Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
12-25-2009, 02:19 PM   #31
Veteran Member
vizjerei's Avatar

Join Date: May 2007
Location: Toronto
Photos: Gallery
Posts: 1,418
QuoteOriginally posted by gazonk Quote
From my son's translation, I understood that they actually said that the release of the 645D won't delay the further development of the existing 2 lines. I took that as a strong hint that we'll see updates of both K-x and K-7 no later than Photokina.
When is that Photokina?

12-25-2009, 02:21 PM   #32
Senior Member




Join Date: Jul 2008
Location: Warwickshire, UK
Photos: Albums
Posts: 196
QuoteOriginally posted by vizjerei Quote
When is that Photokina?
September 2010
12-25-2009, 03:55 PM   #33
Veteran Member
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,863
QuoteOriginally posted by Nubi Quote
To me, it looks like Pentax is stalling to see if the cost of manufacturing the sensors will come down.
In some other thread, I computed the manufacturing cost for APS-C, FF and MF sensors based on data published by semiconductor analysts. I don't remember the exact figures but if I remember correctly, they were such that camera body street prices of 500$ for APS-C, 1000$ for FF and 2000$ for MF are perfectly feasible -- for a company controlling CMOS manufacturing.

Now, we almost have a chicken and egg situation: camera vendors stick with APS-C and charge premium for FF as long as the market follows. Which in turn keeps the volume of the FF sensor market low and FF sensor purchase prices for camera vendors high. The FF sensor market has formed a price bubble where price is disconnect from cost. I predict this bubble to burst soon.

Ironically, Pentax creates pressure onto this price bubble (with an MF camera cheaper than some FF cameras ...) and as soon as it bursts, it will have become impossible to sell 1000$ cameras with anything less than FF. Pentax needs this $1000+ market and will react accordingly. Question is if they are really prepared as the bubble could burst as soon as late 2010.

I predict the bubble to burst by Canon attacking Nikon to regain lost market share.
12-25-2009, 08:15 PM   #34
Nubi
Guest




QuoteOriginally posted by falconeye Quote
In some other thread, I computed the manufacturing cost for APS-C, FF and MF sensors based on data published by semiconductor analysts. I don't remember the exact figures but if I remember correctly, they were such that camera body street prices of 500$ for APS-C, 1000$ for FF and 2000$ for MF are perfectly feasible -- for a company controlling CMOS manufacturing.

Now, we almost have a chicken and egg situation: camera vendors stick with APS-C and charge premium for FF as long as the market follows. Which in turn keeps the volume of the FF sensor market low and FF sensor purchase prices for camera vendors high. The FF sensor market has formed a price bubble where price is disconnect from cost. I predict this bubble to burst soon.

Ironically, Pentax creates pressure onto this price bubble (with an MF camera cheaper than some FF cameras ...) and as soon as it bursts, it will have become impossible to sell 1000$ cameras with anything less than FF. Pentax needs this $1000+ market and will react accordingly. Question is if they are really prepared as the bubble could burst as soon as late 2010.

I predict the bubble to burst by Canon attacking Nikon to regain lost market share.


Wow. MF for $2G? That can cause some serious unrest. Pleasure to watch, I am sure. But your line of reasoning makes sense, if it really is feasible in the very near future. Very exciting. What this means is that we the so-called serious photographers or high end amateurs ought not to buy another APS-C??? Would they replace k-7 before they come out with 645D????

12-26-2009, 04:49 AM   #35
Veteran Member
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,863
Cost of CMOS sensors, various sizes

QuoteOriginally posted by Nubi Quote
But your line of reasoning makes sense, if it really is feasible in the very near future. Very exciting. What this means is that we the so-called serious photographers or high end amateurs ought not to buy another APS-C??? Would they replace k-7 before they come out with 645D????
No, the 645D will be next and the K-7 replacement will still be APS-C, I am pretty sure.


