Home

Please support mflenses.com if you need any graphic related work order it from us, click on above banner to order!
SearchSearch MemberlistMemberlist RegisterRegister ProfileProfile Log in to check your private messagesLog in to check your private messages Log inLog in

New Sigma SD1 - soon
Post new topic   Reply to topic
Home   Official News and Announcements
View previous topic :: View next topic  


PostPosted: Wed Sep 22, 2010 2:30 pm    Post subject: Reply with quote
Anu wrote:
Orio wrote:
calvin83 wrote:

I just found something interesting: The design of the X3 sensor seems has close connection to the Chromatic Aberration of the microlens.


Do you think that this is the reason why the 5D Mark II constantly gives me more lateral CA than the original 5D mounting the same lens?


How do you measure CA? Relativce to the image or relativce to pixel?


Just... visually, I would say... in a perceptual way, not measuring. It's more noticeable. I mean not just for size also for the colour intensity. I first blamed the lenses but then verified that the same lens does much better on the old 5D. In my ignorance of the technical matter, I thought that this was caused by the tight proximity of the 5DMkII sensor cells.


PostPosted: Wed Sep 22, 2010 2:43 pm    Post subject: Reply with quote
Anu wrote:

This is where you are totally lost, sorry. 7D is a 18 megapixel camera, Sigma is a 15 megapixel camera. Photosite and pixel are different. While the 7D photosites only record about half the light that hits them (the other half is blocked by they bayer filter as you know - I say about half since Canon has notoriously weak color separation), the Foveon pixels don't record too much more more data (or to be more precise, signal/noise) in normal photographic conditions.

Quote:

7D's sensor collects 18 milions of color subpixels. It can be said, that each of the photodiode captures one-third of brightness data. So if you refuse to call Foveon 46 megapixel sensor, you should also refuse to use similar naming regarding common cameras.


There are no such thing as color subpixel. The Canon, as the other bayer-filtered cameras record imperfect data for all the pixels, in this case for 18Mp. The Foveon also records imperfect data, though for different reasons - for example the three photosites stacked on top of each other do not capture different wavelenghts anywhere near perfectly - the colour separation is actually worse than it is with the bayer sensors - I think the fellows behind the Foveon sensor said there would be needed about 6 layers of photosites for similar coolour accuracy to typical Bayer-separates sensor. You conviniently igonore this and other sources of imperfectioin for Foveon pixels and give each of them a imaginary weight of 3 units and each imperfect Canon pixel weight of 1 unit - what I mean that you somehow think that a Foveon pixel is three times "better", or more valuable than a Canon pixel, therefor you can calll Foveon a 46Mp (or whatever). But this is just silly, as in reality a Foveon pixel may be even less information containing than a Canon pixel (depending on shooting conditions and ISO used). Using your logic the number of pixels in a sensor should change with the used ISO etc. Smile

Thanks for this long explanation Anu. I wanted to answer more or less the same thing, but you did it with much more accuracy and details that I would have done...


PostPosted: Wed Sep 22, 2010 3:13 pm    Post subject: Reply with quote
Orio wrote:
calvin83 wrote:

I just found something interesting: The design of the X3 sensor seems has close connection to the Chromatic Aberration of the microlens.


Do you think that this is the reason why the 5D Mark II constantly gives me more lateral CA than the original 5D mounting the same lens?


Since the Foveon pixel structure consists of three photosites in different depths, it is much more sensitive to the angle the light hits the sensor than a traditional Bayer-sensor.

Orio wrote:
Anu wrote:

How do you measure CA? Relativce to the image or relativce to pixel?


Just... visually, I would say... in a perceptual way, not measuring. It's more noticeable. I mean not just for size also for the colour intensity. I first blamed the lenses but then verified that the same lens does much better on the old 5D. In my ignorance of the technical matter, I thought that this was caused by the tight proximity of the 5DMkII sensor cells.


I was thinkin that if you see the CA covering more pixels, it would be due to higher resolution of the 5D II, but if you see increase at the image level as well, it might be either because of different processing (especially in jpg-workflow, but also raw), or because of diferent microlens or photosite design of the newer sensor.


PostPosted: Wed Sep 22, 2010 3:45 pm    Post subject: Reply with quote
Anu wrote:
This is where you are totally lost, sorry. 7D is a 18 megapixel camera, Sigma is a 15 megapixel camera. Photosite and pixel are different. While the 7D photosites only record about half the light that hits them (the other half is blocked by they bayer filter as you know - I say about half since Canon has notoriously weak color separation), the Foveon pixels don't record too much more more data (or to be more precise, signal/noise) in normal photographic conditions.

