Canon and the Apple in the old days (Dxomark)

[Babarous]

A few years ago, Apple used to make their own propriety CPUs for Macs. Does anyone remember that? Then they had the "brilliant" idea: "why are we competing against Intel" in CPU technology? That's a losing battle. So they dumped their own CPUs and started using the intel chip in Macs.

Canon is in a very similar situation right now. Nikon buys the best sensor from whereever they can find (Sony, Aptina, Renesas) and uses them in their own cameras, concentrating instead on the other aspects of a making a good camera (something that is Nikon's speciality).

Canon, on other hand, (just like Apple was against Intel), is stuck with "competing" in semiconductor technology against electronic companies like Sony (D800 sensor) Renesas (D3s sensor) and Aptina (Nikon 1 sensor). Good luck with that.

This is 100% guranteed a losing battle.

Canon should seriopusly consider this: when upgrading 7D, put the best sensor in it, even if you have to buy it from a third party.

Or stay at the bottom of dxomark for the next several decades ... What's the probability Canon will start beating Sony (who makes 30 million sensors a month) in semiconductor technology? Nill. I actually predict the gap would get even bigger in future.

 

[Musouka]

Speaking of DxO, they seem to think that camera phones have better pixel sensitivity:

"If you scale down the quality to the sensor size, today the [phone] cameras and sensors are better than the SLR sensors," Guichard said. "In the end, the image quality is not as good because it's smaller. But if Canon were able to put the technical quality of a 2012 phone camera on full-frame sensor, they would win about 1 stop more [in image quality]. It's a big difference."

Source: How DxO Labs tests hot cameras like Canon's latest SLR

That said, Apple weren't exactly making their own CPU's before their transition. Yes, they were involved in the PowerPC alliance with IBM and Motorola. They simply switched to Intel after they got dissatisfied with the lack of progress in the architecture (and to gain better compatibility with PC-related software architectures and the like).

Whatever Canon decides to do will be dictated by the market, if they kept selling strongly then why bother? If they started bleeding market share then they might consider changing their strategy.

Apple have been doing incremental upgrades themselves with the iPhone 4S and the new iPad but still sell strongly. There were even rumors that there were going to completely switch to ARM and start making their own processors (sparked by their purchase of P.A. Semi several year ago).

 

[Babarous]

Dxomark results are confirmed by actual samples

http://www.fredmiranda.com/5DIII-D800/index.html

The DR difference is really there.

 

[Musouka]

I wasn't disputing the findings. I was just quoting DxO Labs' chief scientist.

 

[7enderbender]

Not sure about this. More competition is always better and I have no complaints about what Canon has done with their sensors so far. Plus don't they bring a lot of expertise to the table anyway that spills over from their professional (non camera) business for science and industrial applications?

I don't know what triggered the Apple decision. It certainly made Macs look (slightly) more interesting to us PC/Win users since some of the compatibility issues could be addressed. That doesn't apply here - at least not until there is some kind of open camera platform. That would be ideal anyway if sensors were standardized and exchangeable. I wouldn't mind having something like the upcoming b/w Leica sensor available for a 5D-type camera.

But there is also a risk. Just look what has happened to monitors since "HD" TV screens have become the norm. Now there is an opportunity for Apple to change that right now...

 

[NormanBates]

I don't think this is as bad as the apple situation was: Canon has a big enough market that it could go on developing its sensors and be competitive; but with this generation they screwed up: they made a mistake (my guess is that the ADC are the weak link), and they'll either adjust the price, or just be relegated to a corner and suffer

 

[pdirestajr]

A few years ago, Apple used to make their own propriety CPUs for Macs. Does anyone remember that? Then they had the "brilliant" idea: "why are we competing against Intel" in CPU technology? That's a losing battle. So they dumped their own CPUs and started using the intel chip in Macs.

Canon is in a very similar situation right now. Nikon buys the best sensor from whereever they can find (Sony, Aptina, Renesas) and uses them in their own cameras, concentrating instead on the other aspects of a making a good camera (something that is Nikon's speciality).

Canon, on other hand, (just like Apple was against Intel), is stuck with "competing" in semiconductor technology against electronic companies like Sony (D800 sensor) Renesas (D3s sensor) and Aptina (Nikon 1 sensor). Good luck with that.

This is 100% guranteed a losing battle.

Canon should seriopusly consider this: when upgrading 7D, put the best sensor in it, even if you have to buy it from a third party.

Or stay at the bottom of dxomark for the next several decades ... What's the probability Canon will start beating Sony (who makes 30 million sensors a month) in semiconductor technology? Nill. I actually predict the gap would get even bigger in future.

But wait. Aren't the 5D mkiii's OTHER FEATURES (besides the sensor) it's major advantage over the D800?

So if Nikon just "slapped a sensor" into their camera, what exactly were they doing the whole time? I don't think the the D800 is groundbreaking in any way besides the sensor. Based on the OP's theory, shouldn't they have had more time to perfect an ergonomically brilliant, award winning piece of design? That camera looks squished & melty and exploding with buttons, ports & dials to me.

Canon already had a pretty great sensor in the 5Dii, why is it so wrong that they chose to focus on ALL the other aspects of photography & not the "film". IF photographers can't make a decent exposure from that sensor, well, there are bigger problems.

 

[smirkypants]

A few years ago, Apple used to make their own propriety CPUs for Macs.

Your understanding of history is a bit off... Apple never made their own CPUs until the boondoggles known as the iPhone and iPad. They only switched because IBM basically left them with no choice.

 

[PhilDrinkwater]

This is 100% guranteed a losing battle.

I disagree. There are definitely advantages, but there are disadvantages too:
* You lose some control over your direction
* Someone else benefits from your money
* Someone else can take your business from you
* Someone elses business can go in a different direction, leaving you high and dry

.. and so on.

I couldn't say which is the right route, but it's most certainly NOT a 100% guaranteed losing battle when you consider the long term of that business decision.

Don't forget that Canon aren't on their knees. I know a lot people seem to think that the sensor situation is going to kill Canon virtually overnight but I guarantee that it wont. Most people who own a camera never even go to DXO. They've never even heard of a sensor test. Many photographers don't even know what DR is.

As an example, dpreview.com 10-15 times the level of traffic that nikon.com gets (and 80% of those are to the main site, not the forum). Nikon.com gets 3 times the level of traffic that dxomark.com gets. Canonrumours.com and nikonrumours.com didn't even feature on this scale at all.

We are in a very niche area of the industry on this site and a number of other similar sites. People are much more likely to see what dpreview say. So far they've not said anything much the 5d3 (that I've seen) but they certainly don't torture test sensors (again, that I've seen). They present a fairly balanced view rather than just crushing on a sensor.

A camera is more than a sensor...

 

[Loswr]

A few years ago, Apple used to make their own propriety CPUs for Macs. Does anyone remember that?

No, I don't, except for the newer mobile devices as Smirkypants indicates. I remember them buying CPUs from Motorola. Seems like a problem with your cerebral RAM to me...

 

[kennykodak]

question:
for days i have about the disbelief that DXO rated Nikon's 800 so much higher than the 5D3.
how of you use DXO products?

 

[Babarous]

question:
for days i have about the disbelief that DXO rated Nikon's 800 so much higher than the 5D3.
how of you use DXO products?

See http://www.fredmiranda.com/5DIII-D800/index_controlled-tests.html

The shadow noise difference between Sony and Canon sensor is clearly present, confirming Dxomark result. Canon even with fullframe sensor can't compete in read noise against APSC sensor that Sony made 5 years ago (12 MP APSC in D300/D90).

I really think this will continue to hurt Canon constantly, if, aside from competing against Nikon in camera specs, they also have to compete against electronic companies in semi-conductor technologies.

 

[Musouka]

So, if they were losing the battle 5 years ago... why haven't they been hurt all this time?

 

[nitsujwalker]

A few years ago, Apple used to make their own propriety CPUs for Macs. Does anyone remember that? Then they had the "brilliant" idea: "why are we competing against Intel" in CPU technology? That's a losing battle. So they dumped their own CPUs and started using the intel chip in Macs.

Canon is in a very similar situation right now. Nikon buys the best sensor from whereever they can find (Sony, Aptina, Renesas) and uses them in their own cameras, concentrating instead on the other aspects of a making a good camera (something that is Nikon's speciality).

Canon, on other hand, (just like Apple was against Intel), is stuck with "competing" in semiconductor technology against electronic companies like Sony (D800 sensor) Renesas (D3s sensor) and Aptina (Nikon 1 sensor). Good luck with that.

This is 100% guranteed a losing battle.

Canon should seriopusly consider this: when upgrading 7D, put the best sensor in it, even if you have to buy it from a third party.

Or stay at the bottom of dxomark for the next several decades ... What's the probability Canon will start beating Sony (who makes 30 million sensors a month) in semiconductor technology? Nill. I actually predict the gap would get even bigger in future.

But wait. Aren't the 5D mkiii's OTHER FEATURES (besides the sensor) it's major advantage over the D800?

So if Nikon just "slapped a sensor" into their camera, what exactly were they doing the whole time? I don't think the the D800 is groundbreaking in any way besides the sensor. Based on the OP's theory, shouldn't they have had more time to perfect an ergonomically brilliant, award winning piece of design? That camera looks squished & melty and exploding with buttons, ports & dials to me.

