M8 versus EOS 5D

[gwebster]

Hi Chris

 

Sure! Since I published this phrase on photo.net it's (presumably) public domain anyway, but thanks for asking. I don't imagine that photo.net would claim any copyrights to it either but I could be wrong.

[chris_tennyson1]

cool thanks!

[jtdnyc]

<Since I published this phrase on photo.net it's (presumably) public domain... >

 

From Photo.net's Terms of Use:

 

"You retain copyright to images and text uploaded to photo.net."

 

By posting, you've granted Photo.net a license to use you posted material on this site, but my understanding is you retain the copyright to the material you post. Whether a short phrase of that nature is copyrightable in the first place is another issue.

[bobatkins]

Much of the Leica mythology is faith based. You can't argue logic against faith.

<p>

You'll never convince a Hindu that Buddhism is the one true faith and you'll never convince a Leicaphile that an EOS 5D is a better camera. Whether it is or isn't is immaterial. It's a matter of faith. If you don't believe, you'll never understand. You can't argue with <a href="http://www.bbspot.com/News/2005/11/faith_based_firewalls.html">faith</a>

[fred_c1]

Not just any faith -- it's <i>religious faith</i>. ;-)

[httpwww.photo.netphoto1664880414]

you can look at it as faith, and that is a fun thing to do. But the other side of the coin is that Canon are getting toppled here. Perhaps not today, and not by M8, but that day is coming. Camera development in the things that matter has been slow for some 5 years. But this year is going to be a very competitive year in which Canon can only lose sales, although most likely not position.

 

I can see the more minor players pressing development hard, and something will have to give from all that. It is plain that Canon havnt been progressing their R&D in pace with others, and now the 'in house' CMOS might become somewhat more of a burden. Providing less flexibility for the future.

 

Riley

[Brad_]

<I>I can see the more minor players pressing development hard, and something will have to

give from all that. It is plain that Canon havnt been progressing their R&D in pace with

others, and now the 'in house' CMOS might become somewhat more of a burden.</I><P>

 

A burden, having your own silicon foundry and proprietary semiconductor processes and in-

house design capability? Please explain. <P>

 

Also, how is it plain that "Canon havnt been progressing their R&D in pace with others?"

Citations would be great.

[harvey_edelstein1]

Canon is a big electronics company that is vertically integrated, they have a fast new product cycle. I think there are some Canon haters here. Leica is a camera and lens company its design has basic features not all those options some digital users want. I don't know if 10mp is enough to fully show the great Leica glass' full potential. I feel that all electronics of digital has lessen Leicas advantages of mechanical excellence. No longer having need for smoothness wind and rewind and being battery dependent. And with high iso being better on the 5D and it having great dynamic range, FF and image stabilized and nearly silent AF lenses, low light work at slow shutter speed is not a weakness of the Canon 5d as it was on old Film slrs. Some of these old axioms of RF vs. Slr need to be updated in todays DRF vs. DSLR world.

[httpwww.photo.netphoto1664880414]

With the point and shoot market this year expected to top out, growth for 2007 lay within 1 of 2 strategies. Expand into dSLRs or stand your ground in P&S. No surprise that new entrants are appearing in the dSLR market then, each of Pentax Sony Matsushita Samsung. In March 06 Nikon predicted 30% increase in dSLR sales, and gradually released detail of what we now know is D40. Canon responded to this onslaught with a defeatured xti, thus explaining competition in low end dSLRs is determined by price. While Sony has already taken the Olympus #3 spot in the japanese market, it will press on into prosumer dSLRs worldwide. And lets not forget that Sony also manufacture chips for Nikon dSLR and Canon P&S. Who will they share their development capitol with ? Who will lose sales ?

 

It needs to be remembered that P&S are about 89% of the market, and around 70% of income. The true sales leaders have been Kodak and now Sony. These are the people with the R&D capitol, and there faith is pressing CCD development. CMOS is by no mens a dead ball, but it is losing development pace against CCD market dominance. Oh and Brad.... case in point Fuji F30 P&S with impressive iso performance on a tiny 1/1.7" sensor. How long is it since CMOS made a 2-3 stop jump in iso performance?

 

Given this obsession by some for FF sensors, notably lacking in edge sharpness at wide and especially ultra wide, due to more divergent/less telecentric light properties of very wide glass and strong AA filtering. Why not go medium format if size is everything?