You are oversimplificating some facts and overcomplicating anothers.

17 megapixel bayer-based sensor doesn't capture 17 milions of pixel, but 17 milions of color-subpixel (= partial pixel data). In terms of color photography, placing bayer-filter over 17 megapixel BW sensor turns each photodiode into color subpixel.

In theory, 1/3 of color information is captured, the rest isn't and it's later guessed by bayers interpolation. The output is 17 megapixel, but 2/3 of the data wasn't captured by the sensor - just interpolated by software.

SD1's foveon sensor captures 46 milions of color subpixels. Each subpixel contains 1/3 of color data - just like bayers color subpixel. 7D captures 18 milions of color subpixels, it's processor guess another 36 milions of subpixels by interpolation and combines it to create the resulting image. SD1 captures all the 46 milions of color subpixels, combines them and creates the resulting image.
Anu wrote:
There are no such thing as color subpixel.

Shocked

Quote:
Typical color image sensors used for digital cameras employ an array of color filters, to divide the light intensity at a pixel into its RGB primary color components prior to opto-electronic conversion. Traditional Bayer array has a pixel divided into 2x2 subpixels. One of the four subpixels is devoted to observation of the R component, another observes B, and the other two observe G.

Tomohide Someya, Tomoko Yokokawa and Masaru Kamada
Department of Computer and Information Sciences, Ibaraki University
http://tinyurl.com/384ub7v

Anyway, back to your opinion. I don't understand, why are you talking about color separation, which isn't directly related to resolution.

Color separation is a biggest issue of Foveon, but it isn't related to resolution and even not to final color accuracy, but it's related to noise. Large portion of data had to be removed to obtain correct colors and this makes Foveon 3-4 less sensitive than typical bayer-based sensor. But resulting problem is high noise level at high ISO, not resolution nor color accuracy.

Anu wrote:
I think the fellows behind the Foveon sensor said there would be needed about 6 layers of photosites for similar coolour accuracy to typical Bayer-separates sensor.

I have always heard this argument only from you. Could you please tell me, what is the source (link would be fine). It doesn't make much sense, at least without some context.

The biggest problem is, that significant portion of light passes through without being captured and other significant portion of light is captured near the surface. More layers wouldn't solve this problems.

Anyway, with the SD15 it was more or less proved, that color accuracy was problem of imperfect processing of the older cameras. SD15's color accuracy is better than on many current bayer-based cameras.


PostPosted: Wed Sep 22, 2010 4:04 pm    Post subject: Reply with quote
no-X wrote:
SD1's foveon sensor captures 46 milions of color subpixels. Each subpixel contains 1/3 of color data - just like bayers color subpixel. 7D captures 18 milions of color subpixels, it's processor guess another 36 milions of subpixels by interpolation and combines it to create the resulting image. SD1 captures all the 46 milions of color subpixels, combines them and creates the resulting image

8 Mpixels is enough for most of us
we don't care much if the sensor is ccd, bayer of foveon
if image is nice, we are happy
good luck to all of them Very Happy


PostPosted: Wed Sep 22, 2010 8:27 pm    Post subject: Reply with quote
Dear no-X,

I wrote a very long answer to your previous post, but considering the origiinal topic of 46 Mp vis-a-vis 17 Mp (or whatever the actual pixel counts were), and the discussion started wondering, I decided to just grab one paragraph of my answer for you to read as it holds pretty much what is relevant. Others can read it too Smile

Anyhow, here were go:

You might want to talk about photosites - the Foveon has 46 million of them, Canon 17 million. And then please notice that not all photosites are equal. While each Foveon photosite tries to record either Red, Blue or Green, always a total of three of thse photosites share the very same spatial position. On the other hand each Bayer photosite tries to record not only either Red, Green or Blue, but also ties that data into one unique spot on the sensor. This is important to understand. Not all photosites are equal any more than all pixels are equal. While a Foveon pixel in principle records more data, each of it's photosites record less.


PostPosted: Fri May 20, 2011 3:02 pm    Post subject: Reply with quote
Quote:
Sigma Corporation of America to ship SD1 camera in early June

Ronkonkoma, NY, May 20, 2011 – Sigma Corporation of America, a leading researcher, developer, manufacturer and service provider of some of the world's most impressive lines of lenses, cameras and flashes, is pleased to announce that its flagship DSLR, the Sigma SD1, will be available for purchase for the MSRP of $9,700 in early June.


