This post focuses on the light metering of the A99, A900 and A77 and other measurable values like file sizes. Since most of the testing I wanted to do was qualitative, I figured I would get the quantitative parts out of the way, right off the bat. But as I’ve found out, the numbers may not tell the whole story…
A99 w/ 85mm f/1.4 ZA, A77 w/24mm f/2 SSM ZA, A900 w/ 70-300mm f/4.5-5.6 G SSM
I wanted to get a shot of the three bodies side by side. Unfortunately I had no good camera left to shoot with, so I used my cell phone to get them all in one shot.
I took all of the best untouched RAW files, changed nothing on them, and I exported them using identical settings in Lightroom to JPEG format. Then I used Windows Explorer to add the columns with the data I wanted to display, sorted by file size, took a couple of screen shots and added some lines in Microsoft Paint. The horizontal yellow line is above the first appearance of a body/file type combo reading top-down. The horizontal green line is the first appearance of a body/file type combo reading bottom-up. The extreme quickest and slowest shutter speeds for each body are in the orange boxes. The A900 has an ISO range of 100-6400, where the A77 can go 50-16000 (I only tested up to 12800), and the A99 can go 50-25600. The A900 had shutter speeds ranging from 1/160th of a second at ISO 6400 and f/5.6 to 8/10ths of a second at ISO 100 and f/8. The A77 had speeds ranging from 1/400th of a second at ISO 12800 and f/5.6 to 1.3 seconds at ISO 50 and f/8. Finally, the A99 had speeds ranging from 1/500th of a second at ISO 25600 and f/5.6 to 2 full seconds at ISO 50 and f/8.
As you can see from the table, the A900 is the winner when it comes to files taking up the most disk space, with the A77 shortly behind. (Update: I now have a pretty good idea why the A900 shots are larger) The surprise winner of the lightest files is the A99. Surprising, to say the least. The two sharpest A99 files took up the least amount of hard drive space in JPEG format. The A900 actually had smaller JPEG exports at ISO 100, despite the extra pixels. Apples-to-apples, the A77 has the biggest JPEG exports at ISO 100 through ISO 6400 versus both bodies and larger than all of the A99 ISO 12800 and 25600 with the A77 ISO set at 12800. The ISO 12800 export of the A77 at f/8 is around 34 megabytes. Here are the extreme high ends of all three bodies, all at f/8. The A900 at ISO 6400, A77 at ISO 12800, and A99 at ISO 25600.
This isn’t a fair “apples to apples” comparison. The ISO values are not equal, but this example shows me a couple things. The first thing I noticed was the A77 depth of field is much wider than the two full-frame bodies. Much more of the paintbrush grain is evident in the shot. The words on the rightmost yellow crayon are barely distinguishable on the A900 sample, completely blurred on the A77 sample, and you can make out individual letters, even a diacritic on the A99 sample. A fiber present between the middle two crayons is evident on the A99 and A900 sample, but you really have to look on the A77 sample. On the middle yellow crayon, the A99 and A77 samples show the word “lemon” where the A900 sample is a bit bright.
Out of curiosity, I am putting the f/5.6 A77 sample here to show a shallower depth of field.
The f/5.6 sample still shows more of the paintbrush grain, though to my eye, the word “lemon” is a bit brighter.
For anyone annoyed by the noise I have made you witness, here is some eye relief.
Hopefully your eyes feel better.
Breaking the files down by shutter speed (or Exposure time as Windows Explorer states) shows something totally different.
Admittedly, there is a lot going on here, and I was nice enough to provide a key this time. The A77 had the quickest exposure times all-up. Between the A99 and A900, there were a few ties, however the A900 was faster a few times. The A99 never experienced the fastest shutter speed. I imagine since the A77 is at a different focal length the discrepancy is explained. I am more interested in seeing the difference in exposure between an A99 and A900 tie versus when the A900 is faster than the A99.
Here is the tie at ISO 200 f/8. Admittedly, I may have been slightly out of focus on the A900, or experienced a little bit of mirror slap as the A99 shot just murders the A900 shot in sharpness. But the point of this exercise is exposure. I also cropped less of the rook on the A99 shot so it appears larger when downsampled to 320px wide. Again, the exposure is the key. At the same shutter speed, I’d say the A99 has a slight edge between these two in brightness if you look at the orange wood blending into the blue clear plastic, but the white of the magic marker looks brighter on the A900 sample.
For completeness here is the same shot on the A77
To me it is inconclusive. Maybe I need to look at all three side-by-side.
Ahh. Now I can tell. The A77 is a tad darker than both, it seems. Maybe the A77 tried to be a little too fast.
This sample shows again I may have been out of focus on the A900 a tad, but the point is not necessarily the focus more than the exposure. At 1/30s, the A99 is slower than the A900 at 1/40s. But you can see the A99 is much brighter. And at ISO 1600, I am thrilled with the noise. Just for kicks here is the A77 sample at the same metrics, ISO 1600, f/5.6 but at 1/50s.
To me, the brown in the background is slightly brighter, so the A77 may, in fact be overexposing these shots. Here is another example at ISO 800 and f/5.6. The A99 and A900 tied at 1/20s where the A77 shot at 1/25s.
On this shot, the A77 is a tad brighter than the A900, which seems to be brighter than the A99. Now I am completely confused.
So this wraps up part II of this excursion. I’ve learned I may need to reshoot with the A900 and compensate for mirror slap, and I may need to put the A99 in Aperture Priority, take note of the metering and then put the A900 and A77 in manual and meter accordingly. The next post in this test will be the ISO noise test. I have a feeling I know who the winner will be here, however I think there may be some surprises along the way.
I am planning on posting the full size samples shortly. I need to find a place to put the files, they are very large.