|
This will be the normal running review of the Sony A900. As I learn more about the camera, I'll provide
the information. I should have a good jump on things by national boss day (gag), or by Wednesday or Thursday. I
used the A900 on the left, and the A700 on the right for the product comparison shots below. Look
and feel. The A900 is very similar to the A700, especially the backside layout, where all the buttons are the
same, and placed in the same area. The only major difference is the large viewfinder rising noticeably higher than the
A700. On top you have a small LCD that shows basic camera settings, much the same as the later film cameras. There
is also a button that will illuminate the display for about 5 seconds, (orange color) then it will automatically go off, or
you can hit the button again and it will go off. It has the same buttons on top as the A700, they're just placed
differently to accommodate the LCD. The sides are slightly rounded on the A900, but have the same features, as in the
memory card slots on the right side, (you still can't write to both cards simultaneously) and all the connector ports
on the left side, the only difference here is the flash sync terminal is on the front now instead of on the side. The
right side grip is not as deep as the A700, so if you have long fingers, the A700 may feel more comfortable and less restrictive,
this is a surprise. My finger tips jam hard against the A900 camera body. Also, there is no more eye-start auto-focus,
(good, as it was a worthless feature IMHO) so no sensors on the grip. The weight and size differences are not a problem,
and are not very noticeable unless held side-by-side. The color of the A900 body is more of a gloss black instead of
the flat black, with a little spatter thrown in. It matches the lenses better. The last item of note is the rubber
insert around the tripod socket on the bottom of the camera, which offers a good non-slip surface with less possible twist
as compared to the smooth bottom A700, see the image below. In a nutshell, if you've been using an
A700, you'll be very comfortable with the A900. It almost could be an upgraded A700, though you'll be missing
a pop up flash, but gaining a full frame sensor. File sizes in "fine" quality
mode the average is 7mp. Extra fine averages nearly 20mb, I've had some at more than 30mb! Using "Vivid"
picture style will up the file size about 10% based on my testing. RAW+JPEG runs in the neighborhood of 45mb-50mb, and
the access light is on for about 10 seconds for each shot. TIFF conversions are nearly 70mb!
|

