Welcome to our community

Be a part of something great, join today!

Register Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

First Helium 8k footage

Zeb B said:
If the two solar panels have the same stated Wattage Output . . . then they would output the same Wattage
If two photo sensors have the same stated Signal to Noise ratio . . . then they would have the same Noise level

It's not magic - it's physics that I am trying to understand. Without in-camera gain / noise reduction and with the same stated SN ration, how can a sensor, with much smaller photosites, have less noise?
Click to expand...


See my previouse explanation. Wattage and snr are apples and oranges. A sensir can be more sebsutive while having the same signal to noise ratio, and therefore less need to turn up gain to induce more niose. Snr is likely a max but drops with gain, or exposure time (sort of).
 
Zeb B said:
My understanding as well. So question remains: how can you reduce noise with a much smaller photosite, with no in-camera gain or NR, when the laws of physics remain constant? Not trying to start anything here - just trying to understand.
Click to expand...
If the signal to noise ratio is the same the noise level should be close. But it is the smaller pixel that has the same signal to noise ratio so that shows that the technology has progressed.
Just to be crystal clear- for the same technology the noise level remains the same. With smaller pixels the fill factor (i.e signal) is less. So theoretically the signal to noise ratio for smaller pixels with the same tech should be lower. Since it is the same the only logical explanation is that the noise level on the helium sensor has been lowered with better tech.
That should be clear enough now....;-) I hope
Add to that the weaker olfp since the pixels are smaller, black shading improvements, debayer improvements. Of course based on pure physics smaller pixels create less electrical charge so on a sensor level the power consumption, and therefore heat generated is less. That affects the noise as well
 
Zeb B said:
Thank you - thank makes sense
Click to expand...

It is not that simple. Circuits in the quantum domain produce noise, as circuits randomly fluctuate and leak so your sampling circuit converting charge to data is going to be subject to.this improving that sampling improved the snr. Modern chip technologies go.towards reducongnthis leakage, and it seems that the technology that Intel chips were based on might be coming out of patent, and foundries are suddrnly offering similar low powered schemes.

Most top sensors should be 70%+, QE. The highest QE I saw around 11 years ago was over 80%, maybe near 90%. So that is old in technology terms. Now, raising a pixel from 70% to 100% is not going to double efficiency in the same size let alone half the size, and by what was previously said, 100% QE should have 0% noise, which as we can see is not the case. 70% efficence would imply around 4db sihnal to noise ratio if QE was the sole factor on noise, and 99% efficiency would imply less than 42 db, while sensors with I think 35% achieved around 42 db 12 years ago. However, by redoing the sensor, we can get more light into an efficiency. The reviled back side illumination is a way to do this, also changing other front of sensor pad technologies, which could account for this increased light gathering technology.

Modern research seems to be moving towards amplifying the incoming light instead, once again reviled but obviouse the only way forwards. So electron multiplication allows a single photon to move 100 electrons with no extra noise, dropping the relative noise floor compared to signal by over 40db. So 80db would become 120db. Obviously.this is not that case here, but it doesn't mean that there is not some over 100% QE electron multiplying technology going on. The original was a fairly passive technique or maybe 150% QE from memory, 20+ years ago I think. That would be beyond a 20 year patent and a possible candidate. I have my doubts about the 10,000% CE (or was that 10,000 electrons) technology used in deep field space cameras, as they ate meant to multiple very few photons compared to the many in a normal scene.

Now, all I have said is simplistic understanding, there is bound to be more to it, but beware of evrn more simplistic understandings, and be content with the statement Zeb, that it is because technology has advanced that newer smaller pixels match older larger ones. I'm sure Red can send a sensor scientist in here to explain fully why their new sensor is better than their old one and fill up a few pages with stuff we could barely understand, but do we want that.
 
Nice article Edward.
 
You must log in or register to reply here.
Back
Top