I googled my contribution where I computed the above numbers. They have been challenged by some. But in no serious manner. Here we go:Revisiting all facts again, the respective costs (APS-C/FF/MF 5641.5mm (note that digital MF seems to be cropped instead ...)) are:

300mm Wafer: $2500 (cost including factory depreciation, source ICKnowledge)
Chips / wafer: 164/66/20 (source University of Hamburg)
Yield rate: 85%/68%/35% (source ICKnowledge (defect data) and myself (calculus))
-------------------------
Cost per chip real estate: $18/$56/$350
Cutting, testing, packaging: +50%
-------------------------
Cost per chip real estate: $27/$84/$525
Development: $500,000
Volume: 500,000/40,000 (5 models for 2% 10 million DSLR/y)/3,000 (guess)
Development depreciation: $1/$12/$165
-------------------------
Total: $28/$96/$690
Approx: $30/$100/$700


Assuming a street price where sensor cost is fixed to be 5% of camera body street price, we get:
$560 / $1900 / $14000
which is pretty much what we see.


BUT

we pay 20x the manufacturing cost of the sensor. This is OK for 30$ sensors. With all the extra value added by the camera body. But with the larger sensors, a lot of profit is aggregated down the "value chain". So much that I call it a prize bubble.

Note that the above costs are those for, e.g., Kodak. The CMOS makers don't necessary sell them for a price close to cost. Not without pressure from the market. One source says that Kodak MF sensors sell for $3500 which would be 5x the cost. So, CMOS makers charge 4-5x premium and camera makers another 4-5x premium.

I am sure that small sensors (phones) have become a low margin business.

Therefore, the industry (CMOS and camera makers together) fight hard to keep the large sensor market high margin. Canon even prints broschures giving false information why full frame is expensive. The industry either makes large format expensive, or tells us they won't produce it or we won't need it.

It is the end user market which needs to create counter pressure and it does. E.g., FF is the single most discussed topic in this forum.

Last edited by falconeye; 12-26-2009 at 05:02 AM.
12-26-2009, 05:03 AM   #36
Veteran Member




Join Date: Jul 2008
Location: Colorado USA
Posts: 1,337
I wonder if Kodak CCD Medium Format 50 mp dropped in retail price ?

I wonder if Kodak CCD Medium Format 50 mp dropped in retail price over past 18 months ? Or if they sell them at $3,500 a piece still ?


EEProductCenter.com :: Kodak delivers next-generation 50-MP CCD image sensor

And the "Bubble Prize" 50mp Hassy for $25,200 the 18month old $3,500 retail cost kodak CCD sensor is installed in:

http://www.bhphotovideo.com/c/product/644448-REG/Hasselblad_70380542_H3DII_5...LR_Camera.html

Thanks for the sensor manufacturing costs insight Falk, nice to have it in one handy place.

>>>"

Rochester, N.Y. — Eastman Kodak Company has claimed the industry's first 50-megapixel CCD image sensor for the professional photography market, capturing digital images at unprecedented resolution and detail thanks to a newly designed 6.0-micron pixel and several other advances. With an 8176 x 6132 pixel array, Kodak said the KAF-50100 50-megapixel sensor provides the highest resolution available in the popular 48 x 36-mm optical format used in medium format photography.

Working with leading camera makers, Kodak's Image Sensor Solutions group focused on several key areas of improvement with the new CCD image sensor design including higher resolution while maintaining dynamic range, reducing noise and faster capture. The result is the company's new TRUESENSE 6.0 micron Full Frame CCD Technology Platform, which increases both the resolution and camera performance. The KAF-50100 is the first Kodak sensor to use the company's new technology.

Based on a 6.0-micron pixel, the platform provides increased data throughput for a faster frame rate, a reduced "click to capture" time for improved camera response, lower power consumption for improved battery life, and improved color fidelity, while still retaining key performance parameters available from the larger, 6.8-micron pixel used in current products.

