Pages: 1 ... 7 8 [9]   Go Down

Author Topic: 50MP Canon 5DS/R spec leaked  (Read 81616 times)

bcooter

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1520
Re: 50MP Canon 5DS/R spec leaked
« Reply #160 on: February 10, 2015, 01:28:41 pm »

I've never owned a camera that didn't require a compromise.

the Nikon d3 I gave up detail and more exacting color for high iso and excellent focusing.

The Canon 1dx I gave up extreme detail for robust ethernet tethering and canon color which I find more pleasing than most.

With the Leica S2 I lose high iso and lightning fast focus for a more traditional camera, that shoots a very nice file.

In Motion it's the same.

With RED I give up small portability for a professional cinema look.

With the gh3s I lose log files for a small package and the Canon 70d I lose 4k, for excellent lifestyle mobility and thoughtful and accurate autofocus.

Would I like all the best attributes of the cameras I just mentioned in one package?  Sure and of course I'd love it to be $4,000, but life just doesn't work that way.

The point most are missing is Canon has a huge user base, especially in the professional ranks.

They are also the only company (other than Sony) that makes professional still cameras, decent post production software with a lot of cross platform use with a professional cinema line.

Unlike Sony Canon is very deeply embedded in the professional ranks, with rentals or purchase on almost every metro area.

So if you work in advertising and mixed media, a case of Canons will pretty much cover everything I could/would/plan to shoot.

Is it free or cheap? 

No, but professional equipment rarely is.

As I mentioned before, nearly all higher end digital cameras are very good, in both still and motion and the shoot budget, or talent of the artist will limit the production much more than a camera.

Internet conversation tends to lean to the negative.   Don't know why, maybe it's just the human condition and you see it in this thread and that negative Leica thread, but personally I see things in a much more positive light.

I also know in the professional world, image creators don't grind on about cameras.  Most have their preferences, most use what works until it either stops or something forces them to move upstream and if they can't afford to buy, they rent.

But nobody shoots poorly because of the camera and no great image is produced because of the camera.

I have and do own Nikon, Canon, Leica, RED, Panasonic, Olympus, Sonys and I've very, very rarely started production without a case of Canons and consequently rarely produced a project of any size that didn't have some or all content produced with a Canon.

I don't think anybody here needs to worry about Canon, Nikon or Leica's P+E.   They're all fine.



IMO

BC
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: Can was/is a relatively small upstart in sensor design and electronics
« Reply #161 on: February 10, 2015, 03:39:02 pm »

Do you think it's the design technology, or merely the processes that are available to the designers? I can't remember the numbers, but last time I looked at the Canon chips, the feature sizes were really large, larger than you'd want if you were trying to get the fill factors up and the pitch down.

Jim
I have no idea of the details (designers vs equipment), and even if it is designers, it can be a matter of quantity, not quality: Sony can probably justify putting more engineers on the task of designing the next generator sensor cell or whatever, on the expectation of greater total sales.  I only see the overall result; that bigger electronics operations have a natural advantage in pushing the performance of their sensors forward.

It could well be that Canon insists on using fab. equipment that it makes and owns itself, and int fab business, Canon has fallen well behind industry giant AMSL (and probably behind the industry's #2, Nikon Precision Equipment.)  Contrary to what some suggest, I am sure that Sony is happy to buy its fab. equipment from whoever offers the best tool for the job, even it means using AMSL gear to make sensors for Nikon.
Logged

eronald

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6642
    • My gallery on Instagram
Re: Can was/is a relatively small upstart in sensor design and electronics
« Reply #162 on: February 10, 2015, 04:57:47 pm »

I have no idea of the details (designers vs equipment), and even if it is designers, it can be a matter of quantity, not quality: Sony can probably justify putting more engineers on the task of designing the next generator sensor cell or whatever, on the expectation of greater total sales.  I only see the overall result; that bigger electronics operations have a natural advantage in pushing the performance of their sensors forward.

It could well be that Canon insists on using fab. equipment that it makes and owns itself, and int fab business, Canon has fallen well behind industry giant AMSL (and probably behind the industry's #2, Nikon Precision Equipment.)  Contrary to what some suggest, I am sure that Sony is happy to buy its fab. equipment from whoever offers the best tool for the job, even it means using AMSL gear to make sensors for Nikon.

