Sigma Rumors posted a rumor saying that Sigma will make a Micro Four Thirds camera with Foveon sensor. One of my trusted sources actually told me that the rumors is at “least partly true”. There will be definitely a new Sigma interchangeable and mirrorless camera in 2014. But it’s yet not sure if it will have an MFT mount or a new mount!
I would really like the idea to get a Foveon sensor in a Micro Four Thirds camera and you?
164 Responses
“least partly true” hmmmmm…. so that could mean anything.
“at least”. Quote is in the wrong place.
Which would either mean PanOly will use a Sigma Foveon sensor, or Simga will make a Foveon sensor camera, but isn’t MFT.
I guess it means that Sigma will release a mirrorless camera but with their APS-C sensor.
Well, a µFT sensor is sort of “contained” in each APS-C sensor, so even an APS-C Sigma camera would be in a way “partially” a µFT. Well ok, the bayonet would eventually be different 😀
Frankly, I do not care about Foveon at this stage of the development; the sharpness is A+ but the hi ISO is less than mediocre. A X-trans µFT would be better 😉
I understand your concern, but I am confident that there is a demand for Foveon.
Well, your confidence is thus based on nothing more than your opinion. The facts would indicate otherwise. The Foveon experiment has only garnered marginal sales, sub-marginal reviews, and an utter lack of interest by any other camera manufacturer for that technology.
Except every major camera manufacturer has a patent for similar technology – so there is SOME interest. They just don’t want to be beholden to Sigma for it.
WWith such high price, it’s hard to sell in number, and until récently with the art lines, Sigma lenses means budget lenses. Foveon is not video, not for high speed action and not for high iso, but it’s very good at color reproduction 🙂
It means sigma will lunch a new foveon camera, but mount is unkown
Sigma has already its own mount – SA mount. Also, Sigma is primarily lens manufacturer. I very very much doubt that the cam would have a m43 mount. Their ultimate goal wouldn’t be to sell more cams – but to sell more lenses.
That would be cool to see. .. And might turn out to be an interresting camera. 🙂
Very interesting indeed. Foveon deserves another format and another chance to make it big.
Camera making is hardly within Sigma’s core competence.
I wish that Sigma is smart enough to get another company to make the n4/3 camera that will use the Foveon Sensor. A Sigma branded GX-7 or replacement model for the E-M5 would be perfect.
While at it, Sigma might as well make some really nifty f/1.8 or f/2.0 zoom variants to supplement Panasonic’s 12-35-100mm f/2.8 and the Olympus’ 12-40-150mm f/2.8.
As much as I would like Sigma to make use of an existing m4/3 camera design from Panasonic or Olympus to the recipient of the Foveon Sensor and perhaps re-brand this as a Sigma, the worst that will likely happen is that the Sigma Merrill will be modified to incorporate an interchangeable m4/3 lens mount (and possibly the Sony E lens mount as well). Halfway between the first and this worst case scenario is that Sigma will make it own m4/3 camera that will be better than the Merrill in every way.
Whatever happens, this will still be a win for Sigma and the m4/3 system. I am looking forward to this!
If it’s turns out true, it would be a dream! Beside a Ricoh gxr mFT module, a foevon sensor is on my wishlist since I had the dp1.
But I wouldn’t bet on sigma really joining mFT. Maybe they finally sell the foveon to other manufacturer? Out the start a own mount with mFT sensor size?
That would be great as well-because then they would sell the small lenses with mFT mount a well 😉
Sigma already is part of the (m)43 alliance.
Hmmmmmm…..
A Ricoh m43 module sounds very nice.
It is good that Sigma might be planning a mirrorless camera but not if it turns out to be yet another mount (that would be bad for the entire industry), we are facing the risk now of too many different systems on the market. I don’t think Sigma would be that foolish, however we have to seriously consider their existing path of lens design, whilst they do produce some lens’s for u4/3, let us not forget these are NEX designs but then we need to consider what Sonys plans for NEX are, are they committed to it or looking at A7 as their true mirrorless path. We need to look at this rumour with wide open eyes and not expect a new member into the fold that easily, wonderful if it does happen but be wary.
I will definitely pick up one if they use an MFT mount.
Hope this comes true. What a great addition to the µ4/3 format.
I don’t care about Foveon, but I’d like another serious manufacturer in the system. Moreover, if Sigma started making native m4/3 lenses, that would be great. Their last couple of lenses are excellent.
Definately would be great if they made native m43 lenses. I hope that happens.
Please Please Please Sigma ! I need This for my M43 Lens and complementary of my Em-1 , could be very useful for landscape and architecture work… those field are the weakness of M43… with that foveon camera no more need to get FF.
I’ve looked at the Foveon cameras that Sigma has out. And I concluded that they were worthless unless/until Sigma shares the tech-details with Adobe so they can be utilized in Lightroom, Adobe Elements, and Adobe Photoshop. Using their own branded RAW processor and then transporting pictures to other processors is to clunky to be dealt with. OTOH, if Sigma _were_ to merge with the 21st Century, a m43 Foveon camera would, indeed, be a very nice thing to have.
Great! Really hope Sigma does make a m43 camera. Or that Olympus or Panasonic will use the Fovon sensor in a m43 camera. And fix the high ISO color issue of the sensor. Black and white photos are very good though.
