Lytro Illum – Revolution in photography or expensive toy?


When the first Lytro lightfield camera was announced in 2011 the buzz about it was incredible. A camera that could allow you to refocus a photo after you’ve taken it! No more blurry pics. It was every shit photographer’s dream.

Of course the original camera was a bit of a disappointment to say the least. It had a resolution of 10 MegaRays, which means absolutely nothing to anyone anywhere, and Lytro closely guarded the real resolution of the camera as if it was a state secret. But more on that later.

So, what is a lightfield camera? Essentially it’s a normal digital camera sensor with a special filter covered with microlenses on top, These microlenses capture light coming in at different angles so the final image records light at different angles allowing it to capture depth information along with the image. I won’t pretend that I know how everything is processed into the finished result, but the end result is what Lytro call a ‘live image’, which is one that can have focus and a limited amount of 3d perspective movement altered in post-processing.

But the original lytro camera with a fixed lens in a tiny square tube was a bit of a toy. We all wanted to wait to see if they would produce a DSLR-type version with a real lens and high resolution. Well… they did. Sort of.  And this is it, the Lytro Illum.

Recently there have been a bunch of price reductions on the camera, and finally it got to a price point which I figured was worth giving it a try, so I ordered one.

I thought about doing an unboxing video to show you the cool box it’s packed in. Someone obviously wanted to put this box design on their CV before applying for a design job at Apple. But watching unboxing videos never appealed to me, it always felt like the time when you’re forced to sit and watch at Christmas as your siblings open their presents while you’re really just thinking about what you’re getting next.

Still, for those of you who are fascinated in watching other people unbox their cameras, here is one someone else did. Bear in mind that I didn’t have the patience to watch it through the end. Maybe he drops it. Maybe it spontaneously combusts, I didn’t check.


I am not sure if the same designer who crafted the box worked on the actual design of the camera. It’s certainly different. It does somewhat resemble a fat iphone welded at an odd angle to a can of soup. Still, there’s no doubting that it’s distinctive.

In real use the camera can be a bit awkward. It’s not a DSLR, there’s no optical or electronic viewfinder, just the display on the back which can be tilted up and down but not left to right. I don’t like using displays like this in daylight but, in general, this one isn’t too bad.

To be fair the distinctive look means you are always going to get noticed when using it, and the admiring glances of technogeeks who know what you have in your hands.  The lens is a crazy zoom with an equivalent 35mm zoom of 30-250mm af f/2.0. YES! an f/2.0 zoom from 30-250mm. Except of course it’s nothing of the sort. In reality its a 9.5mm – 77.8mm zoom. The barrel has two rotating rings for zoom and focus with two nice blue rings aro…. WAIT A MOMENT. DID YOU SAY FOCUS RING? Why do you have to adjust the focus on a light field camera? Well, because of the zoom range of the camera you can’t actually have everything in focus, you have effectively a band of focus that you define – so it’s like you have an f/2.0 lens but in reality you have f/16 of focus to play with – so you aim your focus point to the center of the focus range you want to capture.

And here is where the clash of marketing and technology hits us like a brick to the face. You see, someone came up with the idea during design that this really needs a zoom lens and a 9.5mm – 77.8 zoom range was just what was needed.  Except there is absolutely no point to it. When zoomed to full zoom there’s almost no situation where you get enough separation between your foreground objects and background to make a good live photo. Maybe if you were to catch a low-flying bird going in front of the moon… 99% of the time you’ll be using it at the widest possible settings, which makes me wonder whether a smaller and more portable non-zoom lens would have been better all along. Still 30-250mm f/2.0 sounds impressive and when I tell that to all my camera-geek friends I know it will impress them – isn’t that the whole point?

Now, back to the tricky subject of what resolution the camera really is:

Wikipedia says the real resolution of files created by the original Lytro was 1080×1080 pixels, but that doesn’t really tally with reality. Someone has reverse engineered the file format which shows the original lytro has a 3280 x 3280 pixel sensor – around 10.7 megapixels  or close to the 10 megarays figure that Lytro originally quoted.  But – and this is the big thing, the image is actually composed of a honeycomb of circles containing the image from each little microlens.

lytro copy

These are about 10 x 10 pixels across  so one could argue that if you take each microlens as being a single pixel (and if you aren’t you’re going to lose sharpness as the intermediate pixels have to be interpolated) then you’re REAL maximum sharp resolution from the original Lytro camera is a whopping 328 x 328 pixels, or 0.1 megapixels (it’s actually not that straight forward because as you can see each alternate line of lenses is offset, and each microlens has vignetting). But there’s no way you’re going to get a sharp image at 1080×1080 out of the original Lytro.

