While I am still madly in love with my Q2M, I occasionally think about a FF camera with interchangeable lenses. The M10M/M11M would fit the bill if it weren't for their lousy IR response. So, what are my options? Am I right in assuming that I have to get a color camera and then have to get it modified for mono and IR?
And what are my options regarding lenses? I am willing to sacrifice autofocus, but I do want a 28+90 mm combo with zero hot spots and no corner smearing. Is there a database that shows which lenses work on which mono/IR-converted sensors? Does the removal of the thick Sony filter stack automatically guarantee that classic RF glass will NOT suffer from corner smearing?
That said, much of it is dependent on your type of conversion. My bare sensor conversion isn't prone to hotspots at all so long as I'm using a front IR filter (I do still get reflection problems with certain lenses, though, like the Nikon AF 24/2.8D, where the reflections are internal to the lens). And I have used the cheapest filters, including Fotga's 550-750 variable filter you can pick up off Amazon for $25, with fantastic results across multiple lenses. I've tested the combo with some of the toughest film-era lenses for digital use (Biogon 21/4.5, the first version Elmarit 21/2.8 with the .5m MFD, Biogon 28/2.8, Elmarit 28/2.8 vIV, and Summicron 28/2 vI) with results that match their film performance on an M7. Getting the best out of film-era lenses was the entire point of the type of conversion I chose, and I haven't been disappointed.
If you plan to go with a conversion that puts the IR filter in the camera or use clip-in filters, that's when you need to start looking at hotspots. Unfortunately, you're going to be hard-pressed to find broad M-mount testing for those specific filter arrangements, so there will likely be some testing necessary on your part to find the best 28/90 combo available.
I have an IR-converted Canon 5D MkII which I use less and less these days after I vested into a Leica M 246 monochrome camera. This camera is suitable for good IR photography with simple R72 filter attached to the lens. I can handheld take IR photos with this setup using higher ISO between 1600 and 3200 which the M 246 can easily handle without much noise creation.
M lenses have from my experience two advantages for IR photography, too: First, there is barely any focus shift with M lenses. This is different from other lenses I have used in the past for IR photography where the focus needed to be readjusted in IR light (either in LiveView focusing or by adjusting on the red IR mark on the lens distance meter). Second is that I have not seen a M lens yet in my experience which does not work in IR light for example by creating hot spots - also much more common with other lens types.
More modern cameras might use a stronger IR absorbing layer on top of the sensor which makes it more difficult to use for IR photography. Not sure if this is still the case in Leica monochrome cameras where the Bayer pattern is removed. In case of IR absorbing layer present on top of the sensor, older digital cameras have an advantage since this layer is thinner and allows more IR light to pass through (shorter exposure times, lower ISO, more whiteish foliage instead of grey in the final B&W IR image).
My understanding is that the M10M and M11M do not have much of an IR response (though I am having trouble finding the link right now), but I know for a fact that my Q2M is practically blind with an R72 filter. Interesting to see the M246 is working fine with an R72.
I guess one advantage with staying with Leica is that corner smearing will be a non-issue, so I can focus on finding hot-spot free lenses. I don't mind testing a couple of lenses ... odds are I would start with lenses that are readily available on the used market.
theHUN wrote:
My understanding is that the M10M and M11M do not have much of an IR response (though I am having trouble finding the link right now), but I know for a fact that my Q2M is practically blind with an R72 filter. Interesting to see the M246 is working fine with an R72.
I guess one advantage with staying with Leica is that corner smearing will be a non-issue, so I can focus on finding hot-spot free lenses. I don't mind testing a couple of lenses ... odds are I would start with lenses that are readily available on the used market....Show more →
Thanks for adding here - the fact that also the Q2M is "blind" with R72 filter on the lens indicates that my assumption was correct that there is a stronger IR-cut off filter still used on top of the sensor. I am a bit surprised by this since these kind of cut-off filters make sense for color photography but not for pure B&W - maybe to ensure that dark foliage appears also dark and not a bit lighter due to passing IR light maybe? Some B&W film is also receptive to lower wavelength IR which looks actually quite appealing IMO in the final B&W photo.
