Read this Article about Sensor development and why canon has a hard time

JonSnow

EOS 80D
Sep 10, 2018
147
100
https://www.foto-schuhmacher.de/artikel/hardware/sensor-dilemma.html

the article is in german but you can use google translate.

https://translate.google.com/translate?hl=de&sl=de&tl=en&u=https://www.foto-schuhmacher.de/artikel/hardware/sensor-dilemma.html

it shows why canon has a hard time competing with sony.
and why it is a business decision, not some malicious intend, why canon sticks to sensor tech longer than sony.

it looks at the past, the current situation and the future.

i think it is well worth a read.
it gives some insight why canon decided in 2016 to sell sensors to other companys and how development of smartphone sensors paved the way for sony to dominate the market for ILC senors. it also touches the dangers for other company to rely on sony sensors.
 
Last edited:

JonSnow

EOS 80D
Sep 10, 2018
147
100
in short the article has one important point, that canon stopped being at the top of sensor development 10 years ago.

sony has the benefit of testing new technology (BSI, stacked sensors etc) in small sensors for smartphones.
these sensors sell in the millions. and R&D cost is therefore distributed over a ton of sensors.
than they only need to adapt the technology for bigger sensors. this is much cheaper than the initial development.
you have a smaller number of APS-C / FF sensors you can sell, but the adaption cost less than the original development.

you may noticed that new technology is today first seen in small sensors (BSI, stacked cmos was for years in smartphones before it came to big sensors).

  • It follows that all basic research and development of photo sensors has been taking place in these markets since at least 2012 and not for photographers. At best, photographers will receive the waste products of other markets years later . So BSI sensors have been produced since 2007, optimized since 2009 by Sony 2010 discussed everywhere and shortly thereafter used in smartphones (iPhone4s, etc.), but were only hesitantly and years later moving into the larger sensors of the classic photo cameras and came not until 2015 with Sony's a7RII and 2017 at Nikon's D850 in the full-frame. The already developed in 2012 for smartphones Stacked CMOS sensor was also only in 2017 at Sony's A9 use.
canon did not have these benefits. they do not sell smartphone sensors or sensors in such huge numbers to other companys.
canon tried to changed that in 2016. selling sensor to other companys makes sensor R&D more affordable.

  • Canon itself has already recognized that their own R & D is not as efficient as the one at Sony. But in my opinion, efficiency improvement programs in the Group are missing out on the central weak point. The main problem is that Canon needs to do a lot of research and development on its large sensors (mostly APS-C and full-frame), while Sony can do this on very small sensors for the smartphones. The latter is easier, cheaper and faster to perform. Subsequently, Sony adapted this only for large sensors. Canon is missing not only the sector of small sensors for testing, but Canon are also missing the huge economies of scale and profits from that sector. Therefore, I think it is out of the question that Canon - efficiency enhancement programs / austerity programs or not - can ever achieve the same sensor development efficiency as Sony. It follows, in turn, that Canon ultimately has to spend a lot more on R & D for the same image quality. That's a strategic disadvantage.

  • Solution options are only available in a few directions:
    • Cooperation with other camera / sensor companies to distribute costs, but also to let other companies participate in their knowledge. This was addressed in 2016. So far, however, demand has remained limited.
    • Resignation with a slightly lower sensor quality at the same product price. This seems to me currently the Aussitzpolitik, which one believes to be the market leader to afford. For some customers, the quality already achieved is indeed good enough.
    • Significant increase in investment in R & D with (hopefully) consequent increase in sensor quality, ultimately increasing product prices. In the current photo crisis, however, price increases are at least not sales-promoting. As the price-limiting size of cameras is added that the next competitor - Nikon - currently advertises with rather low prices for the pure housing at least comparable overall quality, and the sensors slightly higher image quality.
    • Dodge to other, more lucrative future markets. Canon seems to at least partially pursue this with video. But even there, sensor quality will play a decisive role very soon, so that the problem area is only shifted for a short time instead of being solved in the long term.

unfortunately canon still has not the numbers sony has. so it will be very hard to match sonys R&D.

sony will soon release a new generation of sensor. the EOS R as latest example basically uses the 5D MK4 sensor.
that 5D MK4 sensor has not catched up to sonys latest generation and will soon see another generation of sony sensors as competition.

the 18 MP sensor was another good example. it was basically in all canon rebel models for years.
they needed to sell a lot of these big aps-c sensors to get R&D cost back.
if canon would have benefited from the huge demand of smartphone sensors... you can be sure we would have seen more sensor models.

maybe we will see some positiv effects in the future. now that canon sells sensors for security cams, industrial cams etc.
but that canon will take the lead again has a small chance.

but depending on sony as sensor supplier has some negativ effects too, as the article points out.
 
