Camera Nitty Gritty - The Series Overview and Conclusions

Published by at

Over the last six months, I've tackled a number of phone camera-related issues head on and tried to draw conclusions, even though they sometimes fly in face of phone marketing. In this final part to the popular series, I'd like to provide an index to all the articles and also summarise some of the individual article conclusions.

In each case, following the bulleted link will take you to the full original article:

  • Part 1 - I looked at exposed camera glass on some phone models and investigated whether scratches really make a difference
     
    Surprisingly, I found that, for average camera phone use, even quite bad scratches on the outside camera glass make negligible difference to the quality of the finished photos - the reduction in light levels just isn't significant and the focussing point is far, far beyond the scratches themselves. The only caveat to this is when shooting into the sun, when the scratches cause lens flare and possibly unwanted lighting effects.
     
  • Part 2 - I investigated the pros and cons of LED/dual-LED/Xenon flashes in camera phones - does Xenon or dual LED flash make that much difference?
     
    I demonstrated quite conclusively the power of Xenon. Yes, dual LED flashes were naturally brighter than single LED, but even so they're many times weaker than Xenon. Secondly, the long shutter time needed for working with LED flash means that subjects in motion end up badly blurred - anyone taking a LED flash photo at a party or disco will know bad this blurring can get. In contrast, Xenon flash (seen on the Samsung G810 and Nokia N82) 'freezes the instant' superbly.
     
  • Part 3 - I investigated the fabled 'Megapixel myth' with an objective eye
     
    All other factors (light, etc) being identical, for most people's purposes, there's seemingly little point in chasing after higher megapixel numbers - at typical display or print dimensions, you simply won't be able to tell the difference. 3 or 5 megapixels seems to be a sweet spot here - 8 megapixels and above simply aren't needed for average users.
     
  • Part 4 - I focussed on the difference between good and poor optics (at the same camera resolution)
     
    Surprisingly, I found only a negligible difference in the quality, with the sensor fitted and the software algorithms used playing an equally important part. In summing up, I said that branded optics might be a clincher when agonising over two phones of similar camera spec, but not to use this as the be all and end all of choosing a camera-toting phone in the first place.
     
  • Part 5 -  I pitched all the top smartphone cameras head to head, in detail
     
    An atypical part in the series, I wanted to test the current state of play in terms of the best phone cameras out there. My comments on the 'megapixel myth' above notwithstanding, it seems that the 8 megapixel Samsung INNOV8 (i8510) is currently the pick of the phones tested, for semi-pro use, with an awesomely good sensor and optics. Even with its Xenon flash, the Nokia N82 couldn't match it in terms of ultimate performance.
     
  • Part 6 - AAS guest writer Dirk Snoyt took up the theme of camera phone flash research and got all technical on the theme of colours...
     
    The main point to take away from this point is, again, how superior Xenon flash is, this time in terms of the colour spectrum of the light it puts out - using an LED flash can result in different forms of colour skewing of the subject - which then have to be put through the mangle in photo processing software later in order to correct.
     
  • Part 7 -  I returned with a look at the difference optical zoom makes and asked the question "Is it better to have optical zoom or just much higher resolution?"
     
    The answer, it turns out, is that neither is 'better'. Zooming in by 3 times with a 3mp camera produces detail roughly equivalent to that from a 8 or 9mp camera unit. However, the use of optical zoom means lower light levels (one extra piece of glass in the way of the photons), greater expense in terms of hardware build costs and a much greater susceptibility to 'camera shake' - all reasons why, despite the apparent desirability, we probably won't see many camera phones with optical zoom in the future.
     
  • Part 8 - I looked at phone camera video capture potential
     
    Perhaps not surprisingly, the Nokia N82 and N95 took centre stage here with their great VGA capture, but they were pipped to the post by the Nokia N93, which gathered extra points for having stereo microphones as well. Tellingly, the Samsung INNOV8 again starred, with an initial focus in video recording mode, let down here only by an audio sync bug - with this fixed, it would win this test and thus have walked away with best phone camera for both stills and video....
     
  • Part 9 - I looked at video focussing differences in Nokia's Nseries
     
    I was particularly interested in the way Nokia preset the focus in video mode differently for different devices. The N82's camera seems to shoot video focussed nicely at a metre or so, whereas the more recent devices, such as the tested N96, are somewhat bizarrely focussed at infinity, meaning that they're not very good at shooting people.
     
  • Part 10 - I looked at the difference sensor and aperture size can make, for both stills and video.
     
    Comparing the new Nokia 5800 with an N95, I showed how a smaller aperture and correspondingly smaller sensor lead to dramatically different levels of digital 'noise' in low light conditions. Quantum physics even gets a mention! These effects are even more dramatic for video capture, incidentally, where getting as many photons of light to register per 1/30s frame is vitally important.
     
  • Part 11 - I looked at ways of using high Megapixel counts for intelligent cropping of photos and for Megapixel Microscopy.
     
    Inventing a new technique (or so I like to think), I show how the use of a high megapixel camera in macro focussing mode can lead to some fabulously detailed photos - when cropped, the result is effectively microscopy and very educational!
     
  • Part 12 - I looked at the various Scene modes and additional settings in Nokia's Camera application - are any of them needed?
     
    Actually, it turns out that almost none of them are needed. I demonstrated benefit from using 'Sports' mode and from tinkering with 'White balance', but for most practical purposes you need never feel guilty about simply leaving everything on 'Automatic' all the time.
 

So there you have it - my Camera Nitty Gritty series in a nutshell. With the advent of the Nokia N86 with variable aperture lens and 'third generation' LED flash, maybe I'll need to add to some of the parts above? And Dirk Snoyt and I still long for a Symbian OS device manufacturer to produce a really decent camera smartphone with Xenon flash - maybe Sony Ericsson's Idou will eventually turn out to be the one we've waited for?

Could we one day see a phone with 8mp sensor, branded optics, Xenon flash, focus in video capture and enough quality to finally silence the critics of phone cameras? Who knows, maybe 2009 is the year?

Steve Litchfield, All About Symbian, 22 Feb 2009

INNOV8 camera