Now, the new Lytro Illum claims a ’40 megaray’ sensor. Which presumably means it’s using a 40 megapixel sensor underneath the bubblewrap coating. The claimed resolution of this is 2450x1634px which gives us a 4 megapixel image. Assuming the microlenses are the same size I’d expect the real useable resolution to be 816 x 544 pixels, which when I compare to the real results I get out of the camera seems about right. But the sensor size is tiny – 1/2″ (6.4 x 4.8 mm) according to Wikipedia and a more sensible 1″ (12.8mm x 9.6mm)  according to more reliable sources.   A 40 megapixel sensor within this size range might sound unlikely, but such a sensor was launched in the Nokia Lumia 1020 phone in 2013 – so maybe the sensor in the Lytro is related to this mobile phone sensor.

Still 816×544 pixels relates to just over 0.4 megapixels, a far cry from the 4 megapixel images we were promised! This means if we wanted a TRUE 4 megapixel image out of a light field camera we’d need a camera sensor of 24500 x 16340 pixel resolution, or a 400 megapixel sensor. Plus we’d need to shrink the microlenses down and I’m not even sure the optics would allow that.  So let’s assume we have to have at least a full frame sensor – meaning an even more huge lens than we have now. Plus file sizes would be the same as a 400 megapixel raw file. Your 16GB memory card would hold the same number of shots as a 35mm film roll. So don’t hold your breath for a very high resolution version of the Lytro, it’s not coming any day soon.

Of course, Lytro say, none of this is really relevant because I’m completely misunderstanding what this camera is for. it’s not for crap photographers to refocus their bad shots, it’s to create these new live images to share with the world. Like this one I prepared earlier of a Hermit Crab at Bird World (yes, I know it’s not a bird) …


Click on it, drag it around and see the perspective. And then get bored and go and do something else. Because yes, for all the wonder of these live images they are not going to change the world of photography for the very important reason that you can’t really do much with them except stick them in a blog or on Facebook. There are some crazy people who seem to spend half their lives putting things on blogs or Facebook (ahem) so maybe for us  maybe for them it’s a handy tool.

My big concern is that this only works if Lytro host the images for you. Once they go out of business (or get bought out for their patents and closed down) all our lovely creations will disappear from the web – forever. If you’re reading this article in ten years time you may be wondering what the grey box above is. It was a live image, except it’s no longer living.

If you’ve read this far you may be mistaken for thinking I dislike this camera. If you haven’t read this far then you really should go and read it in order because otherwise my comments won’t make much sense.  But, getting back to the point, I don’t dislike this camera. Actually I’ve enjoyed using it. But I can’t find any real serious use for it.

But there are some great non-serious uses for it. For example, you can create animations from your live images. Here’s a beautiful live photo of a glove stuck on a pole.

It’s kind of fun and this way at least you are in charge of your images, and they’ll remain online as long as you want. Or as long as YouTube survives, if you stick them on there.

But for me the real attraction was in creating 3D images, you see it can export a stereo pair 3d image from a single live image. And this is quite handy. Except for the most part my little Fujifilm FinePix Real 3D W3 camera does the same job better. Go back and look at my crab live picture. Notice how parts of the antennae are cut out because the camera couldn’t tell the difference between the antenna and spots on the crab behind it. You don’t get those sorts of problems with a two-lens 3D camera. And if I really wanted to do it properly I’d get a couple of cheap 2nd hand EOS 500D cameras, stick EF 50mm f/1.8 II lenses on them both and create a rig where they both fire at once.

Also, I’m mostly using the 3D images to add to my mineralogical website, and it seems that despite me adding support for showing these images as true stereo images on a 3D TV or 3D monitor I’m probably the only person actually using it.