Canon already had a pretty great sensor in the 5Dii, why is it so wrong that they chose to focus on ALL the other aspects of photography & not the "film". IF photographers can't make a decent exposure from that sensor, well, there are bigger problems.

"IF photographers can't make a decent exposure from that sensor, well, there are bigger problems."

I really agree with this! All this hype about the sensor being inferior (not placing a vote one way or another on this) makes me questions...How did all the pro wedding photogs, landscape photogs, and others use the 5d2 to make GORGEOUS photographs?? 5d3 won't magically make better photographs--nor will the d800--nor will some magical 16 stops of dynamic range.

 

[smirkypants]

"IF photographers can't make a decent exposure from that sensor, well, there are bigger problems."

I really agree with this! All this hype about the sensor being inferior (not placing a vote one way or another on this) makes me questions...How did all the pro wedding photogs, landscape photogs, and others use the 5d2 to make GORGEOUS photographs?? 5d3 won't magically make better photographs--nor will the d800--nor will some magical 16 stops of dynamic range.

I think what has most people's underwear up in a bunch, when you get right down to it, is that the 5D3 costs more than a machine that appears to beat it on most levels. Now if the 5D3 were $2799, you'd hear a lot less pissing and moaning.

 

[nitsujwalker]

"IF photographers can't make a decent exposure from that sensor, well, there are bigger problems."

I really agree with this! All this hype about the sensor being inferior (not placing a vote one way or another on this) makes me questions...How did all the pro wedding photogs, landscape photogs, and others use the 5d2 to make GORGEOUS photographs?? 5d3 won't magically make better photographs--nor will the d800--nor will some magical 16 stops of dynamic range.

I think what has most people's underwear up in a bunch, when you get right down to it, is that the 5D3 costs more than a machine that appears to beat it on most levels. Now if the 5D3 were $2799, you'd hear a lot less pissing and moaning.

Yeah, I can understand that.

 

[Musouka]

Also, correct if I'm wrong, but didn't Nikon design their own sensors for the D700 & D4 (among others)? That design was then manufactured by Renesas (or whoever). And didn't Canon use Sony's CCD sensors in their PowerShot lineup?

Another thing is to consider is that Canon might actually posses the technical know-how necessary for designing better sensors but they might be limited by patents owned by Sony/Nikon or others. They need to work around those limitations or end up licensing them (if the other companies were willing to license, that is).

In the end, the decision to develop, license or buy is purely market-driven. It might be that they are trying to see the impact this will have or buying time while they develop a better technology. Heck, they might have already decided that the impact is minimal. Whether they are right or wrong remains to be seen.

 

[psolberg]

A few years ago, Apple used to make their own propriety CPUs for Macs. Does anyone remember that? Then they had the "brilliant" idea: "why are we competing against Intel" in CPU technology? That's a losing battle. So they dumped their own CPUs and started using the intel chip in Macs.

Canon is in a very similar situation right now. Nikon buys the best sensor from whereever they can find (Sony, Aptina, Renesas) and uses them in their own cameras, concentrating instead on the other aspects of a making a good camera (something that is Nikon's speciality).

Canon, on other hand, (just like Apple was against Intel), is stuck with "competing" in semiconductor technology against electronic companies like Sony (D800 sensor) Renesas (D3s sensor) and Aptina (Nikon 1 sensor). Good luck with that.

This is 100% guranteed a losing battle.

Canon should seriopusly consider this: when upgrading 7D, put the best sensor in it, even if you have to buy it from a third party.

Or stay at the bottom of dxomark for the next several decades ... What's the probability Canon will start beating Sony (who makes 30 million sensors a month) in semiconductor technology? Nill. I actually predict the gap would get even bigger in future.

IMO canon is still hanging on to the idea of the golden days when they were the top dog and will insist holding on to their old ways. The sony/Nikon partnership has already robbed canon from the majority marketshare in many areas. They are still #1 overall in some markets but are no longer over 50% as they were 10 years ago and continue to lose share to the Nikon/Sony tidal wave. Hubris is getting the better of them. They are out-pricing their base and trying to make traditional photographers into Hollywood film makers if they like it or not.

Add to that the increased popularity of Mirrorless which even in its infancy has hugely eroded entry level DSLRs which have been a big profit center for canon and yet they have failed to evolve into this market.

Sony's manufacturing arm which is comparable if not larger than canon's provides Nikon with an effective way to nullify canon's larger size. Sony gets Nikon's IP and expertise in sensor design and more importantly PATENT SHARING. Those two companies are the perfect storm together and I wouldn't be surprised if Sony eventually eats Nikon up they way they did minolta to form the largest camera company in the world by marketshare. that would keep some lights on at canon.

I'm absolutely LOVING the nikon/sony team's work and look forward to future products from those two. It has given canon much needed competition. What a difference a decade makes.