The 46mm Hoya +1 option looks good. Somehow I missed this option when browsing through the thread. For the 46mm Hoya +1 I hope Contax 46mm UV filter frame work:
It was standard accessory for all sold G45 and G90 (maybe other G lenses as well?).
I don't even dream on using 55mm (or 46mm) polarizer and ND, I have no problem stepping up to 58/67/82mm when filter is needed. I don't think I have ND filters other than 58/67/82mm. So as long as I can attach step-up ring I'm fine. Step-up ring adds ~1.25mm height, and the inner diameter reduces 1.95mm (46mm) or 2.25mm (55m).
GMPhotography wrote:
I'm wondering if your ELPro retaining ring is too thick. I have 46-55 and a thin empty 55 with retaining ring built in it . I'll get a link to that up. I actually think I can add a CPL with it. I just don't have anything in 55 to try. I'll take a shot of my setup and I'm going to count how many shims are removed as well, I hit infinity right at the hard stop with a Fotosy adapter.
It might be - at the moment it's only retaining ring I own, which will hold Proxar lens. Proxar diameter is 51.0mm. Elpro retaining ring inner diameter is 49.1mm, so it could be ~1.5mm larger in diameter and still keep Proxar in place. For example Contax UV filter retaining ring is 51.65mm and other 55mm filters I found from my boxes and drawers are about the same => they will not hold Proxar in place.
GMPhotography wrote:
I have my Hawk on my ZM 35 and another Fotosy on my ZM 85. These cheap adapters from Amazon are actually quite nice. So all three lenses have a adapter so I don't have to carry the Zeiss M rear caps with me. So everything is FE caps. Love this. I hate caps to begin with.
Most M-adapters are nice - much bigger quality differences in C/Y adapters for example. Few year ago purchased cheap ones have light leaks, but as I have strict "adapter for each lens"-policy that is no problem; I just add black tape to bottom of the lens/adapter. New ones I have recently received from eBay have been very good and no light leaks.
---------------------------------------------
GMPhotography wrote:
Okay first things first Shims. . Taken out of the front 2 Gold and 1 Bronze. I believe I have a silver in the front. On the rear i took out the Bronze left the rest in there. ZM 25
Based on these color definitions I had 3pcs "Gold" and 1pcs "Bronze", and I assume I can put one of them back and still reach infinity.
---------------------------------------------
artur5 wrote:
...if all these shims stacked are 0.71mm, according to theory that's enough for reaching inifnity with a +1 front element and any adapter that focuses normally up to infinity or a bit past it.
That was just one lens sample - the shims are installed by Zeiss on factory to match some tolerances => your lens may have different shims. Before attaching the filter and removing shims the lens did focus quite much over infinity, I forgot to write down (and my memory is worth maximum for some minutes...) where real infinity focus was on focus scale - between 5m and infinity for sure, but no idea was it closer to infinity or 5m. Next time I open the lens I'll add one of the shims back and see if it still can focus to infinity, now infinity is between 5m and infinity.
artur5 wrote:
I believe that the most convenient setup for using the ZM25 without additional filters is what Guy did: get a 46mm +1 Hoya HMC close up filter..Remove the glass from the mount (it's too thick and vignettes a bit with a 25mm lens). Relocate the glass to an empty ring from another 46mm. filter (UV or similar. Those come in thin mounts that won't vignette.). If possible, use a filter with a retaining ring system, as they're easy to remove and reinstall.
This way you can use the original hood on the lens. Step-up rings will not allow that, except maybe a 46-49 ...Show more →
Yes, I somehow missed this option - I assumed Hoya doesn't even make close-up filters on this rare filter size. However if also Hoya doesn't solve boke issues, only provides nice infinity performance, then I anyway need to try the plano convex lenses - which will cause need for similar mounting mechanism, and makes choice of retention rings even more narrow as OptoSigma lenses are 1mm smaller diameter than Proxar lenses. One option glue; attach 46-67mm step-up to lens and glue the lens to the step-up adapter. Then just add empty filter to attach filters to the 67mm side (depending on filter it will touch the plano convex lens, most of my filters would touch).
