My infrared converted GX1 continues to draw me in and has become somewhat addictive. I often see comments from people saying getting an old camera converted is making good use of it. If you have an old camera and are considering getting it converted I would however suggest thinking twice. Not because it’s addictive and will move you away from standard photography, but because I don’t think old cameras are really that usable when converted. I have three reasons for this:
Shooting infrared becomes much easier when you are able to see the IR results in live view, something that a lot of older cameras don’t have. Trying to compose an image through a viewfinder often leaves too much to luck as you attempt to maximize the IR effect. Live view is a real bonus in this respect and I would consider it almost essential for IR work.
The sensor quality of the older cameras is not really great for IR. Most of the data in an IR image is captured in the red channel with the other channels being interpolated from this by the camera and RAW converter. This means your image quality can be much lower than expected so using an old low resolution sensor can lead to disappointing results.
Simply using an old camera can be frustrating. It’s quite amazing how much the technology has moved on in the past 5 years and how much you will miss some of the features you now consider standard.
The purpose of this post however is to share the image above of Whitby Abby and mention a strange effect I have noticed. When shooting Infrared I find I can often use the camera handheld when a traditional camera would struggle with low light levels as was the case with the image above. The other thing I have noticed is that there is less contrast in shadows and this allows the Infrared camera to reveal more detail. Certainly in the image above a standard GX1 would have shown dark shadows and bright highlights. With the image above I actually needed to boost the contrast as the shadows were too light.
I don’t know if this information will be useful to anyone but I thought it worthwhile sharing in case.
I am and always have been a fan of panoramic photography. I’m not sure why but the format (usually somewhere between 2:1 and 3:1 ratio) really appeals to me and makes sense as the way I see the world. Unfortunately, to create good panoramic images you need additional equipment beyond just the digital camera and this tends to go against my lightweight ethos.
Typically to make a good panoramic I need a tripod, panoramic head and stitching software. Unless you are prepared to spend tens of thousands of pounds (or dollars) on a digital panoramic camera, panoramic images need to be shot as a sequence of overlapping images which are then stitched together in specialist software.
I should at this point mention that I am a real stickler for quality so if any aspect of my images is lacking (in my mind) I will not be satisfied with the finished panoramic. This means that I don’t like to handhold my camera when shooting image sequences and always try to mount my camera on a tripod. I would also like to use a panoramic head to avoid problems of parallax error where objects in the various images don’t align correctly, as the stitching software will either distort the images to make them align or leave “ghosting” traces of objects. I do have a panoramic head for my tripod but it’s so heavy and bulky that I seldom take it with me.
This combination of problems means that I need to rely on stitching software to do a good job of aligning and merging images. Until now I had been using either Photoshop or Panorama Factory to complete my stitching. Photoshop seems to do a reasonable job but feels a little clunky and doesn’t give me the fine tuning/image optimisation that I want. It also has a habit of distorting images when I don’t want it to and not aligning all the objects along a stitching seam correctly. It’s usually close but not quite good enough.
As I have never really felt completely satisfied I tried out and invested in Panorama factory. This does a nice job of aligning the elements of the image as well as offering lots of power, but really does need a panoramic head to work properly. It often leaves some areas which are not quite sharp e.g. where fine details such as grass didn’t align exactly between images. My solution to date has been to output the panoramic image as a layered Photoshop file. This allowed me to fine tune the blending to remove blurred areas by adjusting the masks in the layered file. This is time consuming and quite complex even when you know exactly what you are doing with Photoshop masks.
If I am totally honest with myself I shouldn’t have invested in Panorama Factory if I wasn’t prepared to use a panoramic head but I was swayed by the cost. You see, when I did my testing a few years back PTGui was really the best option given how I wanted to shoot but I was put off by the cost. PTGui is only a graphical front end for Panorama Tools (which is a freeware package) and I just wasn’t prepared to pay a hefty license fee for something built around the genius of another’s work. This was the wrong decision and I think I should have purchased PTGui.
As I have now decided to make panoramic work a major feature of my photography I have recently downloaded the trial version of PTGui and PTGui Pro once more. I have to admit that I am very happy with the ease of using the software which can be highly automated saving me time and effort. This sits well with my lightweight philosophy. The only problem; I am still reluctant to pay the license cost as I would need the Pro license.
