In depth review of the Lytro Light Field Camera | Hacker News Reader

In depth review of the Lytro Light Field Camera

dpreview.com

32 pointswxs15 comments

I'm trying to think how I would use it. I, like everyone else, use shallow DOF to emphasize different aspects of the image.

If I want to emphasize background and foreground in different shots I just take two shots at different focus points.

If I want everything to be in focus I stop down the lens.

I have to agree with the dpreview: what artistic problem does this solve?

I can think of a contrived example where there is no time to make two shots (a basket ball player shooting a hoop) and where you want two parts of the image at different distances both in focus and the rest blurred. You could do it with this device, but I would just use image manipulation.

What I would really like is the blade runner camera (or rather dataformat) since the blade runner picture seems to be a composite of many shots.

I believe the target market here are amateurs who don't have the skill or reflexes to focus properly ahead of time. I certainly fall into that category.

My point and shoot camera has an auto everything mode. It figures out what scene to use (eg night versus landscape versus portait), where to do metering and what to focus on. I've never had a problem nor needed any skill. You can press the shutter half way down and it will follow focus whatever is appropriate showing you what it is doing.

The replacement model even has a touchscreen so if you want you can simply touch whatever it is you wanted focussed. Again absolutely no skill needed.

Reflexes is a little more interesting. Digital pictures are essentially free so there isn't any harm in taking lots of them. Amongst the specs listed on comparison sites are times to first shot and burst modes, the combination of which means you are likely to get something good despite your reflexes. I can't find any mention of the lytro having a burst mode.

It doesn't look like the lytro has any benefit compared to a point and shoot for conventional picture taking including focus and reflexes. But as we all know from The Innovator's Dilemma that isn't what matters - look to where things are different. One thing that is immediately apparent is it can build a 3D map of the image as it knows the distance to every pixel.

Your point and shoot doesn't do depth of field very well, let alone let you adjust it after the fact. It's also likely that focusing is much much slower for you than the lytro.

In any case, the lytro probably isn't for you right now.

Focussing on the point and shoot is 0.2 to 0.4 seconds. (Latest model is 0.1 seconds). Note the comment I was replying to claiming the lytro market is amateurs with slow reflexes. I don't think the lytro is even close to a good fit for that market even for amateurs with slow focussing and reflexes. An auto everything point and shoot is far superior n every aspect. Note how the review does mention you have to contrive shots to make the lytro functionality worthwhile.

I agree with you in that I don't see the "living pictures" aspect as particularly useful, I think you'll probably have a given point you want to be in focus, and rarely will want to switch between foreground and background.

I see it as more useful in that it eliminates the issue where you take a great photo and then realize that the focus was off. It's a shame the resolution is so low, but I think this is a very interesting first commercial foray into the area.

If it was possible to do your own processing of the raw data, some very cool effects could likely be achieved. After all, this camera is capturing precise info on the distance to every pixel.

Examples: realistic atmospheric effects, adjustment of color/contrast based on distance (so you could emphasize what you want), correction of color casts in mixed-lighting scenes, motion blur where the length/direction of the blur was a function of distance, synthesized 3d/stereo views.

Figuring out exactly what to do with the data in these kinds of images is going to take years of experimentation and creative effort. At the moment, they aren’t that interesting for non-technical users who just want to make pictures.

>I have to agree with the dpreview: what artistic problem does this solve?

Sloppy focusing. Or maybe just a quick shot where you didn't have time to properly check focus? Or your AF wasn't fast enough or accurate enough?

If this worked in 'real' cameras, I'm sure sports and nature photogs would be happy about it.

I've been crowing about these issues for months now. They have built something that solves a problem (getting the important thing in the image in focus) that has been better solved by multiple other means (AF, hyperfocal distance + high DOF) without the extremely adverse side effects (super low resolution, bad low light performance from the sensor). Then they invented a new problem (the desire to switch between two important subjects after the fact), that for some reason gadget bloggers fell in love with. Gimmicks should not masquerade as revolutions.

Ya, but this is a holographic camera. The stupid thing actually captures honest to goodness holograms -- not even just stereograms. Why they don't play this up more is beyond me. Maybe the holographic viewing angle is too narrow to be useful beyond viewing the image head-on?

Summary: how to ruin a great idea with bad software and a bad business model. Who wants a digital camera that can only work with one company's website, and then only if you have a Mac?

This isn't a summary, this is your own takeaway. I don't know what article you read, but the OS and site limitations were a passing mention in this one, and though they didn't like this, it didn't even end up in their own conclusion.

It's probably not a killer limitation to require a Mac for a revolutionary photography device.

The problem is that it's not wonderful enough to qualify as a thousand-dollar specialist camera, it's not cheap enough to be a nifty holiday gift ($50-150) and it isn't widely usable enough to be a killer point-n-shoot replacement.

The post-processing on Lytro's own system though... that makes it aggravating. I'll bet 50 cents that Marcus Ranum's nudes are verboten by their policy.

I look forward to the next generation of the technology, but I'm not convinced that Lytro is going to be around to produce it.

It used to be that if you wanted to use your camera's RAW files, you were locked in to one piece of software or had to wait forever for support in other software. This has gotten better, but it's still a problem. RAW support for my camera didn't come for a month and a half after it was released.

Can we get people to post their design ideas for this product to improve screen size and general sexiness?

Reply on news.ycombinator.com