---------------------------------------------
polylepis wrote:
Effusive thanks Samuli for addressing most of my basic questions about modding the ZM25. It is just so helpful to have this information collected in one place. 24-28 is my preferred FL for landscapes and an excellent, compact lens has been elusive for use on the Sonys. The many knowledgeable contributors on this thread have really moved this discussion forward. Hats off to everyone!
And as you see from discussion the "article" is not ready yet. I'll update once I get 46mm Hoya +1 and OptoSigma 1500mm = in few weeks.
Here is the two setups I used with my ZM 25, regardless of the setup you will need to remove a number of front shims from the lens as mentioned throughout this thread in order to reach infinity, the number to remove will vary from lens to lens.
1: With the Proxar 1m I used a Fotodiox 46-55 step up ring, inserted and centered the proxar in that, next I added a Fotodiox 55-52 step down ring and inside of that ring a 52mm retaining ring.
Pluses: Does not vignet and allows you to use a front cap if you like
Negatives: You won't be able to use the oroginal lens hood with this setup.
2: With te Hoya HMC +1 filter the filter housing is actually too thick so you will experience vignetting without modifying the housing. There is two ways to approach this, you could remove the lens from it's original filter holder and install it on a thin 46mm filter housing that has a removeable retaining ring like Guy did or you could gring the original Hoya housing down by a couple of millimeters to "thin" it down to prevent the vignetting. I think Guys approach is easier. I took the more difficult approach because I didn't have a 46mm filter lying around, I ground one of the housings that came with the Hoya set and mounted the +1 Hoyal lens on it.
Pluses: Don't need to purchase additional filters to mount the lens, absolutely no vignetting at all and the biggest plus is you an now use the original Zeiss hood on this setup.
Negatives: The ground filter holder will prevent you from mounting an additional filter due to the missing filter threads and you won't be able to use a lens cap for the same reason. Here are some photos with and without the hood:
Hoya +1 with "thinned" housing DSC00127 by Jose Raposo, on Flickr
Note: in the photo above you'll notice that the hosing says +4 however the lens that is mounted in that housing is the Hoya +1, I damaged the original +1 housing during the removal of the +1 lens so I used the +4 housing to mount it.
artur5 wrote:
That depends a lot on how the adapter is built. There're ways to retighten the fit after shaving the front plate, if we don't exceed 0.2-0.3mm. On some models, it can be done enlarging those wedges made in the cams of the adapter that press the rear of the lens against the camera mount. Other adapters employ a leaf spring system to perform the same task. The springs allow for some leeway in the thickness of the front plate before the lens starts to wobble. The best brands have more margin than cheapish units with weaker springs.
In order to use it with a front filter modded Zeiss M lens, I bought recently a Fotassy adapter that uses the leaf spring system. I grinded it about 0.2-0.25mm. and there's no difference in the lens fit after the mod. No wobble at all.
Of course, instead of modifying the adapter, it's better to remove lens shims but sometimes this is very difficult or impossible.
If you get an adapter that's already very tight on the camera, shaving just a fraction should not make it wobbly. It all depends on the amount of shaving necessary to achieve infinity.
Yes, It's a close shave.
With the ZM25 +1 m front filter you need to remove around 0.6mm. so grinding a bit the adapter isn't enough, but it may help if your lens has fewer shims inside than Samuli's, for instance.
In the case of the ZM35 + PCX 5000, shaving 0.2mm might be enough in order to avoid removing shims when the user feels uncomfortable messing inside the lens (although it's very simple task if you have a spanner wrench)
Anyway, I advice to try the shim solution first. Only if the lens has no shims inside or it's very complicated to disassemble you should try grinding the adapter. It's a very tedious task. You need lots of patience and a good caliper. Not easy at all to achieve an even thickness around the ring. Only 0.02 mm. off and your corners will never look like Choppy's.