That’s when I came across Hugin which is also a graphical front end for PTGui. It feels remarkably similar to PTGui in terms of operation and it appears to be just as capable with very similar features to the Pro version of PTGui. The image above was created from 4 images captured on a Panasonic GX1 which was tripod mounted. I tried the stitching in Photoshop and Panorama Tools but I could see problems. PTGui did a great job but so did Hugin with an almost identical result even down to how it determined the stitching – not surprising given they both have the same stitching engine.
Now you might be thinking that I am about to repeat my past mistake of not paying for the best tool because of the cost. I don’t however know at this stage which is the best tool. And, Hugin is freeware so there is no cost other than in my time to learn and experiment with the package. I might still decide to invest in PTGui but so far Hugin is doing a great job and meeting all my requirements. If you are interested in shooting and stitching panoramic I think this is a great package that’s well worth investigating and it’s free.
I’m now back from a trip to the US and thought I would restart blogging with some images from the trip.
The first area I visited was Death Valley and the photograph shown above was taken at Zabriskie Point in the valley. It’s a bit of an odd place to visit in terms of Landscape Photography as the clear sky tends to limit when and how you shoot. My own preferrence when shooting landscapes at sunset is to have plenty of broken cloud which will colour up with the low sun. Here however the sky is clear much of the time so you don’t get the colourful sky. You can however achieve rather dramatic side lighting as shown on the hills here. In case you are wondering, these hills are just mud and gravel but they are rock solid and painful if you happen to slip on them.
The image was captured using my Panasonic GX1 and 14-45mm lens which was tripod mounted. There was plenty of light around so it wasn’t necessary to tripod mount the camera but I didn’t want to take any chances. I also think tripod mounting works well in any light and ensures very sharp images.
More photographs will follow once I have had the opportunity to download and sort them.
It’s been a very busy month and its set to get more hectic for me. I have quite a lot on in the run up to Easter so I am going to take a short break from the blogging for a couple of weeks. Rest assured I will be back soon with new material and lots of opinions about Light Weight Photography.
So you don’t forget me I thought I would leave you with a nice image of an interesting tree shot in Infrared.
In my last post I discussed some of the choices we photographers now have in choosing paper surfaces for printing. Someone raised a question that I responded to about how to get the colours accurate but I think this subject deserves a more in depth answer; so here is a little more on the subject.
The secret to getting prints to look the way you want them to is all wrapped up in Colour Management. This can be a very simple process but it can quickly become a wide discussion with lots to confuse. I will try to keep this simple and discuss two areas of colour management that are essential to achieving accurate colour (and for that matter black and white) prints. These are:
It’s necessary to have an accurately calibrated screen so that you have confidence the colour you see on screen is the colour of the image. This is really vital because you could make your image look great on screen only to find you have compensated for all kinds of colour shifts and contrast problems with your monitor.
There are software solutions that allow you to calibrate your screen visually but these will never be as accurate as a hardware solution that measures the colours on screen, creating a specific profile for your monitor. There are a number of solutions available that will quickly allow you to generate a bespoke profile for your monitor and I suggest you invest in one of these. Its money well spent. Which model you invest in will depend on your budget and the approach you intend to adopt for printing.
When printing, it’s necessary to have your printer calibrated to the specific ink and paper. This allows you to ensure colours and tones are accurately reproduced for a given paper. There are three options here:
Download and install the ICC colour profile for your printer and the paper you will be using. This is good but not as good as having a custom profile created for your printer.
Have a custom profile created using a profiling service. This involves printing out a target image which is then measured to generate a profile. You can then install and use the profile for printing. There are a number of such services advertised in the back of photography magazines. Alternatively if you purchase Permajet or Fotospeed papers, they offer this service for free.
Invest in a hardware solution that allows you to generate your own profile. This is good if you have a lot of profiles you want to create or switch papers often. This is the option I have chosen and have purchased a ColorMunki Photo tool. This allows me to calibrate both my screen and printer. So far the results have been exceptional.
Even if you intend to send your prints away to a printer, it’s still necessary to calibrate your monitor.
When it comes to printing, the printer and paper combination you have selected may not be able to represent accurately the range of colours and tones you see in your image. To compensate for this the printer will adjust the colours so that they fit within the Gamut that the printer can handle. This can affect how some colours appear as well as the contrast level in the image.