I haven't encountered any M lens with hot spot issue so far - maybe I am just lucky with my lens choices? Some older LTM lenses also work, but some have a red IR dot to adjust the focus manually a bit. But some of my LTM lenses also performed great in IR light. Maybe this is because of a smaller lens array in rangefinder lenses leaving less space inside a lens to potentially give IR light reflections between lens elements opposite to SLR and DSLR lenses?
rollei35_warton wrote:
dont do anything on your M bodies.
Just get a modified DSLR or even cheap mirrorless like fuji x-t30, you would be much better off.
I have a nikon D810 fully converted to full spectrum, and use some old lenses like DC 105f/2 and 35/1.4D, very happy about the result.
Leica M is just not built for IR purpose.
I do not mind vintage glass, but I was hoping for a modern (40+ MP) full frame sensor. Which options guarantee no hot spots and no corner smearing with a 28 mm lens?
My thinking was that if an M monochrome were to have its IR filter removed, then the odds would be high that a native 28 mm would not smear the corners. But it is an assumption, and perhaps I am dead wrong.
rollei35_warton wrote:
Leica M is just not built for IR purpose.
Actually Leica M has a big advantage for IR photography no other camera has - the rangefinder. You can have a pitch black R72 filter on your M lens mounted to a M camera and still compose and focus through the separate viewfinder. With other cameras you need to apply either LiveView for accurate focusing, have a mirrorless camera modified for IR photography to see the image through the EVF as such, or you have a (D)SLR without LiveView where you need to compose before the IR filter is attached to the lens, and then you need to adjust the lens focus for IR light which is a hit or miss to get it right.
IR photography works decently on older unmodified digital color based M cameras, but it requires longer exposure time and higher ISO setting. Better option with at least 1-2 stops improvement over the color M sensor is with monochrome M cameras. That no lens focus correction is needed with M lenses in IR light is another advantage (even when faster f-stops are used!).
An example how "bad" unmodified Leica M does is shown below.... That IR photography is not commonly practiced with M cameras doesn't mean they are not made to use for this purpose.
retrofocus wrote:
Actually Leica M has a big advantage for IR photography no other camera has - the rangefinder. You can have a pitch black R72 filter on your M lens mounted to a M camera and still compose and focus through the separate viewfinder. With other cameras you need to apply either LiveView for accurate focusing, have a mirrorless camera modified for IR photography to see the image through the EVF as such, or you have a (D)SLR without LiveView where you need to compose before the IR filter is attached to the lens, and then you need to adjust the lens focus for IR light which is a hit or miss to get it right.
IR photography works decently on older unmodified digital color based M cameras, but it requires longer exposure time and higher ISO setting. Better option with at least 1-2 stops improvement over the color M sensor is with monochrome M cameras. That no lens focus correction is needed with M lenses in IR light is another advantage (even when faster f-stops are used!).
An example how "bad" unmodified Leica M does is shown below.... That IR photography is not commonly practiced with M cameras doesn't mean they are not made to use for this purpose.
Theoretically, a camera sensor sees much more infrared light than visible light
So its best to get a full spectrum conversion than to pay the premium for a “monochrome” cam
Although the M Monochrome pics look stunning, the effect is more drastic in the higher nanometer IR range.
So best to take off that filter on the sensor and use dedicated filters (i have a kolari converted A7iii with clip in filters - just note that the hot mirror kolari clip in sucks a lot of light bumping your ISO if you ever want to use it in the visible light spectrum - this would be more extreme on a higher resolution sensor because it needs more light - my non-converted A7Riii significantly bumps the ISO when adding filters on top of the already stacked sensor)
850nanometer and above (i.e. 900nm) filter works best for B/W because theres hardly any (visible) color left after you do the white balance flip (shoot in raw, many cameras cant get low enough on the WB to adjust in camera - also Capture One can get your whitebalance lower than other apps in a simpler way)
For lenses Ive found the Sony 55/1.8 and Tamron 28-200 to work flawlessly without hotspots.
