Infrared performance of Fujifilm X Mount Lenses
These test images were taken on a modified infrared only Fujifilm X-E1 with a Schott Glass internal 830nm infrared pass filter. They show how well each lens works in the infrared only spectrum where problems often occur with hot-spots - bright areas in the centre of the frame.
Hot-spots have several causes. Some are caused by internal reflections inside the lens elements or between the infrared pass filter and the rear element of the lens. In visible light these internal reflections are reduced by 1/4 wave multi coatings on the elements but these coatings are ineffective at wavelengths beyond 800nm and can sometimes cause increased internal reflections at infrared wavelengths. Others can be caused by internal reflections off the aperture blades of internal lens mechanisms - again the anti reflection coatings may not be fully effective at IR wavelengths. A third type is caused by the lens multi coating transmitting variable amounts of IR depending on the angle of incidence - these most often effect very wide angle lenses.
Generally hot-spots are worst at smaller apertures (larger f numbers). However softness caused by diffraction in infrared generally means that the optimum balance between depth of field and sharpness is between f/8 and f/11 for and infrared only camera and between f/5.6 and f/8 for a false colour infrared camera.
Whilst these tests do not guarantee the lens performance on a false colour infrared camera with a lower pass filter of 740nm, in my experience lenses that perform well in pure infrared generally perform well in false colour. However they do not necessarily show how well the lens will perform on an unmodified camera with an infrared filter mounted externally as the external filter can also be the cause of hotspots.
Please note that the conclusions are my own subjective opinion of how well each lens performs in infrared and should only be used for guidance. All images were shot as RAW files and converted into JPGs using Adobe Lightroom using the Adobe Standard profile. The test images are intended to show the presence or absence of hot-spots rather than overall sharpness and are all scaled to 1000px wide to save bandwidth.
For those who want the quick answer my favourite lenses for IR shooting are the XF14mm and XF35mm primes and the XF18-135 WR Zoom - both fantastic performers in infrared!
Coming soon for 2020 - the results of testing I haver carried out on all Fujinon lenses in conjunction with ACS and Fujifilm UK - watch this space!
Fujinon XF Prime Lenses
XF 14mm f/2.8 R Excellent IR performance at all apertures
The 14mm prime is a superb performer in IR at all apertures and never exhibits hot-spots. As expected it shows some edge softness at f/11 and above but is extremely sharp wide open.
 |
| 14mm at f/2.8 |
 |
| 14mm at f/11 |
XF 16mm f/1.4 R WR Excellent IR performance at all apertures
Tested by IanT with excellent results - very slight hint of a hotspot at f/16 but below that it is perfect - more test images at https://imgur.com/a/8LgyKoG
 |
| 16mm @ f/1.4 |
 |
| 16mm @ f/8 |
XF 18mm f/2 R Not suitable for IR
The older 18mm prime does not perform well in infrared. It is very soft at f/2 and has increasingly bad hot-spots at all other apertures.
 |
| 18mm at f/2 |
 |
| 18mm at f/4 |
 |
| 18mm at f/8 |
 |
| 18mm at f/16 |
XF 23mm f/1.4 R Excellent IR performance at all apertures
Like the 14mm, the 23mm prime is very well behaved in IR showing no hot-spots at any aperture. It is sharpest at around f/5.6 with diffraction well controlled at f/16
 |
| 23mm at f/1.4 |
 |
| 23mm at f/5.6 |
 |
| 23mm at f/8 |
 |
| 23mm at f/16 |
XF 27mm f/2.8 - Only usable at f/2.8 in IR
This ultra-compact pancake lens is one of my favourites for street shooting however in infrared it exhibits a sharply defined hotspot dead centre at all apertures other than wide open at f/2.8
(images to follow)
XF 35mm f/1.4 R - Good IR performance at f/11 and below
The older 35mm prime is another good infrared performer. It has a very slight hot-spot at f16 but at all wider apertures it shows no hot-spot and excellent sharpness even at f/1.4
 |
| 35mm at f/1.4 |
 |
| 35mm at f/2.8 |
 |
| 35mm at f/8 |
 |
| 35mm at f/16 |
XF 56mm f/1.2 R - Only usable at f/2.8 in IR
Sadly the 56mm prime, a stunning lens in visible light, does not fare so well in IR. Wide open it is quite soft, sharpening up nicely at f/2.8 with no hot-spot. However by f/4 a hot-spot is clearly visible which worsens as you stop down. The APD version of this lens has also been tested by Bertrand T and has the same IR performance as the standard version.
 |
| 56mm at f/1.2 |
 |
| 56mm at f/2.8 |
 |
| 56mm at f/5.6 |
 |
| 56mm at f/16 |
XF 60mm f/2.4 R Macro - Only usable wide open in IR
The macro 60mm is sharp wide open and shows no hot-spot. However as soon as you stop down the hot-spot starts to appear rendering it unusable at f/5.6 and above.
 |
| 60mm at f/2.4 |
 |
| 60mm at f/4 |
 |
| 60mm at f/8 |
 |
| 60mm at f/22 |
Fujinon XF Zoom Lenses
XF 10-24mm f/4-5.6 R OIS - Not suitable for IR
The 10-24mm wide-angle zoom is another excellent visible light lens that doesn't work so well in infrared. Between 10mm and 20mm it shows a soft but visible hot-spot at all apertures, only giving useable results wide open at 24mm, with the hot-spot returning as soon as you stop down.
 |
| 10-24 at 10mm f/4 |
 |
| 10-24 at 10mm f/11 |
 |
| 10-24 at 18mm f/5.6 |
 |
| 10-24 at 18mm f/11 |
 |
| 10-24 at 24mm f/5.6 |
 |
| 10-24 at 24mm f/11 |
XF 18-55mm f/2.8-4 R LM OIS - Not suitable for IR
Like the 10-24mm, the 18-55mm standard range zoom is not a lens to use in infrared. At 18mm a hot-spot is visible at all apertures. At 35mm it is just about usable wide open but the hot-spot appears as soon as you stop down. At 55mm it is again fine at f/4 but exhibits a hot-spot at all other apertures.