I have a Sigma dp2 Merrill with Foveon sensor, its image quality is outstanding at ISO 100-200. Check this nice review of the camera
http://www.luminous-landscape.com/reviews/cameras/sigma_dp2m_review.shtml
A m43 with Foveon sensor will be excellent for landscape photography.
Partly true could mean :
1/ Fixed lens, regular 4/3 sized sensor
2/ Fixed lens, foveon 4/3 sized sensor
3/ Fixed lens, foveon whatever sized sensor
4/ µ4/3 mount, regular 4/3 sized sensor
5/ µ4/3 mount, foveon 4/3 sized sensor
6/ µ4/3 mount, foveon whatever sized sensor
The two options making the most sense for Sigma from my point of view are 2/ and 5/.
2/ would be a new camera in the DP line, this time with a zoom lense, the 4/3 sensor format would help designing a more compact zoom without sacrificing IQ too much and would help recoup the R&D costs by reusing the design as a µ4/3 lens.
It could also be a very compact DP camera, but I somehow doubt it.
5/ would be a new line to use with the already released µ4/3 lenses, which could mean more µ4/3 lenses, this time truly built for the format, it’s the least likely of the two I’d say, but Sigma has proved they are not afraid of going with very niche products and supporting them even when other companies would have called it quit.
Going with a new mount would be insane, but then again, it’s Sigma.
Let’s not forget Sigma is a member of the original Four Thirds consortium. It would make sense them to make a M4/3 camera. If true, and the poor high iso is fixed, that would be a great addition to the system.
Whether it will be a m43 mount or just a 43 foveon sensor, what’s mandatory (to me) is a way better performance at medium ISOs. Right now only base ISOs are good (great), and I haven’t considered to buy one for landscaping simply because of its limited performances.
I really would like to see it.
I don’t I’d buy such a body, but I hope it becomes true. One more heavy contender for the m4/3 system can only be a good thing. Specially one that that besides bodies can also bring very good lenses, along the ones they already have.
Really hope this comes to be true.
Foveon M4/3 would be an IQ killer machine. Image quality that is slightly below FF sensors and a company that can produce quality glass. And there is other good quality glass in the market! And a Japanese company who thinks different from the stogy likes of Cannon and Nikon! And Sigma has upped their game by improving quality too! In a small footprint that is in a light and easy to carry package. Brilliant if and only if Sigma solves the slow processing the DP Merrell series has.
Currently shooting D7000, Nikon 1V1, 4×5 sheet film Lumix G3 and a ……..DP2 Merrill……….
Look out fanboys there is a new sheriff in town!
Foveon IQ is good in only one respect, resolution. In all other regards the current lot of Foveon sensors struggle to match the IQ that was common in APS-C cameras in the mid-2K’s.
Foveon gives you two things, good resolution and no colour aliasing. The cost is horrible noise performance, lousy colour accuracy (there’s massive overlap in the sub-pixel’s colour sensitivity), extremely poor DR and high processing costs (too much data for a given amount of performance).
Note that since the current Foveon sensor just about matches a 24MP Bayer sensor in resolution, you can get far better all-round performance with no tradeoffs just by buying a D7100 or K-3.
The one thing that appeals to me about the Foveon cameras is how they do green leaves. I’ve never see greens look so good in any other camera. The Fuji’s have the opposite problem, at least according to DPR (great high iso, smudgy greens).
FCUK lab colour accuracy!
Get a Sigma out on a sunny day and it will smash any bayer sensor with all its colour nuances… Bayer might be accurate but its flat and nuance blind…
Who cares about your cynical opinion anyway? Go and be happy with D7100 then…
Colour nuances? You mean all the edge cases where the Sigma simply blows the colour because there’s too much cross-talk between the colour channels?
The Sigma’s do pleasing colour when said colour has strong components in R, G or B. It’s absolutely craptacular when you get colours which ride between R/G or G/B transitions. The term for this is metamarism, and Foveon is utterly crippled by it. It’s inherent to the technology (you cannot have good rolloff on diffusion filters, so you are either going to waste a lot of light or have lousy metamarism, and Foveon can’t afford to waste light)
If you want really good colour, get one of the high-end Sony FF’s (A99, A7 A7r) or a Nikon D3X or D800. Either option when paired with good glass simply destroys the Foveon’s in terms of colour. And the D800 or A7r will utterly stomp the Foveon’s in every regard including resolution.
What about the fact that 2/3 or so of colour in a Bayer sensor is made up? That is a lot worse than bad rolloff…
Sure it might perform accurately in a lab, but it is flat, dull and lifeless in the real world where it matters…
And that is why you need Photoshop so badly… To make those “accurate” colours somehow pleasing again…
Have you ever owned a Sigma by the way?
If the Foveon sensor catches 100% of blue, red and green as the graph on the top of the page says, why is the High ISO results so bad?
Because these are stacked layers with some light loss and that becomes a problem at lower light. Higher sensitivity makes more noise.
plus it would consume even more power
The two main reasons are the large spectral overlap of the different layers and that they top two layers have a relatively low ‘QE’ (where letting, eg, some ‘blue’ light through the top layer lowers the number of photons ‘counted’). This results in:
1) In general, the blue layer lets some ‘blue’ light through and absorbs some ‘red’ and ‘green’ light and the red layer lets some ‘red’ (and probably ‘blue’) light through. So, for once the first two layers capture less ‘blue’ and ‘red’ light respectively compared to a Bayer sensor and secondly the two bottom layer receive less ‘red’ and ‘green’ light than in a Bayer sensor (this could alternatively also be described as spectral sensitivity curves that have lower peaks than in a Bayer sensor, aka, some light of layer’s colour passing through, and that don’t go down to zero for a wide range).
