Dante_Stella
Rex canum cattorumque
I was thinking about converting my old X100 to full-spectrum (UV, visible, IR) (yes, first-world problems) - and I had a few questions:
1. Has anyone actually done this? Any issues? I'd probably go with LDP because I'd rather pay more than deal with dust problems.
2. Can anyone comment on B+W 092 and 093 filters on a converted camera and in particular, how the 092 compares to a 720nm filter? And can you use the EVF/contrast detect AF with the 093?
3. In terms of front filters for visible light, I have identified two possibilities. One is something based on the Schott BG39 (ionically colored glass, abosrption type), but it doesn't really block UV - and it is susceptible to rare earth corrosion. I also can't tell if the Heliopan versions of this are coated (which could mitigate the corrosion issue).
The other is a B+W 486 (interference type, MRC coated), which in my testing doesn't seem to lead to shaded corners on the X100 - without any filter on the stock camera, I see a slight color shift (1-2% toward red and green in the corners); with, I see just the opposite (about 1% toward blue and green). This seems to mesh with the specs on interference filters, which generally seem to indicate that a 35mm FOV should be within bounds. It's also consistent with my use of 486s on the M8 with a 28mm (no ill effects there). Anyone have a contrary result?
Thanks,
Dante
1. Has anyone actually done this? Any issues? I'd probably go with LDP because I'd rather pay more than deal with dust problems.
2. Can anyone comment on B+W 092 and 093 filters on a converted camera and in particular, how the 092 compares to a 720nm filter? And can you use the EVF/contrast detect AF with the 093?
3. In terms of front filters for visible light, I have identified two possibilities. One is something based on the Schott BG39 (ionically colored glass, abosrption type), but it doesn't really block UV - and it is susceptible to rare earth corrosion. I also can't tell if the Heliopan versions of this are coated (which could mitigate the corrosion issue).
The other is a B+W 486 (interference type, MRC coated), which in my testing doesn't seem to lead to shaded corners on the X100 - without any filter on the stock camera, I see a slight color shift (1-2% toward red and green in the corners); with, I see just the opposite (about 1% toward blue and green). This seems to mesh with the specs on interference filters, which generally seem to indicate that a 35mm FOV should be within bounds. It's also consistent with my use of 486s on the M8 with a 28mm (no ill effects there). Anyone have a contrary result?
Thanks,
Dante
GaryLH
Veteran
Interesting idea on the x100. Please keep us informed on your results.
Not sure if. Noticed my thread about my new to me Olympus epl1 concerted to ir.
http://www.rangefinderforum.com/forums/showthread.php?t=138502
Post number 11 in that thread I tried different ir filters on my already converted ir camera. If I had to do over again I would get a 650nm converted camera and use 720 and 830 nm filters..
Keith in that same thread basically is doing the same thing w/ his 650nm converted camera.
I have not had opportunity to try any of the filters u mentioned.. I find the Hoya ones good enough for my needs. the b&w mrc and heliopan ones are my preferred filters, but since I was experimenting and not knowing if ir was something I was going to stick w/, the Hoya one was my good enough for me. I actually bought it for my unconverted x100 about a year ago to c how the x100 would do.. I used it initially on my camera to check the accuracy of the conversion. There was just enough of a difference between my 715nm and the 720nm filter that some shots u can c the difference.
Gary
Not sure if. Noticed my thread about my new to me Olympus epl1 concerted to ir.
http://www.rangefinderforum.com/forums/showthread.php?t=138502
Post number 11 in that thread I tried different ir filters on my already converted ir camera. If I had to do over again I would get a 650nm converted camera and use 720 and 830 nm filters..
Keith in that same thread basically is doing the same thing w/ his 650nm converted camera.
I have not had opportunity to try any of the filters u mentioned.. I find the Hoya ones good enough for my needs. the b&w mrc and heliopan ones are my preferred filters, but since I was experimenting and not knowing if ir was something I was going to stick w/, the Hoya one was my good enough for me. I actually bought it for my unconverted x100 about a year ago to c how the x100 would do.. I used it initially on my camera to check the accuracy of the conversion. There was just enough of a difference between my 715nm and the 720nm filter that some shots u can c the difference.
Gary
Overkill-F1
Established
For some reason, my attempts to include a link will not work.
There is a successful conversion discussed at Fuji X-Forums.
It is in the 'X100 and X100S Discussions'
fujix.com
...Terry
There is a successful conversion discussed at Fuji X-Forums.