|
A900 box and contents |

|
A900 with Sony 50mm F/1.4 lens |

|
Front |

|
Bottom, notice rubber insert around tripod socket, A900, left |

|
Top side |

|
Backs look the same |

|
Left side |

|
Right side |
Let's check out the ISO of the new A900 and compare it to the A700.
First, I'll tell you what the settings where on the cameras. Both were in "fine"
quality mode, both were in "Vivid" picture style, both were set at "low" for high ISO noise reduction,
and both were using the original firmware. The colors of the A900 are a little more bold in the "vivid" mode
than the A700, but don't worry too much about the colors being exactly the same. Pay more attention to loss of details
and background appearance; such as colors visible in the dark area above the subjects. I moved the cameras
around so I could capture the same crop area, since the A900 has no crop factor, and is 24mp, while the A700 has a crop factor
of 1.5x and is 12mp. Everything is pretty close in the crops above side-by-side, but they aren't exact. The A900
was about 8-9ft (2.5m) away from the subject. That's pretty good detail capture at ISO 6400. The subjects
are a baseball glove I bought in 1945 when I got home from the war, and a standard Sony neck strap. I had a larger,
and more interesting display, but the size of the crops and my page layout made me choose what you see above.
The
images at ISO 100-400 look pretty close to me, it looks like the A900 sensor may be a little sharper, but I'll have to
do more testing to confirm that. It looks like the A700 starts to smear more at ISO 800 and has slightly more chroma
noise. At ISO 1600, the same story, then by ISO 3200, the A900 clearly pulls ahead, with much more detail present, even
more so at ISO 6400, the fabric lines are still visible on the strap. It appears Sony is using less luminance noise
reduction here with the new A900, as you see much less smearing of details compared to the old yesterdays news A700(!)
Each line in the neck strap above is about 1/16" (1.6mm) wide. ISO 3200 comparisons below.
Above are crops of ISO 3200 from the A900 and A700. The A900 crops were taken from the 13mp setting,
the A700, from the regular 12.2mp setting. They both look the same to me, the A900 has a slight advantage in pixels,
but the noise reduction looks to be nearly the same, and dropping the A900 down to 13mp doesn't offer much better control
of noise, I thought it would, silly me. Below are some noise reduction settings.
|
I'd probably use the "low" setting for everyday use, or maybe "off" if you don't
mind a little clean up work later. I would definitely stay clear of the "high" setting, as that produces a
smeared mess and is not recoverable later. Photos above were taken at ISO 6400. Below are
samples of creative picture styles.
|
Standard
|
Standard +3
|
|
|
Vivid
|
Vivid +3
|
|
|
I've added a few samples of creative style images above. The "standard" mode is a
little boring, but +3 adds quite a bit of zing. Vivd looks good to me, and I most often find myself using it for landscape
shots. It's more saturated than the A700 vivid setting. Vivid +3 is over the top, and blows some detail, so
I would stay away from that setting. Below are detail comparisons between the cameras.
This represents the real difference in detail with an additional 12mp and larger sensor. I've
upsized the A700 to match the A900 and cropped a portion of both. Printed as you see it here would show the entire image
as over 5' wide, (1.52m).
The images above represent the real difference in detail when you adjust the distance so the crops come
out the same. In other words, you have to back off the A700 a factor of 1.0736 to get the same crop area. If the
A900 is 100' (30.48m) from the subject, you would have to move the A700, or other APS-C camera back 7.36' (2.24M)
to get the same crops. Most people think you use the standard 1.5x factor, but you can't when you have the additional
megapixels. All camera settings are identical for both, including exposure.
Now we have actual, same distance crops for you, meaning both images were taken from the same spot, where
the A700 has a 1.5x crop factor advantage. The same lens was used for both cameras, and all settings and exposures were
the same. I see more detail in the A700, but not as much as I would have guessed.
The same images here, but I adjusted the A900 size to match the A700. I see a difference, but not
very much. I thought I'd have to haul out the APS-C camera to take the best telephoto shots, but it looks like I
can leave it at home. Macro crops next.

|
A900 F/11 |

|
A700 F/11 |
Above are macro images from the A900 and A700. As you can see, there isn't a whole lot of difference
between the two. I used the KM 28-75mm F/2.8 for the shots, though I haven't tested this lens yet. Next,
some APS-C on-off mode examples.

|
Sigma 10-20mm @ 10mm |

|
Sigma 10-20mm @10mm with APS-C crop on. |

|
Sigma 10-20mm @ 10mm APS-C crop off and cropped manually from top shot. |
Here's a good way to pack extra stuff in your image when using a super wide angle APS-C lens.
If you use an aftermarket lens like the Sigma 10-20mm F/4-5.6, the camera won't read the "DT" code to automatically
crop the image, so it thinks you're using a regular full frame lens, and you naturally wind up with heavy vignetting.
If you set the camera to APS-C "on" you'll notice the camera crops off a little more than necessary, but if
you leave it off, you can do your own careful cropping, and you'll end up with about 14mm equivalent, instead of 15mm,
it makes a noticeable difference. There is still some vignetting in the bottom photo, but that small amount would be
hidden behind the picture frame. The last shot (full resolution) is 4545x3030, or 13.7mp. The middle image is
the automatic crop of 3984x2656 or 10.6mp, Sony calls it 11mp on the camera. The top shot is the full 6048x4032. Check
out the nasty issues with the KM 17-35mm F/2.8-4.