Kolari has a hot spot killer (HSK) filter that does suck a lot of light also. Also all my old Takumars work like a charm in IR.
Most of my CV manual Focus lenses (both M and E mount) work fine up until about f/8 or slightly below regarding hotspots.
Also: adding IR Filters to stock camera sensors is like using a STRONG ND filter. Any IR Filter on a full spectrum camera sensor sucks up much less light - so youll be able to work easy in liveview with very short hand-holdable shutter speeds.
freaklikeme wrote:
It surprises me that you're finding M lenses don't require focus correction for IR. That is not the case on my set up.
The M lenses I have used so far with R72 filter attached and which did not require any focus adjustment in IR:
Leica 50/2.0 Vers. V
Leica 35/2.0 Vers. IV
Leica 90/2.8 Vers. I
CV 75/1.5
I haven't tested others since I only have Hoya R72 filters in 39 and 58 mm diameter size. Other M lenses I own require different filter sizes, and I haven't gone there yet to vest in additional filters. Which of your tested M lenses required focus adjustment in IR?
Mr Leica also has a video on YT comparing the M11 and M11M where he says there is little difference once the M11 files are converted to monochrome.
I don't want a color camera. I want a monochrome with its IR filter removed so I can slap on an R72 (or something even wider in the IR). Anyone claiming there is little difference between a color sensor and a native B/W sensor is not helping themselves or others.
retrofocus wrote:
The M lenses I have used so far with R72 filter attached and which did not require any focus adjustment in IR:
Leica 50/2.0 Vers. V
Leica 35/2.0 Vers. IV
Leica 90/2.8 Vers. I
CV 75/1.5
I haven't tested others since I only have Hoya R72 filters in 39 and 58 mm diameter size. Other M lenses I own require different filter sizes, and I haven't gone there yet to vest in additional filters. Which of your tested M lenses required focus adjustment in IR?
I've only tested for differences between a UV/IR cut filter (B+W 486) and two IR filters (Coklin 720 and 850) on the ones we own. So, the three in my kit:
Biogon 25/2.8
Biogon 35/2
Ultron 75/1.9
That's where I planned to stop, but then I got obsessed and had to test Rolf's kit:
What I found, in general, was that the below 50s with the 720 filter at landscape distances and stopped down to f/4 or slower can be set to hyperfocal without much difference in the results. There's enough visible light coming in to soften the results regardless, so shifting focus slightly has the most visible impact on foreground elements. Focused at full-body portrait or closer distances, the more important precision becomes. From 50 on with the 720, the longer it gets, the more important precision becomes at any distance.
With the 850 filter, things get more brutal so far as focus precision if you want sharp results. It's never much of a shift, but it is important. Additional notes from the testing; the Zeisses and the Cron 28 all have minor central hotspots that don't really become problems unless I'm using a strong filter like the 850 and stopping down beyond f/5.6. Otherwise, the hotspots just blend into the already considerable vignetting.
I did all this before we acquired the APO-Lanthar 28 and pre-A Summicron 90 and before I started using the variable filter regularly, so now I have an itch in the back of my brain to redo the whole thing. I'm trying to ignore it, though, because my standard IR kit these days is the Canon 24-70/4L IS, Minolta APO 80-200/2.8, and Minolta MD 100/4 Macro, or, alternately, some trio of Minolta primes when I don't want to carry the big heavy zooms. I have to remind myself that not all itches need to be scratched.
theHUN wrote:
Word on the street is that the IR filter on the M10M and M11M uses a more permanent glue than other cameras, hence the lack of mod options.
