Monday, February 12, 2018

The Future of Silicon Wafer technology in Digital Photography

Silicon wafer technology is reached to its maximum limit now. There is nothing new to offer by all these camera companies when it comes to sensor technology to be honest. Number of Mega pixels are pretty much limit on both APS-C and Full Frame now. You cannot cramp more than 24MP in APS-C sensor size which ultimately gives you around 54-56MP over the Full frame if you keep the same pixel density. Why i am saying this? Reason is, if you have more then 24MP on APS-C size sensor, the Dynamic Range and SNR performance of that digital sensor degrades exponentially. So 24MP is the optimum threshold and all camera companies are making DSLR around it. Only Samsung have one model which got 30MP i guess on APS-C size sensor and its DR and ISO performance is way worse.

Now i have used a word "same pixel density". What it means that both APS-C and Full Frame having same number of pixel per square mm. Best examples are Nikon D7000 (16MP) which gave us D800 (36MP). Then we witnessed 7D MK II (20.2MP) which ultimately leads to Canon 5DS/5DSR. And then in last the world witnessed the Nikon D500 which leads to Nikon D850. All these cameras used the same pixel pitch which means the size of the pixel on APS-C and Full Frame remains the same. So what exactly makes any Full Frame camera works better then APS-C if both are having the same pixel pitch? We will discuss that in some future blogs. Lets stick to the sensor technology future right now.

Here is the funny part. The more pixels you going to push into the sensor (regardless its APS-C or Full Frame), the more DR you are going to lose. It means pixel are becoming smaller and smaller day by day and their ability of light capturing the light is becoming limited. The more pixel is going to be smaller, the quicker it will fill up with light which ultimately going to reduce the ability of recovering shadows or reducing highlights in post processing. The only salvation is nothing but "SNR firmware updates" which going to make the performance difference and help these companies to offer the new model with improved ISO performance based on keeping the same 24MP count. I guess I have unlimited times mentioned in many posts about this fact ISO has nothing to do with exposure because it is APPLIED Gain which applies on the "capturED" image after the sensor. All these camera manufacturing companies releasing different models based on the different SNR software. Best example is Nikon D750 which used the old Nikon D610 sensor but due to its strong SNR firmware its performance is still unparalleled in the category D750 stands in.

Now on horizon, there are 2-3 new tech coming. As now we have learned that you cannot have more than 24MP in APS-C sensor size, so what is selling point of future DSLR? Increased number of pixels or improved ISO perhaps !!! The only way to increase the pixel density is by changing the architecture of the sensor which is ...... BSI sensor. (Please google about BSI sensor if you don't know what it is). In brief, the traditional digital sensor pixel wiring from the top of the photosite which blocks the light by a huge amount. In BSI (Back Side Illumination) the wiring of the photosite is underneath the sensor which make photo receptor to capture more light then before.

By wiring the photosite (AKA pixel) from the back of the sensor we are actually having around roughly 30% more light coming into the same tight pixel which means we will be having the better native gain. Now in order to take advantage of having this extra gain, what these companies are working on right now is, they are cramping more pixel into the same tight APS-C size sensor and then using the same SNR technique which they are using on other camera models, they are maintaining the same (probably better) ISO performance on the APS-C size sensor by having MORE then 24MP. This is how companies are using SNR as an extra leg of the exposure triangle and playing with the sensor technology and making money at the same time.

Soon we will be having around 28-30MP max keeping the same Dynamic Range & ISO performance variance on Fujilfilm X-T3 just like X-T2, but this time, mega pixels will be more then X-T2. Just like X-T1 when it had 16MP but they released the X-T2 with more mega pixels and better SNR software which makes X-T2 a legendary camera status. Now history is going to repeat itself for X-T3 and again we will be having more mega pixel on X-T3 (thanks to BSI sensor) and with improved SNR. Other camera companies will be doing the same.

So BSI sensor is the next big thing which is already available now and giving these camera manufacturing companies to WORK AROUND their way (as silicon wafer technology is already hitting the ceiling/threshold of 24MP) by using BSI sensor. There are around 50 new ways of making any sensor efficiently using different architecture and sensor manufacture techniques and BSI sensor is just one example.

In future, you will be seeing curved sensor (which required total new lenses line up) and "Adaptive resistance" sensor which will be making HDR image right from the camera, and i am not talking about that HDR option in your DSLR which only works on JPG. in Adaptive resistance, it will be applying directly on RAW.

Time to wrap up this small blog showing the limitation and workaround of any digital sensor coming in modern day DSLR. Hope you like it.

Happy Learning guys.



  1. This is the first time I read such a remarkable article from the Pakistani Photographer.

    Best of luck

    1. Thanks for the kind words of appreciation. You will find more such information at my blog. Stay tune.

  2. This is very informative, I think I don't need to explore other articles by keep following this blog.