No way to overstate how cool this thread is, and how interesting. I hope at some point you guys do a cliff notes version, a synopsis and make it a sticky in the Sony forum.
It probably does not need to be said, and a well modified ZM lens should be worth more than a stock one: but it's critical to disclose attempts to modify any M lens in a sale. If some of these copies get out into the wild and it's not understood what has been done, it will waste the time and money of multiple photographers.
You guys might even consider a mark. Etch the mount ring or something. I don't worry much about the people posting here, but imagine once a lens has passed into multiple hands. A film shooter won't realize he has a calibration problem for weeks perhaps months. Then a return is problematic.
Please don't take this as discouragement. If I also shot natives on the Sony I would be doing these lens mods myself. What a great option to give many A7x shooters, please keep going
Well, the mods suggested here are not irreversible or damaging at all ( if properly done, of course).You can replace those shims any time you want and a careful test will assess if the lens works again as it should -i.e focusing exactly to infinity when there's no front filter mounted. I don't see the resale value dropping much for that reason. Personally, I'd never advertise such lens with terms as 'mint' or 'practically new' even if it was absolutely unmarked outside and working perfectly, but then I never do that with a used lens.
artur5 wrote:
That depends a lot on how the adapter is built. There're ways to retighten the fit after shaving the front plate, if we don't exceed 0.2-0.3mm. On some models, it can be done enlarging those wedges made in the cams of the adapter that press the rear of the lens against the camera mount. Other adapters employ a leaf spring system to perform the same task. The springs allow for some leeway in the thickness of the front plate before the lens starts to wobble. The best brands have more margin than cheapish units with weaker springs.
In order to use it with a front filter modded Zeiss M lens, I bought recently a Pixco adapter that uses the leaf spring system. I grinded it about 0.2-0.25mm. and there's no difference in the lens fit after the mod. No wobble at all.
Of course, instead of modifying the adapter, it's better to remove lens shims but sometimes this is very difficult or impossible.
Brilliant suggestion, Artur5. I expanded the springs on the underside of the front plate of the adapter, and now I can get a solid attachment to the lens. Photos looks much better than before. Tomorrow, I will try to run a proper comparison of front end modified Ultron 35 and ZM 35.
uhoh7 wrote:
No way to overstate how cool this thread is, and how interesting. I hope at some point you guys do a cliff notes version, a synopsis and make it a sticky in the Sony forum.
HaruhikoT has updated the original post, Phillip has article about this on his website. In addition I made webpage for this as it was impossible to follow what filter works with what lens, so I read whole thread and collected info about what works to page and now I update it when new information appears.
However this far all the effort has been testing infinity. As I do >95% of something else than stopped down infinity shooting I'm not yet very convinced yet about this technique. First tests I did (ZM25) indicated that yes infinity performance is good as reported here, but there are issues on other scenarios. I don't know is it somehow on the setup or in the methodology.
If this technique is just for landcape shooting, I'm not sure is sticky justified for this. For landscape shooting set of Loxias is much easier solution (well, depends on how lucky one gets on lens lottery...might be pretty difficult to find symmetrical 21 if you are picky). I would not choose these "character lenses" if I would primarily shoot landscape. Your case is pretty unique as you shoot both systems.
PS. Thanks for the Kolari ultrathin filter testing & reporting, very interesting. I might consider it, but I'm not very keen on repeating the disaster with shipping companies and Finnish customs (of course all went fine with Kolari) - it might actually be cheaper and easier next time just to fly to NYC, rent a car and fetch the camera from Kolari...
uhoh7 wrote:
It probably does not need to be said, and a well modified ZM lens should be worth more than a stock one: but it's critical to disclose attempts to modify any M lens in a sale. If some of these copies get out into the wild and it's not understood what has been done, it will waste the time and money of multiple photographers.