1b) And possibly even when neglecting that some of the ‘blue’ light is passing through the blue layer, the blue layer has a low QE and a higher read noise (I expect that getting moving to translucent light sensitive layer results in a worse performance than a normal, non-translucent light sensitive layer in a standard pixel).
2) After starting with a higher noise level in each recorded channel, because of the overlapping spectral absorption and the partial translucency, to convert the green signal to something a human would perceive that wavelength range and correspondingly for red and blue, some moderately complex math has to be performed and even with that it doesn’t get a very good match, ie, bad colour accuracy. Adding these calculations, the noise in input signal gets amplified (think of subtracting to somewhat similar numbers from each other when each number is noisy, the noise to signal ratio on that smaller delta value will be noticeably higher).
If you could create a sensor that:
a) has spectral sensitivity curves for each layer very close to that of the human eye, and
b) has similar QE as existing Bayer pixels,
then a Foveon-type sensor would have a better performance than a Bayer sensor (about as much better as a pure B&W sensor is).
The problem is that even if b) is let’s say half full-filled (which should give the sensor a small advantage over Bayer sensors), and about the same for all three layers (which is tough, some green light will always be lost in that it not arrives at the last layer, you would need to make the top layers 100% transparent for the wavelengths of the lower layer(s) ),
full-filling a) is equally tough because you would need to make the top layers completely insensitive to the wavelength of the layers below them.
if they create yet another mount theyre crazy
so the question is
‘which mirrorless format is best placed for Sigma to adopt?’
Answer that would sensibly be NEX, more in keeping with the DP format and also the lens’s used for u4/3 are originally designed for NEX, only adapted for u4/3, so the logical option is NEX. However it is Sigma and they might just see u4/3 as the better option being considerably more mature with some top quality glass already there to be used.
The NEX system is not an open system like m4/3, so to make cameras in that system, they’d have to get Sony’s permission. With that mount developed to fit FF sensor now, it’s a lot more complicated.
I think you’re wrong, at least the specification is open, unlike Canikon.
NEX’s E-mount is only open to lens manufacturers, not camera manufacturers. IOW, to Sigma, who is already member of 43 consortium, the m43 is more open if they want to build a cam.
Otherwise, I think Sony wouldn’t mind Sigma building an E-mount camera, since it is very doubtful that Sigma would pose much competitive threat to Sony’s own cameras. And most Sigma lenses are APS-C anyway. APS-C would make a lot more sense to them. Even more: NEX might get even more lenses from Sigma.
There is of course also the Samsung NX, but since Samsung is Korean and Sigma is Japanese, no way anything would happen there.
PENTAX did work with Samsung, using their sensors and putting their label on lenses/cams
Neither m43 nor E-mount are open. Olympus or Sony decide who gets the spec or doesn’t
Would be amazing for landscapes, light and small, yet amazing performance. Only hope they go with something more close to 16:9
Several Olympus 16mp mft cameras can shoot at 16:9 and 1:1 at 12mp or 3:2 at 14mp. It’s better to have a camera that can shoot different aspect ratios because you can compose better.
http://digital-photography-school.com/aspect-ratio-what-it-is-and-why-it-matters
http://blog.mingthein.com/2012/07/31/aspect-ratios-and-compositional-theory/
I mean native resolution of course, I can crop in post processing too.
Everybody can crop in post processing but the results are not as good as shooting from the begining and seeing in the EVF/LCD screen what the picture will look like from a composition point of view (shooting raw will let you see and use even the uncropped version). Composition is the base of the photographic technique.
For me doesn’t really matter and I’m not restricting myself to any particular size. All I’m saying that for landscapes would be better to have more pixel for width, than height, since anyway most of the time it’s getting cropped out. Best solution was multiaspect ratio sensor of course, something like GH2
Rules are meant to be broken, I have gone back to post-processing cropping. Eg, BIF can make all the difference if you stick with 43, as you don’t necessarily have the time to compose exactly in the VF
Would be amazing for shooting artworks.
Makes sense to put already available APS-C sensors in it. A lot of room to crop from for PanaOly lenses, and Sigma’s lenses will use full image circle.
Sigma..Please make it happen.
And some of that lens goodness that you have been dishing out to the APS-C cameras. Give us some of that love.
And turn that 18-35mm into its equivalent 12-50mm f1.8
sure, with 200 g and for 100 bucks? If someone could imagine how huge and expensive a 4x f/1.8 zoom for µFT would be, he’d never ask for it 😉
12-35 f1.8 then?
Surely, it can’t be bigger than the 40-150 f2.8 from Olympus.
Even a 12-35 f/1.8 would be huge; think of a Pana 12-35 f/2.8 and imagine that the front lens for constant f/1.8 aperture should be that big, that it can capture 4/3 EV more light (that is more than 2 times more, which translates to an at least 40% larger diameter) than the Panny. So even that would be a chunky, heavy and expensive lens. a 2x zoom should be easier to construct but even that one would be quite large for µFT standarts
why not just get the Olympus 14-35mm F2? ok not 12mm or F1.8 but very close…
I have the Panasonic 12-35 f2.8. But f1.8 is f1.8.
