Compact Camera vs DSLR – a silly comparison?

Most keen photographers have always faced a dilemma – their DSLR (or SLR for those who still use film) and standard zoom produce very good results, but carrying one all the time is a pain and opportunities are everywhere! A small camera is the solution, but small digital cameras are usually compromised by limited ISO performance, they’re not often that small and even their best results aren’t as good – at least that’s what I’ve found having used several (small film cameras are a different matter). A test is in order…

So to see if things have changed here’s a test between a two-year old mid range DSLR with an upgraded kit lens against a new top of the range compact. Not a fair test on the surface, but who said anything about fair? The differences in size and weight are obvious but the results are a bit of a surprise….

The APSC sensor format Canon 60D with interchangeable EF-S 15-85mm (24-135mm equiv) lens on the left, Sony RX100 with smaller one inch sensor on the right with a fixed 28-100mm lens. The 60D boasts 18Mp, the RX100 20Mp – a negligible difference in practice.

The Canon has been used consistently for over two years, and has never failed to impress over ten thousand images with a wide variety of lenses. The Sony is relatively new (three months)  but is up to 1000 shots already. It’s images are more ‘consumer’ oriented with brighter colours and what looks like more sharpening, but very good nevertheless.
The Sony’s lens is a bright f1.8 to f4.9 across it’s zoom range, the 15-85mm a more modest f3.5 to f5.6. I’ve no complaints about the handling of either camera, neither having any irritating quirks which would drive you mad. My personal choice for useable maximum ISO is 800 on the RX100 but the Canon can be pushed further to 3200 in an emergency.

The Sony is doing a lot of processing to work around the design compromises of fitting such a tiny fast lens into a small body. Here’s a close up (ish) wide angle image with distortion correction on and off (done in the Sony Raw converter). Although the correction is done very well my initial thoughts would be that this much correction must result in poor edge performance – we’ll see! It’s worth stressing that this correction has to be explicitly switched off in the RAW converter to see this – you won’t see it on the camera’s replay function or in JPGs or RAWs by default.

Sony -distortion correction on and off

The main ‘problem’ with the Sony is the colour rendition – reds, greens and yellows are all a bit ‘off’ for my taste, but shooting in RAW and using a correcting colour profiles in ACR (see Maurizio Piraccini’s website here) fixes the problem to give a more subtle result.

Corrected colour – here a red postbox, the corrected on the left and the straight RAW to JPG result on the right. DPReview found the same thing in their (much more scientific and exhaustive) test.

So – on to the mini test and it will be familiar to anyone who’s read the film and lens test from earlier in the year – there’s a lot more vegetation now though! All shots in RAW and converted to JPG using the supplier’s RAW converter. The Canon’s ISO setting was 100, the Sony’s 125 (it’s native ISO). I haven’t worried about colour here as it’s important to compare default outputs.

Wideangle on both lenses – the 15-85mm Canon is a bit wider than the Sony – 24mm vs 28mm, but not significant for these tests.

Starting at max aperture, this definitely a surprise and a significant difference. The Sony is producing very sharp results (it’s sharpening is at a higher level by default), and the edges which have been heavily ‘corrected’ aren’t bad at all.

Canon – 15mm setting f3.5. Centre and top right crops.

Sony 28mm setting f2. f1.8 would gave resulted in more overexposure.  Centre and top right crops.

At mid apertures things are much more even – mid apertures usually produce the best results in all lenses.

Canon – 15mm setting f5.6. Centre and top right crops.

Sony 28mm setting f4.5. Centre and top right crops.

On to approximately a 50mm setting :-

Canon at 50mm f5 (max aperture at this focal length) and excellent.

Sony at around the same focal length (not exactly – hence the slightly different edge crop – apologies). This is good too!

Finally at tele setting – 135mm equiv on the Canon, 100mm on the Sony.

Canon at 85mm f5.6 (max aperture at this focal length) – bit soft at the edge but OK.

So definitely a surprise. I checked then double checked that the images were correctly attributed, but it was right first time! The little Sony is matching or even exceeding the Canon 15-85mm in terms of sharpness and detail, as well as having a wider maximum aperture. As I remember this lens alone cost as much as the compact camera!

The differences are probably down to the default sharpening parameters in the Sony, and highly polished image optimisation for a fixed zoom lens – the Canon can have hundreds of different lenses attached and can’t optimise images ‘in camera’ for all of them.

The Sony isn’t a replacement for the 60D – far from it. There’s no optical viewfinder for a start (composing on an LCD in bright sunshine is pure guesswork), the lens is fixed and the 60D’s sharpness and colour rendition is much more neutral and allow more latitude in post processing. Having said that, the RX100 is producing very impressive results without any work in terms of sharpness, and the ability to tweak the results in pp means that the gap between DSLR’s and compacts has definitely narrowed and I can use the RX100 with confidence in most situations.

Hope you find this useful – thanks for looking!

p.s I’m, not (unfortunately) sponsored by Canon or Sony – just using the cameras….

Advertisement