Throwback Thursday: Sigma SD1

https://www.dpreview.com/forums/post/60217597

I just shot some high ISO photos today with my SD1 Merrill. I was actually surprised how reasonable they looked. Here's an ISO 3200 shot processed from raw:

Oct 8, 2017permalink

https://www.dpreview.com/forums/post/60217196

Here's a whole thread I posted, showing some samples from my SD1 Merrill at ISO settings from 100-6400:

Oct 10, 2017permalink

dprived prev

Nov 22, 2017*permalink

Dougbm_2

'Announced in 2010' but came out in 2012.
It's not THAT old.

Oct 8, 2017permalink

It's older than Nikon's D800.

Oct 8, 2017permalink

mgblack74

And people complain about Nikon's 2 month D850 backorder.... :-)

Nov 1, 2017permalink

Dougbm_2

I had one and loved the acuity or 'natural' detail. But sometimes the level of detail was too distracting. Colours (especially the level of natural saturation - like film - there I said it!) were often impressive but occasionally were not quite right and there was no way to adjust colour temperature in the camera or in SPP with is a major failing. The kit 17-50 lens was outperformed by the sensor and was soft at the edges. As result my old 5D MK1 with a 70-200 IS L lens out performed the SD1 with the kit lens. Ok I could have/should have bought a decent Art lens but the colour temp thing was the main reason I sold it. I had no problem with af and the wait time for processing didn't affect shot to shot times at all, until you filled the buffer and then you may have to wait 2 minutes. The other factor that caused me to sell was the very average viewfinder (narrow eyepoint). No better than an entry level DSLR. Disappointing. Great image quality. Disappointing flaws.

Oct 8, 2017*permalink

You can change between about six different presets for color temperature, and then there is the custom color temperature capability that camera has. The 17-50mm f2.8 EX OS lens is a spectacular lens. If that's the lens you're talking about, you must have missed out on the fact that it has a severe field curvature, which makes it perfect for shooting interior shots. For landscapes a lens with a flatter field would be better. The 18-35mm f1.8 Art would be great for landscapes.

Oct 8, 2017permalink

Dougbm_2

No Kelvin adjustment. 17-50 lens only good not great.
Curvature makes it perfect for interiors?!

Oct 8, 2017permalink

Yes, curvature makes it perfect for interiors. When shooting interiors, the stuff at the edges of the frame are almost always closer than the stuff in the middle. Think about a room with a couch against the far wall. If you shoot the couch in focus, with a normal lens you'll get that back wall in focus, but the walls to the left and right will be out of focus. The curvature of field caused by the 17-50 will mean that the table and chairs to your left will be in focus, as will the painting on the wall to your right.

As far as Kelvin adjustment I understand what you're saying, and fortunately Sigma has fixed this with the Quattro cameras. Still, you can make fine tonal adjustments using the color adjustments tool in SPP for any of the Sigma raw files going back to the SD9. I'm sure you knew that though, right?

Today's Sigma SD Quattro H is an amazing camera, and I urge those who don't care about shooting at settings above ISO 1600 to try it. Some want to shoot at ISO 6400 though.

Oct 10, 2017permalink

fishy wishy

Same problem with the SD10, common to the Foveon design. The color will get progressively out of whack as you go up the iso scale. Possibly the SD1 extends the useful range by up to one stop over the SD10 to a debatable 800, but that's it.
The reason this x3 sensor design is not taken up by other manufacturers is not some dastardly conspiracy cartel. It's the inherent weakness of two of the color layers only receiving the remnants of the light that isn't filtered by the first color pixel. On most sensors all the red green blue get first crack at the light which makes for lower noise, more secure color, and you can design sensors with at least 50% more pixels this way- at the expense of the color resolution being technically much lower.
Any Sigma Foveon x3 should be considered a base iso camera and you are compromising color and noise potentially fatally by raising the iso. And yes, I regularly got shots where I could not control the color properly.

6 months ago*permalink

Jesse_Just_Him

When do we have a wood version of the D850?

Oct 2, 2017permalink

entoman

I'm just wondering what sort of person would want to walk around with a camera in a polished wood case! I'm sure it's extremely well made, but er.....

Oct 1, 2017permalink

adegroot

Throwback? I'm still using mine. A lot.

Oct 1, 2017permalink

terence_boylen

I would've given my left nut for one of these!

Oct 4, 2017permalink

digi2ap

Yes, same here, what a great camera.

Oct 7, 2017permalink

zodiacfml

Don't like the image quality from the Merrill generation. The Quattro is a bit better. They do really have to chase the resolution spec, sacrificing shooting speed and dynamic range?

Sep 30, 2017permalink

Mr. Yamaki was impressed by the details in photos from an 8x10 camera. I believe that since the beginning of their digital camera line the detail in images have been what Sigma is all about. The EX lenses and the Foveon sensor lead toward better and better capture of detail. The Art series of lenses are now the best lenses available at any price for capturing detail. The SD Quattro H is capable of taking advantage of those lenses, and it pretty much matches any 35mm DSLR for the ability to capture ultimate detail. I believe that eventually Sigma will make a full-frame Quattro camera. If they do, it will be equivalent to an 80 MP Nikon or Canon, and it will capture more detail than all but the very best medium format cameras available today. It will beat the new Fuji and Pentax cameras, while being less expensive and more portable. The SD Quattro H seems to be pretty popular, so hopefully Sigma will make a full-frame camera soon (announcing it early next year?).