The 14-35 is in fact optically f1.4, it’s constrained to f2 by firmware (this is easily testable, if the lens is mounted to a dumb adapter it will meter 2/3rds of a stop faster than f2, the other 1/3 of a stop is the difference between the f1.4 optical speed and the T1.6 effective speed).
So a 14-25/1.8 (the actual equivalent to the 18-35) would be smaller & lighter than the 14-35.
be nice if they did a fw update to get the f1.4 even if there might well be IQ issues… let the consumer decide what IQ they want!
rather odd move by olly?
I would love to have 14-25mm f/1.8
At least, Sigma please make lenses for m4/3 system, especially the Art series. And will be nice if they bring out the Foveon sensor for m4/3 and use the focus system from either Panasonic or Olympus. It will make the m4/3 stronger!
The “partially true” does not correspond with the very clear language in the source:
“We’re told Sigma may be developing a Micro Four Thirds, mirrorless interchangeable lens DP camera with Foveon sensor”
That does not sound partial, it sound like they will use MFT sensor and mount.
Yes please! 90% of my shooting is landscapes at base ISO, this would be perfect. Given how cheap used Panny/Olympus bodies are I’d be happy to keep a second body for occasional high-ISO work.
What Sigma should really do:
Make a APSC -sized Foveon camera with interchangeable mount adapters.
Then we can attach all kinds of Fuji/Nex/Mft/Legacy Lenses to the Foveon camera, depending on the fitted mount adapter. Kind of like the Ricoh GXR system, but the sensor is built into & remains in the camera.
Would be the optimal solution, currently the foveon cameras are useless due to the camera part and the MFT cameras due to the sensor.
mft is useless because of the sensor? LOL
right, if you want to compare penis size
Every new MFT camera is a good camera for the system as such, and with Sigma it would mean more potentially interesting lenses, too. Which is what I would welcome most. Personally, I don’t need much more than the existing 16 MP Olympus sensor, even though I see why many people find the Foveon technology attractive.
It will probably be APS-C, because then they can easily make versions of their lenses for it for Sony, Fuji, Samnsung, Canon, and MFT.
So it will mean more and cheaper lenses for MFT, but they won’t be as compact as those by Olympus and Panasonic.
First of all…I don’t believe (although I would love to see Sigma make an MFT Camera) that Sigma will make an MFT Camera, as they have not produced one native MFT lens. Their lenses have all be adapted APSC lenses.
Secondly, the Foveon Sensor, while exciting, (I would love to see that become a full-range sensor)…. is extremely limited regarding ISO range.
If Sigma has had a breakthrough with the sensor, it makes no sense that the company would introduce it in a new format other than APSC as they have already produced many APSC lenses (they are a lens company after all).
I would love to see Sigma launch the incredible Foveon MFT camera that comes to mind in my head when reading this rumor…but my gut tells me that this is VERY unlikely. 🙁
…but…YA NEVER KNOW!
I wonder if this means Sigma will introduce a 1″ sensor with a m4/3 mount, similar to the Kodak branded camera.
Isn’t it a Polaroid instead? http://www.43rumors.com/we-have-a-new-mft-camera-polaroid-micro-four-thirds-camera-now-at-amazon-us/
It’s not instead. I was thinking of the Kodak one, http://www.43rumors.com/kodak-mft-camera-release-postponed-to-next-year/. Forgot about the Polaroid one.
Where does it say the Kodak will have a 1″ sensor?
This is about Sigma. Way to focus on what doesn’t matter.
Yeah, I though so too: What does this have to do with the Kodak S1?
This is about what Sigma may do and you chose to focus on whether it should be polaroid or kodak? quite sad.
Ì wouldnt be surprise if they do m43 mount…better then create own mount and then had to start build lenses from scratch
m43 has most of the native lenses from ILC producers available …so its natural they consider m43
but I think they will go APSC … because their target audience would be those who want “bigger senzor”
Bull….Sigma should make a FF Foveon if they want to make some money.
This could be nice 🙂
I agree with a few earlier posters, my guess is that sigma will release a mirrorless camera, but it will probably be aps-c and use their sigma mount. I’d love if they joined m4/3 though- a new manufacturer would really solidify the future of the format.
Best part- that would mean native sized sigma m4/3 lenses. They’re putting out nice products for reasonable prices right now, so the competition would be great!
Another aps-c player: no thanks (there are enough on the market)!
A 43 size sensor with a new mount: non one will buy it because of lacking of lenses choice.
A 43 size sensor with a m43: make sense.
m43 mount (i forgot a word)
Interesting prospect. I recall reading how the Foveon sensor rivals anything out there in FF for image quality. I believe some even comparing to medium format. However, at higher ISO it deminishes quickly. While it would be cool to see them go with m43, I feel the format is not ready to take a step or two backward in ISO performance to make a leap forward in the future. People scream image quality all the time, but what they are really saying is they want noiseless high ISO capability….
I am not sure if the desire for always-higher ISO really interests all that many photographers. The latest arms race among camera manufacturers is ISO, just as it was megapixels 10 years. Joe and Buffy Suburban may be interested in high ISO for the kid at the football game, but many photographers who take landscapes, architecture, and commercial work do fine with ISO below 10,000 or even 1,600.
