Still no news about a Canon shift in sensor fabrication?

[jrista]

Thank you, Ankorwatt. Guess I missed this one. So I hope, they come up with something similar at a lower cost and even more improved within the next 4 to 5 years, once the 5DIV or the 5DV (or whatever they may call it by then) is released. What is the reason for the higher internal cost at Canon? March 2011 hurricane?

Keep in mind, ankorwatt only presumes that Canon cannot produce a FULL FRAME sensor on a 180nm fab process. As far as I know, Canon does have the ability to manufacture APS-C and smaller sensors in their 180nm fabs on smaller wafers (8in, rather than 12in). Technically speaking, one could produce FF sensors on smaller wafers as well, it would just be terribly inefficient and therefor prohibitively expensive.

I would not be surprised if the 70D sensor (an APS-C, not FF, sensor) WAS produced on 8in wafers with a 180nm process. They would certainly have to do something in order to be capable of producing 40 million pixels, along with the extra transistor logic to support simultaneous single-readout as well as binned readout. That is a lot more transistors per pixel than Canon ever used to have. I would be rather surprised if they are capable of doing that with a 500nm process.

Canon's primary low-iso DR problem has less to do with fabrication process than it does with the fact that they use off-die, high frequency ADC (and when parallel DIGIC chips are used, that increases the chances of strong vertical banding with an even pitch.) If Canon has finally moved to a 180nm process, I think there is more hope of them finally moving to an on-die parallel ADC solution. Who knows if they will also move to digital readout like Exmor, but I think at the very least moving to on-die hyper-parallel ADC will help their read noise issues.

 

[David Hull]

The big unknown at this time is whether Canon will use the same sensor in the 7DII or not.

If the 7DII has the same sensor as the 70D, Canon will have a hard time, IMO, convincing people to spend more on the 7DII.
The 70D specs are quite good already, so why bother?

It would be a completely different story, though, if the 7DII has better image quality than the 70D.

So, it makes more sense for Canon to put a better sensor in the 7DII - but who knows what they will actually do.

I have long thought (and I have seen you post along the same lines) that the cost of the external AFE would drive Canon to integrate (or partner with a sensor vendor who does) but this hasn’t seemed to be the case. Their cameras seem to sell well despite somewhat poorer performance on the low ISO end. My gut feel is that the 5DIII is outselling the D800 (for example) despite the higher (MSRP) price and the lower base ISO performance.

I think that the bulk of camera buyers outside of guys like us that hang out on gear forums grousing over DxO results don’t pay much attention to the sensor details. Even I, having participated in these types of discussions for the last few years, bought a 5DIII. I did this with full knowledge of the test data and full understanding of the potential implications having seen Horshack’s demos, and the endless postings from Mikael and others. One of the main influences on my decision was the fact that I used the 5DII for several years and never found any IQ deficiency that wasn’t relatively easy to work around. My biggest complaint was the AF system which they fixed.

I still think that ultimately it will be the cost that makes them improve the sensors but apparently for now their cost model seems to be working. From an IQ perspective there really is no difference until you push it into the corner and most users never do that.

 

[jrista]

The big unknown at this time is whether Canon will use the same sensor in the 7DII or not.

If the 7DII has the same sensor as the 70D, Canon will have a hard time, IMO, convincing people to spend more on the 7DII.
The 70D specs are quite good already, so why bother?

It would be a completely different story, though, if the 7DII has better image quality than the 70D.

So, it makes more sense for Canon to put a better sensor in the 7DII - but who knows what they will actually do.

I have long thought (and I have seen you post along the same lines) that the cost of the external AFE would drive Canon to integrate (or partner with a sensor vendor who does) but this hasn’t seemed to be the case. Their cameras seem to sell well despite somewhat poorer performance on the low ISO end. My gut feel is that the 5DIII is outselling the D800 (for example) despite the higher (MSRP) price and the lower base ISO performance.

I think that the bulk of camera buyers outside of guys like us that hang out on gear forums grousing over DxO results don’t pay much attention to the sensor details. Even I, having participated in these types of discussions for the last few years, bought a 5DIII. I did this with full knowledge of the test data and full understanding of the potential implications having seen Horshack’s demos, and the endless postings from Mikael and others. One of the main influences on my decision was the fact that I used the 5DII for several years and never found any IQ deficiency that wasn’t relatively easy to work around. My biggest complaint was the AF system which they fixed.

I still think that ultimately it will be the cost that makes them improve the sensors but apparently for now their cost model seems to be working. From an IQ perspective there really is no difference until you push it into the corner and most users never do that.

Well said.