The solution to this problem is to use soft proofing for your images. Photoshop, Lightroom 4 and other packages will support soft proofing. This involves selecting the profile you are going to use to print your image and then the software tries to represent the image on the screen as it will ultimately appear on paper. There are lots of solutions here so your best option is to look up how to perform soft proofing for your chosen package. I would even recommend you soft proof your image when sending them off to a third party printer.
So, in summary:
Profile your monitor and set it to use the customer ICC profile (most calibration units do this step automatically for you).
Print using a profile generated specifically for your printer and paper combination. You should then use this profile when printing so you will need to print from an ICC aware application such as Photoshop or Lightroom. I use a package called QImage for reasons I won’t go into here other than to say it makes the job easy.
Check your image using soft proofing before you printing to see if you want to make any further adjustments before you print.
Follow these steps and you will end up with accurate colours and tones in your prints.
I have just answered a question posted on yesterdays blog and it made me realise I am missing the obvious again. This time it was that most people reading my blog have probably never seen a direct conversion from a RAW infrared file. Neither will you probably know what all this problem is with white balance and why I was so concerned. Here then are some examples that hopefully will put this right. All are created from the same RAW file used for the image at the top of the page and I have done no other adjustments to the files beyond what I explain below.
This first example is what you get in Lightroom when you set the camera white balance to AWB and all the other sliders are at 0.
The next example shows what you get in Lightroom when you set the camera white balance correctly for infrared light. Better but still not correct.
Here is the same RAW file in SilkyPix using the correct Infrared white balance. Notice the difference in colour from the Lightroom image. In this image there is much more information in all three colour channels and it makes for a better conversion to black and white.
The next example shows what happens in SilkyPix when you set the white balance using the white balance picker on the grass near to the castle. This is how the file should look before converting to black and white. This gives a nice spread of information in all the channels and makes for a high quality conversion. Even though a lot of the information in the B and G channels is probably interpolated by the RAW converter, it still appears to be a better conversion and that’s what I am interested in.
In this final example, I have done a channel swap between the Red and Blue channels to create a false colour. Not to everyone’s taste but it can be quite effective.
I am always on the hunt for new ways to do things that can improve my photography. You only need to achieve a few small quality improvements and it can quickly add up to dramatic improvements in your work. One area that I had been exploring was RAW converters and some of you might have seen an earlier post I made about Photo Ninja which looked very promising but was quite expensive. My decision was therefore to stick with Lightroom which performs well.
More recently with my new found passion for Infrared photography, I have identified that Lightroom isn’t good for processing RAW files from my converted GX1. The quick explanation of this is that Lightroom can’t set a white point for the infrared image and you end up with an image which is red. This then prevents you from using a technique called channel swapping to produce false colour but it also appears to detract from the quality of the final image when working in black and white. My search was on then for a RAW converter to use with my infrared images.
I returned initially to Photo Ninja which did a good job and allowed me to set a correct white balance. Unfortunately the cost put me off although I did come close to making a purchase and probably would have if it wasn’t for RAW Therapee. This is a free RAW converter which performs well and has some nice features such as allowing me to do a channel swap during the RAW conversion. The only problem is that it’s tricky to use.
I moved on to search for another free RAW converter “Bibble” which it seems has been purchased by Corel and is now sold as Corel Aftershot. This works well enough and is good price. I just had a niggling feeling that I shouldn’t make a purchase just yet.
After a lot of searching and experimentation I remembered that Panasonic Cameras capable of shooting RAW images are packaged with a special version of the SilkyPix RAW converter. This is version 3 of the SilkyPix Developer Studio that has been limited to only working with Panasonic cameras. Whilst it won’t convert my Canon and Sony files it will process my GX1 and LX5 images fine.
After initial experimentation with the latest version I found the images to be super quality, containing lots of detail, appearing nice and sharp, with good colour and being free from noise. They are in my opinion better than those from Lightroom (both traditional and infrared images). Here is a section of the image above at 100% (click the image to zoom in).
The real decision maker for me is when I received an email the day after offering an upgrade to version 5 developer Studio (still limited to Panasonic cameras) for JPY3800 (about £24). Decision made!
I am now therefore using SilkyPix for my infrared RAW conversions and am extremely pleased with the results.