 

Who is forgetting that Leica began 35mm in the 20's, and should it say something if given their options in 2006 they went with a crop sensor. They did this to maintain quality despite the availability of an off the rack FF from Kodak, despite the willingness of the Leica club to accept FF; and it appears to be making headway, based on reports that they are printing HQ 30x40 prints from the M8. Anyone who still believes that despite the IR problems, the M8 has inferior resolution seriously needs to question continuing that ambition.

 

Canon has to hold the fort across the entire dSLR spectrum, and maintain development of CMOS virtually alone, Leica has to do one thing, build and sell the unique M8 (Epson's RD1 will likely dissapear soon). Will CCD technology step up the ladder because of this....you bet. Will Canon be distracted further by the overheated dSLR 'pro' sumerism, thats a walk on.

 

Riley

[Brad_]

<I>These are the people with the R&D capitol, and there faith is pressing CCD

development. CMOS is by no mens a dead ball, but it is losing development pace against

CCD market dominance.</I><P>

 

That is just silly. CCD imagers are based on a bipolar semiconductor process technology

that has been around for 30-40 years. There are no new bipolar foundries coming online

to manufacture CCD sensors. Talk about a dead-end technology. That's not to take away

from CCD, but it is certainly not growing on a technology and features basis. Kodak has a

ton of patents going back 30 years in this area. That's eons ago technology-wise. Because

of their shaky financials and them bleeding huge amounts of money every year, they have

little ability ramp up CMOS development, let alone build a new CMOS foundry which costs

in the neighborhood of US$4B.<P>

 

<I>Will CCD technology step up the ladder because of this....you bet. </I><P>

 

How? Bipolar processes are limited. No new foundries, no ability to integrate other analog

and digital processing onto the die. And most of all, little $$$ spent for R&D on process

improvements (which are limited to begin with).<P>

 

<i>case in point Fuji F30 P&S with impressive iso performance on a tiny 1/1.7" sensor.

How long is it since CMOS made a 2-3 stop jump in iso performance?</I><P>

 

Again, silly. The noise "performance" comes due to post-processing in firmware, not due

to any breakthrough in CCD development. Characterizing it as a 2-3 stop jump is also

stretching it - that performance comes at a price in ultimate picture quality due to post-

processing.<P>

 

 

<I>Canon has to hold the fort across the entire dSLR spectrum, and maintain development

of CMOS virtually alone, Leica has to do one thing, build and sell the unique M8 </I><P>

 

Somehow, I think canon will survive... And "maintaining development of CMOS alone" is

what has made them a market leader. They're highly vertically integrated, doing

everything in-house, from semiconductor process R&D, to sensor design/manufacturing,

to design/manufacturing of bodies and lenses, to designing their own custom

postprocessing ASICs. That gives them a huge competitive edge. It's laughable to think

they're at a disadvantage being burdened with proprietary technology, design capability,

and silicon foundry.<P>

 

 

<I>Will CCD technology step up the ladder because of this....you bet.</I><P>

 

How will that happen? Do you really think leica's tiny yearly demand for CCD sensors is

going to influence CCD development? It's a drop in the bucket. They're just along for the

ride.

[httpwww.photo.netphoto1664880414]

"Again, silly. The noise "performance" comes due to post-processing in firmware, not due to any breakthrough in CCD development. Characterizing it as a 2-3 stop jump is also stretching it - that performance comes at a price in ultimate picture quality due to post- processing."

 

you mean like Canons Digic chip doesnt? ... so being thats the case, and not from me to be ignorant of the novel structure of this particular Fuji chip, why dont canons P&S outperform it ? The S3 is worse at iso800 than F30 at iso3200 (nows the time to remember 70% of the income is from P&S) but ofcourse canons S3 is a Sony chip isnt it, their new competitor, be a fly on the wall with those developments.

 

"Somehow, I think canon will survive... And "maintaining development of CMOS alone" is what has made them a market leader. They're highly vertically integrated, doing everything in-house, from semiconductor process R&D, to sensor design/manufacturing, to design/manufacturing of bodies and lenses, to designing their own custom postprocessing ASICs. That gives them a huge competitive edge. It's laughable to think they're at a disadvantage being burdened with proprietary technology, design capability, and silicon foundry."

 

the vulnerability is because they have to work inhouse, they have no choice, do they? add to that the pressures from new dSLR entrants down the scale which you also ignore, is the focus going to be on high end development...when all the action is in the more competitive lower end ? very doubtfull.

 

"How will that happen? Do you really think leica's tiny yearly demand for CCD sensors is going to influence CCD development? It's a drop in the bucket. They're just along for the ride."

 

the existence of the sensor, the further development for new CCD sensors, for other customers, work by other chip makers like Fuji too, all canon competitors, thats how. you seem not to appreciate that kodaks chip has any resolution at all, these are developments that Canon wont have access too, and couldnt use if it did. Sony will be sure to make it hurt at the P&S end to give themselves an edge and pressure canons P&S income

 

Heard about the deal between Sony and Kodak yet, laugh about that :)

 

Riley (cheese)

[Brad_]

<I>why dont canons P&S outperform it ?</I><P>

 

That's easy. Image quality. It's really easy to beat down noise if you're willing to take a hit

on IQ. Canon apparently chose not to. The market sweet spot for digicams is not

ISO800-1600 shooting, it's image quality<P>

 

<I>the vulnerability is because they have to work inhouse, they have no choice, do they?

</I><P>

 

Hardly a vulnerability. That's what IP and in-house capability is about - having proprietary

technology available to no one else enabling you to bring great performing products to

markets less expensively than your competitors. That's what capital investment is all about

and what fuels technology in general. Not buying commodity components off the shelf -

and letting others make the money, similar to your competitors.<P>

 

<I>the existence of the sensor, the further development for new CCD sensors, for other

customers, work by other chip makers like Fuji too, all canon competitors, thats how

</I><P>

 

The mere existence of a sensor based on 30 year old technology that hasn't change

remarkably does not explain how. Feel free to talk about the breakthroughs (in detail,

please) about these new CCD sensors.<P>

 

<I>add to that the pressures from new dSLR entrants down the scale which you also

ignore,</i><P>

 

Who's ignoring. Of course there's competition. That's what being in business is about.<P>

 

<I>you seem not to appreciate that kodaks chip has any resolution at all,</I><P>

 

Resolution is just one metric of a sensor. And it's not limiting performance. Resolution is

relatively cheap.<P>

 

<I>Heard about the deal between Sony and Kodak yet, laugh about that :)</I><P>

 

Yes, it concludes their on-going patent litigation through cross-licensing. <P>

 

Heard about Kodak's huge losses, declining revenue, and monstrous layoffs?

[httpwww.photo.netphoto1664880414]

well you just dont seem get it do you, allow me to illucidate more acutely

 

"Hardly a vulnerability. That's what IP and in-house capability is about - having proprietary technology available to no one else enabling you to bring great performing products to markets less expensively than your competitors. That's what capital investment is all about and what fuels technology in general. Not buying commodity components off the shelf - and letting others make the money, similar to your competitors."

 

70% of their income comes from a sensor owned by somebody else. That someone is Sony, the competitor that has now signed cross licensing deals with Kodak, the other competitor. Canon have concentrated on the CMOS, and left their 70% of development income at risk through not having inhouse CCD development. Already 2 stops behind Fuji, thats vertically integrated....i dont think so

 

"The mere existence of a sensor based on 30 year old technology that hasn't change remarkably does not explain how. Feel free to talk about the breakthroughs (in detail, please) about these new CCD sensors."

 

yes you know, the ones Canon buys from Sony for 70% of income, you got it now? Off the rack CCD's

 

"Resolution is just one metric of a sensor. And it's not limiting performance. Resolution is relatively cheap."

 

it is a rather important parameter though isnt it, actually primary isnt it, moreover, noise is a background element next to resolution isnt it. And cheap, well I dont agree, within a systematic approach its very difficult to pick up on your losses. Add to that losses within a system cannot be recovered.

 

"Yes, it concludes their on-going patent litigation through cross-licensing.

 

Heard about Kodak's huge losses, declining revenue, and monstrous layoffs?"

 

yep, since they killed dSLR manufacture and shifted wholy to sensors i guess they accessed some flexibilty, and the cross licensing, kodak got the better of sony there....guess whos next :) no-one is talking money but its a safe bet that there is either a keep off the grass agreement or a fair amount of monetary compensation, but that we may never hear of given the nature of these two organisations.

 

second bet, Canon axe some P&S models...soon

 

Riley

[Brad_]

Riley, seems you nothing about the semiconductor industry. If you do, speak up with your

credentials.

 

The reason canon may use a commodity sensor from sony is because it makes sense to

financially in the commodity digicam market. Has nothing to do with cmos vs bipolar

(CCD).

 

With respect to mere resolution, it is cheap. A sensor die is relatively bare. It's super easy

to put down a few million extra gates to get higher resolution. Doing that without a hit on

noise performance is the difficult part

 

With respect to the Fuji 2-3 stop advantage, again, it's ONLY an advantage if the bulk of

your market is willing to give up IMAGE QUALITY for higher ISO - like the f30 does. That

is certainly not canon's digicam market. It may be OK for you and others who are not too

fussy about image quality. I really think canon know how to make the proper trades for the

markets they target. If you think otherwise, perhaps you should join their board and set

them straight on their mistakes.