 

[Rienzphotoz]

sony is making 63% of it´s profit from insurance these days.

the electronic part of sony is a money grave.

sony may has invested the most in manufacturing plants... but it does not PAY.


http://www.nytimes.com/2013/05/28/business/global/sonys-bread-and-butter-its-not-electronics.html?pagewanted=all&_r=0

Interesting

 

[jrista]

sony is making 63% of it´s profit from insurance these days.

the electronic part of sony is a money grave.

it´s more and more likely sony will split off or even sell the electronics department.
canon would be dumb to depend on sony for sensor manufacturing in the long run.

and i have not much trust in the sony DSLR system either.

sony may has invested the most in manufacturing plants... but it does not PAY.


http://www.nytimes.com/2013/05/28/business/global/sonys-bread-and-butter-its-not-electronics.html?pagewanted=all&_r=0

you are wrong, Sonys sensor department is healthy, they earn money from the mobile sensors up to 24x36mm
that other departments such as television, consumer electronics, etc. are not going so well is another question

Sorry, but the linked reference begs to differ. This is what I've been saying for months, if not a year. Sony has had to go so far into debt to build their electronics industry over the last decade that their debt costs too much. They pay an inordinate amount of money to service their debt, and the income they make off of their electronics products isn't enough to cover it. Electronics, especially today, are low margin products...its tough to make meaningful profit on any electronic device...be it TV, phone, or CMOS chip. The problem becomes even worse when Sony undercuts themselves while trying to undercut the competition (i.e. $399 for PS4 in order to undercut Microsoft's $499 XBox One). Just "being top dog" isn't enough if you aren't making any money on the product. Sony may have won some hearts with a $399 PS4, but if it doesn't pay the bills...will there actually BE any PS4s for those heats to buy when the time comes? Exmor is an amazing CIS, but if Sony isn't around in a few years to actually continue supporting a highly competitive environment because they folded under too much debt and not enough revenue, what good is it?

 

[x-vision]

Wise words as usual David. Hope you have a nice summer as we have.

+1

 

[David Hull]

I for my part think that Canon made the switch to a new process with the 70D sensor. Canon admited, that 18MP was the limit with the process they had then. So if they now make an APS-C sensor with 40million photodiodes (there are two diodes that can be read seperatly in every of the 20.2MP), then I think they definitly made the move to a new process.

For those who think that Canon has a problem with high ISO noise, just go over to DXO Mark and check the graphs (not their rubbish ratings, but the real measurments). Take the Nikon D600, Canon 6D and Sony a900; look at the graph, and tell me which camera you would take for low light photography. I think we have a clear winner with the 6D. The 6D looses out in dynamic range at low ISO, and has worse color sensitivity, but DR and SNR are better than the rest at high ISO.

Don't get yourself fooled by the DXOMark Rating, it shows not the whole picture. It is also difficult to judge what a 5 or 10 point difference means exactly.

I'm looking forward to the DXOMark measurments of the 70D; I think we will see some surprises, not that the 70D will storm to the top, but just a different behavior then the recent Canon sensor (which was rather predictable).

if you take a look at all parameters you se that the canons sensors are behind in color depth,DR, and also in high iso.
this is a Toshiba sensor

How does any of this affect a real photograph? Let's just take one of them, colour depth. The difference between 22 and 24 bits is going to be indistinguishable in any type of real world use. Can you show us a photo where this difference is evident?

 

[David Hull]

I for my part think that Canon made the switch to a new process with the 70D sensor. Canon admited, that 18MP was the limit with the process they had then. So if they now make an APS-C sensor with 40million photodiodes (there are two diodes that can be read seperatly in every of the 20.2MP), then I think they definitly made the move to a new process.

For those who think that Canon has a problem with high ISO noise, just go over to DXO Mark and check the graphs (not their rubbish ratings, but the real measurments). Take the Nikon D600, Canon 6D and Sony a900; look at the graph, and tell me which camera you would take for low light photography. I think we have a clear winner with the 6D. The 6D looses out in dynamic range at low ISO, and has worse color sensitivity, but DR and SNR are better than the rest at high ISO.

Don't get yourself fooled by the DXOMark Rating, it shows not the whole picture. It is also difficult to judge what a 5 or 10 point difference means exactly.

I'm looking forward to the DXOMark measurments of the 70D; I think we will see some surprises, not that the 70D will storm to the top, but just a different behavior then the recent Canon sensor (which was rather predictable).

if you take a look at all parameters you se that the canons sensors are behind in color depth,DR, and also in high iso.
this is a Toshiba sensor

How does any of this affect a real photograph? Let's just take one of them, colour depth. The difference between 22 and 24 bits is going to be indistinguishable in any type of real world use. Can you show us a photo where this difference is evident?

yes I can, and there are difference between my Canon and Nikon cameras, especially in colors as green.You can also se it in lower levels where Canon can not reproduce the same colors as Nikon.
I have been working with colors, colors reproduction, color profiles since my first scanner in the beginning of the -90
and can invite a color specialist who can easily can explain how the different cameras CFA works and how steep these filters are and what it means, Canons color filter are thinner by the years to gain light which you can se by comparing 1dsmk3 and the 5dmk3 or mk2 .

I don't really need an explanation of how the CFA works; I have been reading that stuff for years. Nor do I need someone to tell me that Nikon and Canon render colors differently, that fact is well known and has also been discussed ad nauseum here and elsewhere.

The reason I responded to your comment is that I continually see people trotting out these DxO numbers claiming that they represent some sort of scathing indictment of Canon technology yet when I look at images produced by each technology, the results are pretty much indistinguishable in terms of IQ. About the only one of these that can be demonstrated in an actual photograph is the oft discussed read noise and that requires some pretty serious “image abuse” to do.

To me, the acid test will be to perform a double blind experiment where several photographers walk through an exhibition of displayed prints and correctly identify which camera shot them. I have never seen this done (or even attempted) but if all of this DxO stuff really carried any real world significance, identifying the superior performing technology in such a test would be a “slam dunk” and… I think we both know that is not going to be the case.

The proof lies in the images themselves, if dramatic differences not evident in the images, then there has to be some question as to the real world relevance of the measurements that are supposed to be indicative of image quality.

 

[northbyten]

I just want to see what Canon's next FF sensor can do.

 

[jrista]

This is a iPhone picture, and people have let me know here at CR that they think it is good picture regarding sun set

Hmm...first things first....any "good" landscape photo, sunset or otherwise...would have a flat horizon! (This one is wildly tilted...)

As for good...try this (taken with a Canon by Marc Adamus):

http://500px.com/photo/2905633

 

[David Hull]

I for my part think that Canon made the switch to a new process with the 70D sensor. Canon admited, that 18MP was the limit with the process they had then. So if they now make an APS-C sensor with 40million photodiodes (there are two diodes that can be read seperatly in every of the 20.2MP), then I think they definitly made the move to a new process.

For those who think that Canon has a problem with high ISO noise, just go over to DXO Mark and check the graphs (not their rubbish ratings, but the real measurments). Take the Nikon D600, Canon 6D and Sony a900; look at the graph, and tell me which camera you would take for low light photography. I think we have a clear winner with the 6D. The 6D looses out in dynamic range at low ISO, and has worse color sensitivity, but DR and SNR are better than the rest at high ISO.

Don't get yourself fooled by the DXOMark Rating, it shows not the whole picture. It is also difficult to judge what a 5 or 10 point difference means exactly.

I'm looking forward to the DXOMark measurments of the 70D; I think we will see some surprises, not that the 70D will storm to the top, but just a different behavior then the recent Canon sensor (which was rather predictable).

if you take a look at all parameters you se that the canons sensors are behind in color depth,DR, and also in high iso.
this is a Toshiba sensor

How does any of this affect a real photograph? Let's just take one of them, colour depth. The difference between 22 and 24 bits is going to be indistinguishable in any type of real world use. Can you show us a photo where this difference is evident?

yes I can, and there are difference between my Canon and Nikon cameras, especially in colors as green.You can also se it in lower levels where Canon can not reproduce the same colors as Nikon.
I have been working with colors, colors reproduction, color profiles since my first scanner in the beginning of the -90
and can invite a color specialist who can easily can explain how the different cameras CFA works and how steep these filters are and what it means, Canons color filter are thinner by the years to gain light which you can se by comparing 1dsmk3 and the 5dmk3 or mk2 .

I don't really need an explanation of how the CFA works; I have been reading that stuff for years. Nor do I need someone to tell me that Nikon and Canon render colors differently, that fact is well known and has also been discussed ad nauseum here and elsewhere.

The reason I responded to your comment is that I continually see people trotting out these DxO numbers claiming that they represent some sort of scathing indictment of Canon technology yet when I look at images produced by each technology, the results are pretty much indistinguishable in terms of IQ. About the only one of these that can be demonstrated in an actual photograph is the oft discussed read noise and that requires some pretty serious “image abuse” to do.

To me, the acid test will be to perform a double blind experiment where several photographers walk through an exhibition of displayed prints and correctly identify which camera shot them. I have never seen this done (or even attempted) but if all of this DxO stuff really carried any real world significance, identifying the superior performing technology in such a test would be a “slam dunk” and… I think we both know that is not going to be the case.

The proof lies in the images themselves, if dramatic differences not evident in the images, then there has to be some question as to the real world relevance of the measurements that are supposed to be indicative of image quality.

I have done double l blind test regarding resolution,letting people look at 1m wide prints from 5dmk1 and 5dmk2 and asked them to give their votes about which print from which camera regarding resolution , a skilled person can se the difference immediately, he or she knows what to look after. Only 3 person pointed out the right pictures, but this 3 persons was highly skilled people
The same with color reproduction,color nuances and DR and especially if the image has been changed in programs as Photoshop, lifted shadows etc etc . Many people are happy with results from iPhone or compact cameras , I also when I have handle the camera right and in right circumstances.
This is a iPhone picture, and people have let me know here at CR that they think it is good picture regarding sun set

Looks nice to me. It may not hold up when pushed to a meter wide though.

 

[David Hull]

This is a iPhone picture, and people have let me know here at CR that they think it is good picture regarding sun set

Hmm...first things first....any "good" landscape photo, sunset or otherwise...would have a flat horizon! (This one is wildly tilted...)

As for good...try this (taken with a Canon by Marc Adamus):

http://500px.com/photo/2905633

It's that "Dutch angle thing

 

[jrista]

Regarding DXO numbers, I have only one comments, the resolution test= cameras+ lenses and the scores, I do not understand them.
Regarding DXO tests of DR I have no problem with them, I have by my self done a lot of DR tests and they are simple to explain

It is not that the DR tests are hard to understand. It is that they hold far too much weight on the final camera score that they skew the results, making some cameras seem "bad" or even "terrible" when they are nothing of the sort.

 

[Skulker]

........ Sonys sensor department is healthy, they earn money from the mobile sensors up to 24x36mm
that other departments such as television, consumer electronics, etc. are not going so well is another question ......

Do you have any justification for this statement? or is it just plucked from the depths of your mind?

 

[meli]

........ Sonys sensor department is healthy, they earn money from the mobile sensors up to 24x36mm
that other departments such as television, consumer electronics, etc. are not going so well is another question ......

Do you have any justification for this statement? or is it just plucked from the depths of your mind?

They sell to everyone -including canon-, they're bound to be profitable.

"Sony, the third largest semiconductor firm in Japan, divides its semiconductor business into in-house use and outside sales. The fourth-quarter revenues from outside sales reached JPY90 billion, representing a sequential increase of 19%. The firm has reported revenue growth for two consecutive quarters and continues to stay in profit."
from :http://www.digitimes.com/news/a20130325PD206.html

 

[jrista]

........ Sonys sensor department is healthy, they earn money from the mobile sensors up to 24x36mm
that other departments such as television, consumer electronics, etc. are not going so well is another question ......

Do you have any justification for this statement? or is it just plucked from the depths of your mind?

They sell to everyone -including canon-, they're bound to be profitable.

"Sony, the third largest semiconductor firm in Japan, divides its semiconductor business into in-house use and outside sales. The fourth-quarter revenues from outside sales reached JPY90 billion, representing a sequential increase of 19%. The firm has reported revenue growth for two consecutive quarters and continues to stay in profit."
from :http://www.digitimes.com/news/a20130325PD206.html

Profit is a relative and often misleading term. When you are tens of billions of dollars in debt, a significant portion of which was required to build the fabs that make Sony sensors, as well as put into the R&D and patent purchases...well, profit simply means deferred payments on debt. The electronics divisions of Sony, of which IC fabrication is a part, have been in the red for years. Even if some of the subdivisions are making some money, overall Sony is hemorrhaging from their electronics division (and they STILL have immense piles of debt at high interest rates.)

 

[Don Haines]

........ Sonys sensor department is healthy, they earn money from the mobile sensors up to 24x36mm
that other departments such as television, consumer electronics, etc. are not going so well is another question ......

Do you have any justification for this statement? or is it just plucked from the depths of your mind?

They sell to everyone -including canon-, they're bound to be profitable.

"Sony, the third largest semiconductor firm in Japan, divides its semiconductor business into in-house use and outside sales. The fourth-quarter revenues from outside sales reached JPY90 billion, representing a sequential increase of 19%. The firm has reported revenue growth for two consecutive quarters and continues to stay in profit."
from :http://www.digitimes.com/news/a20130325PD206.html

Profit is a relative and often misleading term. When you are tens of billions of dollars in debt, a significant portion of which was required to build the fabs that make Sony sensors, as well as put into the R&D and patent purchases...well, profit simply means deferred payments on debt. The electronics divisions of Sony, of which IC fabrication is a part, have been in the red for years. Even if some of the subdivisions are making some money, overall Sony is hemorrhaging from their electronics division (and they STILL have immense piles of debt at high interest rates.)