The MSRP of SD1 is about 6.5 times of SD15. Shocked Shocked Shocked

http://www.sigmaphoto.com/news/sigma-corporation-of-america-to-ship-sd1-camera-in-early-june


PostPosted: Fri May 20, 2011 4:27 pm    Post subject: Reply with quote
well for 10 grand hopefully they made the viewfinder actually useable as opposed to the long dark tunnel of the sd14. put me down for a dozen. Laughing

what a joke...$10,000. imagine this conversation: 'hey honey, we could buy that new car we need or i could get that sigma, cause i think they fixed the VF!' should go over well, no?

the only thing that might make as much sense is spending $5000 on a cam, another couple grand on some of the best lenses in the world to go with it, only to find i had to put filters on those lenses in order to render black...oh wait, that happened too! Laughing


PostPosted: Fri May 20, 2011 4:38 pm    Post subject: Reply with quote
... and it is still not FF and ISO only till 6400. I am not saying ISO 6400 is not enough though. I don't understand the pricing strategy though.


PostPosted: Fri May 20, 2011 5:20 pm    Post subject: Reply with quote
The last zero must be a mistake! Or not? Do they really want to sell an APS-C camera body at the price of Pentax 645D body? Shocked


PostPosted: Fri May 20, 2011 5:28 pm    Post subject: Reply with quote
must be foveon $
give them 10$ and tell they worth 100$


PostPosted: Fri May 20, 2011 5:49 pm    Post subject: Reply with quote
iso 6400??? the sd14 went to 3200 but was unuseable over 400! so i guess the extra $9000 for this one gets you useability to 800!


PostPosted: Fri May 20, 2011 5:59 pm    Post subject: Reply with quote
rbelyell wrote:
iso 6400??? the sd14 went to 3200 but was unuseable over 400! so i guess the extra $9000 for this one gets you useability to 800!


+1 I believe this will be last Sigma if they try to sell on this price.

There is many damn fine full frame camera from well known makers like Canon, Nikon , Sony , Leica ... Sigma sound still means crap for most people and try to sell crap for $$$ quiet impossible.


PostPosted: Fri May 20, 2011 6:56 pm    Post subject: Reply with quote
i agree with attila, though i will say i loved the IQ of my old sd14, but only in bright light when i managed to nail the focus with that horrible vf! but if everything clicked correctly, the IQ was better than my 5d when everything clicks. problem was that given the crappy hardware and horrible iso and vf, only one of twenty shots 'clicked'!

so i would be a lot more welcoming of the new sigma at $1000 aimed at correcting its hardware and iso faults. or a more expensive FF came that also addressed these. but like a $5000 leica that needs filters to render black, a $10,000 apsc cam is a nonstarter to me.


PostPosted: Sat May 21, 2011 3:39 pm    Post subject: Reply with quote
That kind of pricing for a marginal performer in the market place is the same as shooting themselves in both left feet. Sounds to me like the death knell for foveon, which is a shame. But who knows - maybe one of the big boys will snap up a failing enterprise?


patrickh


PostPosted: Sat May 21, 2011 8:36 pm    Post subject: Reply with quote
I'm not sure what's more confusing - the 46MP thing and subsequent 2 pages of argument or the fact that Sigma have just taken a decision to price out all of their remaining loyal customers Laughing Laughing

I can only imagine that the time and cost to develop and produce the new model has forced Sigma to slap such a crazy price on the camera. The specs alone (e.g. low res LCD screen) show that it's been in development for a while. It will be interesting to see how much Sony charge for the a700 replacement, as that model appears to have had a similar amount of time spent on it.


PostPosted: Wed Feb 08, 2012 4:40 pm    Post subject: Reply with quote
SD1 is rebranded and is now known as SD1 Merrill. Also, the new MSRP is US$3300.

In order to compensate those who bought SD1 (at $10,000), the company is offering SD1 owners points which they can use to exchange products.

Full info: http://www.dpreview.com/news/2012/02/08/Sigma_SD1_Merrill_price_drop


PostPosted: Thu Feb 09, 2012 1:51 pm    Post subject: Reply with quote
"Points" or not, I would be pissed off by Sigma if I had bought an SD1!!
I can understand every SD1 owner who is in a real peeve!
Post new topic   Reply to topic
Home   Official News and Announcements
Goto page Previous  1, 2
Page 2 of 2