You guys might even consider a mark. Etch the mount ring or something. I don't worry much about the people posting here, but imagine once a lens has passed into multiple hands. A film shooter won't realize he has a calibration problem for weeks perhaps months. Then a return is problematic....Show more →
Atleast here in Finland it's pretty difficult to get same money from modified lens than you get for non-modified. I can't imagine if somebody would pay more for it - unless there would be some real labor intensive modification e.g. similar what Derek has done to Contax G-lenses (moved them to other lens body and give them real helicoid focusing etc.). Or if someone has converted dead system lens (e.g Contax N, Contax 645) to some current system (e.g. Sony FE). I'm going to put the shims back to lenses if I ever sell these lenses.
"Modification" mentioned here is to adjust the focus with shims as Zeiss has intended them to be used. Of course there is risk that people skrew up and adjust some shims, which are not supposed to be adjusted - e.g. the easiest accessible shims in 2/35 adjust rear lens group distance from front group = no effect to infinity position (or atleast I didn't notice any). I can't imagine any rangefinder shooter who would not calibrate the secondhand lens to camera body before start shooting. It's also possible that Zeiss has adjusted lens to specific Zeiss Icon body, and lens doesn't have enough shims in place to make it work for specific Leica (=nothing to do with this "modification"). Of the ZM lenses I have experience this far 2/35 didn't focus to infinity (it was not much short, focused to ~20 meters, you can barely see that it doesn't focus to "infinity"), 2.8/25 goes over infinity and 4/18 stops exactly at infinity when tested with same adapter and camera body - I hope that rangefinder shooters don't assume that they don't need to calibrate or they shoot infinity f/8 stuff only, where it's not so accurate do you focus little back or forth.
Samuli Vahonen wrote:
However this far all the effort has been testing infinity. As I do >95% of something else than stopped down infinity shooting I'm not yet very convinced yet about this technique. First tests I did (ZM25) indicated that yes infinity performance is good as reported here, but there are issues on other scenarios. I don't know is it somehow on the setup or in the methodology.
Nice finding someone as skeptical as me.
I have run tests at different distances with my ZM 35mm 1.4 + 5m PCX and found the behaviour pretty similar, regardless of distance. Same goes for the VM 35mm 1.7.
But I also can't say these solutions are perfect for every scenario:
On the ZM 35 a midzone dip is introduced (or better: exaggerated) which I think the floating elements design is the reason for
(if you set the lens to infinity and focus by a helicoid adapter the midzone dip gets worse, so I assume the change in flange distance by removal of shims is to blame).
On the VM 35 the midzone dip is hardly noticeable, but the corner improvement is also not as dramatically.
Still: for what it is the front filter solution can yield pretty astonishing results and that filter won't be taken off of my ZM 35.
Bokeh issues are also resolved with this filter.
one issue i've found with the g21 and reversed pcx 1.5m filter is that i seem to get central degradation at smaller apertures (only with the filter reversed). i suspect it's a kind of hot spot effect, need to check if it goes away with less light directed at the lens. doesn't seem to be present with the g28 and the same filter.
BastianK wrote:
Bokeh issues are also resolved with this filter.
Could you elaborate on this comment? From Haruhiko's 2 meter and 4 meter photos, I don't see much difference in field curvature (exists with/without PCX) as the focus sharpens going towards the edges and corners.
JimBuchanan wrote:
Could you elaborate on this comment? From Haruhiko's 2 meter and 4 meter photos, I don't see much difference in field curvature (exists with/without PCX) as the focus sharpens going towards the edges and corners.
Please have a look at these shots from my upcoming 35mm comparison: https://www.flickr.com/photos/140854680@N04/sets/72157680901728016/with/33104186551/
These shots were taken at the most critical conditions regarding the influence of field curvature on bokeh: focus near border, focusing distance ~6 m.
You can choose either pair, FLE, VM or ZM, outcome is always the same, you might see the Loxia 35mm 2.0 as reference, as it only has very minor field curvature.
I will now quote myself from this post:
"It [the front end filter] will help and it is actually the main reason I was initially interested in these front end filter solution.
BUT you may not get the results in the end that you are hoping for.
Most fast rangefinder lenses have severe vignetting (especially compared to the DSLR counterparts)
which will also effect sharpness (more) and blur (less) towards the borders.