 |
| 18-55 at 18mm f/2.8 |
 |
| 18-55 at 18mm f/11 |
 |
| 18-55 at 35mm f/3.6 |
 |
| 18-55 at 35mm f/11 |
 |
| 18-55 at 55mm f/4 |
 |
| 18-55 at 55mm f/16 |
XF 55-200mm f/3.5-4.8 LM OIS - Usable below f/8 at all focal lengths
After disappointing results from the 10-24mm and 18-55mm zooms this lens was a real surprise. At all focal lengths it is free of hot-spots wide open and generally useable up to f/8. Beyond f/8 a gentle soft-spot appears in the 55mm to 100mm range, but this is markedly reduced at longer focal lengths and by 200mm the lens performs well at all apertures.
 |
| 55-200 at 55mm f/3.5 |
 |
| 55-200 at 55mm f/8 |
 |
| 55-200 at 100mm f/4 |
 |
| 55-200 at 100mm f/8 |
 |
| 55-200 at 200mm f/4.8 |
 |
| 55-200 at 200mm f/11 |
XF 18-135 f/3.5-5.6 R LM OIS WR - Excellent performer at all focal lengths
After preliminary testings with a prototype 18-135 I have now been able to shoot extensively with a production version of this lens and can confirm that it is a superb all round performer in Infrared with practically no evidence of hot-spots at any focal length or aperture below f/11 and only the slightest hot-spot appearing at f/22 @ 135mm. Sample images coming soon.
XF16-55/2.8 R LM WR - Mixed reports from users - caution advised...
XF 50-140/2.8 R LM OIS WR - Excellent performer at all focal lengths
Tested by Bertrand T - Fine across the range at all apertures
Fujinon XC Zoom Lenses
XC 15-45mm f/3.5-5.6 OIS PZ - Not suitable for IR
XC 16-50mm f/3.5-5.6 OIS - Not suitable for IR
XC 50-230mm f/4.5-6.7 OIS - Not suitable for IR
Fujinon Teleconverters
TC1.4 and TC2 - Not suitable for IR
Zeiss Touit X-mount Lenses
Touit 1.8/32 - Not yet tested
Touit 2.8/12 - Not yet tested
Touit 50/2.8 Macro - Usable below f/8 at all focal lengths
Tested by Bertrand T - Perfectly fine until f/5.6, barely visible hot-spot at f/8 (still very usable aperture) and then severe worsening.
Samyang X-mount Lenses
Samyang 8/2.8 fisheye: - Mixed reports from users - caution advised...
Tested by Bertrand T - Fine at all apertures...
Also tested by Gary - only really usable wide open...
Totally agree with your assessment of the 10-24, 14, 18--55 and 55-200.
ReplyDeleteI have added the 16-50 for IR which behaves similarly to the 55-200...ok to f8
Awesome, thank you for your work
ReplyDeleteZeiss Touit 50/2.8 Macro: Perfectly fine until f/5.6, barely visible hot-spot at f/8 (still very usable aperture) and then severe worsening.
ReplyDeleteSamyang 8/2.8 fisheye: Fine at all apertures.
Fujinon XF 50-140/2.8 R OIS: Fine across the range at all apertures.
ReplyDeleteFujinon 56/1.2 R APD: Unsurprisingly similar to the regular version, with a hot-spot from f/4 onwards (although, in my opinion, the hot-spot is still quite discrete at f/4 and this aperture could be used with caution).
Thanks for this! Might try a 16-50 ;-)
ReplyDeleteForget the 16-50 mm XC it doesn't focus Ir and it has a hotspot at all apertures all FLs. Fact is you get better than the Fuji 14mm f2.8 and Fuji 35mm f1.4. I can also add that the new 16mm f1.4 has hotspots - just about fixable at faster than f2.8.
ReplyDeleteVic
Although I have yet to check this lens for a second time I should say that my above comment was based on using it with an R72 lens filter and unconverted body. I've since bought and had an XE1 converted to 590nm IR and I now find that lenses which performed well at (say) up to f8 now perform well up to f11/f16. Other comments here with a converted camera indicate the 16-50 XC is a good performer so my advice now is - no good with lens filter only but apparently worth checking out if used on a converted camera.
DeleteBelow is a link to some of my IR images taken with 590nm converted Fuji XE1. Most colour and black and white are taken with a Hoya R72 (720) filter on the lens in addition. Those with very strong blues and yellows are with no added filter. Lens used is about 90% Fuji 14mm and 10% Fuji 23mm f1.4, 35mm f1.4 and 55-200mm (deer and swan shots). I also use Oly Zuiko 50mm f1.4 and 28mm f3.5, 24mm f2.8 Minolta and occasionally Minolta Rokkor 135mm f2.8 for IR. I could easily manage with just the 14mm f2.8 set to f5.6/f8 and hyperfocal as mine are shot. Equally, one should check for manual focus legacy lenses since IR is not for the fast shooter and there are many fine MF lenses to be had for good prices and old film glass will have distance and depth of field markings to make shooting at hyperfocal settings a lot easier.
Deletehttps://goo.gl/photos/UGkenCK8MV9qL1XF9
Edit should read "can't get better than 14, 35mm."
ReplyDeleteThanks Vic and others - your comments have been added into the mix - keep them coming !
ReplyDeleteWow, great research. Thanks for saving me some money - I was getting ready to get the 10-24mm, now after reading your article I'll be getting the 14mm. I searched out your article on google after shooting test images with my 18mm f/2 and seeing the crazy hotspot. My buddy has all three Zeiss Touit lenses and I'm going to test the 12mm and 32mm when he gets back in town... the used Touit 12mm is only a bit more than a used 14mm.