 

For other products, like their dSLRs, where performance is king (in addition to IQ), using

their own proprietary technology lets them differentiate based on performance, and on

cost due to not having to pay a premium to someone else. Their foundry runs 24 hours/

day. There's no doubt it's an asset rather than a liability - both from a cost and market

leadership position...

 

Once more what are your semiconductor credentials?

[httpwww.photo.netphoto1664880414]

"Riley, seems you nothing about the semiconductor industry. If you do, speak up with your credentials."

 

thanks for your kind offer but im not seeking alternative employment right now

 

"The reason canon may use a commodity sensor from sony is because it makes sense to financially in the commodity digicam market. Has nothing to do with cmos vs bipolar (CCD)."

 

actually, an aside from the discussion about the business aspects i alluded too, CMOS have proven to be less successful in smaller applications right? so canon are pretty well locked into a situation where they must use someone elses CCD in small P&S applications, ~ and with that the hereto issues i identified earlier

 

'With respect to mere resolution, it is cheap. A sensor die is relatively bare. It's super easy to put down a few million extra gates to get higher resolution. Doing that without a hit on noise performance is the difficult part'

 

the noise is everything, resolution is cheap is going to wash, all but 11% of photographers access iso beyond iso1600, still a lesser 4% beyond to 3200. If a sensor can acutely perform at iso1600 it has 89% market useability. what i would like to see are iso beneath iso100, iso50 and iso25

 

"With respect to the Fuji 2-3 stop advantage, again, it's ONLY an advantage if the bulk of your market is willing to give up IMAGE QUALITY for higher ISO - like the f30 does. That is certainly not canon's digicam market. It may be OK for you and others who are not too fussy about image quality. I really think canon know how to make the proper trades for the markets they target. If you think otherwise, perhaps you should join their board and set them straight on their mistakes."

 

it sounds like you have not looked at the Fuji sensor array and seen what the architecture of this chip looks like, i suggest you do so and then come back and tell me how superior the canon aka Sony CCD is in these sizes.

 

"For other products, like their dSLRs, where performance is king (in addition to IQ), using their own proprietary technology lets them differentiate based on performance, and on cost due to not having to pay a premium to someone else. Their foundry runs 24 hours/ day. There's no doubt it's an asset rather than a liability - both from a cost and market leadership position..."

 

i quite recognise that the dSLR market has a position of 'auto everything", but i have never thought this is a justafiable pursuit. I wouldnt be alone in that position. Certainly photographers need to be able to access exposure support, but there are to my mind many features that artisans of photography prefer to do without. that said, it depends some good way on your mission.

 

Even among rangefinders, AF is quite uncommon and rightly so, what this particular segment of the market seeks is IQ above all, and that is expressed both in the images rendered, and the sizes they can be pushed out too. Further to this, it is this particular luxury that will imprint on the minds of photographers, for IQ in its most stunning beauty cannot be superceded by auto BS, and represents the true art in photography. Everything else is in the photographers hands, which for a master craftsman is just as it should be.

 

To restate, Canon do not own the sensor technology to the CCD's that fuel 70% of the market, moreover, they do not have any control of their future in this technology. That is a weakness and refutes your 'verticaly integrated' claim

 

and btw, running a foundry 24/7 is quite the only way to run them, even if for quite some periods they are actually idle. such is the cost of shutdown and restarting them.

[Brad_]

<I> thanks for your kind offer but im not seeking alternative employment right now

</I><P>

 

Ok, we'll take that as having no or little experience in semi design and

manufacturing.<P>

 

 

<I>actually, an aside from the discussion about the business aspects i alluded too, CMOS

have proven to be less successful in smaller applications right? so canon are pretty well

locked into a situation where they must use someone elses CCD in small P&S applications,

~ and with that the hereto issues i identified earlier</I><P>

 

Where has CMOS **proven** to be less successful? That's where your ignorance of

semiconductor process technology shines through. Canon is hardly locked into a situation.