Oh, I agree completely! I never shoot above 1600, if I can help it. I actually prefer not to go above 400. However, “Joe and Buffy Suburban” are exaclty the people m43 has to cater to in order for this format to survive. They have failed miserably at that. The lack of viewfinders, speed and responsiveness lasted too long, not to mention poor advertising. They need to get the word our that these cameras are wolves in sheep’s clothing. Finally, these cameras can keep up and in some cases exceed the competion (Canikon). I held on for a long time, and it was frustrating for some of my needs, but it has finally paid off.
Some of us never shoot above ISO 400.
When I shot film it was usually ISO 50 and if I was feeling daring I would use ISO 100.
I agree. I did most of my work with Kodachrome 25. But I was not a sports photographer, so high speed work was not my activity. And I still use Panatomic-X at ISO 20, developed in Rodinol.
Interesting, if
Sigma can get rid of the noise problems with higher ISO and
with mFT mount only – NO new system!
Wow! this is excelent news if true!
If sigma do this m43 will be very established as a competative force to the establishment – many factors would help establish this.
slightly cheaper and easyer to produce sensors… being smaller than aps-c etc.
4 companys making cameras with m43 mount.
Established market for adaption to other formats (if not electricly yet)
Larger lens range spanning several companys
this is headding for the closest we have ever had to an open standard… although its not open! but still better than a mount per manufacture
“I would really like the idea to get a Foveon sensor in a Micro Four Thirds camera and you?”
Depends on the price really. Since we haven’t heard any news about Foveon technological advancements, I have to assume it is the same sensor as in the older models. Frankly, I can’t see me paying much for a camera with it. 400-500€ may be. More than that – very doubtful.
But it would be very interesting to pair the cam with say 1.8/45mm or 2.0/50mm and see the portraiture capabilities.
OK, I admit I am intrigued.
Still, Sigma are fools. They’ve waited to jump into the mainstream with foveon until its irrelevant. Sensors have improved to the point that there is very little advantage to this technology. They had a window over the past 5 years to make massive inroads simply by licensing the tech out and getting royalties or selling sensors to 3rd parties.
>I would really like the idea to get a Foveon sensor in a Micro Four Thirds camera and you?
Yes! It’s going to be tough to choose which body to go with, with all these great choices µ43 is amassing, but it will make the choice of the system easier. 🙂
Talking of universal mount m4/3 is the only one, that can do it, if you think. Any APS is proprietary, m4/3 isn’t, in practice.
This in the past was the greatness of M42 and K mount in the past. For Small MFR it is the only way to have a share of the pie. I think that Leica should do it too, but then there is pride of the brand.
Sigma would be the perfect match. I use daily Sigmas and they are there with the best for sharpness, colour signature, and other optical virtues.
Sigma of course will make some market research. Sony however has all but killed NEX, I don’t see Fuji owners, those silly snobs, taking an interest in Sigmas, Sigma’s own mount is dead – so there is hope.
Keep in mind that for landscape, m4/3 offers a much better ratio size/distance to flange and thus evenness across the frame. Sigma lenses are so good because they are telecentric.
I might be wrong about the death of NEX, but it’s really a matter of image: Alphas have obscured it to the point that at DPR’s NEX owners are crying rape 🙂
I can already see the ad with “bragging rights”
This is the highest resolution mft camera EVER!! 45MP!!!! LOL
What about m43 mount but bigger Sensor? They already have several nice lenses for APS-C sized sensors, so maybe that makes sense too. Their range of fixed lenses for m43 is wasting alot of possibilities.
So basically what u get is a “larger than m43” sensor and with Sigma lenses it runs at full potential, but with m43 lenses u can run kind of a crop-mode, which would be still freaking great in terms of resolution.
It may not work because of the diameter of the opening in the m4/3 mount. A long time ago, m4/3 probably should have chosen a larger mount just in case it wanted to scale up the senor size.
The GH1 had a slightly larger sensor. The issues is that the standard specifies the size of the image circle cast by the lens. Even if a larger sensor would fit, current lenses (designed according to the standard) would not necessarily cast a large enough image circle to cover them.
Yes exactly. But the question is, do the Sigma APS-C designed m43 lenses cast a bigger cycle inside the body, so that there would actually fit a APS-C sized sensor (or close-to) inside?
And all other lenses could be used with crop-mode and therefore lower resolution.
Yeah, that’s what would likely happen, like the way Nikon handles its FX and DX lines. One other option while staying within the standard would be to make a multi-format sensor which could cover different aspect ratios within the four thirds image circle. Any larger sensor would enable this feature, even for existing lenses. And the standard allows for it. And 4:3 really isn’t the most area-efficient format in this system.
I don’t relish the fragmentation a larger image circle might bring to the system. My feeling is that it really wouldn’t be worth it to just move up to APS-C if it would complicate the ecosystem with different formats. APS-C is really not a significantly larger format (especially Canon’s size) and current APS-C sensors, aside from Nikons, aren’t really significantly better. And Nikon’s sensors are better right now because they are currently using newer technology, and not primarily because of their size.
In my opinion its not the size advantage that matters, its the already available APS-C sized Foveon sensor that would make his way into the new camera. So Sigma has basically no additional R&D costs. If they would need to crop the sensor to 4/3 size, that would mean a lot of costs. Even they are not obvious in the first place.