ReplyDeleteThanks Britt - glad it was useful. Would be really interested to see how you get on with the Zeiss lenses in IR - it would be great to add some info for the 12mm and the 32mm in my roundup...
DeleteWill be interested to see how you like the Touit 12mm, looking around I have seen some IR shots on Flickr from a photographer "Cruiser223" who uses it on an X-E1 and X-Pro1 and it doesn't appear to have hot spots.
DeleteVery useful info, thank you. Any data about the new XF 90mm f/2 ?
ReplyDeleteI can report that for the X100T, both bare and with the 35mm "tele" conversion lens, a hot spot and a halo begin to appear around f/5.6. Below that I find the camera very usable for IR (using an R72 filter).
Oh and the XC 16-50 is just terrible. Stay clear.
ReplyDeleteCorrection on the X100T: should have said "to appear around f/4".
ReplyDeleteThnx for all the useful info!
ReplyDeleteI'm using a full manual 28mm f2.8 nikon AI lens on a fuji XE-2 , wich seems to do the job great.. although I haven't done an all aperture test.
Do you know anything about focus shifting, cause the nikon has a red dot for IR on it but do you need to adjust the focus on a x-trans sensor?
Hi Pascal
ReplyDeleteThe red dot on your Nikon AI lens shows you the shift between the plane of focus in visible light vs infrared light. This was necessary when shooting with an SLR and IR film because you viewed and focused visible light through the optical viewfinder and then had to adjust by the focus shift before exposing the film.
On the XE2 the viewfinder image is from the sensor itself so what you see is what you get in IR and the focus shift is no longer necessary - this is one of the huge advantages of shooting IR with mirrorless cameras. It also means that if you use an autofocus Fujinon lens the AF will work perfectly in IR without adjustment.
Thank you Simon for these very useful resources. I have the XF 16-55 and I try it with my new IR72 filter (16mm @ f/8). And I experienced a big red circle at the middle of the picture that now I know it's a hotspot (link to the images here post-processed with LR and PS with swap channels: http://adobe.ly/1RlTVW1)
ReplyDeleteSorry, It's not the best shot neither the best test (I'm completely new to the IR photography). It's only one shot, with a classic X-T1 and a filter. Anyway, I will wait until you have the possibility to test properly this lens :-)
Cheers,
I just tried out some 720nm colour IR shots with the 18-135, at f8, and was disappointed to see that I was getting a "hot spot" with a marked colour shift, that rendered the raws unusable for colour processing, while still being fine for mono. Somewhat disappointing; shall experiment further with a 590nm filter (once I have one; I've come to prefer it for colour IR - but don't have one the right size), and wider apertures.
ReplyDeleteHi Tom - I have used the 18-135 extensively with my modified X-E1IR without any hotspot problems. I know that using an external filter can give rise to hotspots on any lens, but one thought: did you remove any protective filter you have on the lens as this may be the culprit?
ReplyDeleteCheers - Simon
A wealth of information here for the aspiring IR photographer.
ReplyDeleteTo add to my recommendations above, Minolta 24mm f2.8, 28mm f2.8 and 50mm f2 all gave me excellent results with no hotspots and of course, being old 135 film lenses they all have distance scales and depth of field scales which are very useful for IR. I've kept them even though I have most covered in native Fuji lenses now because they are much lighter and more compact when I want to travel light - they are also more easily covered by a single filter and step-up rings.
I'm currently waiting for an XE1 to return from 590nm IR conversion in USA chosen because of good feedback and prices that are less than half of other conversion companies even after 2 way postage. You'll find him on ebay with a link to his business if you prefer to save even more. The name is Isaac Szabo but just search for XE1 Infrared conversions - he does all makes and models. I'll monitor replies in case anyone wants more details.
FUJIFILM recently announced the launched of two Zoom Lenses: FUJINON UA13 x 4.5 wide angle Ultra HD (UHD) lens and UA80 x 9 1.2 EXT high magnification zoom lens fujifilm zoom lens
ReplyDeleteI can now add Olympus Zuiko 50mm f1.8 and f1.4 to the list of good IR performers at all apertures, also surprisingly good is the Oly Zuiko 70-140mm f4, not that I personally use long Fls for IR. BTW, the Fuji 90mm f2 is a good'un for IR.
ReplyDeleteI have had my Fuji X-E1 back from conversion to 590nm for a couple of weeks and I'm really enjoying the freedom of working without a tripod. I use it at 590nm and also with a screw-in Hoya R72 and a cheapie Neewer 850nm filter for BW IR. Shutter speed is unaffected with the R72 and only loses about 1.5 stops with the 850nm - so still easily handholdable. This of course is because all of the IR light can now reach the sensor plus some visible light while the 850 blocks virtually all visible light but remains handholdable at medium apertures by doubling ISO.
I briefly tried some night time "Dark light" shooting with a very cheap, old, small and low powered Sunpac flash circa 1980s. Even in the open space of my back garden (so no reflecting of light) it stretched beyond the 70 ft of my boundary with the R72 over the flash. With the 850nm covering the flash a little less light passed through. Electronic flash seems to emit more IR light then visible light thus increasing the flash range. While the R72 allowed a little glow of flash light to show through the filter I held in front of the flash, light through the 850nm was hardly visible. I'd hoped to rig the camera to take time lapse of what animals cross my garden at night but sadly the X-E1 doesn't have a time lapse facility but anyone with a suitable converted camera could have fun with a similar setup. Anyway it's worth remembering you can use dark light for IR at night if you get bored with conventional Daytime IR. I also took a few pictures without flash of my garden at night (dark but reflected light pollution from overcast sky - S.London suburbs) and there was a surprising amount of IR and a little visible light recorded.
I read Simon Weir's post with interest since I've found fewer and weaker hotspots using a converted camera so it's just possible that some lenses listed as having a hotspot at (say) f5.6 might stretch to f8 on a converted camera compared to an on lens filter.