Again, one more time, there is an abundance of cheap commodity CCD on the market. For

inexpensive commodity products where there is little differentiation and performance is

not crucial, using a commodity sensor makes the most sense. Do you really think canon is

stupid in how they fuel their cameras? And you know more?<P>

 

You're talking in circles about resolution.<P>

 

<I>it sounds like you have not looked at the Fuji sensor array and seen what the

architecture of this chip looks like, i suggest you do so and then come back and tell me

how superior the canon aka Sony CCD is in these sizes.</I><P>

 

OK, you;'re the expert on the Fuji sensor. Fire away. Again, the ISO performance is not in

the sensor, but in the noise processing that occurs afterwards. If you knew anything

about bipolar and CCD designs, that would be apparent. But one just needs to look at the

Fuji IQ to see what's going on. The same can be done in post with a variety of noise

reduction, and with the same compromise in image quality.<P>

 

<I>i quite recognise that the dSLR market has a position of 'auto everything", but i have

never thought this is a justafiable pursuit. I wouldnt be alone in that position..</I><P>

 

A great non-sequitur - something not in the argument.<P>

 

<I>for IQ in its most stunning beauty cannot be superceded by auto BS, and represents

the true art in photography. Everything else is in the photographers hands, which for a

master craftsman is just as it should be.</I><P>

 

Question: Are you a photographer? Providing a link to your images would help me

comment on the above.<P>

 

<I>To restate, Canon do not own the sensor technology to the CCD's that fuel 70% of the

market, moreover, they do not have any control of their future in this technology.</I><P>

 

OK, you're right on that. Canon is in a perilous position. Will probably go out of business

at any moment. I'm sure sony and kodak, two great companies making tons of money

right now are plotting that as I type. Should be easy, right? Just turn of the flow of

commodity sensors and canon would not know what to do, and will be lights out in a

month.

<P>

<I>That is a weakness and refutes your 'verticaly integrated' claim</I><P>

 

OK, keep believing canon is not vertically integrated.<P>

 

<I>and btw, running a foundry 24/7 is quite the only way to run them, </I><P>

 

You're stating the obvious. And the fact that canon is running 24/7 only aligns to the fact

they've made the correct decision on owning their technology and manufacturing.<P>

 

But what's really telling, is your position of superior knowledge on how canon should be

running their operation and their impending doom.

[httpwww.photo.netphoto1664880414]

"Ok, we'll take that as having no or little experience in semi design and manufacturing."

 

Quite wrong, I dont owe someone like you an explanation of who I am, Im not the least bit interested in who you think you are for instance.

 

"Where has CMOS **proven** to be less successful? That's where your ignorance of semiconductor process technology shines through. Canon is hardly locked into a situation. Again, one more time, there is an abundance of cheap commodity CCD on the market. For inexpensive commodity products where there is little differentiation and performance is not crucial, using a commodity sensor makes the most sense. Do you really think canon is stupid in how they fuel their cameras? And you know more?"

 

CMOS lower fill factor goes against it in small sensor applications, that would be why your verticaly integrated canon structure decided to buy sensors from what turns out to be a competitor. Its not for nothing that Canon use CCD's for smaller installations, or are Canon wrong in your eyes too? I dont think they are stupid, I do think they made a strategic mistake that isnt going to be easy to recoil from.

 

"OK, you;'re the expert on the Fuji sensor. Fire away. Again, the ISO performance is not in the sensor, but in the noise processing that occurs afterwards. If you knew anything about bipolar and CCD designs, that would be apparent. But one just needs to look at the Fuji IQ to see what's going on. The same can be done in post with a variety of noise reduction, and with the same compromise in image quality."

 

It completely escapes you that this sensors architecture is quite different, which points to how uninformed your knowledgebase is. Without exception, all manufacturers process jpegs internally with NR at the detrement of detail. Just take a moment to compare the S3is with the F30 at iso800 and tell me you think the S3 is better. Or maybe you will discover that a more natively noise free design needs less NR and hence gets better results for the same iso.

 

"Question: Are you a photographer? Providing a link to your images would help me comment on the above."

 

Answer: you can find images from M8 in dozens of places, I suggest you use www.Google.com and actually research something for once before you critisize it.

 

"OK, you're right on that. Canon is in a perilous position. Will probably go out of business at any moment. I'm sure sony and kodak, two great companies making tons of money right now are plotting that as I type. Should be easy, right? Just turn of the flow of commodity sensors and canon would not know what to do, and will be lights out in a month."

 

thats you talking not me, what Im saying is that their R&D at the top end is going to suffer fro the reasoning that I previously offered, all but a nine-iron pratt could figure that out

 

"But what's really telling, is your position of superior knowledge on how canon should be running their operation and their impending doom."

 

more of the same nonsense, were done here

 

Riley (say cheese)

[harry_zet]

http://www.outbackphoto.com/reviews...masHoepker.html

 

that says everything