The flare from the 7-14 on my E-M5 is no longer purple

[Anders W]

So what did I do?

Before

After

 

[texinwien]

Anders W wrote:

So what did I do?

Your yellow filter finally arrived, and you used it?

tex

 

[Martin.au]

PHOTOSHOP!!!

Kidding

 

[Anders W]

texinwien wrote:

Anders W wrote:

So what did I do?

Your yellow filter finally arrived, and you used it?

No idea how you could guess that tex. I thought I had kept my filter fiddling strictly confidential.

But yes, the Kodak Wratten 2A gelatin filter that I bought on ebay nearly a month ago has finally arrived. And my little theory as outlined here

http://www.dpreview.com/forums/post/50983159

does seem to work out as I hoped it would. So far I have just held the filter in front of the lens. Still remains to be seen how it works when affixed, one way or the other, behind the rearmost element.

 

[Anders W]

Mjankor wrote:

PHOTOSHOP!!!

Kidding

Straight out of LR with everything but lateral CA-correction at default. Promise.

 

[texinwien]

Anders W wrote:

texinwien wrote:

Anders W wrote:

So what did I do?

Your yellow filter finally arrived, and you used it?

No idea how you could guess that tex. I thought I had kept my filter fiddling strictly confidential.

But yes, the Kodak Wratten 2A gelatin filter that I bought on ebay nearly a month ago has finally arrived. And my little theory as outlined here

http://www.dpreview.com/forums/post/50983159

does seem to work out as I hoped it would. So far I have just held the filter in front of the lens. Still remains how it works when affixed, one way or the other, behind the rearmost element.

I'd probably make a good detective / spy. Oh no, I've said too much

Looking forward to finding out how you affix the filter, and how well it works, affixed. I'll have to add one of these to the list of filters I plan to purchase.

This is excellent news for those having trouble with the combination of the 7-14mm (and some other lenses) together with the new Sony m43 sensor. Thanks for doing the brainwork to come up with the theory and the legwork to test it out, successfully!

Congrats!

tex

 

[Anders W]

texinwien wrote:

Anders W wrote:

texinwien wrote:

Anders W wrote:

So what did I do?

Your yellow filter finally arrived, and you used it?

No idea how you could guess that tex. I thought I had kept my filter fiddling strictly confidential.

But yes, the Kodak Wratten 2A gelatin filter that I bought on ebay nearly a month ago has finally arrived. And my little theory as outlined here

http://www.dpreview.com/forums/post/50983159

does seem to work out as I hoped it would. So far I have just held the filter in front of the lens. Still remains how it works when affixed, one way or the other, behind the rearmost element.

I'd probably make a good detective / spy. Oh no, I've said too much

Looking forward to finding out how you affix the filter, and how well it works, affixed. I'll have to add one of these to the list of filters I plan to purchase.

This is excellent news for those having trouble with the combination of the 7-14mm (and some other lenses) together with the new Sony m43 sensor. Thanks for doing the brainwork to come up with the theory and the legwork to test it out, successfully!

Congrats!

Thanks tex! I hope there are no hidden catches lurking on the last few steps towards a workable solution, but I take nothing for granted just yet.

Optical1 has been kind enough to offer to help out with shipping the gelatin filter holder designed for the Panasonic 8 mm fisheye (but hopefully fitting the 7-14 too) from the US to Sweden. But as far as I know, the part is still back-ordered. Don't know yet whether I will wait for it or if curiousity gets the better of me and I'll try to tape the filter somehow as a short-term solution.

 

[JamieTux]

Good work Anders

Can you take a normal image (with some purple in) with the filter up too or are we not at that stage yet?!

 

[Michael J Davis]

Anders W wrote:

Mjankor wrote:

PHOTOSHOP!!!

Kidding

Straight out of LR with everything but lateral CA-correction at default. Promise.

Excellent! Does it work equally on the G1, or does it over compensate?

Mike

 

[Anders W]

JamieTux wrote:

Good work Anders
Can you take a normal image (with some purple in) with the filter up too or are we not at that stage yet?!

Thanks Jamie. I hope to be back with some more realistic/practical examples as soon as I can. But I have to find a practical way of affixing the filter behind the rearmost element first so it might take at least a little while.

 

[Anders W]

Michael J Davis wrote:

Anders W wrote:

Mjankor wrote:

PHOTOSHOP!!!

Kidding

Straight out of LR with everything but lateral CA-correction at default. Promise.

Excellent! Does it work equally on the G1, or does it over compensate?

Haven't had the time to compare with the G1 yet, but as far as I can tell at this point, the filter means the 7-14 on the E-M5 will work just the way it does on the G1. What the Wratten 2A filter does is essentially to strengthen UV filtering to approximately the same level as that of the UV filter already fitted on the sensor in many/most/all Panasonic bodies. As far as I can tell at this point, the reason why the E-M5 has the purple-flare problem whereas the G1 doesn't is that the E-M5 has a weaker UV filter on the sensor. Some of the flare/reflections from the 7-14 are precisely at the border between UV and longer wavelengths, and the E-M5 UV filter lets them pass whereas the G1 UV filter doesn't.

 

[kenw]

Great work Anders! Glad to see the Wratten 2A does the trick.