They needed additional resolution that directly drives the requirement for a smaller pixel, moving from a 6.8-micron pixel for the 39 MP sensor to 6.0-micron pixel for the 50 MP device, said Michael DeLuca, marketing manager for Image Sensor Solutions at Eastman Kodak Company. This is a 30 percent change in area resulting in a 30 percent change in resolution, he said.

In addition to increasing resolution, they wanted to maintain all other specs including dynamic range. In order to retain the 70-dB dynamic range spec for the device, Kodak had to implement a few other redesigns in the sensor. DeLuca said it required a new amplifier design to help reduce noise and a new four-output architecture, compared to two outputs used in the 39-MP sensor, to help manage the increased quantity of data available from the image sensors, enabling sensor operation at one frame per second (fps) for a 50-megapixel device, and to gain additional bandwidth. This also helps to lower the master clock, which helps further reduce noise, said DeLuca.

Kodak also implemented several other performance improvements. A new global reset function, which allows the entire sensor array to be reset using only a single clock pulse, reduces both power consumption (improving battery life in the camera) and the "click to capture" time (now measured in microseconds instead of milliseconds) for improved camera response.

Kodak also refined the spectral characteristics of the red pigment to improve color fidelity, allowing sensors based on this new platform to provide richer, more accurate colors.

The 50-MP CCD image sensor may be used in other applications including aerial photography. Case-in-point: With a 50-megapixel camera, in an aerial photo of a field 1-1/2 miles across, you could detect an object about the size of a small notebook computer (1 foot x 1 foot).

The new 50-MP CCD image sensor has already been designed into Hasselblad's new H3DII-50 high-end DSLR camera system, touting the 1 fps capture rate due to the 4-channel readout structure. The H3DII-50 will be available in October 2008.

Availability: Engineering grade devices of the KAF-50100 are currently available, with volume production planned for fourth quarter 2008.
Pricing: $3500 in volume.
Product information: KAF-50100

<<<"

Last edited by Samsungian; 12-26-2009 at 05:25 AM.
12-26-2009, 04:17 PM   #37
Nubi
Guest




What this tells me is that once 645D comes out I should wait a bit, because sensor prices should still be on the way down when it comes out. Maybe ogl is right, in that there are 2 MF models. One close to 8G with bells and whistles, and another close enough to compete with D3s or x or whatever it was.

How much more improvement in APS-C can be expected? Is it worth spending more R and D to squeeze out the last drop of what that format can offer? If the price of MF is to come down significantly, then I ain't going to buy another APS-C camera( or lens es for that matter), and start buying 2nd hand MF lenses on ebay.

I think that DA lenses and ltds won't be a waste in the end because I know that eventually we will have a set up like EVF to mount it on.
12-26-2009, 07:47 PM   #38
Veteran Member
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,863
QuoteOriginally posted by Nubi Quote
How much more improvement in APS-C can be expected?
Current CMOS can improve a bit (in any size) but not too much:

1. Quantum efficiency can improve 2-3x.
2. Loss due to Bayer filter theoretically can be avoided with prism or some other more elegant technology, another factor 3.
3. Fill factor may improve by 20%.
4. Read-out noise may improve to the point where it is marginal.
5. Full well capacities may improve significantly, providing lower ISO and higher DR.

But that's it. Beyond this point, no further progress is possible within the constraints of laws of physics. Dedicated cooled monochrome astro cameras have almost reached this point already.

This means that the APS-C sensor of the year 2050 (or 40,000 A.D.) performs about as good as the MF sensor of the year 2009.

12-27-2009, 12:26 AM   #39
ogl
Pentaxian
ogl's Avatar

Join Date: Dec 2007
Location: Siberia
Photos: Gallery | Albums
Posts: 7,254
QuoteOriginally posted by Samsungian Quote
I wonder if Kodak CCD Medium Format 50 mp dropped in retail price over past 18 months ? Or if they sell them at $3,500 a piece still ?


