A57 or A65?

Started May 27, 2012 | Discussions thread
Tom2572 Senior Member • Posts: 1,129
Re: A57 or A65?

philbot wrote:

theswede wrote:

philbot wrote:

there will be a tipping point at which it stops holding on to high contrast detail and apply smudging.

While this is true ...

This is largely because of the physics, the 24MP sensor is going to start off (at lower ISO) with more image detail, but it decreases as you move up the ISO scale, crossing over the 16MP at some point, and ending up a little bit worse at extreme ISO's.

... this is not true. There is no law of physics which mandates this.

At a given sensor area low light performance depends on the total light gathering capability and read noise of the photosites covering that area. There is no reason to conclude that a higher photosite density will necessarily affect the noise per area either up or down. It is perfectly possible to create a sensor with higher photosite count and equal light gathering capability per area (in fact it will probably be a slight bit better, as long as photosite quality can be maintained) and comparable read noise performance.

On 1:1 viewing the image from the higher photosite sensor may have more visible noise than the image from the lower photosite sensor, that is to be expected, but there is absolutely no necessity for this to be true when viewing them at the same size relative to the sensor size.


I agree with your theory, which is pure theory, but I think it isn't what was being discussed..

The discussion was about a general statement that a 24MP sensor always produces a better image then a 16MP sensor, so immediately I am only considering like for like, i.e. comparable technology with the only difference being pixel size.

In which case, you have to consider all the non-linear side effects that each smaller site has with regards to noise (based on identical technologies, so read noise and shot noise constraints and parameters are identical).. as well as other non linearities due to structural / physical elements such as microlenses..

I think it's OK to state the laws of physics mandate this, but it has to be in context, which I hope I've just added.. if not, then I retract that, but the principle I am sure is correct.. You can't base a theory on having lower relative read noise on one vs the other, because you'd have to say why less dense sensor wouldn't use the same lower relative read noise design, and the same for shot noise/electronics noise etc..

I largely used http://www.clarkvision.com/articles/digital.sensor.performance.summary/index.html as my source of information many moons ago..

Seems to me it's you that needs to learn how to read. I absolutely did not make a general statement that a 24mp sensor always produced a better image than a 16mp sensor, The general statement that I made was that the a65 will always produce a better image if you are willing to process RAW files yourself than an a57, and I will continue to stand behind that one. And again, I'm standing behind that statement as an a57 owner who also thinks that the a57 produces better in-camera JPEGs than the a65.

Just for grins though, how does your law of physics explain the D800? Is Nikon governed by a different set of laws than Sony is?

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow