2019 global camera market share numbers are out, Canon leads the way

[SteveC]

OK, how is the sensor cost different? The die for each is the same size, I am assuming that they are both made on the same fab line, in the same process node, etc. In which the cost won't be much different. Semiconductor cost is determined by the finished cost of the wafer divided by how many chips you get off of it (with adders for packaging, test etc.). So, if the finished cost of a wafer is $2500 and the yield is 25 chips then the cost per chip is $100 whether is has 25 MP or 50 MP. What you charge for it is a different story.

I still maintain that an EOS R6 and an EOS R5 are almost identical with a few differences (some of which I listed). I don't think there is a $1000 difference. That doesn't stop them from pricing it that way -- assuming that the customers will pay it and the competition doesn't come along offering something as good for a lower price. This appears to be what happened to the EOS R. I am guessing that there is room to move in the R6 as well.

Here are a couple different questions: Would they have priced it higher if the higher definition video modes actually worked well? Also, what is the use case for the video as it is? If you can only shoot for 20 minutes (and even that is qualified), after which you have to power the thing down and let it cool for 2 hours (?), in what situation is that actually usable? Can you still shoot stills if it has overheated like that?

Quite possibly the sensor IS more expensive to fabricate, using as it does smaller circuitry sizes. Or it may make no difference; I don't know if the same foundry makes them both with the same equipment. But you do have to amortize development costs, too: the R5 is the first camera with this sensor, the R6 is using a previously used sensor.

But that's not the only aspect of it; you have to be able to handle more data, more quickly in the R5 than in the R6 (and that, indeed, is part of the overheating issue), which requires more processing power, likely more memory for the buffer, and on and on and on. And this will more than likely apply to the autofocus system as well. It's not just a matter of dropping a different wafer onto the spot that says "put sensor here."

I had no difficulty ponying up the extra cash for the higher-res sensor, video modes or no video modes.

 

[melgross]

Well, then there is nothing to complain about then. In your view they gave you a better camera for the same price, and threw in the video for free so who cares how well it works? I will point out though that the only thing different about the R5 from the R6 in terms of actual components is the VF, the CFE slot, a couple connectors and a slightly better chassis. Yet the R6 is $1k less. IMO, one of the things they are counting on to justify the higher price is the video capabilities.

You’re missing a number of things that make a large bit of the price difference. Why?

the 45mp sensor costs considerably more than the 20mp sensor. The higher rez EVF costs considerably more as well. A difference in bodies. The R5 has an all magnesium body. The R6 uses a polycarbonate front. The R5 is better weather sealed than the R 6. The 5 is sealed to the standard of the 5Dmk IV, and the 6 to the standard of the D6.

those are a lot of differences, and they’re all significant. There are some smaller differences as well. How you can pretend they don’t exist, or matter, only you can know.

 

[melgross]

OK, how is the sensor cost different? The die for each is the same size, I am assuming that they are both made on the same fab line, in the same process node, etc. In which the cost won't be much different. Semiconductor cost is determined by the finished cost of the wafer divided by how many chips you get off of it (with adders for packaging, test etc.). So, if the finished cost of a wafer is $2500 and the yield is 25 chips then the cost per chip is $100 whether is has 25 MP or 50 MP. What you charge for it is a different story.

I still maintain that an EOS R6 and an EOS R5 are almost identical with a few differences (some of which I listed). I don't think there is a $1000 difference. That doesn't stop them from pricing it that way -- assuming that the customers will pay it and the competition doesn't come along offering something as good for a lower price. This appears to be what happened to the EOS R. I am guessing that there is room to move in the R6 as well.

Here are a couple different questions: Would they have priced it higher if the higher definition video modes actually worked well? Also, what is the use case for the video as it is? If you can only shoot for 20 minutes (and even that is qualified), after which you have to power the thing down and let it cool for 2 hours (?), in what situation is that actually usable? Can you still shoot stills if it has overheated like that?

high rez sensors do cost considerably more, like it or not. They always have, and they always will. The reject rate is higher with more dense chips. That’s just one area. But the lenses are much smaller and more expensive. The RGB filter is more expensive. The positioning of these parts needs to be considerably more precuse, and leads to more rejects.Pretty much everything regarding it is more expensive. They cost more than $100, several hundred.

 

[SteveC]

high rez sensors do cost considerably more, like it or not. They always have, and they always will. The reject rate is higher with more dense chips. That’s just one area. But the lenses are much smaller and more expensive. The RGB filter is more expensive. The positioning of these parts needs to be considerably more precuse, and leads to more rejects.Pretty much everything regarding it is more expensive. They cost more than $100, several hundred.

The man is convinced he's being forced to pay extra for video features he didn't want. We've been over and over and over that and it never does any good.

 

[David Hull]

Quite possibly the sensor IS more expensive to fabricate, using as it does smaller circuitry sizes. Or it may make no difference; I don't know if the same foundry makes them both with the same equipment. But you do have to amortize development costs, too: the R5 is the first camera with this sensor, the R6 is using a previously used sensor.

But that's not the only aspect of it; you have to be able to handle more data, more quickly in the R5 than in the R6 (and that, indeed, is part of the overheating issue), which requires more processing power, likely more memory for the buffer, and on and on and on. And this will more than likely apply to the autofocus system as well. It's not just a matter of dropping a different wafer onto the spot that says "put sensor here."

I had no difficulty ponying up the extra cash for the higher-res sensor, video modes or no video modes.

The man is convinced he's being forced to pay extra for video features he didn't want. We've been over and over and over that and it never does any good.

Not exactly, read my post more carefully. It is one thing to pay for something you don't want or don't need. We do that all the time. You can probably say that about the whole damn camera ;-). However, in this case, it seems that we are being asked to pay for something that isn't really very practical. What is the use case for the 8k mode when you have to cool the camera down an hour+ after shooting video for 20 minutes. Even as a stills shooter, I find the ability to grab stills shots out of the 8k stream interesting. I can think of shots where that might be usable. However, if I overheat the camera, can I drop back to stills or do I have to put it back in the pack and go have a couple beers.

it was more of a comment or observation rather than a complaint. I still think that the R5 is a really great camera for stills shooters, it checks all of the boxes that many Canon users have been wanting but, i don't think that the HD video features are going top prove that useful. I would be interested in hearing from others what they thing the use cases for a video camera with these sorts of limitations would be.

 

[David Hull]

high rez sensors do cost considerably more, like it or not. They always have, and they always will. The reject rate is higher with more dense chips. That’s just one area. But the lenses are much smaller and more expensive. The RGB filter is more expensive. The positioning of these parts needs to be considerably more precuse, and leads to more rejects.Pretty much everything regarding it is more expensive. They cost more than $100, several hundred.

Yea, you have a point. I wasn't thinking about the lens and filter part which might impact final yield. I come from some 30 years in semiconductor manufacturing for large SoC's where it pretty much works as I described -- chips that consumed the same wafer area tended to be the same cost at that point. Then there is packaging and testing.

 

[Michael Clark]

Well, then there is nothing to complain about then. In your view they gave you a better camera for the same price, and threw in the video for free so who cares how well it works? I will point out though that the only thing different about the R5 from the R6 in terms of actual components is the VF, the CFE slot, a couple connectors and a slightly better chassis. Yet the R6 is $1k less. IMO, one of the things they are counting on to justify the higher price is the video capabilities.

I'm not complaining, I was responding to you who was complaining that the enhanced video features added $1000+ to the price. That's simply not true, based on the price history of 5-series cameras.

You're also forgetting the difference between 45MP and 20MP...

 

[Michael Clark]

OK, how is the sensor cost different? The die for each is the same size, I am assuming that they are both made on the same fab line, in the same process node, etc. In which the cost won't be much different. Semiconductor cost is determined by the finished cost of the wafer divided by how many chips you get off of it (with adders for packaging, test etc.). So, if the finished cost of a wafer is $2500 and the yield is 25 chips then the cost per chip is $100 whether is has 25 MP or 50 MP. What you charge for it is a different story.

I still maintain that an EOS R6 and an EOS R5 are almost identical with a few differences (some of which I listed). I don't think there is a $1000 difference. That doesn't stop them from pricing it that way -- assuming that the customers will pay it and the competition doesn't come along offering something as good for a lower price. This appears to be what happened to the EOS R. I am guessing that there is room to move in the R6 as well.

Here are a couple different questions: Would they have priced it higher if the higher definition video modes actually worked well? Also, what is the use case for the video as it is? If you can only shoot for 20 minutes (and even that is qualified), after which you have to power the thing down and let it cool for 2 hours (?), in what situation is that actually usable? Can you still shoot stills if it has overheated like that?

Yield will not be the same for sensors with radically different pixel densities. There will be more errors on the smaller pixel densities, which will cause a higher percentage of the chips made from each wafer to be rejected.