Another problem (and I feel bad for adding more into this review) is that the file sizes are huge. Remember the raw file is  a raw file from a 40 megapixel sensor, so you’re dealing with raw files not far off the size of those from a 5DS/5DSR. And when you import images it has to process them to convert them into a live image, and even on my Mac Pro 2013, which is no slouch, this takes 20-30 seconds per image.

I really, really wanted to be impressed by the Lytro Illum. And I keep thinking that maybe I just need to figure out exactly what I need to do with it and how to use it. There’s certainly a possibility of creating some interesting live photo animations of action scenes, when used at a wedding for example, but I see this more as something that could be thrown into a video edit than anything else – at least here the resolution is not so much of an issue.

I do keep wondering what the camera would be like without the microlenses, a futuristic 40 megapixel camera with a 30-250mm f/2.0 zoom. Of course it’s stuck at f/2.0, there’s no aperture control at all. And the optics are all designed to focus on the microlenses so it probably wouldn’t even work. But when you start wondering whether the best thing to do with a light field camera is rip out the light field system that’s not a good sign about the future of the technology.

It’s amusing reading articles gushing with superlatives about the Lytro Illum, such as this one which proudly claims “this is the camera of the future” but I can’t keep wondering which I’d rather look forward to, a 1 megapixel light field camera or a traditional camera with the 100 megapixel sensor that would be needed to make it work.

I’m going to keep playing with the Lytro – and playing is exactly the right verb to use for this product – because I really, really want to find more positive things to say about it. I will report back!

This article has been read 1119 times.


You may also like...

4 Responses

  1. Steven Rialto says:

    I don’t think you are the only one that feels this way about the camera, and the price tag is certainly high. Where there is an advantage over the Fuji W3 is in the production of output that can be used to make lenticular prints, either through the Lenticular export function in Desktop or by exporting a depth map + color image for use elsewhere. I detail the process here: . But even this is limited, you can only create so much parallax with a single depth map before things get weird. Thanks for the review.

    • jolyon says:

      Thanks for your comments – the price of the camera has now fallen significantly (at least here in the UK) – but still I think to myself that I paid as much as a cheap Canon ‘L’ lens for this, and I know which I’m likely to use longer.

  2. mark says:

    What do you think of the new Light Camera? How does it compare to the Lytro?

    “Light L16: multiple images from different lenses combine to give wider dynamic and focal range than a single camera can capture (Credit: Light)” News article is here: “”

    Anyhow, I ordered the original cube style lytro camera at the last price drop on Amazon. Had a lot of fun with it. Pain in the but though to take pictures with, but got good results.

    I am surprised this technology hasn’t blossomed for on-line advertisers. My guess is because Lytro hasn’t released/licensed their image viewer to the world, so advert companies (and even their clients) probably don’t want to trust their proprietary images to a 3rd party.

    If I was in advertising, I would be eating this up. You could seriously highlight different aspects of your product with this thing. You could view your product in a whole new way.

    Instead of a static image, advertisers could have a dynamic image that moves and can be manipulated to some degree by the viewer.

    I know there are similar things like this out there, but this puts it all into one tiny package with the capability of not really needing any post-processing other than to link it to the Lytro site and to publish it where-ever you want.

    Anyhow, this camera makes me wish I was involved in advertising. I think it could be huge.

    I am also interested in learning about the 16-lens LIGHT camera too. Their website is here: The LIGHT camera seems like it has a lot of potential too. However, I am not sure it has the perspective shift that the LYTRO camera has.

    Here are some early pictures that I took with the cube style Lytro camera. Have to find time to take more shots. Wishing I had the $400 to get the big Lytro today on Amazon. It is today’s (Nov. 4, 2015) deal on Amazon for $399. (Although I think B&HPhoto is selling it now for $599 retail).

    Anyhow, thanks for listening. Hopefully you will reply with some of your insights.


    • jolyon says:

      Well, I was quite sceptical about this device when I first read about it, but I’ll look forward to seeing what a real photographer thinks about it when it comes out! Or maybe I’ll get a chance to review one!

      Very disappointed to hear about the Lytro Illum price decrease, was hoping to sell mine!

Leave a Reply to mark Cancel reply

Your email address will not be published. Required fields are marked *