And don't underestimate the impact of profit on a product line. Take a look at Olympus 4/3 cameras ( 4/3, not micro 4/3). The line was not profitable so they let it drop. If Sony cameras are not profitable, expect to see them dropped.

 

[Skulker]

........ Sonys sensor department is healthy, they earn money from the mobile sensors up to 24x36mm
that other departments such as television, consumer electronics, etc. are not going so well is another question ......

Do you have any justification for this statement? or is it just plucked from the depths of your mind?

They sell to everyone -including canon-, they're bound to be profitable.

"Sony, the third largest semiconductor firm in Japan, divides its semiconductor business into in-house use and outside sales. The fourth-quarter revenues from outside sales reached JPY90 billion, representing a sequential increase of 19%. The firm has reported revenue growth for two consecutive quarters and continues to stay in profit."
from :http://www.digitimes.com/news/a20130325PD206.html

The whole point of the nytimes article was that the electronic division did not make money. In fact over the last decade it had lost $9.8 billion (from memory, the actual number is not too important ) the company overall made money. But about half of what it made from films and insurance was lost again on electronics.

It is in such a bad way one significant shareholder is saying get rid of it. And the electronics business might be worthless according to the nytimes article.

The nytimes article talks about the electronics division and the digitaltime talks about the semi conductor business. Are they the same, I think not as the first includes such as the playstation.

Ankorwatt specifically claimed the sensor department was healthy. My question was did he have a justification for that claim. I asked for two reasons. First it would be interesting if true as Sony make some good sensors, and maybe they could sell it off if it really was healthy. Although that is unlikely I accept. Secondly Ankorwatt make some claims I find strange to say the least, so I was wondering whether it was worth taking any notice of his claim this time. Normally I find his rants very ignorable.

 

[J.R.]

Ankorwatt specifically claimed the sensor department was healthy. My question was did he have a justification for that claim. I asked for two reasons. First it would be interesting if true as Sony make some good sensors, and maybe they could sell it off if it really was healthy. Although that is unlikely I accept. Secondly Ankorwatt make some claims I find strange to say the least, so I was wondering whether it was worth taking any notice of his claim this time. Normally I find his rants very ignorable.

Ankorwatt may not be wrong. This is an extract of a recent press release of Sony's Corporate Strategy Meeting of 2013 -

Imaging Businesses

Placing image sensors, a particularly strong category for Sony, at its core, Sony is concentrating the focus of its imaging businesses on creating value-added products, while aggressively exploring new applications for its imaging technologies in both the consumer and professional markets. In terms of image sensors, the Company will continue to commercialize new sensor technologies capable of differentiating finished products, for use in a range of consumer and professional applications. The Company also plans to engage in aggressive capital investment in order to meet the robust demand for these components. At the same time, Sony is also developing technologies that further expand the range of sensor applications, including sensors capable of sensing beyond the visible light spectrum, and sensors capable of detecting and categorizing different types of information. For the professional market, Sony will continue to reinforce its professional camera lineup centering on 4K-compatible cameras, as well as cameras for cinematography. The Company will also target further business growth by extending the scope of its digital imaging technologies to new business areas such as security, sports and medical, and will reallocate resources accordingly. In the consumer market, where business conditions continue to shift rapidly, Sony aims to expand sales of value-added compact digital still cameras by introducing models that leverage Sony’s image sensor technologies to further enhance image quality, and also incorporate feature enhancements such as reduced size and weight, and higher-powered zoom. Sony will also seek to firmly maintain its number one global market share in the growing mirrorless lens camera category. Through these measures, Sony will target sales of 1.3 trillion yen and an operating profit margin of more than 10% across the image sensor, professional, and consumer categories by FY14.

The full text is available here -

http://www.sony.net/SonyInfo/News/Press/201305/13-065E/

 

[Skulker]

Ankorwatt specifically claimed the sensor department was healthy. My question was did he have a justification for that claim. I asked for two reasons. First it would be interesting if true as Sony make some good sensors, and maybe they could sell it off if it really was healthy. Although that is unlikely I accept. Secondly Ankorwatt make some claims I find strange to say the least, so I was wondering whether it was worth taking any notice of his claim this time. Normally I find his rants very ignorable.

Ankorwatt may not be wrong. This is an extract of a recent press release of Sony's Corporate Strategy Meeting of 2013 -

Yes he may be right, that's why I asked him. : : it would be nice to hear from him, if he has any support for his claim I'm sure he will reply.

But I hope it's based on something a lot more substantial than a press release from a strategy meeting. :