A77 Gets DxOmarked - Things Looking Very Good for the NEX-7 Sensor!

[Amin Sabet]

   ---            

DxOmark today published their sensor tests for the A77 SLT camera, which has the same 24MP sensor as the NEX-7. According to the DxO data, the A77/NEX-7 sensor has essentially no deficiencies when compared to the latest 16MP Sony sensors, upping the resolution while maintaining parity or better in signal/noise, dynamic range, tonal range, and color sensitivity! Quite an impressive accomplishment from Sony.

I suspect that all the negative talk about A77/NEX-7 low light high ISO performance will be squashed when these cameras are out in photographers' hands and have RAW file support by Lightroom and other popular processing applications.

Link: DxOmark A77 Report via DPReview forums

 

[Y-V-E-S]

This is awesome news!!

 

[Y-V-E-S]

I'll be looking forward to purchasing the NEX-7! Thanks for the news Amin!

 

[Armanius]

That's good to know. Let's hope so!

 

[Vorkhosdn]

It's interesting the high colour accuracy it gets, and I was also expecting less dynamic range. High ISO score will be a bit better on the NEX-7 because of the lack of translucent mirror but it's quite clear that it's not the strenght of this sensor. For example, SNR at pixel level is lower than any current m4/3 sensor, even the 6yo 12mpx one. Let's hope NEX-5N will get tested soon: it will blow them away.

I keep thinking that for amateur users, 24mpx is way too much. 8mpx are enough for high quality A3 prints and most people they even don't print. For me it's the most negative thing of the camera. Moving a 5 layers PSD would require such an amount of RAM memory I would have to buy a new computer and an external hardrive. There is the option of binning, yes, but then I would be paying an extra amount of money for the sensor technology and engine processing power I won't use. The NEX-7 with the 5N sensor would have been "the camera".

 

[Amin Sabet]

For example, SNR at pixel level is lower than any current m4/3 sensor, even the 6yo 12mpx one.

Why look at SNR at the pixel level? That has no bearing on how the sensor will perform at the image level (ie, displayed final image on screen or in print). Seems to me that high ISO SNR is in fact a strength of this sensor. The folks at DxOmark think so too: DxOMark - Sony A77, measurements and review of the world-first 24 MP APS-C camera

 

[Vorkhosdn]

I said "at pixel level" just as a measuring of efficiency of each photosite. For example, Panasonic, when moving from the old 12mpx sensor to the one on the G3, was able to increase both the pixel count and the pixel quality, and it seems it's not the case with this NEX sensor. Of course, once resized to the same size, the NEX-7 image will look better in terms of noise than any m4/3 image. But I keep betting it won't beat the NEX-5N on high ISO prints.

 

[Amin Sabet]

But pixel level doesn't correlate with sensor efficiency per unit area of sensor. Based on the A77 testing, it looks like the NEX-7 should be at the level of the NEX-C3 on high ISO prints, which is much better than most 12-16MP APS-C sensors even if it turns out to be not quite as good as the 5N.

 

[Vorkhosdn]

I agree, the NEX-C3 is a good comparison model for noise on prints. I'm reading what I wrote and it seems somehow I'm saying the NEX-7 sensor is not that good, when I think it's just the opposite. Its overall quality is at least up to the NEX-5N, Pentax-K5, Nikon D-7000, etc, if not better. The main difference is that it's focused on resolution rather than on high ISO performance. Good news is that the latter it's still very good, while for the first is just the best APS-C option available and probably for a normal shooting, the resolution is more useful than a clean ISO 6400.