Looks like I will go the Sony route, simply because I already have an a7r3 that spends most of its time collecting dust.
I wouldn't argue against the rIII at this point, having used the rII for years now and never having shot it in a way that would trigger the problem, but you should know...
"WARNING:? This camera contains an internal diagnostic IR Led. This can cause light leaks when using very long exposures and high ISOs in the H-alpha and IR enabled conversions, in the range of 30 second exposures at 6,400 ISO."
That comes from Kolari, but I think MaxMax or Lifepixel also did a write up on it.
freaklikeme wrote:
I wouldn't argue against the rIII at this point, having used the rII for years now and never having shot it in a way that would trigger the problem, but you should know...
"WARNING:? This camera contains an internal diagnostic IR Led. This can cause light leaks when using very long exposures and high ISOs in the H-alpha and IR enabled conversions, in the range of 30 second exposures at 6,400 ISO."
That comes from Kolari, but I think MaxMax or Lifepixel also did a write up on it.
theHUN wrote:
So, what are my options? Am I right in assuming that I have to get a color camera and then have to get it modified for mono and IR?
A little off the beaten path, but the Sigma Foveon and Quattro cameras are *excellent* for monochrome work, even if you shoot in color and do the mono conversion in Sigma Photo Pro.
Further along that path, these cameras have a removable dust protector, and now you are in "full spectrum". Do note that a replacement dust protector (hot mirror) isn't exactly easy to get if you break one, and that a typical Bayer sensor hot mirror will NOT work properly.
The SD1M does well with a 580 nm filter or a 720 nm filter.
The Quattro (SDQ) (IMHO) does not do well with anything other than a 580 nm filter or in full spectrum mode, likely due to the different sensor type compared to Merrill. However, the Quattro has live view and the SD1M does not.
The elephants in the room are the Sigma software, generally poor autofocus on the SD1M and spotty to poor autofocus from the SDQ unless one is using Sigma Global Vision lenses (can be spotty with GV, poor with others like non macro EX lenses). Both cameras use the SA mount, which is discontinued.
Both can be very frustrating to use, but when it works, it works.
These are not Bayer sensor cameras, so the IR false color thing as shown on IR conversion websites is a whole different ball of wax.
None of the Sigma Foveon or Quattro cameras are "full frame". Closest would be the SDQ-H. However, IMHO, a Sigma SD1M or Quattro can stand up to a Canon 5DsR.
The newer Sigma cameras have Bayer sensors (fp; fp-l; bf) and would require conversion.
Update:
I picked up a second hand A7R4A, and have sent it out for monochrome + full spectrum conversion. I have also picked up a Voigtlander 35mm f/1.4 II SC, 35/2 Biogon, and Voigtlander 28/2 Apo. I hope one of them performs well enough at IR and visible. Will report back. As for the 90 mm option, I will try my existing Hasselblad V 100/3.5 and 110/2 first. If they are truly awful I will give the Voigtlander 90/2 Apo a try.
theHUN wrote:
Update:
I picked up a second hand A7R4A, and have sent it out for monochrome + full spectrum conversion. I have also picked up a Voigtlander 35mm f/1.4 II SC, 35/2 Biogon, and Voigtlander 28/2 Apo. I hope one of them performs well enough at IR and visible. Will report back. As for the 90 mm option, I will try my existing Hasselblad V 100/3.5 and 110/2 first. If they are truly awful I will give the Voigtlander 90/2 Apo a try.
Nice. I'm a bit jealous of your EVF. I might have to look at the rIV when my rII gives up the ghost.
The Loxia 85, Minolta MD Tele-Rokkor 100/2.5, and MD 100/4 macro have worked out well for me at those focal lengths. I've got the CY Sonnar 100/3.5 on my list of lenses to check out when I come across one at the right price.
. That IR photography is not commonly practiced with M cameras doesn't mean they are not made to use for this purpose. 