So the best solution would be, to simply take the existing sensor and all work is done.
Yes, that’s true. Interestingly, their interpretation of “APS-C” makes that rather easy. Their sensor is almost identical in height to the four thirds one, so they could simply crop it in software and call it a multi-aspect sensor, capable of supporting several aspect ratios.
Actually, the difference between four thirds and any APS-C isn’t huge. But Sigma’s is the closest.
I’m pretty excited about the possibility of having this sensor available for use with some of the incredible Micro Four Thirds glass I’ve come to love.
From what I’ve seen, the big benefit of this sensor is the incredible dynamic range, color reproduction and sharpness, especially at base ISO. So far, I don’t believe that there are any expectations that it will deliver superior high-ISO performance, but maybe some day.
Sigma DP Merrill battery life is horrendous. Battery life is about 100 or less (real world shooting). Sigma includes two batteries in their retail package because of this.
To keep speculating, the m4/3 advantage is not only common lenses across brands, but also to share common bodies.
So Sigma could sell lenses AND bodies separately, or match them, to the same m4/3 consumers.
IMHO because of Sony’s move, APS mirrorless sales, starting from NEX are due to crumble. The only viable alternative is m4/3.
I would watch sales in Japan in a few months to see how Sony is faring, compared to m4/3. You might be surprised, because I read that there buying of the As is v. cautious.
Clearly they are no headless chickens, and wonder about the future 🙂
For tripod-based work, I can see this as a major plus to the system.
lets hope is not M43 with thats msaller sensor whats the point is better with biggest sensor possible M43 already show that is fail and soon will die so why is pointless apsc ok FF even better
Thanks for letting us know that you use the entire frame of your 4/3 sensor. And not 3:2, or 16:9, or 1:1. Full Frame! I love it. 🙂
with terrible aspect ratio
amalric all native M43 lenses relay in software correction basically all are terrible
As usual, you have no idea what you are talking about.
No. You are wrong.
You’re more than wrong. Many full frame lenses rely on software correction because of the significant peripheral illumination problems that exist with large sensors. Designing lenses without these problems for full frame is extremely costly and the resulting lenses end up being much larger than would generally be considered acceptable.
Micro four thirds has a huge advantage if you want to design a well corrected lens because such a lens would not be unacceptably large. While many small lenses, mainly pancakes, do rely on correction, high grade lenses like the 75mm f/1.8, the Leica 25mm f/1.4, the constant f/2.8 zooms from Olympus and Panasonic and the Voigtlander f/0.95 25mm and 17mm do extremely well without it. Even the 12mm f/2.0, renowned for it’s incredible sharpness, is only corrected for distortion, and you simply can’t design a 24mm equivalent for full frame with that kind of performance, and certainly not at a reasonable size and weight.
I usually leave peripheral illumination correction off on practically all my glass. It’s really not needed. And Olympus didn’t even implement CA correction until the most recent generation of cameras.
I have been researching the DP Merrills for many months and am seriously considering the DP1M. Those who use these cameras are aware of their shortcomings and there are a few that need to be considered:
1. Battery life in a DP Merrill is limited to an average of around 70 shots. Sigma is aware of this and each kit comes with 2 batteries. The batteries are small, but the processing of the RAW files is what seems to drain the batteries quickly. An average file size is 50MB for each RAW file.
2. Write times are long and slow. This also seems to be a limitation with writing such huge files. I am certain that with current hardware, write times could be reduced, but they will never match a late model mirrorless camera.
3. Sigma software must be used to convert the RAW files. There is one 3rd party program that can also read the files, but the end results are not the same. Adobe and Apple have not included any functionality to decode Sigma RAW files and therefore Sigma software must be used to export the files as TIFFs in the first instance. This adds another level in the PP workflow.
4. The best results from a Foveon sensor are in the low ISO range, preferably at 100 or 200. The latest Sigma software can convert files to monochrome and the results are amazing; showing wonderful tones. Significantly higher ISO settings can be used for monochrome, but with some limitations. Due to the fact that the blue sensor layer is on top, this layer receives the most light and signals. Therefore, to use higher ISO a large percentage of blue tones must be used in monochrome conversions to achieve good results.
5. I am no expert, but I believe there is a reason that Sigma cameras do not have in body stabilisation or lens stabilisation. I believe the sharpness created by the foveon sensor would be degraded by adding any form of stabilisation.
If one understands the limitations and can work with them, the results are wonderful. If I was to purchase a SIgma merrill, it would be a 2nd camera specifically for landscapes and possibly some experimentation with monochrome street photography. For all other genres,I consider other current technologies superior in operation and speed.
This was a nice summary of Foveon sensors. Thanks for writing it down. I stayed away from Merrils mainly because of the PP extra work, limited use at low ISOs and huge files. I think Adobe could work on Sigma’s RAW files, and ISO performance could be increased too.
These two things would turn a new Foveon sensor camera interesting enough to me, at least when it comes to landscape photography.
If they get clean ISO 400 and good ISO 800 then its good for me… If its close to the EP1 quality for hi ISO then it will be very usable!
Still not sure whether the Foveon technology is the really promising one for future sensors or not. The Panasonic’s splitting beam technology is better in all respects but resolution and demosaicing artifacts.
yeah right are talking of microwaves technology?