BJL, Jim,

 AFAIK, sensors like memory are mainly repetitive structures, and when I was a student, received wisdom was the design itself is a matter of having a few really good designers to create the elementary cells, rather than an army of average ones. On the other hand, the device itself is often at the upper limit of economically viable yield, and so process control is crucial.

 I talked to the Aptina guys a couple of years ago, they thought cellphone tech is now driving the sensor business - this is where the most R&D dollars get invested. Of course Sony is a very big player in cellphone modules which may explain why its tech has advanced so quickly.

Edmund

Logged
If you appreciate my blog posts help me by following on https://instagram.com/edmundronald

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: Can was/is a relatively small upstart in sensor design and electronics
« Reply #163 on: February 10, 2015, 05:46:36 pm »


 AFAIK, sensors like memory are mainly repetitive structures, and when I was a student, received wisdom was the design itself is a matter of having a few really good designers to create the elementary cells, rather than an army of average ones. On the other hand, the device itself is often at the upper limit of economically viable yield, and so process control is crucial.


I think with analog designs -- and most of a sensor is analog -- it was ever thus. You're probably too young to remember rock-star analog IC designers like Bob Widlar.

Jim

eronald

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6642
    • My gallery on Instagram
Re: Can was/is a relatively small upstart in sensor design and electronics
« Reply #164 on: February 10, 2015, 06:57:49 pm »

I think with analog designs -- and most of a sensor is analog -- it was ever thus. You're probably too young to remember rock-star analog IC designers like Bob Widlar.

Jim

Jim, you made my day with that reference!

Quote
However, the story about Widlar bringing a goat to trim the lawn in front of his office, retold by The New York Times after his death,[8] was incorrect.[11] It was a sheep, not a goat;[71] Widlar brought her in his Mercedes-Benz convertible for just one day, which included a photo op for the local journalists.[11] According to Pease, Widlar abandoned her in the nearest bar;[11] according to Lojek the sheep was "mysteriously stolen".[71]

This rings a bell with me, because I remember how astonished I was, walking across the Cam bridge to the back gate of King's when I saw this flock of sheep grazing inside the grounds. So I asked around and was told that the college rented the meadow out to a farmer :)

I find it incredible that in an industry that thrives on reverse engineering people found it so difficult to replicate his designs.

Edmund
« Last Edit: February 10, 2015, 07:10:51 pm by eronald »
Logged
If you appreciate my blog posts help me by following on https://instagram.com/edmundronald

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: Can was/is a relatively small upstart in sensor design and electronics
« Reply #165 on: February 10, 2015, 07:10:28 pm »

I have no idea of the details (designers vs equipment), and even if it is designers, it can be a matter of quantity, not quality: Sony can probably justify putting more engineers on the task of designing the next generator sensor cell or whatever, on the expectation of greater total sales.  I only see the overall result; that bigger electronics operations have a natural advantage in pushing the performance of their sensors forward.

Sony semi-conductor is basically the leader in every segment of consumer imaging sensors, including mobile phones, compact, 1 inch, APS-C, FF and now [smallish] MF.

They probably sell about as many APS-C/FF sensors as Canon does (most probably at higher average prices since only a subset of their sales are internal), but then also 10 times more compact camera sensors and probably 30~40 times more sensors overall when you take into account mobile phones.

As always in consumer electronics, the larger the volume the higher the technological investments that can be done. Sony can probably spend 10 times more in R&D than Canon can without taking any risk from a profitability standpoint.

So it would take incredibly talented engineers on Canon side to compensate for this.

A person I know reported 3-4 years ago that Sony already had at that time samples of non linear sensors behaving like negative film in terms of highlight response... They haven't had to release such technologies to the market yet.

Cheers,
Bernard

eronald

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6642
    • My gallery on Instagram
Re: Can was/is a relatively small upstart in sensor design and electronics
« Reply #166 on: February 10, 2015, 07:13:10 pm »

Sony ...

They haven't had to release such technologies to the market yet.

Bernard


Yes. Sad innit?

However, I think it was Noyce at Intel who said that the only real strategy of survival as a market leader is to obsolete your own products by means of new introductions as quickly as possible ... before your competitors do it for you.