Certainly the converted camera gives more regular and predictable results "out of the box". Perhaps to extent that some fun is lost in processing but I'm now concentrating on weakening the IR effects to get a more ethereal feel. I have 2 IR galleries: This is my IR with lens filter gallery: https://goo.gl/photos/wQ27DAeQcsy8CVQ79 and this more recent with 590nm conversion: https://goo.gl/photos/UGkenCK8MV9qL1XF9
This comment has been removed by the author.
ReplyDeleteI've tested the Zeiss Touit 32mm/1.8 and had hotspots at all apertures above f/2.8 on a converted (665nm) XE2. Same with the Fujinon 90mm/2. I had great success with a Nikon 28mm/2 AI-S
ReplyDeleteI have a IR Fuji X-A2, and I use the XC 16-50, and I made some tests with it.
ReplyDeleteThere are some hotspots at f11 and up. Superbly sharp af f8 and below.
See results at the bottom of this page:
http://www.ericconstantineau.com/photo/review_fujixa2_en.html
I have converted a X-E1 to 820nm and on the basis of above reviews bought a Samyang 8mm fisheye. Unfortunately it is unusable at and above f8. Larger aperture OK. Maybe better with other cameras.
ReplyDeleteLate with this reply but anyway, the 8mm on XE1 focused at about 4ft needs to be open no more than f4/f5,6 for sharpness from 2ft to inf.
DeleteHi Gary - you will find most lenses begin to show hot spots at f/8 and over but more importantly they will also very quickly become soft because of increased diffraction at IR wavelengths - I almost never shoot stopped down beyond f/8 even on known good lenses for this reason. I know a couple of people using the 8mm Samyang lens and have seen decent results so suggest you persevere !!
ReplyDeleteThanks for the follow up Simon. My post was to advise that my experience was not what was said on your post; "Excellent IR performance at all apertures".
ReplyDeleteI found this page by being top of the search "infrared lenses fuji" and assume others will also follow. Hopefully my post will help clarify through my experience that with a 820nm conversion to my X-E1 that this lenses is only hot-spot free when shot wide open.
Thanks for putting the page together and allowing updates. It should be a useful resource for others.
Cheers Gary - I have updated the page to "Caveat Emptor" for this lens !!!
ReplyDeleteFollowing Gary's comment, I've double checked the Samyang 8/2.8 stereographic fisheye on my converted X-E2 (820 nm) and I stand by my assesment that the lens is fine for IR work, with no hint of hotspot at any aperture.
DeleteBut this lens is very sensitive to flare, in IR as well as in visible light, from light sources within or just outside of the (obviously large) field of view. At small apertures (f/8 onwards), the flare appears as a bright sharp-edge hexagon. At wider apertures the hexagon is softer and more diffuse, so that at f/2.8 or f/4 it actually covers most or all of the image and rather appears as a general loss of contrast. This might be the reason why Gary finds the lens "unusable". But with careful framing and/or shading against sources of flares, there is no issue.
For those of you interested, I did a quick review of the Fuji XF 35mm F2 on a Fuji X-T10 Full Spectrum Camera with the Hoya r72 and the B+W 403UV filter to check for any hotspots. Basically no signs of hotspots from what I could see. I haven't tested a 830nm filter yet (because I don't have one at the moment).
ReplyDeleteCheck out my blog post here:
https://kiwiinkyushu.wordpress.com/2016/08/28/fuji-xf-35mm-f2-ir-uv-hotspot-test/
Anyone news how is the FUJI 16-55 2.8f?
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteThanks for your post.
ReplyDeleteI've bought an X A-1 with the XC 16-50 kit lens and let it converted to infrared. The 16-50 is sold and I'm looking for a better standard lens from Fuji. At the moment I'm using only adapted lenses from Nikon and Beroflex.
Up to the 16-50: One of the worst lenses I've used for infrared. A hot spot and very poor borders. I posted some example photos here:
http://www.dslr-forum.de/showpost.php?p=14146541&postcount=189
http://www.dslr-forum.de/showpost.php?p=14149097&postcount=193
If you can't read German, don't worry. The photos say enough.
Thanks Johanna
DeleteSeveral readers have reported the 16-50 kit lens as being poor in InfraRed so I am not surprised !
If you want a zoom I suggest you try and get hold of a XF 18-135 f/3.5-5.6 R LM OIS WR as this is a fantastic IR performer.
Happy shooting - Simon
Fujinon XF 16/1.4 R WR: Similar to the XF 14/2.8 and XF 23/1.4, the 16 mm is an excellent IR lens at all apertures.
ReplyDeleteAnd, for the fisheye lovers out there, I'm happy to report that the Yashuara Madoka-180 (a circular fisheye with orthographic projection) also works fine in IR.
As for my previous reports, the lens were used on a 820 nm-modified X-E2.
I have used the 18-55 lens on my X-T10 with no sign of a hot spot, using both a IR 720 and IR 850, so not sure what is going on here as the tests above say it will not work, but it does !
ReplyDeleteHi Bob - thanks for your comment!
DeleteI suspect you have been lucky !! My tests with a modified IR X-E1 showed consistent hot spots at 18mm with some improvement at longer focal lengths but only when used wide open.
The situation may be better with an external IR filter and an unmodified camera but other users have reported problems with lens as well.
If you only shoot with the aperture wide open and tend to use 35mm or more then this lens should be OK but I cannot recommend it for general IR use.
Best wishes - Simon
I just tried the 18-55mm on an X-T2 with IR 720. Hot spot in center of image.
DeleteHi Jeffrey - thanks for your comment - exactly what I would expect from that lens !!!
DeleteLife Pixels will convert a camera to full spectrum. My question is if a camera is converted would using a external visible bandpass filter give you as good a picture as the original that comes with the camera? I have the Fujifilm X-T2 on order.
ReplyDeleteHi there - interesting question... - I haven't tried this so I am using my experience (and physics degree!) to answer !