Last edited:
Reactions: AlanF and Click

JonSnow

EOS 80D
Sep 10, 2018
147
100
one more thing.

stacked cmos sensors are more important then most people think today.
basically it will be possible to control every single pixel with future stacked sensors.

that means features are possible with these sensors (in camera) that are only possible in post today.
the bad thing is... only sony has this technology ready today. and is working on it for a long time.
it would mean a billion investment for canon to close the gap.

things that could be possible in camera with future stacked cmos sensors that can control every single pixel.

way better lens correction.
ISO values on pixel level (individual pixels getting different ISO values).
backlight motives could see faces lightend and the back darkend.
cameras that can shoot infrared/ultraviolet and visible light.
pol and ND filter on sensor level. even ND grad filters.
higher DR and lower noise on pixel level
fake DOF (like you can do in post today when you have a z-depth mask. depth mask could come from the AF phase detection.)
etc.

this stuff is in the future but it would not be possible with traditional cmos technology.

will we see it... i think so.
but i guess first in smartphone sensors and first for features that work for bigger areas of pixels.

think of it like the evolution of TV panels.
first we had EL wire, CCFL lamps behind the LCD panel, than we controlled regions with edge lighting, than even more control of smaller areas with full array backlighting (FALD ) and even better quality came with control of single pixels (OLED, MLED).

we may also see a decrease in megapixels with those sensor. as the electronics to control the pixels of course need some space.
 
Last edited:
Reactions: MayaTlab

privatebydesign

Would you take advice from a cartoons stuffed toy?
Jan 29, 2011
7,422
443
119
The premise is not very accurate.

First, the 135 format sensor with the maximum DR is currently a Canon, the 5D MkIV when developing the two RAW files contained in the single dual pixel RAW format, which rather destroys the argument the Canon can't compete. They lead!
Second, Sony sensors are not improving, they plateaued, indeed their highest performing sensor when measuring DR is two generations old, this rather contradicts the point that Sony are better at R&D.
Third, the technologies you mention have not been proven to scale well. BSI made a huge difference to small sensors, FF sensors not so much. This means the point about being able to scale features developed in smaller sensors has limited value.
 

JonSnow

EOS 80D
Sep 10, 2018
147
100
Second, Sony sensors are not improving, they plateaued, indeed their highest performing sensor when measuring DR is two generations old, this rather contradicts the point that Sony are better at R&D.
the main focus of the A9 was speed not image quality.
i don´t say sony can do miracles. :)

but you only saw the start of this technology yet. the sony A9 was the first camera of this kind (FF camera, lets ignore the sony RX 100 Mark IV).
it is correct that sony had to go a step back because of miniaturization issues. but canon is not there at all.
today you can´t cramp more and more features into the sensor (faster and more accurate AF etc) and at the same time enhance image quality (well you could but it would be expensive).

but this technology will enhance. and faster than traditional cmos sensors.


Third, the technologies you mention have not been proven to scale well. BSI made a huge difference to small sensors, FF sensors not so much.
image quality or dynamic range is only one part of the equation.

from DR to readout speed canon is not able to compete (head to head) and i fear what the next generation of sony sensors can do.

will the image quality difference be so huge that it will make a difference in real life?
as the article says, for many the current canon sensor technology is good enough.

and a sensor does not make a camera.
but sony also learns to make better cameras. thought slower than they learn to make good sensors....

5D MkIV when developing the two RAW files contained in the single dual pixel RAW format, which rather destroys the argument the Canon can't compete. They lead!
that again is misleading. sony could also just start to stack images.
not the same as reading both halves and on half of the dual pixel and combining them.
i know... but there are ways to increase DR with sonys new quad pixel sensor design.

i own the 5D Mk4 and i never really used this feature. i played around with DPRSplit and other tools but never made real use of DP raw.

also you can gain 0.8-1 stop if i am not wrong. that would make it barely equal to sonys best sensor.

the inherent advantages of BSI and stacked sensors are not something you can talk away.

it is correct that smaller sensors benefit more from BSI than bigger sensors. when we talk just about image quality.
but that is kind of missing the other benefits of stacked sensors.
readout speed for example is already an issue for canon sensors i guess.

