This post may contain affiliate links(s). An affiliate link means I may earn advertising/referral fees if you make a purchase through my link, without any additional cost to you
. It helps to keep this site afloat. Thank you in advance for your support. If you like what we do here, maybe buy me a coffee.
A document of a Canon 63MP full frame sensor has appeared - legit or hoax?
A document from EOSHD appeared describing a 63MP CMOS Sensor of Canon manufacture.
This could be a legitimate sensor, or it could be a hoax. It should be mentioned that the user on EOSHD did accurately leak some Sony sensor data in the past.
A few things I find odd about this or reasons why I don't think it'll end up in a camera;
- The user also mentions it was made with a 300nm process. Canon has used 130,180 and 500nm processes that I'm aware of. I would fail to see the reason why Canon would upgrade some equipment to an interim step of 300nm.
- The read noise of this is far worse than for instance an 80D sensor scaled. At worst case, this sensor should be no worse than the 80D sensor, as it's the same pixel pitch as the 80D sensor.
- The DR matches the 6D Mark II DR curve and not the current generation 5D Mark IV or even 80D DR curves.
Calculated DR for ISO's
This is exactly like the 6D Mark II DR curve. (btw, if you ever wanted proof that the 6D Mark II used on sensor ADC's. here it is)
- If you look at Canon's actual real datasheets for sensors (go here https://canon-cmos-sensors.com/) they have revision marks and notes in the footers and a big red "specs are subject to change without notice" in the header. This document does not have either.
- The pixel pitch should have been 3.6um, not 3.7um (35mm / 9740 full width )
- It's too slow (even by Canon standards) 5.2 fps at 12 bit depth means it would be around 1.3 fps at 14 bit assuming it's using slope ADC's (which canon uses). Canon releasing a 1.3 fps stills camera? Not only that but DPAF takes around twice the read time so if you want AF you're looking at .6 FPS or a glorious 2.6 fps in 12 bit ("speed mode")
- We won't even mention the rolling shutter effect that this camera would show but with a 188ms sensor read time (in the "speed mode" no less), it will be at least twice as bad as the EOS R in full sensor readout electronic shutter.
It's entirely possible, sure. It's full frame, it's DPAF. It ticks off those boxes.
However; faced against some of the new Sony sensors that are coming out - this one would be the 6D Mark II equivalent of a high MP camera from Canon. Literally. This would have the same DR as the 6D Mark II from ISO 100 to ISO 3200. Exactly the same. You thought the internet crashed when the 6D Mark Ii was released? Hah. It will be a distant fond memory when the firestorm of outrage occurs if an EOS Rs comes out with this sensor. I'll probably be out there with a torch and pitchfork with the rest of you.
Personally, I think this legitimate sensor document, even with the errors. But I also think it is an internal use datasheet from Canon about a product that never came to market, which is using a prior generation of technology. They may release it as another sensor they are selling, or it got scrapped. If you notice the sensors that Canon released ie: the 120MXS and 3U5MGXS. They never ended up in an actual Canon camera, and the 35MMFHDXS_A ended up in this. Let's face it. Canon doesn't need to create a datasheet unless they are planning on selling it externally. It should also be noted that none of the three sensors Canon has for sale externally would be considered the same generation as the 5D Mark IV or equivalent sensor.
There's just too much "wrong" with this sensor as far as using it in a high-resolution camera as far as I'm concerned.
At least I hope so ;)
blog comments powered by