DeleteI think it will all boil down to the quality of the filters and their coatings (if any). When you have a camera converted with an IR bandpass filter installed in front of the sensor you are replacing a very sophisticated multicoated IR blocking filter with a plan uncoated IR pass filter. As the transmitted wavelengths are all within the IR band the lack of visible light multi coatings is not an issue.
In a full spectrum conversion the multicoated IR blocking filter is replaced with an optical glass flat that transmits all wavelengths but generally has no multi coatings. Its optical quality should be every bit as good as the original glass but the lack of coatings will mean there is a slightly higher risk of internal flare reducing contrast - this is however minimal as almost all modern lenses designed for digital cameras ensure that the light exits the lens perpendicular to the sensor avoiding any internal reflections.
If you then add a visible light bandpass filter to the front of the lens you are at much greater risk of flair reducing contrast because of the lack of multicoatings on what is now the "front element" of the lens. This is just the same as putting a £2 eBay UV filter on the front of an expensive lens - we all know it is going to be the weak link in the optical chain...
Plenty of manufacturers make normal clear or UV filters with multicoatings as sophisticated as the actual lens coatings but I am not aware of anyone making band pass filters (or IR filters for that matter) with multicoated surfaces...
So that is the theory - what about the practice?
Well I would suggest that under normal lighting conditions the difference will be minute but as soon as you have the potential for flare you will see the difference, especially with wide angle lenses.
Hope this helps - would be interested to see the results.
Best wishes and a Happy New Year - Simon
I too would be interested to hear from someone who has done this. My feeling is that jpgs won't be "Fuji jpgs" but Raw is what you make it yourself.
DeleteI have done exactly that with an X100T: full spectrum conversion, and I use a Hoya HMC UV+IR filter, which is multi-coated. It mostly works well except that occasionally I still get some flaring.
DeleteThis comment has been removed by the author.
DeleteThis comment has been removed by the author.
DeleteHi, I got my X-T10 converted to Full Spectrum by LifePixel and I also purchased their Visible Bandpass filter to take normal photos.
ReplyDeleteI'm not sure if the colours are still the same 'Fuji' colours, and sometimes you have to set the wb manually as auto wb doesn't work sometimes.
But i'm interested in shooting more Ir photos than normal photos, so i'm happy with the bandpass filter.
In the future when I get time (and some money) i'll try renting a non modified fuji and shot the camera side by side with my full spectrum fuji with the bandpass filter and compare the photos.
Fuji XF 50mm F2 - tested for hotposts at my for those of you interested. Basically no signs of hotspots from what I could see.
ReplyDeleteCheck out my blog post here:
https://kiwiinkyushu.wordpress.com/2017/03/21/fuji-xf-50mm-f2-ir-uv-hotspot-test/
Thanks for sharing this information Simon, i bought 14mm lens for replace my XC16-50 ois ii that really bad IR performer.
ReplyDelete14mm work very well for my XT1 converted with 720nm.
But somehow i miss the 10mm ultra wide lens for landscape & marital Photography, anyone here have some info what manual ultra wide lens good for IR?
When processing your IR images, does anyone use SilkyPix to set the gray point? I am setting a custom white balance in camera, it was suggested to me by LifePixel, the company that converted my Fuji X-Es2 to a 720nm filter that I have to use the SilkyPix software to set the gray point before I process in LR or Photoshop. Any thoughts on this subject? I would prefer not to have to go through so many processing steps to get the results that I want to accomplish.
ReplyDeleteThanks in advance for your suggestions and thoughts.
Hi Wayne.
DeleteThere shouldn't be any need to go through SilkyPix to set the white balance - Lightroom is more than capable of doing that just as well.
The in camera white balance is crucial though (set with a patch of sunlit green grass usually instead of a grey card).
If you are processing in B&W then my workflow is:
Open Raw file in Lightroom using custom WB as set in camera - set black and white points to maximise dynamic range without any clipping - export as TIFF to Silver Effex Pro 2 and process to taste - enjoy !
Cheers - Simon
Just wanted to chime in with a recent test of the 16-55 f2.8 on a 590nm converted X-T1. I am seeing significant hotspots at everything over f5 unfortunately. I'm guessing false colour filters exaggerate the hotspots?
ReplyDeleteHi Ian
DeleteThanks for that addition - I have had similar problems with that lens on my 830nm camera but need to test it properly to pin it down.
I don't think the false colour will exaggerate the hotspots per say but the post-processing may well exaggerate them as they are only present in the IR wavelengths and should not be occurring in the visible light.
Cheers - Simon
Thanks for keeping the thread alive. Very useful resource.
DeleteXM1 (not converted) with samyang (rokinon) 12mm f2, with 720nm filter leads to hotspot, partly visible at f2, worse after.
DeleteUnlucky....
Just managed to get hold of a reconditioned 18-135 from the Fuji refurb store. Done a series of tests at 18, 35, 70 & 135mm and at f4, 8, 11 & 22. Superb results so can concur with the tests above.
ReplyDelete(Also had great success with a Vivitar 28mm f2.8 M42 fit)
J'ai trouvé ce site car je ne comprenais pas ce qui se passait avec mon 18-55, je suis content d'avoir pu découvrir que c'était une incompatibilité de l'objecif bien que j'aurais préféré un problème venant du filtre... J'ai donc acheté un 14mm f/2.8, génial, les photos sont excellentes comme indiqué dans le test
ReplyDeleteI use a an XE1 590nm (supercolour) conversion. I mostly use it with Hoya R72 (720nm) or cheapo Zommei 850nm filter on the lens. Main lens is 14mm f2.8 but I also use 23mm f1.4, 35mm f1.4 and 55-200mm - all good for for IR shooting. I've also found that my Fuji 16-55mm f2.8 is good up to f8 but better at f6.4 and wider. As the focal length increases, the hotspot appears to diminish a tad so I can use f8 at longer FLs. I've only used this lens with 850-890nm BW filter so I don't know yet how it will perform with a 720nm colour IR filter on the lens.