Furthermode with the subject (and focus) near the borders (where the field curvature has significant influence)
you may indeed end up with a little more blur towards the borders but in fact less blur in the center of the frame when having mounted such a filter."
Corners/borders would be more out of focus with a 5m PCX filter, but keep in mind the limitations described above, they won't be as blurred as the center, technically impossible.
Samuli Vahonen wrote:
PS. Thanks for the Kolari ultrathin filter testing & reporting, very interesting. I might consider it, but I'm not very keen on repeating the disaster with shipping companies and Finnish customs (of course all went fine with Kolari) - it might actually be cheaper and easier next time just to fly to NYC, rent a car and fetch the camera from Kolari...
Samuli
You might be interested to know that now Kolari offers a free duty service for European customers. They have a partner located in in the EU, where you send your camera, thus avoiding to pay VAT and Customs fees. Check Kolari website.
sebboh wrote:
one issue i've found with the g21 and reversed pcx 1.5m filter is that i seem to get central degradation at smaller apertures (only with the filter reversed). i suspect it's a kind of hot spot effect, need to check if it goes away with less light directed at the lens. doesn't seem to be present with the g28 and the same filter.
......
.
Your tests with the G28, confirms my findings with this lens and a +1 meniscus filter. Better than the naked lens, but nowhere as good as what we achieve with a PCX element.
artur5 wrote:
You might be interested to know that now Kolari offers a free duty service for European customers. They have a partner located in in the EU, where you send your camera, thus avoiding to pay VAT and Customs fees. Check Kolari website.
Damn I was one year too early, I should have modified the two A7 cameras this year... I'll have to check it as the ultra thin filter Charlie (uhoh7) has been testing seems really interesting. Thou it may take some time before ultra thin filter becomes available via the European partner: "Right now this offer is only available on DSLR and mirrorless conversions with the AR coating option." as the ultra thin is not even on webshop in US.
------------------------------------------
BastianK wrote:
I will now quote myself from this post:
"It [the front end filter] will help and it is actually the main reason I was initially interested in these front end filter solution.
BUT you may not get the results in the end that you are hoping for.
Most fast rangefinder lenses have severe vignetting (especially compared to the DSLR counterparts)
which will also effect sharpness (more) and blur (less) towards the borders.
Furthermode with the subject (and focus) near the borders (where the field curvature has significant influence)
you may indeed end up with a little more blur towards the borders but in fact less blur in the center of the frame when having mounted such a filter."...Show more →
Yes, this is obvious - specially lens like Carl Zeiss Distagon T* 1.4/35 ZM. I don't think anyone can assume it to never reach uniform boke quantity all over frame because of the gigantic vignetting:
f/1.4 at ~17mm we have 4 stops vignetting
f/4 at ~20mm we have 1 stop vignetting
Vignetting "stops" don't directly translate to "stops" more in DOF, but there is at least 2-3 stops more DOF in corner wide open. So it will not have uniform quantity of boke even with film or Leica sensor.
I'm more worried that the plane of high contrast curves towards infinity. Area being in DOF doesn't automatically mean there is high contrast, sharpness yes. This doesn't happen on film or with some of the Leica sensors. And if front filter really fixes field curvature it should fix this.
There is something evil about Sony sensors - even with Kolari v2 I get "easy" lenses like Carl Zeiss Distagon T* 2.8/28 C/Y to have worse field curvature than it did in Canon 5DmkII (or 1DsmkI/II). Kolari v2 for sure is much better than standard thick filter stack and lens is usable for most of my boke photos. So I doubt the front filter or other solution can fix the issue, but they maybe make the issue smaller. Generally I don't shoot wide open, unless shooting aperture series to get to know how new lens behaves. When shooting seriously I'm always 1-3 stops down to mitigate vignetting caused DOF issues, mitigate boke quality issues and to increase focus plane contrast.