|
No filter, 17mm |
|

|
Tiffen 77CGND6 filter used, 17mm. |
|

|
You just never know what you'll get. 17mm. |
|

|
Tiffen filter and F/22, bad choice. 17mm. |
|
Above are the results (resized full images) of using the KM 17-35mm F/2.8-4 lens when the sun is near,
or in the image. All pictures here were taken at the widest 17mm setting, but there were similar issues all the way
to 35mm. In certain situations when the sun is near the image but not in the frame, you can block it with your hand,
the included hood doesn't help much at all. In the top left shot I probably could've used my hand to block the
big green blobs. With an APS-C sensor, most problems here don't fall within the sensor area, that's one good
reason to stick to those cameras. Check out the RAW conversion samples below.
Our boring (but very convenient) scene shows some blown highlights, and we'll see if we can try and
bring back the details using Sony's Image data converter SR vr 3, and Adobe's Lightroom vr 2.1 with ACR 4.6.
Original image
|
Amateur adjusted JPEG, or one click photo fix
|
|
|
The left shot shows the cropped original image. On the right is what you'd get if the average
person tried to recover from a bad JPEG shot using quick fix buttons. If you use exposure compensation, you shouldn't
be getting shots like this, and wouldn't have to worry about getting yourself out of a mess. Or, if you have plenty
of time and memory, you can shoot RAW, and adjust as necessary. See below.
Sony IDC RAW -2ev
|
Adobe Lightroom RAW -2ev
|
|
|
The two shots above were adjusted by simple dialing in -2.0ev, that's it, no other adjustment or
enhancements. Two things I notice right off are Sony's use of sharpening in IDC, where as ACR is soft, and the very
obvious blown color from IDC. The top arrow points out the difference in cloud tones, the next lower arrow shows the
color from the side of the house is gone with IDC, the arrow by the leaf shows another area of highlight blowout, and the
last arrow shows a line of harsh color transition between shadow areas because of color clips.
Adobe Lightroom RAW -3ev
|
Adobe Lightroom RAW -2ev
|
|
|
I've used -3.0ev on the left crop, and the same -2.0ev ACR image on the right, IDC only goes to -2.0ev.
You can eke out a little more detail by going to -3.0ev using ACR, though I think about 2-2.5 stops is about the limit, especially
when you consider you have to make the image come back to life, the rest of the picture is obviously too dark here.
See the loss of color information in some of the tan blocks, (inset) which are rendered as blotchy pink with a tiny bit of
white mixed in. You could go ahead and repair that small amount of damage, but you also have to try and give the image
some dimension, it's really flat as it stands now. We don't have to worry about this problem with IDC, because
there is no color to work with. In this example, we're really pushing it, a properly exposed shot would be much
easier to save.
Sony IDC final
|
Adobe Lightroom ACR final
|
|
|
Cropped finals above, the whole image below.

|
Original full image |

|
Adjusted full RAW image, Lightroom ACR |
Here is the final quick fixed RAW image as compared to the original up top. The whole process took
less than one minute. You could do even better if you spent some time going over the image carefully. This is
an extreme example with loads of overexposed areas. In real life, you would adjust the exposure to match conditions.
In a case like this I would definitely use RAW. Use exposure compensation for better results.
Sony RAW IDC final
|
JPEG using -1.0ev, adjusted
|
|
|
Here I shot a JPEG fine using -1.0ev, then applied basic levels adjustments to make it match the IDC
RAW image on the left. I probably could've used -2/3ev like I do many times, and it wouldn't be noticeable.
The clouds are slightly different in the right image because I shot this picture a few seconds later than the RAW image used
beside it. I think it should be clear that you would be better off shooting JPEG if you only have Sony Image data converter
SR vr 3. It's really more important to get the correct exposure than to try and clean up a poorly exposed image
later. In severe cases where you think you need the extra range, by all means use RAW, just make sure you process it
with something better than Sony's IDC program. IDC is terrible and shouldn't be used in my opinion. In
some conditions you may be able to retrieve about one stop of highlight information, depending on how over-exposed it is.
In the case here, it brings back some detail, but loses too much color. The color is gone even when you initially process
the image without adjusting the exposure. Note: I used no color enhancements on any images for these comparisons. In a nut-shell. You probably don't need to shoot in RAW mode for normal conditions,
just use appropriate exposure compensation when needed. If you only have IDC, don't worry about RAW, you might end
up with a better picture using JPEG. This concludes the review. I may add some other stuff later, but I'll be sure and post it on the homepage so you
won't miss anything. Thanks for visiting. If I've helped you, please help me and make a small
donation using the links on the homepage or contact page. Thanks.
|
|