ReplyDeleteFujinon 80/2.8 LM OIS WR Macro: Another perfect lens without any trace of hopspot at all apertures. (Tested on 820nm-converted X-T2)
ReplyDeleteThanks for keeping this great ressource online!
Hi Simon,
ReplyDeleteJust an update on the 16mm f1.4 on a converted Fuji X-T1 (590nm). The lens looks fab - just fab. I've been using the 18-135 as a perma-IR lens, but considering how much I love the 16, I may switch after seeing the results.
Images here : https://imgur.com/a/8LgyKoG which you're welcome to lift if you want.
My eyes are getting old now, but I think I can see the beginnings of a hotspot at f16. f8 though (which is where I spend most of my time) is perfect.
Hi Ian
ReplyDeleteTable data updated above - thanks for the update - can't wait to try one of these myself now!
I bought the Laowa 9mm F2.8 -- excellent for IR, no apparent hotspots. Give it a try!
ReplyDeleteHas anyone tested if the the Fujifilm 15-45mm F3.5-5.6 XC kit lens is a decent IR performer, or is it as bad as the 16-50mm/18-55mm kit lenses?
ReplyDeleteJust a quick update - I've now acquired a copy of the XC15-45mm kit lens and it’s IR performance seems to be every bit as good as the XF18-135mm, with no perceptible hotspots at any focal length or aperture. Sharpness is very good too, especially if stopped down 1-2 stops to f/5.6 - f/8. AF is nice and snappy too, used on a 720nm converted X-A2 body.
DeleteTested with 590nm filter.
ReplyDeleteGood
- Fujifilm XF 23mm f/2 R WR
- Fujifilm XF 50mm f/2 R WR
- Fujifilm XF 55-200mm f/3.5-4.8 R LM OIS (good below f/8 at all focal lengths)
Bad
- Samyang/Rokinon 12mm f/2.0 NCS CS (heavy vignetting at f/2 & f/2.8, hotspot starts at f/8)
- Fujifilm XF 18mm F2.0 (confirmed soft and hotspot start at f/8)
- Fujifilm XF 18-55mm f/2.8-4 R LM OIS (hotspots worse at 18mm than 55mm)
- Fujifilm XF 60mm F2.4 Macro (hotspots start at f/4)
Just wondering if you've had a chance to test the Zeiss 12mm f/2.8 lens yet? Thanks.
ReplyDeleteSamyang/Rokinon 8mm f/2.8 Fisheye II (hotspot starts at f/8)
ReplyDeleteSomehow I get minor hot spot or color shift using 35mm f2 with r72+full spectrum converted xt10. But it is very minor and can be fixed easily.
ReplyDeleteI just tested XF 16mm f1.4 with 720nm (R72) and 590nm (25A) with full spectrum converted X-T10.
ReplyDeleteWith 720nm, color shift appears from f1.4, and get worst with higher f. (warmer towards center of image)
With 590nm, there were no noticeable hotspot/color shift for all aperture, which was a surprise.
Few updates : XF35 f2 also has a minor color shift using 720nm, but works fine with 590nm.
XC 15-45 works fine for both 720nm and 590nm.
Pentacon 50mm f1.8 worked fine with 720nm around f5.6 ish.
I haven't tested enough, but Lens Turbo II also creates hotspot with 720nm with Pentacon 50mm f1.8.
It seems like lower nm creates less hotspot problems.
Good work. thank you for such kind of great information. For More
ReplyDeleteHas anyone tried Kaxinda 25mm f/0.95 ? the reviews for normal use is good, and it's quite cheap - 175$ brand new. So i wonder how is it for hotspots
ReplyDeleteHi there.
ReplyDeleteI'm a complete newbie to IR photography but I'm considering buying a second-hand IR converted X-T20. I have no idea of the nature of the IR conversion and I'm not sure the shop will, either. However, I have a couple of questions that I'd be grateful if someone could answer;
1. If the sensor has been converted, is it still necessary to use an IR filter and, if so, which filter spec is best? Sorry if this is a silly question but I thought I'd better get it asked as, if filters are required, I'll need to get some.
2. I have a selection of Fuji X lenses - all of which have been mentioned, so thank you to the tested for their comments. However, I've also got some Zuiko primes; 21mm f3.5, 35mm f2.8, 85mm f2 and 135mm f3.5. Has anyone tested any of these?
Many thanks in advance.
Hi There
ReplyDeleteWhen a camera is converted to infrared the IR filter is installed directly in front of the sensor so you dont need an external filter.
You should be able to tell what filter in in a converted camera by taking some images with the white balance set to auto. A 730nm filter will produce bright red/purple images whereas an 830nm filter produces slightly pinkish B&W images.
No idea how the Zuiko lenses will perform but do report back here when you have tried them out!
All best - Simon
Fuji 16mm f/2.8 WR - Mediocre results with 720nm converted Fuji X-A3. Wide open no hotspot or color shift, but from f/5.6 and smaller, there is a wide faint circular color shift in the center of the image. Almost like a traditional hotspot but not quite so. BW conversion or shooting with 850nm filter reduces the issues to almost nothing.
ReplyDeleteBut that is not all. I do not know if the lens design is just bad, but I could not get sharp edges (not even talking about corners...) at any aperture without extreme focus stacking. Focusing field turns to reverse U-shape or some such warped plane.
Conversion is totally fine on my X-A3, 23/2 WR produces perfect results.
This comment has been removed by the author.
ReplyDeleteUsed at X-T1, filter sensor converted to 720 Full IR
ReplyDeleteXF16-55, not useable, strong hotspot at every aperture
XF10-24, not useable, strong hotspot at every aperture
XF55-200, ok at F8 and below. Above soft to strong hotspot
Samyang 8mm FE @ F11 very strong hotspot, result landscaping below F11 not good
XC16-50 at all AV soft to strong hotspot
XF 35F2, perfect, clean at all AV
Xf 16 F1.4, perfect, clean
DeleteXF23 F2 perfect, clean
DeleteHi just got an X-T2 IR converted to 720nm. Does anyone know how the 8-16mm f2.8 performs for infrared photography please ? Thanks
ReplyDeleteHi Alex
ReplyDeleteLike all FujiFilm's wide angle zooms the 8-16 f2.8 is not a great performer in IR. It is useable wide open but as soon as you stop down a hot-spot starts to form.