It is in the 'X100 and X100S Discussions'
fujix.com
...Terry
Dante_Stella
Rex canum cattorumque
I am going to save most of my comments for my site's writeup on this, but here are some preliminary points after taking a day with the converted cameras (and a snowy day).
1. This was incredibly expensive to convert at LDP. It was $550. It's not cheap at Kolari either, but it really depends on how fixated you are on speed and sensor cleanliness (the X100 has three dust surfaces where most cameras would have one).
2. Full spectrum is the only way you would ever want to go. Contrary to what I have seen theorized on the web, you can't stack an 830nm filter over a 590 or 720 on the sensor without losing exposure or getting unexpected cut points.
3. You would never want to go to 590nm on an IR-only conversion because you are trying to AF on way too broad a range of light. Not sharp. Even at 720, things are much less sharp than at 830. You get a taste for this on full-spectrum cameras by shooting without a filter. You need to use at least a 486 with visible light, if not also a colored filter to help the camera white-balance (with no filter at all, the IR contamination in the other channels is sufficently strong in the Fuji that things come out greener than you would expect - on other cameras, it only seems to accentuate red). That said, the X100's white balance is versatile enough that using the cyan shift, you can get presentable visible light pictures without an aqua filter.
4. Shooting IR in incandescent light gives you a lot of bang for the lumen (you can shoot 830nm in what looks like very low light), but it can be really noisy. Part of it is that the camera gets up to some high ISOs, but the other part is that RAW converters try to reconstruct color data that actually isn't there (at 830nm, the Bayer filter is completely bypassed). Some new converters are in the pipeline that should rectify this.
5. As a platform, the X100 is actually ideal for IR; the framelines in the glass VF dim significantly where the camera thinks there is low light (i.e., little IR light when you use an IR filter). The EVF is useful for checking out tonal relationships (and whether you want to exposure comp up or down). The X100s would actually be a hot mess - because it would be trying to reconcile phase detect AF designed for visible light with contrast-detect AF that can handle IR.
The X100 really shows this paradox: a camera with a good enough sensor to make IR worthwhile is also one that makes for an expensive project.
Dante
1. This was incredibly expensive to convert at LDP. It was $550. It's not cheap at Kolari either, but it really depends on how fixated you are on speed and sensor cleanliness (the X100 has three dust surfaces where most cameras would have one).
2. Full spectrum is the only way you would ever want to go. Contrary to what I have seen theorized on the web, you can't stack an 830nm filter over a 590 or 720 on the sensor without losing exposure or getting unexpected cut points.
3. You would never want to go to 590nm on an IR-only conversion because you are trying to AF on way too broad a range of light. Not sharp. Even at 720, things are much less sharp than at 830. You get a taste for this on full-spectrum cameras by shooting without a filter. You need to use at least a 486 with visible light, if not also a colored filter to help the camera white-balance (with no filter at all, the IR contamination in the other channels is sufficently strong in the Fuji that things come out greener than you would expect - on other cameras, it only seems to accentuate red). That said, the X100's white balance is versatile enough that using the cyan shift, you can get presentable visible light pictures without an aqua filter.
4. Shooting IR in incandescent light gives you a lot of bang for the lumen (you can shoot 830nm in what looks like very low light), but it can be really noisy. Part of it is that the camera gets up to some high ISOs, but the other part is that RAW converters try to reconstruct color data that actually isn't there (at 830nm, the Bayer filter is completely bypassed). Some new converters are in the pipeline that should rectify this.
5. As a platform, the X100 is actually ideal for IR; the framelines in the glass VF dim significantly where the camera thinks there is low light (i.e., little IR light when you use an IR filter). The EVF is useful for checking out tonal relationships (and whether you want to exposure comp up or down). The X100s would actually be a hot mess - because it would be trying to reconcile phase detect AF designed for visible light with contrast-detect AF that can handle IR.
The X100 really shows this paradox: a camera with a good enough sensor to make IR worthwhile is also one that makes for an expensive project.
Dante
Dante_Stella
Rex canum cattorumque
GaryLH
Veteran
It was a nice writeup Dante. Love the shots and examples.
Gary
Gary
Dante_Stella
Rex canum cattorumque
In the next few days, I should be able to get the 040, 090, 091, 092 and 093 filters tested simultaneously. Also, I have managed to construct an infrared flash out of an SB-800. One problem I have not yet been able to conquer is the false noise in RAW conversion (this results from a true monochrome transmission through the Bayer filter being misinterpreted by the camera). Stay tuned.
Dante
Dante