If the rear baffle from the 8.5mm stays on permanent back order and never shows up there is always this option:

http://www.shapeways.com/materials/strong-flexible

Cost is actually quite reasonable, but you'd have to model the part. I keep looking for an excuse to do some 3D printing. Unfortunately I don't have the time. There are a variety of freeware tools that seem to be good enough to do the modeling. Anyway, something to consider if you want to pick up yet another skill/hobby

 

[alolywu]

Just thinking out loud but is it possible to apply a software filter to the RAW files equivalent to the yellow filter you're holding up to the lens?

 

[kenw]

alolywu wrote:

Just thinking out loud but is it possible to apply a software filter to the RAW files equivalent to the yellow filter you're holding up to the lens?

No unfortunately. There is more detail in a previous thread, but the issue here is that what is causing the purple flare is light right on the edge of the UV spectrum and Anders is using a fairly sharp cut-off filter to eliminate it. Once that UV light is combined with the rest of the violet and blue light that reaches the sensor's blue channel pixels it is nearly impossible to separate it in post processing. If it was just purple flare against a black or grey background in theory you might be able to do some ratiometric processing to try and identify it and remove it compared to "real" blue and violet light. In the more practical case where it overlays complicated parts of the scene you've got no choice but the filtering method Anders has developed here.

 

[gollywop]

you're going to have to change your dPreview username to FilterMan.

--
gollywop

 

[Anders W]

kenw wrote:

Great work Anders! Glad to see the Wratten 2A does the trick.

Thanks Ken! Yes, I am really glad to see that it works out as well. I was fairly optimistic after finding rather precise data on the transmission curves of the built-in cut filters of some Oly versus Pany bodies along with the transmission curves of the filters that might potentially fit the bill. That seemed to clinch it. But I know better than to take anything for granted without proper testing.

What is a bit ironic is that UV filtering was one of the first things suggested and tried when the purple-flare problem first surfaced on the forum. An ordinary UV turned out to have no effect so we quickly left the idea behind. But the only problem is in fact that an ordinary UV is some 20 to 30 nm on the weak side. Had we looked just a bit further into the matter at that point, we might have made progress much faster. But then I suspect there aren't a whole lot of people with a set of Wratten 2s sitting around and ready for testing any more, if there ever was.

If the rear baffle from the 8.5mm stays on permanent back order and never shows up there is always this option:

http://www.shapeways.com/materials/strong-flexible

Cost is actually quite reasonable, but you'd have to model the part. I keep looking for an excuse to do some 3D printing. Unfortunately I don't have the time. There are a variety of freeware tools that seem to be good enough to do the modeling. Anyway, something to consider if you want to pick up yet another skill/hobby

That certainly looks interesting Ken. But for the time being, I think I'll stick to the filter holder for the 8 mm FE or double-sided adhesive. BTW: I recently came to think of the fact that adhesive magnetic tape might be a viable option for rear-mounted gel filters for those who want the option of removing the filter without destroying it:

http://www.first4magnets.com/self-adhesive-flexible-magnetic-tapes-38-c.asp

 

[Anders W]

gollywop wrote:

you're going to have to change your dPreview username to FilterMan.

Since, when moving to MFT, I finally dropped the idea of using UVs for protection (as I always did in the past) and since we all know that filters aren't needed for digital, I thought I was free from the filter plague at last. But somehow they still keep haunting me. Can't figure out why.

 

[Skeeterbytes]

Hi Anders,

Am glad to hear your experiments are bearing fruit--well done!

I've lost track of the narrative given the many lengthy related threads, and am wondering whether you (or anybody else) has experienced the issue with the ZD 7-14 on the E-M5? I ask because I shot a lot with the combination while traveling last week and don't see any evidence of the problem it in my shots. The ones that include the sun in the frame, or slightly off-frame, show the familiar string-of-beads flare I see on an E-series camera, but nother more. Is this problem more prevalent or limited to the Lumix 7-14?

Cheers,

Rick

 

[Anders W]

Skeeterbytes wrote:

Hi Anders,

Am glad to hear your experiments are bearing fruit--well done!

I've lost track of the narrative given the many lengthy related threads, and am wondering whether you (or anybody else) has experienced the issue with the ZD 7-14 on the E-M5? I ask because I shot a lot with the combination while traveling last week and don't see any evidence of the problem it in my shots. The ones that include the sun in the frame, or slightly off-frame, show the familiar string-of-beads flare I see on an E-series camera, but nother more. Is this problem more prevalent or limited to the Lumix 7-14?

I wouldn't think the ZD 7-14 is affected by the specific problem at issue here, i.e., flare of a specific kind (mirror image of the light source) changing color so as to become distinctly purple. In all likelihood, the problem with the Panasonic 7-14 on the E-M5 is the combined result of two causes, both of which being necessary ingredients: a) antireflective coatings on some lens elements that do not sufficiently reduce reflections at wavelengths close to the lower end of the visible spectrum and b) a fairly weak built-in UV filter on the E-M5 sensor (and perhaps some other Olympus bodies too).

 

[Surefoot]

every lens is affected, even the mighty M.Zuiko 75mm, although obviously it's a lot more difficult to reproduce due to almost zero chances of getting flare at all to start with.

More to the point, my M.Zuiko 12mm exhibits the same behavior:

Notice the 3 purple reflections, and the purple halo around the lights..

I have other examples with the sun, which show similar purple spots to the 7-14mm. So very probably the ZD 7-14 will be affected in the same way.