don´t forget the A9 sensor is not the end of it, it is the start.
as fast as the A9 internally is (~20 times faster datatransfer than the A7r II) they still need some shortcuts to get that fast readout speed.
for example the banding issues could be fixed with the next generation (it´s because of how the readout works. it is not optimal yet).

prototypes with bigger and more enhanced dram caches already reached nearly 1000 frames a second.
and that is already old news.
it was a 19.3M pixels sensor and could produce 960 fps FHD (1,920 x 1,080) super slow motion video. (120 FPS for all pixel readout).

thought it´s useless when you have no fast storage medium.

i am not saying canon is doing nothing.
far from it. canon is patenting stuff like crazy for decades.
but sensor R&D sure is more expensive for them.... or lets say less profitable.
it´s good that they opend up to other customers, selling sensors to other companys. they had too.


ps:

i don´t want to come around like a sony fanboy.
i own a A6000 and it´s a nice travel camera. but my heart belongs to canon.
but i also don´t stick my head in the sand and ignore certain facts.
 
Last edited:

AlanF

5DSR
Aug 16, 2012
5,023
1,806
Thank you for the heads up on a most interesting article. A lot of it rings true and so some of it makes for some depressing reading: Nikon is in thrall to Sony who are now withholding their best sensors from them as well as from other manufacturers; the lack of transparency of Japanese companies; Canon having to play constant catch up; the increasing dominance of smart phones etc etc.
 

Keith_Reeder

No apologies for not suffering fools gladly...
Feb 8, 2014
687
143
58
Blyth, NE England
; Canon having to play constant catch up
Canon is not "playing catch-up" - it clearly has no inclination to do so, and as the evidence clearly indicates (people are still buying Canon cameras) there's no need to "catch up", as the hype around "superior" sensors does not translate to a Real World advantage for the vast majority.
 

Keith_Reeder

No apologies for not suffering fools gladly...
Feb 8, 2014
687
143
58
Blyth, NE England
Second, Sony sensors are not improving, they plateaued, indeed their highest performing sensor when measuring DR is two generations old, this rather contradicts the point that Sony are better at R&D.
I think that this is really The Penny Dropping with Sony (and Nikon - remember that the Canon 80D's sensor has better low ISO DR than the Nikon D5) that there's more to a good camera than the ability to grossly crank the shadows in badly-exposed low ISO images...
 

JonSnow

EOS 80D
Sep 10, 2018
147
100
Canon is not "playing catch-up" - it clearly has no inclination to do so, and as the evidence clearly indicates (people are still buying Canon cameras) there's no need to "catch up", as the hype around "superior" sensors does not translate to a Real World advantage for the vast majority.
i agree on the image quality side of things.
but going forward the benefits will be seen in other things.

we see it today with readout speed already.
readout speed correlates with AF speed and tracking accuracy.
 
Reactions: MayaTlab and AlanF

AlanF

5DSR
Aug 16, 2012
5,023
1,806
I think that this is really The Penny Dropping with Sony (and Nikon - remember that the Canon 80D's sensor has better low ISO DR than the Nikon D5) that there's more to a good camera than the ability to grossly crank the shadows in badly-exposed low ISO images...
From your point of view and for your needs, Keith, the differences might not be important. But, there are areas other than DR. Quite a few serious birders have gone over to Nikon and Sony because of much better tracking, which can be done because of the much faster data readout of the Sony sensors. The A9 has a stellar reputation for AF and tracking. I have been using an RX10 IV recently that has the A9 AF and can tell you it is amazing - it looks on to an object and can AF at 24 fps. I want Canon to catch up so I can use my superior Canons lenses on a Canon body that can fire away and track like that without being a 1DX series.
 

Bambel

EOS M50
Jun 12, 2018
32
25
http://www.chipworks.com/about-chipworks/overview/blog/full-frame-dslr-cameras-part-ii-canon-stays-course

Older article from 2012 but the interessting thing is that Canon used the same manufactoring process for a very long time. They pretty much squeezed everything out of this process but there is a barrier somewhere. Canon would need to replace it's well known but rather "crude" process (and thus machines) with a much more modern one. That costs lots of money and then they have to understand a new process and have to circumvent lots of patents. Not a good outlook.

On the other hand: would Canon be willing to ask Sony for APS and FF sensors? And is Sony willing to sell these sensors to Canon? Are there other sensor companies that could be a partner for Canon? The only thing that seems rather sure is that Canon on it's own will have a hard time to improve their sensors on it's own. Not so much about image quality but about readout speed that essential to video.