I would recommend using the 14mm or 16mm primes both of which work brilliantly in IR
Regards - Simon
Thanks very much for this advice Simon - further confirmation that there's little point in me shelling out £1500 for the 8-16 then !!! I have the 10-24 for non-IR photography and the 14mm f2.8 prime which is giving me some nice images paired with my 720nm X-T2 so I'll stick with that !
DeleteThere were comments re the XC15-45mm kit lens and how it's a decent IR performer throughout the zoom and aperture range. I'm looking for a usable zoom and prefer this zoom range over the 18-135. Plus it's smaller and way cheaper. Does anyone have any other feedback re this lens? Thanks!
ReplyDeleteHi there
DeleteIn my testing the XC15-45 begins to produce hotspots as soon as you stop down from wide open and by f8 is unusable.
This is a pattern common to all the Fujinon wide angle zooms - the only exception being the 18-135.
I would suggest going with a couple of wide angle primes instead for IR - the 14mm / 16mm / 23mm / 35mm are all excellent - avoid the 18mm and 27mm which are not good in IR
Regards - Simon
Simon, thanks for the info. I'm transitioning from a Canon DSLR with a 24-205 so I'd like to retain access to a zoom. The 15-45 would be ideal (small, cheap) but I guess the 18-135 may be the best choice.
ReplyDeleteAlso, are these lens recommendations for IR filters or sensor conversions? Does it matter? Cheers.
In IR the 18-135 is a really versatile lens so you wont go far wrong!
DeleteAll my testing is done with an 830nm converted camera but the results have also been verified on 740nm and 650nm conversions. Hot spots with external filters are harder to assess as there are so many variables but in my experience any lens that performs well at 830nm will give good results at all wavelengths and with all filter types.
Regards - Simon
I passed on the 15-45 and picked up a used 18-135. I'm now waiting for my converted XT2 to arrive. Thanks again!
ReplyDeleteDid anyone test viltrox 85mm?
ReplyDeleteAnyone heard how the XF 16-80 f4 lens performs in the Infrared?
ReplyDeleteHi wbLo0n
DeleteI haven't yet tested the XF16-80 but based on Fujifilm's other zooms (with the notable exception of the XF18-135) I would not wxpwct it to perform well in IR - you are much better to stay with the primes XF14/23/35/50mm
Regards - Simon
I've tested a decent amount with the 16-80 with various wavelenths (550, 590, 720, and aerochrome), and while the hotspot performance is quite good (I have yet to find one), I find the edge sharpness of the lens to be a BIG drag until zooming to 23mm or more.
DeleteHow is the sharpness of the 16-80 compared to the 18-135 with regards to IR?
DeleteContrary to the above lens information, I find the Fujinon 23mm f1.4 to produce marked highlighting in centre frame. At both wide open and at f8 it is showing the same result. The camera is a Fujifilm X-A5 converted to 720nm. Has anyone else seen this problem?
ReplyDeleteHi there.
DeleteI have tested the Fujinon XF 23mm f1.4 R on both 720nm and 830nm conversions done by ACS and found it to be fine up to f5.6 with a slight hot spot beginning at f8 and severe by f11
Some immediate thoughts as to why it might not be working for you:
Do you have a protective filter on the front of the lens? If so you should always remove it for IR shooting - the coatings on these can cause real problems.
Who did your conversion? Is it a glass or a resin filter and does it have any coating - certainly not all conversions are equal and I have seen some supposedly "IR Coated" conversions that actually perform much worse than a plain Schott glass filter.
Last thought - the X-A5 has a standard CMOS sensor and not the X-Trans sensor which is in the Fujis we test with (X-E or X-T series cameras) - not sure if that might have an effect but it is possible...
Regards - Simon
Hi everyone. I need to know if Fuji 16 f2.8 has hotspots. Is it tested? I have a Fuji X-T1 modified. Thank you very much.
ReplyDeleteHi There
DeleteThe XF 16mm f2.8 RWR is pretty similar to the f1.4 version - hotspot free to f5.6 and shows slight hotspot at f8 - beyond that it is not usable
Cheers - Simon
Hi there :)
ReplyDeleteI have converted my x-t1 to full spectrum. But i have one problem. I cant get infinite focus with my xf 35f1.4.
Anyone knows why?
See below...
DeleteHi Kjetil
ReplyDeleteThis problem is quite common with full spectrum conversions and depends on exactly how the conversion was done.
To achieve infinity focus the optical path must be the same in a converted camera as it is in an unconverted camera. This means that when the IR Blocking filter is removed it must be replaced with another filter that transmits the required wavelengths for the conversion and presents the same optical path to the light.
For IR conversions this means selecting the correct thickness of IR pass filter which is not necessarily the same thickness as the removed IR blocking filter - optical path depends on the material's optical density.
For full spectrum conversions a Quartz filter should be used and again its thickness must be chosen to present the same optical path.
If the quartz filter is too thin (or not there at all) it will be impossible to achieve infinity focus.
If it is too thick infinity focus will be achieved before hitting the focus endstop of the lens.
Most modern autofocus lenses dont have mechanical endstops at infinity but can focus slightly beyond infinity - this allows contrast detect AF systems to work correctly but also gives a little room for manoeuvre in converted cameras.
So I suggest you go back to whoever converted your camera and discuss this with them, suggesting that they revisit the choice of filter thickness...
Hope this helps - best regards - Simon
I just tested an XC 15-45mm on an X-T20 590nm and it works just fine without a hotspot. I was shooting an f8 during my test and performance of my copy was quite admirable for such a low cost lens. 15mm is obviously the weak point but I've seen worse corners from the Canon EF-M 15-45mm and definitely worse from the Sony 16-50mm.