<<<"
645D will be or with old 39 MP sensor, either with new 40 MP Kodak.
12-27-2009, 05:40 AM   #40
Pentaxian
thibs's Avatar

Join Date: Jun 2007
Location: Belgium
Photos: Albums
Posts: 5,267
1.3X crop ? Crap.
12-27-2009, 07:48 AM   #41
Veteran Member




Join Date: Jul 2008
Location: Colorado USA
Posts: 1,337
Kodak Medium Format CCD Sensor specs

Current Kodak Medium Format CCD Sensor specs:

Kodak Image Sensor Solutions - Photography Applications



Interesting old 2006 article on CMOS and CCD sensors:

http://www.luminous-landscape.com/essays/sensor-design.shtml



And my google search string for more articles on Kodak Sensors:

http://www.google.com/search?sourceid=navclient&aq=2&oq=kodak+medium&ie=UTF-...+format+sensor

Enjoy

Last edited by Samsungian; 12-27-2009 at 07:56 AM.
12-27-2009, 09:10 AM   #42
ogl
Pentaxian
ogl's Avatar

Join Date: Dec 2007
Location: Siberia
Photos: Gallery | Albums
Posts: 7,254
QuoteOriginally posted by thibs Quote
1.3X crop ? Crap.
39 and 50 MP is rather slow. 0.9 and 1 fps. I've heard about approx. 1.5 fps in 645D.

50 MP sensor has almost the same specs as 39 MP.
40 MP Kodak is new sensor with ISO50-3200 and better noise performance.
Made on September, 2009 if I understand right.

In specs from Chinese site 645D's jpeg size - 7264 x 5440

40 MP's active area resolution - 7304 x 5478
39 MP's active area - 7216 x 5412


It's time to think.

Last edited by ogl; 12-27-2009 at 09:30 AM.
12-27-2009, 01:23 PM   #43
Loyal Site Supporter




Join Date: Jun 2009
Location: Tumbleweed, Arizona
Photos: Gallery | Albums
Posts: 5,107
Cameras are computers today, thus electronics is king. Everything in semiconductors is about yield, which translates to geometries. If the MF sensor were to be optimized for the best yield from a given silicon wafer/technology combination, and geometries optimized for pixel resolution and applied, then I would think that Pentax could have a pretty inexpensive (comparatively) at a somewhat MF sensor size with a probably non standard resolution (again comparatively) but hugh by APS-C and FF standards. Say 35 to 40+MP in maybe a slightly off nominal sensor rectangle. If all of this lived with in the current 645 lens circle, then you have a pretty good chance at a win, especially at a possibly lower price point.

Also, if you look back at the K7 and KX and consider what was done with these models, and then apply them to a MF product, you may have a sense of the potential future offering may entail. The K7 came up with the single piece strong lightweight and small magnesium alloy shell. The KX came up with some improved battery/power management technology with respect to taking AA batteries and squeezing out about double the number of images than what the KM/2000/K200/K100 had previously.

Then consider the recent Leica S2 MF product -

Leica S2 First Impressions

It's expensive - $23K to 30K, it is small by MF standards - a bit larger than a FF, looks great, sets a new standard, etc. So, you have a price ceiling, and for each $10K you decrease in price your sales potential goes up quit a bit. You bypass the FF competition as has been observed here already. You get to set your own new standard to a large extent. You virtually have no competition in this market - especially with a lower price point.

Also, looking at Leica S2 performance standards, which if were in a K7 would be dead on arrival, however taking the K7 current performance and scaling by 3.5 to 4 and your there in resolution and divide by the same scale and you are also there in terms of FPS, etc. Also, using the same software, but on an advanced processor chip that may take a bit more power from a better power management system, and a chassis that holds a slightly larger battery, you have a very doable combination to bring to market. If Pentax is able to come in with a similar sized small body MF, and WITH image stabilization, which might be an additional feature - a body larger than a K7 even without a long lens handheld, IS or SR might be a welcomed additional capability.