The rendering style offered by Contax C/Y, Contax G and Zeiss ZM is so tempting that I'm willing to make some compromises, and don't expect perfection. Personally I feel that the new lenses (Otus, Milvus, Batis, Loxia, Sigma Art etc.) offer rendering style I'm interested only for landscape photography - they all seem to have "high contrast in boke"-style, which I don't like as much as the "old style" Zeiss used to have.
Sample of outer corner weirdness, which was not there with Canon sensors or film. Bottom corners focus somewhere under the ground, and even I did my best on composition to avoid it to be visible on top corners, I failed a little on placement of top right corner; the holes in the spruce where sky light coming through are in focus!! I do not assume corners to be sharp and nice, but focus curving from few meters to tens of meters drives me insane. In many scenes I feel cropping to 4:3/16:9 just to avoid the outer corner weirdness, or compose so that corners are against background (like tried in this picture) which will not show ultracrisp tree tops tens of meters behind the intended focus plane. - Kolari v2, D2.8/28@f/4:
Samuli Vahonen wrote:
Damn I was one year too early, I should have modified the two A7 cameras this year... I'll have to check it as the ultra thin filter Charlie (uhoh7) has been testing seems really interesting. Thou it may take some time before ultra thin filter becomes available via the European partner: "Right now this offer is only available on DSLR and mirrorless conversions with the AR coating option." as the ultra thin is not even on webshop in US.
Samuli
Yeah, a bit early for taking the plunge. From this info it's not clear if the AR coating option applies also to the thin filter mod or only to IR/Full spectrum conversions. Also, if the partner does the conversion or they simply are an "intermediate station" between Kolari and European customers. I gather that they might do it by themselves. If Kolari performs the task, that restriction for only " AR option " makes no sense.
Last Friday I sent a message to them asking some details about the possibility of doing a Full Spectrum conversion using a ultra thin clear glass. As this glass doesn't need to filter IR or UV, my uneducated guess is that it can be very thin, but I'm no expert so, let's see what they say ( hopefully in a couple of days).
Today we published my 35mm comparison, I don't want to discuss it here in detail as this thread is already loaded with information,
but there are many with/without filter comparisons at different distances for Zeiss ZM 35mm 1.4, Voigtlander VM 35mm 1.7 and Leica Summilux 35mm 1.4 Asph FLE included,
which I think might be of interest for some of you. https://phillipreeve.net/blog/35mm-comparison-voigtlander-zeiss-leica/
BastianK wrote:
Today we published my 35mm comparison, I don't want to discuss it here in detail as this thread is already loaded with information,
but there are many with/without filter comparisons at different distances for Zeiss ZM 35mm 1.4, Voigtlander VM 35mm 1.7 and Leica Summilux 35mm 1.4 Asph FLE included,
which I think might be of interest for some of you. https://phillipreeve.net/blog/35mm-comparison-voigtlander-zeiss-leica/
Bastian,
Have you tried the PCX 5m in reverse orientation (convex facing the camera)? (ZM 35/1.4)
There are some reports that it yields better performance when the regular way (convex facing the subject) but I can't test it right now because of the weather.
Let me know,
Fred
Fred Miranda wrote:
Bastian,
Have you tried the PCX 5m in reverse orientation (convex facing the camera)? (ZM 35/1.4)
There are some reports that it yields better performance when the regular way (convex facing the subject) but I can't test it right now because of the weather.
Let me know,
Fred
Dear Fred,
I tested it a few weeks ago with my at that time decentered ZM, but I couldn't see meaningful differences.
I only tested at ~3m distance though, as I did not want to change the orientation "in the field",
as I am not sure if I can guarantee a good centering when aligning the filter on a bench in the park.
BastianK wrote:
Dear Fred,
I tested it a few weeks ago with my at that time decentered ZM, but I couldn't see meaningful differences.
I only tested at ~3m distance though, as I did not want to change the orientation "in the field",
as I am not sure if I can guarantee a good centering when aligning the filter on a bench in the park.
True. I have the ability to test this at home (backyard). I just need to make sure both ways are perfectly centered so the only variable is the PCX orientation.. Hopefully the weather will cooperate.