ReplyDeleteHi - Just receive my 70-300mm - looking forward to the test reviews? So far it seemed to work well on my XT2-720nm conversion.
ReplyDeleteHave been using an XT30 converted to 590nm and so far have used the following lenses.
ReplyDelete16-80mm - have only stopped down to f8 but it has been working fine for me without hotspot. Corner sharpness at 16mm is worse than with my regular camera. Once you get to about 20-24mm it gets better.
70-300mm - have only had one outing with it but looks to perform fine with no hotspot detected. Sharpness at 300mm was pretty bad for me. I need to do more testing at different focal lengths to confirm.
14mm 2.8 - works perfectly as expected. I did expect perfect corners but not getting them.
35mm 1.4 - works perfectly as expected.
I’m probably going to pick up the 15-45mm again just to keep my setup compact but it’s difficult when the 16-80mm is working out great for me.
Agreed. I absolutely love the 15-45mm on my infrared-converted Fuji bodies. Yes, a hotspot above f8 and beware of awful ghosts/flare if the sun is nearby. But great lens! The 70-300mm is stunning with no hotspots and acts like “normal”.
DeleteHi All,
ReplyDeleteI recently published a Fuji-centric article about How to Choose the Best Lenses for Infrared. It provides my test results which include some lenses not covered here. In addition to 8 Fuji X-mount lenses I also have results for Sony full frame, APS-C (on my blog version), and Micro 4/3. I include some images and cover issues other than just hot spots as well as resources like this website.
You can find it on PetaPixel or my blog:
PetatPixel- https://petapixel.com/2021/08/26/how-to-choose-the-best-lenses-for-infrared-photography/
My Blog- https://joelwolfson.com/how-to-choose-lenses-for-infrared/
I hope you find it useful!
Joel
Tested today a Viltrox 33mm f1.4 with a Zomei 720nm IR filter... Works like a charm with my X-T2. Can share images if needed.
ReplyDeleteI highly disagree with the XC 15-45mm PZ being listed as "Not suitable for IR". I use mine frequently for IR at all focal lengths and its been a great lens for it. No need to push people away from a good budget option
ReplyDeleteI took this 720nm photo on my 15-45 - https://www.reddit.com/r/fujifilm/comments/otl13i/full_spectrum_fuji_xa3_720nm_ir_filter_1545mm_pz/
Agreed. I use the 15-45mm all the time with fantastic results. Nice and sharp. Need to under expose using 720 converted Fuji bodies and beware sun ghosts, but these can be creative. Love the 15-45mm lens for IR.
DeleteThe 16-50 XC lens works perfectly up to f 8.0 on my 720 nm converted xa5s,no hotspot,perfect performance,please note : I only shoot b/ w! In fact this lens is so good,that I sold my 14 and 23 primes! Palmer
ReplyDeleteThis blogs is really helpful for me, keep sharing such blogs.
ReplyDeleteExtendable Pole Camera
Thanks for sharing such informative content. We will also provide the best DSLR camera microphones which are essential for vilogging . I would like to tell you about my blog DSLR Cameras Microphones
ReplyDeleteI had an X-T3 converted to full spectrum and tested out a few lenses. Note I really only shoot at f8.
ReplyDelete18-135mm - while it has no hotspot, the performance of the copy I tried was downright horrible. The weakness at the extreme ends really shined through with some of my images at 135mm were so soft it looked like I missed the focus completely.
16-80mm - faint hotspot at wide end especially against IR reflective objects. Performance is not bad at the middle of the range and just okay at the extreme ends.
14mm f2.8 - no hotspot but corners/edges are soft.
23mm f2 - this lens surprised me in that I expected perfection based on all of the rave reviews and did not get it. Edges are soft while middle is good.
15-45mm - no hotspot for me but again, I haven't shot above f8. Performance it okay but considering the price, I think it's pretty good. This is the lens I am currently using.
55-200mm - no hotspots for me. Performance degrades as you zoom. At 55mm it's really good but at 200 it's just okay. This is the telephoto lens I am sticking with as I like having 55mm.
As a note I am shooting full spectrum with an IR Chrome filter, B+W 090 filter (590nm) or sometimes a Hoya 25a HMC filter and a cheap 850nm filter.
I have several converted Fuji bodies and LOVE using the cheap XC15-45 which performs fantastically though has lots of lens flare if including bright sun. Even then the lens flare can be creative
ReplyDeleteSigma 16-50 works well with 590nm.
ReplyDeleteThe laowa 10 pancake performs very well on my 720 converted bodies in b/w,realy sharp,no hotspot ,vignetting is relatively strong,that‘s the only downside,but what a lightweight....
ReplyDeleteExcellent analysis. Thank you. Couple of suggestions having shot with several converted Fuji bodies for last 8 years. I use the XC 15-45 almost all the time with amazing results. I use mainly the wide end with stunning and sharp results. Need to under -expose quite a bit. There is a noticeable hotspot above f8 and terrible ghosts and flares if the sun is anywhere nearby-but this can be used creatively. I use the 15-45mm almost all the time because i love the wide angle and flexibility of the zoom. I just edit any hotspots. The Fuji 70-300 is absolutely fantastic with no hotspots:awesome results. I’ve tried the 16mm f1.4 and found I had to recalibrate white balance before shooting to get good results. The Fuji 16-80mm is completely useless for infrared. Email me at barrymayesphoto@yahoo.com and I will send you examples.
ReplyDeleteGood job, but I have a question. Does EBC Fujinon 50/1,4 produce any hot spot?
ReplyDeleteI recently bought the new Fuji X-T50 with the also new XF 16-50mm f/2.8-4.8 lens. I thought I would try the lens on my 720nm converted X-T2 camera, and I found no hot spots from F/2.8 to F/16. I was very pleasantly surprised.
ReplyDelete