The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Infrared photography on the 'standard' IQ4 150mp

beano_z

Active member
I did search around on the internet but so far without good solid result. I was wondering whether anyone has tried to shoot infrared on their 'normal' IQ4 150 digital backs? Can the live view still be used or is that absolutely out of the question?

I have an IR720 filter on the way, which is what I'll start off with and I'm a complete newbie to this all, so any advice is more than welcome.

I'm planning to trying this with the XF and assorted SK lenses and I've read all sorts of scary comments on extremely long exposures, hotspots, focus shifts, etc. Of course, I'm not expecting to shoot this combination handheld, long exposures are absolutely fine with me.

I'm no dedicated infrared shooter and want to use the filter just once in a while when the conditions allow.

Thanks!
 

anwarp

Well-known member
I have an IQ 260 I had converted to full spectrum by P1. No live view, so I used a sunny day to calibrate my main lenses for the IR shift. I reload the system settings from two sets of cards for the 2 backs.
I do see hotspots (slight colour cast) with the 55 80 and 110 sk br lenses.

Regarding the vanilla IQ4 150, I need to try it, but I expect live view will be useless unless you have enough IR getting through the IR cutoff filter.

Wet and cloudy over the next few days here in London.

I‘m very interested in getting to know what you discover.

Thanks!
 

beano_z

Active member
I‘m very interested in getting to know what you discover.

Thanks!
Thanks for the feedback, I should be receiving my filter one of these days, but I'll be busy at the office before the weekend during daytime so hopefully I can get some free time in the weekend to do some testing. Since I'm planning on converting everything to b&w, I'm not really worried about colour casts.
 

onasj

Active member
Your question made me curious so I just ran a quick test.
- Phase One IQ4-150 (color) back
- Leica 50 mm Summilux (current version) at f/2
- B+W 830 nm IR cut-off filter
- ISO 5000, 0.4 s

Of course, the mechanical vignetting is from the fact that the Leica 50 Summilux has a much smaller image circle than the 54x40 mm sensor in the IQ4, though it has one of the largest image circles of all Leica M lenses.

It "works" in that you can get a decent image, but you lose many stops along the way. This photo was taken in broad daylight on a clear day at f/2, so what would normally be around ISO 100 and 1/500 s or so now requires ISO 5000 (5.5 stops) and 0.4 s (another 7.5 stops)—so you lose about 13 stops along the way! If you use a lower wavelength cutoff (720 instead of 830) you'll lose less light but get less "IR effect" since reds will bleed into your image.

Being winter here, there aren't a lot of leaves on the trees, but you can still see the IR effect.

Full-resolution JPG: https://www.dropbox.com/s/ap0hr4noce7mxk6/IQ4-150 IR 830 test Leica 50 Summilux at f2, ISO 5000, 0.4 s.jpg?dl=0

No cropping applied.

IQ4-150 IR 830 test Leica 50 Summilux at f2, ISO 5000, 0.4 s-sm.jpg
 

beano_z

Active member
Thanks for the test, very informative!

Meanwhile I've also received my filter and went out for a few quick shots.

Indeed, your shot has more "infrared effect" than mine. I really wish I could find an 830nm filter in 105mm size here locally!

I estimate the 720nm works similar to an 8~9 stop ND filter exposure wise, but I have no experience yet getting the correct exposure on IR, so this number might be off. Happy to say though that the live view still works, albeit a little slow, as can be expected. But at least I won't have any issues with focusing. So far I've tried 3 lenses on the XF, the 35, 80 and 150/2.8 and I don't think they exhibit any hotspots, at least there are no anomalies I can see from my test images.

Anyway, I hope this is of some interest.

P0002839 Infrared Filter Test SMALL - 13-Jan-2021 1.jpg
 

Shashin

Well-known member
i shoot IR on a color sensor. A few things I found helpful. I set the WB using foliage--a patch of grass, for example. That will allow you to better see the exposure compared to the red view of the a standard WB setting. You will also want to over expose. IR shots are very flat and so moving the tones up the response curve will give much more to play with in post. I found I go two or more stops above what the camera indicates. I also use a B&W filter that gets me into the 800-900nm cut-on range to give that IR look (B&W 093 or Wratten 87C).

The next thing that will take a bit of time is understanding the nature of processing IR in post. It really is a high-key type scene. And the tones don't quite distribute the same way as normal B&W. The histogram always looks like a bell curve in normal distribution (even if the results look like a paranormal distribution ;) ).

From an X Pro2 with a B&W 093. Note the hot spot.

 
Last edited:

beano_z

Active member
Wow, thanks for the tips, I've not even thought of white balance, I thought it wouldn't make a difference shooting RAW.

I just went back to check my original file and indeed, there's actually still some information recorded in the green and blue channels, and the red channel does look like a very even normal distribution! It does make the levels adjustment quite straight forward.
 

