Sony A350: Full-Time Live View at 14.2MP
by Wesley Fink on April 3, 2008 3:00 AM EST- Posted in
- Digital Camera
Current Sony Lineup
Below is a comparison of current Sony models with the Sony 18-70mm f3.5-5.6 lens. For those who wish to buy the body alone, only the A350 ($799) and A700 ($1399) are available without a kit lens. A two-lens kit is available for the A200, A300, and A350. The two-lens kit adds the 75-300mm f4.5-5.6 to the A200K and the 55-200mm f4-5.6 to the A300/A350 kits. The two-lens kit adds $200 to the single lens kit prices.
 |
At 14.2MP, the A350 is the highest resolution Sony you can buy until the A900 is introduced later this year. The current top-of-the-line A700 is the only current CMOS model with a lower 12.2MP resolution. Canon will have a similar situation with their announced 12MP XSi entry-level model in their line with the prosumer 10MP 40D. As we have said in many articles, image quality is about more than sensor resolution. The resolution, sensor sensitivity, and sensor noise levels all contribute to the final image quality - all else being equal - and this is often reflected in the fact that top prosumer models often have a lower sensor resolution than less expensive entry-level models.
Facebook
Twitter
RSS
113 Comments
View All Comments
Wesley Fink - Monday, March 31, 2008 - link
The Nikon D300 has an MSRP of $1799, so it is more than twice the price of the A350. As we discussed in the review the price-comparable Nikon to the A350 would be the 10 megapixel D60, which probably uses the same sensor as the Sony A200.dug777 - Monday, March 31, 2008 - link
The D60 is $749 RRP with the kit VR lens (and you can already get it for less), which is rather a gap ;)According to your table this is $899 with a kit lens...head to head with the XSi (Nikon & Canon usually bracket their cameras, both feature & price-wise, rather than going head to head).
Heidfirst - Monday, March 31, 2008 - link
D60 is 10.2Mp with no live view - Sony's competitor to that is the A200 which lists at $599 msrp with kit 18-70mm.Wesley Fink - Monday, March 31, 2008 - link
The A350 WAS compared in features and handling to other entry level DSLR cameras. The only time the K20D and Canon 5D were used for comparison was when it came to the sensor. We were trying to answer questions about the noise and sensitivity of the 14.2 megapixel CCD sensor and nothing else in entry level is higher than 10 megapixel until the Canon XSi (450D) actually ships.The Pentax K20D is similar in resolution (14.6 megapixel) but CMOS so we thought that would be an interesting comparison. The full-frame Canon 5D at 12.2 megapixel is always mentioned for it's low noise at all ISO settings so we thought that comparison might answer some questions on photosite size that inevitably come up when higher resolution sensors are the subject.
We plan to compare the Canon XSi and A350 in more detail when the XSi is available for testing. The price is almost the same which will make the comparison interesting.
haplo602 - Monday, March 31, 2008 - link
I took a look at the EXIF data of the images at ISO 100:sony:
Exposure time: 0.2
F number: 2
Focal length in 35 mm film: 75
pentax:
Exposure time: 0.166667
F number: 1.4
Focal length in 35 mm film: 75
canon:
Exposure time: 0.166667
F number: 1.4
Lens focal length: 50 (of course this is a full frame)
I wondered why is the Pentax shot so much more SOFT !!! Why the hell are you using a different F-stop on comparisons ? They have the same crop factor, so no DOF difference issue (would not be there anyway). Even the Canon was shot with F 1.4 ...
Wesley Fink - Monday, March 31, 2008 - link
This is the first attempt at the crops to compare noise at the range of ISO settings. The procedures will be refined until the procedure better converys the differences in a web view. The full images are linked as they better show what we saw in the review.Focus was manual, but we left exposure on Program assuming exposure would be the same with f1.4 prime lenses. As you point out with your EXIF data, the camera programs make different assumptions about how to handle exposure in our test lighting. In the future we will also manually control the exposure, or at least use aperture priority, to assure the same aperture at all ISOs.
The other issues were the Image Stabilization features of the Sony and Pentax. Both advise turning off IS for tripod shots which is what we did. In the future we will always use remote shutter release since the Pentax "softness" is possibly the result of not having a remote shutter elease handy for the Pentax, combined with unintended aperture variations. In fairness the Canon 5D was shot with a remote cable but it exhibits similar softness on the web screen so we plan to look more closely at the crop method as a potential problem and refine that process. The downloadable full images do a much better job of showing the noise variations.
Maxington - Monday, March 31, 2008 - link
Why were they shot at f1.4 anyhow, all lenses perform worst at wide open. (I assume thats what haplo quoted in exif data)Should be shot at f8 in raw to eliminate as much lens variation as possible between brands, and to avoid narrow DOF focus errors being classed as "soft" images.
Sony already has a slight advantage since its not shooting wide open, for example. Not that it would be noticable really.
Wesley Fink - Monday, March 31, 2008 - link
You are correct. Future comparisons will use the common f1.4 lenses where possible but aperture will be constant. We plan to use two to four stops down at f2.8 to f5.6 depending on the lighting we use for the tests. f8 is a difficult choice for low ISOs with a scene lit by a 100-watt Tungsten bulb, but it would be a reasonable choice for studio lighting.haplo602 - Monday, March 31, 2008 - link
Exactly. I was wondering why the hell is the Pentax shot so soft ...Anyway all the lenses are f1.4, so stopping down to even f5.6 for better sharpness is desired.
If you are shooting for NOISE comparisons, you always dial in the same aperture/shutter speed on each camera after first settling down on an aperture that all lenses can handle and it is not the lowest one (5.6 - 8 are ideal). this way you get a nice constant EV on all systems at the set ISO.
as to the IS being off, man I suggest you read up on the self timer feature of the bodies. I bet all of them have one. thus you don't need a remote.
anyway I'd give the Sony system the credit for choosing the best aperture in the test :-)
I also credit that you are willing to improve the testing methodology, I'd suggest you read up established lens/camera testing sites on their methodology and adapt yo your conditions.
dug777 - Monday, March 31, 2008 - link
No desire to bash you here Wes, we're just providing some constructive criticism :)I think it's great that AT is branching out into this kind of stuff, and it's pedantic people like us who let you know if we think you're not doing it right ;)
Aside, I love my ML-L3 wireless shutter release, incredibly responsive and convenient, and it's as small and simple as you'd expect it to be.