Edmund
« Last Edit: February 10, 2015, 07:15:37 pm by eronald »
Logged
If you appreciate my blog posts help me by following on https://instagram.com/edmundronald

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: Can was/is a relatively small upstart in sensor design and electronics
« Reply #167 on: February 10, 2015, 08:32:53 pm »


I find it incredible that in an industry that thrives on reverse engineering people found it so difficult to replicate his designs.


There weren't a lot of companies interested in analog IC's in those days. Fairchild, National, maybe Signetics, later Intersil, and a few others that my advanced age has suppressed. There were other designers who got a lot of ink in Electronics News and the San Jose Merc: Pease and Dobkin . Both Bobs, and both Widlar disciples. But, to make your point, probably the most successful op amp of all time, the ua741, was designed by a distinctly non-rock star Dave Fullagar working at Fairchild -- after Widlar spurned it and ran off to National. Based on Widlar's LM101 (that was the mil-spec version, the commercial version was the 301) the 741 went on to far outsell its intellectual progenitor.

So the tortoise won out over the hare in the end.

There's a Canon connection, too. From the link above:

"Although the 741 is Fullagar’s most famous design, the one he best remembers was for a Canon: “My favourite product was a camera circuit developed for Canon Camera in Japan. It took the log of three inputs (aperture film speed and light intensity) summed the logs took the antilog and derived a shutter speed output. The whole circuit was temperature compensated and used a total of about ten transistors.”

Jim



eronald

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6642
    • My gallery on Instagram
Re: Can was/is a relatively small upstart in sensor design and electronics
« Reply #168 on: February 10, 2015, 08:54:07 pm »

So the tortoise won out over the hare in the end.

Jim



 Interesting story. Also quite astonishing that the japanese would hire an outside designer to do a circuit for them.

 I just chased down the 0.9V 555 redesign commissioned from Camenzind himself - and something else in the same vein. I guess my son will soon be old enough so I can breadboard some 555 circuits with him ...

 Update - the "led-driver-charger" thingy has already been obsoleted ....Camenzind's low-voltage 555 still exists as  ZSCT1555 - I'd love to get hold of a few samples!

Edmund
« Last Edit: February 10, 2015, 09:18:50 pm by eronald »
Logged
If you appreciate my blog posts help me by following on https://instagram.com/edmundronald

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Can was/is a relatively small upstart in sensor design and electronics
« Reply #169 on: February 10, 2015, 09:38:22 pm »

Hi,

Chipworks discusses it here:

http://www.chipworks.com/en/technical-competitive-analysis/resources/blog/full-frame-dslr-cameras-canon-stays-the-course/?lang=en&Itemid=815

This sentence is interesting:
Quote
While Chipworks didn’t analyze that camera, every Canon FF sensor analyzed since has used the same 0.5 µm design rules. It is a credit to Canon that it has remained competitive by continuing to optimize its pixels fabricated in a relatively mature process.

The article gives Canon a lot of credit for refining the design, and I get the impression that the pixels are quit good but it is the readout that fail. That is the reason that Canon is pretty good on high ISO but fails in DR.

Comparing "tonal range" which I think is pretty close to "photographic DR" the Canon 6D and Nikon D750 are quite close, but Nikon has a great advantage in DR below 1200 ISO.

Well, tonal range is much dependent on full well capacity, while DR (in the DxO sense) is very much depending on readout noise.

Best regards
Erik

Do you think it's the design technology, or merely the processes that are available to the designers? I can't remember the numbers, but last time I looked at the Canon chips, the feature sizes were really large, larger than you'd want if you were trying to get the fill factors up and the pitch down.

Jim
Logged
Erik Kaffehr
 

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: Can was/is a relatively small upstart in sensor design and electronics
« Reply #170 on: February 11, 2015, 02:47:48 pm »

However, I think it was Noyce at Intel who said that the only real strategy of survival as a market leader is to obsolete your own products by means of new introductions as quickly as possible ... before your competitors do it for you.
A good reminder, and the subject of "obsoleting yourself first" is a theme in this book on Intel:
http://www.barnesandnoble.com/w/creating-the-digital-future-albert-yc-yu/1002976578?ean=9780684839882

That is why I laugh when forum pundits declare that a company will not or should not introduce some new product or feature because "it would just cannibalize an existing product".  Some complacent market leaders do hold back like that (Canon and Nikon seem to be in that mode with respect to making ILCs with an EVF in place of the OVF), but others like Intel and Apple are happy to indulge in a bit of cannibalism.
Logged

eronald

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6642
    • My gallery on Instagram
Re: 50MP Canon 5DS/R spec leaked
« Reply #171 on: February 11, 2015, 03:42:56 pm »

btw, if some here orders chips on a regular basis and could add a couple of samples of the ztc low voltage 555 to their next order and dump them in an envelope ... I don't think I can get them here in Paris without some huge overhead.