No. I mean vacuum cleaner technology for sucking out your brain.
Agree – beam splitting is less lossy than the layer absorption…
Have any of the panasonic cameras started using this tech yet?
And just can’t wait for brain vaccume 🙂
i would be happy if they could add tethering and wifi to the dp merrill line..im so bored to have such a beauty dp3 merrill to work with in studio and any tethering !!!
kazuto san..add tethering to dp3 !!!! please 🙂
“It’s not the size, it’s the technology of the sensor, stupid”. Why does it remind me of something? 🙂
size matter man that cant hold a camera is your problem
In your case your are very transparently speaking of your dick’s size.
I propose this for the Dictionary of the Devil:
“Sensor: the bigger it is the easier it will be to wipe your floors with”.
Start now!
sounds too good to be ture
if ture it would mean more sigma mft lens ?
would sigma be able to make a speedboster like adapter for there own mount to mft ?
With over 75,000 followers on Instagram, and as a Micro Four Thirds Guru, I would say that I bought into the right format. I love it. If SIGMA develops a M43 Camera with Foveon sensor, I will forever love there company! Sigma, you rule the world. Check my instagram for examples of my photography in the website field! 🙂
hipstagram please share on realwebsites not hipster things
ypu really are very lame m43 guru ahahahahahhaahha hipster wannabe thats what you are
Hello trollface, trollster, troller, etc… lol I’ve got work on exhibit in Santa Fe. The only thing hipster here is you trolling anonymously on a legit forum. Now buzz off and quit being obnoxious.
no troll you are lame what instagram exibition please your call your self M43 guru Lol u really have attention problems what olympus give a camera strap or key chain ?
Why did you feel the need to point out your numbers? 🙂
I feel mixed about this. The perfect Sigma M43 camera would actually be made by Olympus or Panasonic and take only the Foveon Sensor from Sigma because that’s the best part. If you look at Sigma’s past “mirrorless” cameras (i.e. the DP series), they all have very very slow focus, very very slow user interface, takes >10 seconds to write a single raw file to an SD card, and have extremely poor battery life (less than 25 shots per charge according to some users). Build quality is also extremely inconsistent. Basically everything outside of the sensor and imaging processing sucks.
most of that is hearsay. The battery life, yeah, but 25 is really pushing it. 40 on a bad day is closer. As for the focus speed, well the DP2M focusses quite fast enough for me. And the build quality is superb. The slow write time is only an issue if you need to chimp quickly, or want to shoot a long rapid burst. Otherwise the buffer is large enough so that it isn’t really a big issue in practice. And the look and detail of the output, well, wow.
I would probably buy a sigma body for landscape and architecture photography if the price is right. I wouldn’t mind the low iso or the battery performance but the camera should be more responsive then the current Sigma camera’s
Like others suggest, the sensor and the body would need dramatic improvement for it to be a compelling option, otherwise would be just a cropped Merrill.
I think Sigma could easily figure this out since they are a sensor manufacturer…
I would love to see this and I think it would be a smart move by Sigma. Lots of folks have an investment in MFT lenses, creating a much bigger potential market than they would have if they created a new mount.
I’ll be purchasing a Sigma body if they make a m43 camera using a foveon sensor. As I owned a DP2 and it was a great camera. This is a smart move by Sigma!
just check your hisptragram i have to say man is terrible all with lots of filters with the tipical hipster style no any picture in low light all pictures withh so much PP that look terrible really go drink PRB mate
Big thumbs up to this idea,
A: I’ve seen foveon results , they are fantastic
B: This could mean native-sized sigma lenses as well as the dual-use ones
we’ve been seeing.
Seems that they would want to build a system to sell their own lenses. My bet would be on their own mount and probably aps-c.
foveon is fantastic sensors. DP merill are very good !!
Sigma makes money with lenses, not with cameras.
I doubt that Sigma sees more profit in selling lenses in a mFT system than seelling lenses in a APS-C system + selling to other APS-C systems (+selling to mFT)
It’s also trying to broaden the market for its sensor technology, which, if successful, would make Sigma a very important player in the future of digital photography. By increasing the market for its X3 sensor, it can help get it into the hands of more photographers who will no doubt recognize its benefits.
Micro Four Thrids seems like a fantastic way to do this. It’s an emerging system with some amazing lenses, used by many early adopters who like to experiment with new technologies. They clearly have a stake in the camera market already, and this seems like a great way to diversify that market and bring some new customers into it.
And they will no doubt be releasing more lenses for the system. As a lens maker, micro four thirds is too big a market to ignore, for lenses too. 🙂
M43 rather than APS-C to be paired with the Foveon sensor in a mirrorless makes perfect sense technologically. With its stacked architecture, the Foveon sensor seems likely to be even more sensitive to the inclination angle of incident light than Bayer-pattern sensors with their fotodiodes in a single plane. Given the short flange distances of mirrorless systems, comparatively large inclination angles at the borders of large image circles are unavoidable, unless one wants to accept unnecessarily large telecentric lens designs.
Sigma make the slowest cams.
This is an interesting rumour, and should it materialize, I would look very closely at a micro four thirds mount camera with a Foveon sensor in it. A large part of my shooting is landscapes at base ISO, and I would likely treat such a camera as a landscape/architectural tool. As such, slow autofocus, poor low-light performance, and other typical quirks of Sigma’s Merrils are not a deal breaker. I have other cameras that do well at higher ISOs, that take great video, etc, but just as I have a dedicated infrared camera, I would consider this as a dedicated landscape camera.