Also, I do not think that we have seen the last of Samsung. Its been observed here also, that Pentax would need to have some additional control at the semiconductor FAB level in order to control costs. Samsung is very interested and heavily invested in cell phones. The future of the cell phone is in the camera, and they have the processing internals, however what any cellphone/camera needs is a lens. Sure you can get 14MP on a cellphone but without a collapsible good quality lens, its garbage. A technology swap lens design for semiconductor production (at a controlled cost with improved yields) would be a good marriage. Plus, the K7/K20 sensors would also be proof of concept runs - and MF really do not need extremely high ISO (again see Leica). The quantity runs for a MF would be relatively small to an APS-C sensor, so costs are again extremely important.

I would also think that Pentax would probably take a somewhat new approach to backs and bodies. I am wondering aloud if Pentax were to go the back route, and have the ability to sell lenses, backs and bodies, especially to an customer set who wants a FF.

I would think that $5K for the body alone might be difficult. Under $10K certainly should and for each $1K under that, maybe count an additional 1 to 3K units sold (pure guess) each with a healthy profit margin, might be just the overall marketing opportunity. Might pull a lot of Canon and Nikon full frame users away. A lot of buzz on the street until Canon and Nikon come up with an answer. Also couple this with the existing 645 and 67 lenses already out there to support the new body, Pentax might just have the leg up that they are looking for.

Update - If Pentax were to go the back route and start out at - I do not know - say 35MP initially, and then each of the following years go to 40MP, 50MP and say 60MP, using the same lens, body combination (maybe updated battery technology), then there is a yearly upgrade path to move folks along the curve according to their needs and checkbook balances. Pentax in this market would, at a low price point, essentially have scale of size in their court - compared to the rest.

I am just thinking out loud...

Last edited by interested_observer; 12-27-2009 at 01:36 PM.
12-28-2009, 07:38 AM   #44
Banned




Join Date: Sep 2006
Location: Millstone,NJ
Posts: 6,491
QuoteOriginally posted by noblepa Quote
Perhaps, but Nikon and Canon already have that (very small) market pretty well sewn up. Breaking into that market segment could be very, very difficult.

OTOH, the MF market, at least at the low end, where a potentially $10,000 camera (which is the most commonly rumored price-point that I've heard) would fit, is more open. Pentax might have a tough time going directly against Hassleblad or Mamiya in the high-end MF market, though.

In addition, Pentax has a long history of pro-level MF cameras.
I bet there are at least 10x more people willing to pay $2,000 for a FF Pentax DSLR body to use with their FF K-mount lenses than are willing to pay $10,000 for the 645D body that needs a whole new lens series.

Last edited by jogiba; 12-28-2009 at 07:44 AM.
12-28-2009, 08:01 AM   #45
Veteran Member
Asahiflex's Avatar

Join Date: Jan 2008
Location: Netherlands
Photos: Gallery
Posts: 2,754
QuoteOriginally posted by jogiba Quote
I bet there are at least 10x more people willing to pay $2,000 for a FF Pentax DSLR body to use with their FF K-mount lenses than are willing to pay $10,000 for the 645D body that needs a whole new lens series.
Indeed; it's as easy as that.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
645d, pentax news, pentax rumors, translation
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
A LITTLE HELP with translation please puderse General Talk 16 12-06-2009 02:11 AM
Need help for translation Malsi Pentax SLR Lens Discussion 4 08-19-2009 01:10 PM
Lost in translation Damn Brit Post Your Photos! 8 05-23-2008 01:16 PM
Chinese translation please JCSullivan General Talk 28 05-19-2008 07:05 AM
got a new 21mm, acurate and tight chals Pentax SLR Lens Discussion 1 05-02-2007 11:11 AM



All times are GMT -7. The time now is 10:06 PM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top