Luminous Landscape Forum

Equipment & Techniques => Mirrorless Cameras => Topic started by: Guillermo Luijk on October 09, 2023, 01:49:29 pm

Title: Noise performance test for Sony A7 II - synthetic RAW file
Post by: Guillermo Luijk on October 09, 2023, 01:49:29 pm

I built a synthetic RAW file for my camera (Sony A7 II), to be able to check for noise performance for any arbitrary exposure situation, both visually and with numerical measurements over some patches. It's something like a more complete version of the DPreview ISO invariance test, but in a single RAW file.

Should you be interested in checking out (English translation top right):

https://www.overfitting.net/2023/10/test-raw-de-rendimiento-en-ruido-para.html

(https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgmOzY2fCluj5ozUMpjDc7ZpeolcTBgL1AUUtWp9BCEe58yxS6V-_c0QvwVFIIc7QzYHOW7i1NHAqHh0TX8Xv3ZB31bmgPVl47RykkmlJO4FLvTjcErV7FNJaE5g4iEcZBTz-fRml288LSchVgrT-K9Hyqrb4_87-U5ElhNjWw7AjTogUuXzPDQamtkgaFc/s16000/crop512.jpg)

(https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhEUFtymq5ZNbLbOxqMgRAe7o-iKyLP11mPX5798Bp1NyNiLOdJEX_5nkhzdmisVtYGKj_wF5fGvTGevUMSDodstOl2ofFkCyd7f95aKmP8_Zf0fg7AMUMtIXx3jSC0oG4YsdOci_DtcpC_CVvZe-5D5fJWZ7BH3oK6W339ltTa9jkgWwd-CXUg8IS4yAP_/s16000/isoinvariancetest512.jpg)

(https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEh40ywTaMIKCx8PrA6FVCoJXJjkUiLRM2VK7D1GuSbS-yhIIO8ct5WX3JLzpQ52kvg3I0UQqlfdgiya9RX6G4ZVL9XBNL8ruJpbLPhvQorCYnJXUbELUCS-Iq_AaQEnuOwf48-ueBva40DLpOrc4HiASgVWxIvxMJVLzq6PaxrSJSfgoO7o74Upkm6a_yDH/s16000/instrucciones.png)

(https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEh_mhotH7T3EDbJjJ2X2J-D0tHKRR2aWPgc3EiZgvlgnyuHSeeoAvHAIbqo4LBYMDMPkm_WfUDfu_mf5a5UPAtlU8OxfqYHXOGbTQCyaUjr3C7uT9ENGCd9PUbUeXZUa7DwKrYvlBz1OfZ5vptwsgh6xMg3a_b0OcWjwz3zp2gTUvK91ULRrSgKRAFQf0Ue/s16000/snrisoinvariance.png)

Regards

Title: Re: Noise performance test for Sony A7 II - synthetic RAW file
Post by: jrp on October 29, 2023, 01:48:37 pm

An interesting experiment, as ever.  Thanks.

Title: Re: Noise performance test for Sony A7 II - synthetic RAW file
Post by: Frans Waterlander on October 30, 2023, 11:56:44 am

Guillermo,

Have you ever measured sensor-heating-related SNR degradation caused by switching the camera on in extended photo and video modes? This subject seems taboo for many as it will show a weakness in mirrorless cameras that nobody wants to talk about, but I want to.

Title: Re: Noise performance test for Sony A7 II - synthetic RAW file
Post by: Ray on November 22, 2023, 08:05:11 pm

Thanks for this very detailed analysis, Guillermo. I read your full article and came across the following statement, which I found very relevant.

"However, given the low contrast of our test scene, I dare to doubt that said ISO invariance remains so radical in scenes with a lot of dynamic range, where we will have to "raise" much deeper shadows."

The issue of noise, for me, has mainly been about the noise in the shadows. When you shoot in RAW mode, even with a full 'exposure to the right', the shadows are always very dark untill they are lifted during processing. It's that lifting of the shadows that usually reveals significant noise. Such noise was very obvious with the early Canon digital cameras, which is why I used to set my camera to take 3 shots when shooting landscapes, using automatic exposure settings such as -2EV to +2EV. I would then merge the 3 images in photoshop. The under-exposed shot ensured I didn't blow any highlights, and the over-exposed shot reduced the noise in the shadows.

One reason I switched from Canon to Nikon was to avoid this complicated process. With my latest Nikon camera, the mirrorless Z50, the only reason I feel the need to increase the ISO setting is when  the circumstances require a faster shutter speed, whilst still maintaining a full 'expose to the right'. Of course, to do this, I use manual exposure.