B.
 

dtaylor

Canon 5Ds
Jul 26, 2011
1,192
478
BSI was supposed to improve high ISO. It made no practical difference in FF sensors. Reading forum posts you would think Canon FF sensors must be behind on high ISO. Looking at actual RAW files they're right at the top.

Nikon and Sony's best DR sensors beat Canon's best DR sensor by 1 stop on DR. At base ISO. If you ignore dual pixel RAWs. That 1 stop amounts to a couple ticks on a noise slider when pushing shadows.

Nobody but Canon has DPAF. Which is a rather complicated sensor engineering feat. While I have video spec related complaints about the EOS R, early reviews indicate it's one of the best focusing MILCs out there. Meanwhile Nikon Z models are getting hammered for AF issues in early reviews. Do they not use Sony sensors? I guess you can't blame them. It's only the recently released gen 3 Sony A7 bodies which offer decent FF AF in the Sony mirrorless line.

For all the endless talk online about "superior" Sony sensors the 3 year old Canon 5Ds and 5Dsr still easily tie the D850 and A7r3 for image quality and large prints at low ISO, and are only very slightly behind at high ISO. If you consider lenses as well, the highest IQ 35mm format system in the world is Canon, and you have to go medium format to beat it.

Speaking of old sensors, there are a lot of posts that rag on Canon for modifying and reusing older sensor designs, like the modified 5D IV sensor in the EOS R. But Sony did the exact same thing with the A7r3 and A73. Yet have you read any complaints about it?

4chan/p recently had a post from a guy claiming to be an ex Sony employee. He said Sony pays people to make social media posts praising Sony technology and bashing technology from other brands. I know what you're thinking. It's 4chan. Anyone can say anything. But when I read repeated lengthy posts about Sony technologies which either do not exist yet, or have not made a meaningful impact on photography...all on a Canon forum no less...I start to wonder....
 

MayaTlab

EOS 80D
Oct 6, 2015
185
66
Speaking of old sensors, there are a lot of posts that rag on Canon for modifying and reusing older sensor designs, like the modified 5D IV sensor in the EOS R. But Sony did the exact same thing with the A7r3 and A73.
Too long a post to answer each part of it. So I'll focus on this one. If you knew what you were talking about, you'd know that the A7RII and A7RIII have a completely different sensor. In less than two years of development, Sony managed to cut by half the readout speed of their 42mp sensor while improving DR slightly. The A7RIII's sensor can be read in 14bits as fast as the A7RII could in 12bits, and twice faster in 12bits.
 

sebasan

EOS T7i
Mar 12, 2016
92
22
sebasantos.1x.com
BSI was supposed to improve high ISO. It made no practical difference in FF sensors. Reading forum posts you would think Canon FF sensors must be behind on high ISO. Looking at actual RAW files they're right at the top.

Nikon and Sony's best DR sensors beat Canon's best DR sensor by 1 stop on DR. At base ISO. If you ignore dual pixel RAWs. That 1 stop amounts to a couple ticks on a noise slider when pushing shadows.

Nobody but Canon has DPAF. Which is a rather complicated sensor engineering feat. While I have video spec related complaints about the EOS R, early reviews indicate it's one of the best focusing MILCs out there. Meanwhile Nikon Z models are getting hammered for AF issues in early reviews. Do they not use Sony sensors? I guess you can't blame them. It's only the recently released gen 3 Sony A7 bodies which offer decent FF AF in the Sony mirrorless line.

For all the endless talk online about "superior" Sony sensors the 3 year old Canon 5Ds and 5Dsr still easily tie the D850 and A7r3 for image quality and large prints at low ISO, and are only very slightly behind at high ISO. If you consider lenses as well, the highest IQ 35mm format system in the world is Canon, and you have to go medium format to beat it.

Speaking of old sensors, there are a lot of posts that rag on Canon for modifying and reusing older sensor designs, like the modified 5D IV sensor in the EOS R. But Sony did the exact same thing with the A7r3 and A73. Yet have you read any complaints about it?

4chan/p recently had a post from a guy claiming to be an ex Sony employee. He said Sony pays people to make social media posts praising Sony technology and bashing technology from other brands. I know what you're thinking. It's 4chan. Anyone can say anything. But when I read repeated lengthy posts about Sony technologies which either do not exist yet, or have not made a meaningful impact on photography...all on a Canon forum no less...I start to wonder....
Nailed