

Unread postby mikeriach » Mon Apr 15, 2013 10:24 pm

Well I've had a look at a sample of images shot with the A77 and a variety of lenses, 16-80, 70-200G, 70-400G.
None of them show any ghosting like the sample images. All I can see is a minor colour fringing at high contrast edges which can be neutralised with LR. One was handheld at 400mm, 1/320s, f6.3, ISO 100 with no ghosting, totally pin sharp (a few others in the sequence had camera shake).

In summary I have none of the reported ghosting at all with any of my preferred lenses in my day to day shooting :D

Mike
Some Sony stuff and now some Nikon stuff (shock horror). Sony cannot supply what I want so rightly or wrongly I'm branching out to someone who does.

David Kilpatrick
Site Admin
Posts: 6248
Joined: Sat May 19, 2007 1:14 pm
Location: Kelso, Scotland
Contact:

Re: SLT light show

Unread postby David Kilpatrick » Mon Apr 15, 2013 10:41 pm

I've spent a half hour today with the 70-400mm on A900 (not a combination I regularly used, because the A77 seemed obviously a better choice - even the A580 is a better choice in terms of telephoto resolution). First of all, the bokeh of the 70-400mm at 400mm and f/5.6 does produce a distinct halo from marginally defocused bright objects and in daylight, where the light is of course from above, it's nearly always visible on the top edge. It disappears completely if the focus is perfect - and by perfect, I'm talking a matter of maybe 1 or 2 metres in 100-300mm.

I've been able to tell quickly that my lens on the 'new' ex-demo body before any lens corrections has consistency back, not front, focus tendencies and also by studying 3D detail that it is back focus which creates the most visible focus halo. I will adjust my AF micro setting and hope to eliminate this.

I have also been able to see from some of the complex subjects I've shot that at 400mm and f/5.6 the lens is critically sharp corner to corner - it may not have been focused on my 'layer' of trees and woodland, but the layer immediately behind in depth is rendered to pixel perfect level. Also, there's no trace of halo on absolutely perfect focus. I'll follow this up on the A99 when I have time and also on the A77, but I suspect 'just within depth of field' is fooling the DPReview poster along with subtle aspects of subject and light.

Here is a quick snap taken today of a chiffchaff (I think - smaller than a robin, tiny sharp beak, very yellow, very fast moving) - not a bird I can remember seeing in April in Scotland before - which was darting from branch to branch in a tree over the riverside this evening. Light was low but the excellent ISO 320 peformance of the A900 enabled a 1/1000th shutter speed- still not enough to freeze the extremely fast movement of this tiny bird. I used single shot, AF-S, centre focus and to be honest with the complex web of branches between me and the bird, I was amazed to get all three shots I managed in ten minutes to be sharp, ansd I do not believe there's any way I could have seen and targeted this subject with the EVF. Others will disagree but recognising the bird in the finder and moving the focus target to it before it had flown off to the next branch required concentration and a very clear view of fine detail.

This image is reduced to 50% size, this area was just over 2500 pixels wide in the original shot.

chiffchaff.jpg
chiffchaff.jpg (160.72 KiB) Viewed 1900 times


Let's just say that right now, I'm very happy to have received my ex-demo A900 at exactly one half of the last 'new' price WEX had it listed for, missing only its flash shoe cover, otherwise in excellent order with a tiny mark on the rear screen.

David

Philip
Oligarch
Posts: 202
Joined: Mon Jun 11, 2007 2:23 pm
Location: Looe

Re: SLT light show

Unread postby Philip » Tue Apr 16, 2013 7:44 am

My experience of the 70-400 on the A900 was always the same as yours David - get the focus right and it's sharp from corner to corner even at f5.6 and 400mm. I cannot remember ever having noticed a 'halo' effect on in focus subjects in a variety of lighting scenes. This is the lens/camera combination I most miss!

Philip

User avatar
Winston
Imperial Ambassador
Posts: 481
Joined: Fri Jul 18, 2008 5:29 pm
Location: Boise, Idaho, USA, 3rd Planet

Re: SLT light show

Unread postby Winston » Tue Apr 16, 2013 5:12 pm

I have a 70-400G which I use on two A77 bodies.
No ghosting with either body.
As David noted a near miss in focus can be really ugly.
Winston Mitchell
KM7D, A700, A77, A77M2

User avatar
mvanrheenen
Emperor of a Minor Galaxy
Posts: 1708
Joined: Mon Nov 29, 2010 3:58 pm
Location: Netherlands
Contact:

Re: SLT light show

Unread postby mvanrheenen » Tue Apr 16, 2013 8:51 pm

Nice shot David. As for the identity: it indeed seems to be a chiffchaff :-)

My experience with the 70-300G and 70-400G is that both are fabulous wide open, but the 70-300G has a LOT more purple fringing in high contrast areas than the 70-400G. That's why I always stop down the 70-300G in these situations. Out of focus twigs in the background of a bird as in Davids image is an example of these high contrast situations I encounter on every shoot. I imagine the 70-200G would have the same issue.

Mark

User avatar
Greg Beetham
Tower of Babel
Posts: 6155
Joined: Sun May 27, 2007 3:25 pm
Location: Townsville, Qld. Australia
Contact:

Re: SLT light show

Unread postby Greg Beetham » Thu Apr 18, 2013 5:50 am

I got the impression that it (top edge fringes in those linked examples) was more related to exposure rather than focus but it could be some kind of weird combination perhaps.
The thing is if it’s focus why isn’t the entire image blurred?
How can an image be blurred by missed focus just along a top edge of something in the photo and nowhere else? On top of that even the blurred fringe has a sharpish edge, what’s the deal with that?
Maybe it’s some kind of internal reflection but even that is odd because you would expect other sharp edges elsewhere to have fringes, unless those don’t have the required amount of exposure difference vs. the rest of the subject as well as the oof background/contrast.
The only other thing that comes to mind is some kind of weird processing glitch like a displaced semi opaque layer that got displaced vertically somehow.
Greg

User avatar
bfitzgerald
Subsuming Vortex of Brilliance
Posts: 3572
Joined: Fri Sep 26, 2008 10:48 pm

Re: SLT light show

Unread postby bfitzgerald » Thu Apr 18, 2013 3:04 pm

Just an observation didn't Sony release the new 18-55mm SAM II and in the press release it mentioned ghosting/flare. I've not used the version I enough to see if this a problem (but I'll check when I actually get an A57 and kit lens) Is it possible "some" lenses are prone to this issue with the SLT models?


Return to “Alpha A-mount System”

Who is online

Users browsing this forum: No registered users and 14 guests

cron