Edmund
Logged
If you appreciate my blog posts help me by following on https://instagram.com/edmundronald

eronald

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6642
    • My gallery on Instagram
Re: Can was/is a relatively small upstart in sensor design and electronics
« Reply #172 on: February 11, 2015, 03:52:44 pm »

A good reminder, and the subject of "obsoleting yourself first" is a theme in this book on Intel:
http://www.barnesandnoble.com/w/creating-the-digital-future-albert-yc-yu/1002976578?ean=9780684839882

That is why I laugh when forum pundits declare that a company will not or should not introduce some new product or feature because "it would just cannibalize an existing product".  Some complacent market leaders do hold back like that (Canon and Nikon seem to be in that mode with respect to making ILCs with an EVF in place of the OVF), but others like Intel and Apple are happy to indulge in a bit of cannibalism.

BJL,

 Yes, that was exactly what I meant, in answer also to Bernard"s post about Sony "having" technology but seeing no need to release it "yet".  Interestingly, Panasonic and Samsung who were for a long time "mainstream followers" as an avowed strategy are now leapfrogging ahead of Nikon and Canon with their well-recieved photo/video hybrid mirrorless products. The focus system on the GH4 is fast and novel, it analyzes the OOF zones thereby bypassing the need for special pixels for a phase contrast system ...

 I think Noyce watched IBM commit slow-motion suicide by attempting to hold back the tide of progress, and learnt from that. I still remember talking to a team that designed the first portable PC for IBM. They had an advanced small laptop design. IBM made them throw it away, and insisted on a redesign so marketing an obsolete monstruosity ... of course the japanese didn't see things in the same way. IBM's attitude in refusing to market what they had  technically pioneered was deeply ingrained, and afforded entry points to any small competitor - DEC in particular ate their lunch even before the PC became an issue, but of course once the PC appeared every manager in every company got involved in computer purchasing and the blood in the water was calling out to the sharks.

Edmund
« Last Edit: February 11, 2015, 04:12:14 pm by eronald »
Logged
If you appreciate my blog posts help me by following on https://instagram.com/edmundronald

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Can was/is a relatively small upstart in sensor design and electronics
« Reply #173 on: February 11, 2015, 04:13:53 pm »

Hi,

About 1.5 years ago Sonyalpha rumors had some strong indications that Sony developed a new generation of sensors around 50 MP. The rumor also said it was extremely expensive in production and that it would not come to market in 2014 but hopefully in 2015.

It can just be that the sensor is still not available in feasible numbers at a feasible price.

Best regards
Erik

BJL,

 Yes, that was exactly what I meant, in answer also to Bernard"s post about Sony "having" technology but seeing no need to release it "yet".  Interestingly, Panasonic and Samsung who were for a long time "mainstream followers" as an avowed strategy are now leapfrogging ahead of Nikon and Canon with their well-recieved photo/video hybrid mirrorless products. The focus system on the GH4 is fast and novel, it analyzes the OOF zones thereby bypassing the need for special pixels for a phase contrast system ...

 I think Noyce watched IBM commit slow-motion suicide by attempting to hold back the tide of progress, and learnt from that. I still remember talking to a team that designed the first portable PC for IBM. They had an advanced small laptop design. IBM made them throw it away, and insisted on a redesign so marketing an obsolete monstruosity ... of course the japanese didn't see things in the same way. IBM's attitude in refusing to market what they had  technically pioneered was deeply ingrained, and afforded entry points to any small competitor - DEC in particular,  ate their lunch even before the PC became an issue.

Edmund
Logged
Erik Kaffehr
 
Pages: 1 ... 7 8 [9]   Go Up