A lot of it would come down to value for dollar. I suspect that, being a bit of a niche product, the price point would not come down to where I need it to be for a specialized camera. Still, I would explore the product thoroughly and seriously consider it as a part of a larger system.
Well i have extensively compared many, many landscpae shots done with the DP2merrill and it is clear that clouds look really funny many times. this ssems to be due to people trying to pull detail back out of blown highlights. I rarely came across a nice, natural looking sky. Colours can be strange too (reds especially). Of course the foveon APS-c sensor is incredible in other ways. Really close to MF format. But Dynamic Range seems to be soarly lacking…
Also: Anything above ISO400 is useless.
Also: battery life is crazy. 50 shots at most out of one battery…AF is slow. Really really slow. And looking at your files life takes 10 seconds easily…
Note that one of its strenghts is the combination with a stellar lens (DP2m). Not easy to duplicate on a mirrorless cam.
A m43 foveon sensor is very nice as long as it is placed in a cam made by Panasonic or Oly. And also only, at least for me, wehn the sensor gets dynamic range like the OMDs, GX7 and Gh3. High ISO should be at least like the one in GX1 or G5.
I hope to god Sigma does not sully and besmirch the greatness that is Foveon by making a version of it in the horrible MFT mount version.
Some interesting arguments above, including the stacked sensor. As a counter to larger sensors, m4/3 can rely on increasing processing speeds, so perhaps some limits to Foveon might be lifted now.
A well known limit to m4/3 was noise, but be saw it solved by in camera denoising at faster processing speeds.
The latter also allowed in camera HDR which was once deemed impossible. So having a small sensor can be an advantage, as mentioned above.
OTH having a dedicated m4/3 body for landscape with Foveon would be a far less expensive proposition than getting a Sony Ar, and one of its coming big expensive wide angles.
For Landscape I am also interested in seeing if a prime UWA can be made without FW correction spoiling the corners, and Sigma could give a hand there too, Foveon or not.
The NR time depends on number of pixels and bit depth, not sensor size! Merill sensor has 48M photosites, Sony’s FF have 36M and 24M. And Foveon requires much more complex NR & getting color, than bayer sensors.
Sure, smaller sensor have advantages – they are cheaper. Yet, the E-M1 costs almost as much as the A7R. And SEL35F28 is not big at all
Perhaps because the Sony A are among the slowest dogs, while the E-M1 is one of the fastest reacting cameras around? LOL.
BTW it is interesting to see the reactions of the DPR’s Sigma Forum. They are evenly split between those for and those against, but it emerges that it would be v. easy for Sigma to implement it, with no great loss of resolution and a much greater access to market for its Foveon.
There is also a small corroboration from a Sigma representative – so there is hope. Sigma was also part of the old 4/3 consortium, and even today its m4/3 lenses have lightning fast AF.
I may be wrong on that but i thought that noise was related to how packed photosites are… The more they’re packed, the harder it is to dissipate heat correctly hence causing more noise (that’s why high ISO is tougher with small high density sensor).
At least it’s on par with http://en.wikipedia.org/wiki/Image_sensor :
“For sensors of comparable types, the signal-to-noise ratio and dynamic range improve as the size increases.”
“And SEL35F28 is not big at all”
And its optically quite nice if you only care about the center 10% of the image…
I thought the normal Bayer sensors are far more complex to code colour from… The sigma design is relatively staright forwards – a RGB value for each pixel… Bayer is a mess of interpolation from adjoining pixels…
Not really, since each of the channels in the Foveon receives some of the other colors too. There’s a fair amount of cross talk, with the top layer receiving all colors, the blue penetrating a little into the green layer which would also pick up some red. And the green penetrating into the red layer a little. There’s also significant attenuation of the red and green signals which have to be cleaned up and will be much noisier as a result.
Another commenter called it the “one ISO sensor”. While incredible results are possible at base ISO, when you start applying gain to this arrangement, the red channel especially, gets very noisy very fast. You either gain a lot of noise or lose a lot of saturation even in the first stop above the base ISO.
On the other hand, Bayer filters are complicated to process *well*, but to get an acceptable quality image with balanced colors is really pretty straightforward. You’re getting a fairly strong signal from all the colors and the amount of cross-talk is pretty minimal, even at moderate ISOs.
The Foveon has big advantages in detail rendering at base ISO because of all the stuff you have to do to make Bayer sensors look really good, but getting good color out of the Foveon, especially as you turn up the gain, is really really hard.
That’s what I’ve been asking for a couple of years. “Sigma please make a Foveon MFT camera. And they’ve got some lenses for it already, so much easier to jump on the bandwagon .
I would like to see a monochrome MFT camera
Despite the ISO of Foveon is out of the game, it has a perfect (read “natural”) color representation and the noise similar to that of the film with an equivalent ISO. So I clearly see a Foveon+M43 as a substitute of film (for landscape) as a second camera.
Offer to admin, would you make a POLL about how many readers would like to buy a Sigma M43 Foveon camera (for a reasonable price) ? You can take royalty for Sigma for this study then :).
” You can take royalty for Sigma for this study then 🙂 .”
from Sigma