FloatingLens

Well-known member
i shoot IR on a color sensor. A few things I found helpful. I set the WB using foliage--a patch of grass, for example. That will allow you to better see the exposure compared to the red view of the a standard WB setting. You will also want to over expose. IR shots are very flat and so moving the tones up the response curve will give much more to play with in post. I found I go two or more stops above what the camera indicates. I also use a B&W filter that gets me into the 800-900nm cut-on range to give that IR look (B&W 093 or Wratten 87C).

The next thing that will take a bit of time is understanding the nature of processing IR in post. It really is a high-key type scene. And the tones don't quite distribute the same way as normal B&W. The histogram always looks like a bell curve in normal distribution (even if the results look like a paranormal distribution ;) ).

From an X Pro2 with a B&W 093. Note the hot spot.

Was that Fuji converted to IRand how would a LEE 87 compare to that B&W 93? Are those filter numbers standardized or determined by cutoff wavelength?

Bear with me... this kind of stuff is anything but obvious.
 

Shashin

Well-known member
Was that Fuji converted to IRand how would a LEE 87 compare to that B&W 93? Are those filter numbers standardized or determined by cutoff wavelength?

Bear with me... this kind of stuff is anything but obvious.
No, that is an off the shelf Fuji. I don't do IR that much to have a converted camera.

I am not sure how manufacturers code there filters. Wratten was such a standard for a long time, I suspect many use that to at least decide the filters in their line. B&W says the 093 is equivalent to a Wratten 87C. The 093 has 50% transmission of a bit longer than 800nm and gets to about 88% at 900nm and remains there. Lee states that their 87 begins transmission (cut-on) at 780nm. I am assuming that is 50% transmission at 780nm. They don't give a plot for the transmission and so I can't tell how transmission increases after 780nm.

The great thing about Kodak in the day was they actually release transmission plots for their Wratten filters. The great thing was these plots would not just tell you the information you wanted, for example, the cut-on wavelength, but the entire transmission profile, so you can see, for example, that your IR filter will also transmit UV.

And to illustrate the complexity of transmission profiles, the blue and green filters on the sensor actually transmit in the IR, but don't cut-on until some point after the red (this is why we can get a two-tone effect in color IR with digital cameras). This is from my unconverted X Pro2. It has been white balanced on the foliage, which is very reflective in the IR, but notice the sky, which does not have any real IR signal is red. So the RGB pixels are picking up the IR from the foliage, but only the red pixels are getting higher frequency radiation to image the sky.

 

FloatingLens

Well-known member
No, that is an off the shelf Fuji. I don't do IR that much to have a converted camera.

I am not sure how manufacturers code there filters. Wratten was such a standard for a long time, I suspect many use that to at least decide the filters in their line. B&W says the 093 is equivalent to a Wratten 87C. The 093 has 50% transmission of a bit longer than 800nm and gets to about 88% at 900nm and remains there. Lee states that their 87 begins transmission (cut-on) at 780nm. I am assuming that is 50% transmission at 780nm. They don't give a plot for the transmission and so I can't tell how transmission increases after 780nm.

The great thing about Kodak in the day was they actually release transmission plots for their Wratten filters. The great thing was these plots would not just tell you the information you wanted, for example, the cut-on wavelength, but the entire transmission profile, so you can see, for example, that your IR filter will also transmit UV.

And to illustrate the complexity of transmission profiles, the blue and green filters on the sensor actually transmit in the IR, but don't cut-on until some point after the red (this is why we can get a two-tone effect in color IR with digital cameras). This is from my unconverted X Pro2. It has been white balanced on the foliage, which is very reflective in the IR, but notice the sky, which does not have any real IR signal is red. So the RGB pixels are picking up the IR from the foliage, but only the red pixels are getting higher frequency radiation to image the sky.

Astonishing! Thanks, Will! I had the idea that IR/wood effect style pictures would not be obtainable with a non-converted camera. I'd be interested to try with my Fuji X-T1 myself. So do I understand correctly: depending on transmission profile of the front filter, non-converted cameras don't prevent you from taking real IR images. The IR cutoff filter in the camera just needs to be compensated for with increasingly long exposures (in comparison to a modified camera)?

Cheers
 

Shashin

Well-known member
Astonishing! Thanks, Will! I had the idea that IR/wood effect style pictures would not be obtainable with a non-converted camera. I'd be interested to try with my Fuji X-T1 myself. So do I understand correctly: depending on transmission profile of the front filter, non-converted cameras don't prevent you from taking real IR images. The IR cutoff filter in the camera just needs to be compensated for with increasingly long exposures (in comparison to a modified camera)?

Cheers
That is all it is. The camera's IR cut-off filter does not reduce the transmission to zero.

As far as the camera, each one is different. I have done this with a Minolta 4MP point and shoot, an Olympus EP-1, and a Pentax 645D, along with my X Pro2. Like Binbin stated, it is like having a 10 stop ND filter on your camera.

BTW, it is not as bad as it sounds. In broad daylight, you can crank the ISO and still use your camera handheld. With the XF14mm f/2.8:

 
Last edited:
Top