The actual sensitivity of each indicated ISO is measured using the same shots as are used to measure ISO noise levels, we simply compare the exposure for each shot to the metered light level (using a calibrated Sekonic L-358), middle gray matched. We estimate the accuracy of these results to be +/- 1/6 EV (the margin of error given in the ISO specifications). We found that measured ISO from the Sony SLT-A65 is roughly 1/3 stop higher than indicated across the ISO range - so ISO 100 indicated = ISO 125 measured.
Noise and Noise Reduction (JPEG)
The SLT-A65's image quality is as good as identical to its bigger sister model A-77. At 100% on screen, noise is visible in JPEGs taken across the A65's entire range of ISO sensitivity settings if you look closely at areas of plain tone (like a blue sky for instance) but starts to become really noticeable at ISO 800 and above. Up to ISO 3200, the A65's noise-reduction system keeps things relatively smooth but above ISO 1600 you can clearly see that fine detail starts to suffer. In terms of detail reproduction at default settings there is quite a leap from ISO 800 to 1600 but this is only really apparent when images are viewed like this, critically, at 100% on screen. The noise graph, shown here, doesn't tell the whole story and it is clear that noise reduction is keeping measured noise levels low.
By ISO 16,000, even with noise reduction set to 'low' the medium-contrast detail in this test scene is all but obliterated. It is worth noting though that in 'real world' shooting at this setting, high-contrast scene elements are still relatively well-preserved, which makes images usable (just) for small prints and web use.
That said, the A65 is unequivocally a pretty noisy camera towards the upper end of its ISO sensitivity scale and if you plan on using the A65's highest ISO settings you should keep your expectations relatively modest.
RAW noise (ACR 6.6 Beta, noise reduction set to zero)
The amount of NR applied 'under the hood' is not high, but it does vary by camera (Adobe is attempting to normalize output across different sensors), so we are still looking at a balance of noise and noise reduction, rather than pure noise levels. However, the use of the most popular third-party raw converter is intended to give a photographically relevant result, rather than simply comparing sensor performance in an abstract manner.
Converting the A65 images in Adobe Camera RAW 6.6 Beta confirms the suspicion that we had when examining the camera's JPEG output. The A65's 24MP sensor is producing significantly higher noise levels than the competition, especially towards the top of its ISO sensitivity range. At ISO 1600 and above the disparity is obvious, and becomes more so as the ISO sensitivity increases. Above ISO 12,800, chroma noise from the A65's raw output is very high indeed.
As always, you should keep in mind that this is a pixel-level judgement and these files contain a lot of pixels. In the real world, with judicious use of post-capture noise reduction you can get much more from the A65's raw output than you might expect from the images and graphs on this page. That said, we've found that at high ISO settings in poor light, it is hard to 'rescue' images taken at ISO 6400 and above, even with careful raw post-processing.