Oct 11, 2017permalink

sonoltaprime

@Scottelly
I read that the Foveon sensor has some major technical issue that prevents it from going full frame. Do you know anything about it, and if it can be overcome? I shoot an sd Quattro, and would really like to move to FF sometime, but I can't get away from the Foveon look--it's too good.

Oct 28, 2017permalink

Mescalamba

Technical issue is a lot of processing power needed to chew that files. Quattro H produces enormously big files. In DNG form each pic has 150 MB. Thats rather insane amount of data to put thru camera.

But as technology advances, I think they will make FF eventually. Quattro H is APS-H sized (1.3x crop) at least. :)

10 months agopermalink

PanoMax

I'm a DP3 Merrill user and most of the images in my DPREVIEW gallery are recorded with it. It was destroyed 3 weeks ago in a car crash and subsequent fire so I will be replacing it with an SD1 Merrill shortly.

Sep 30, 2017permalink

Here's an interesting and much more extensive article, which investigates the Foveon Merrill sensor a little bit: https://www.adorama.com/alc/0013973/article/Sigma-DP-1-Merrill-46MB-Sensor-Compact-Digital-Camera-Product-Review

Sep 30, 2017permalink

PhotoKhan

As the comment posts on this article wind down one thing has become clear to me.

Foveon users only (!) claim for this camera is that it excels in good light, at low ISO.

Unfortunately, when clicking on the users handle links, either (1) there are no photos to be seen or (2) the samples from the SD1 that can be seen completely and utterly contradict their claims.

This should be an applied mass-psychology case study....

Sep 30, 2017*permalink

shauncwalsh

Feel free to take a look, a mix of SD1M, SD14, sd10 and Canon images.
https://www.flickr.com/photos/34887841@N00

Sep 30, 2017*permalink

George_pix

This is how it works with the kit zoom:
https://www.dpreview.com/forums/post/59746618
This is my preferred picture, intended to show the level of detail it can provide.
Anyway if your subject has detail, it always behaves this way.
Autofocus is very accurate, I only use the central point, and I know where to set focus.
I use ISO200 under normal conditions, or ISO 640 in low light. If necessary a tripod or the wonderful METZ 45 CT4.

Sep 30, 2017*permalink

https://www.dpreview.com/forums/post/59862042

What a lie! Where in the World did you get THAT load of B.S?

The SD1 Merrill and other Merrill cameras excel at B&W photography at high ISO too. It's not just about good light either. Take a look at this low-light photo from my SD1 Merrill:

Just look at the image comparison here on DPreview, and you'll see the $1,200 Sigma SD Quattro H matches or beats the new Nikon D850, which costs almost three times as much. What does the Nikon do that the Sigma doesn't for all that money? It can do high-ISO shots at ISO 800 and above better, and it captures more dynamic range than the Sigma. It shoots faster and has video capability too. Oh, and it has Wi-Fi and a tilt screen too . . . and for those who prefer optical viewfinders, it just might have the best optical viewfinder available in a camera that isn't medium format. You have to pay a lot for all that though. An SD Quattro H with an 85mm f1.4 Art lens is less than the Nikon D850 body alone.

Sep 30, 2017*permalink

http://www.bigprintphotos.com/img/s11/v29/p2543213142.jpg

Here's a nice low-light shot, though at ISO 100:

I've shot with my Sigma SD1 Merrill at night too. It all depends what you want to capture. If you're shooting at ISO 100 or 200 it works about the same as any other camera. At higher ISO settings, you have to do a little different processing or convert to B&W. I'd say in B&W my Sigma matches my Sony A65 up to at least ISO 3200, and in color it beats or matches my Sony it from ISO 100 to ISO 1600. Focusing isn't as good though, and the A65 is not the king of good focusing. Luckily Sigma has improved focusing with the new SD Quattro and SD Quattro H cameras.

Sep 30, 2017permalink

AnalogPower

I wish SIGMA make a follow-up SD2 DSLR. There is much to be offered from the old foveon sensor. I firmly believe as many others that SD1 is still quite capable camera. According to other reviews of actual users it is a hidden gem. Just googling the name i stumbled upon a fan review appreciating SD1 as many does here. You can find it at:
https://www.classicdslr.com/sigma/sigma-sd1-still-worth-buying/

Sep 30, 2017permalink

Barrie_Grieve

With a Merrill H sensor , 3x20px layers.

Oct 3, 2017permalink

SigmaChrome

I shoot with the SD1 Merrill and have done so for some time. It's not an easy camera to master but once you do, you will never look back. Paired with quality Sigma lenses not many cameras come close for studio, landscape and macro work. The Sigma DP3 and DP2 Merrill is in the same class. These are special cameras; work slowly in good light and reap the rewards.

Sep 29, 2017permalink

DarnGoodPhotos

Wow, the wood veneer edition.

Sep 29, 2017permalink

Wye Photography

Yes, very Hasselblad.

Sep 29, 2017permalink

shauncwalsh

I shoot with both Sigma and Canon.
The Canon 7D is great for low light and action. It produces acceptable images most of the time.
The SD1 Merrill is not good in low light (although I have been surprised how good it can be at lower resolution with pixel binning). Focusing is variable and the buffer takes way too long to clear. But in good light it is capable of taking images that make me realise why I put up with all its shortcomings.

Sep 29, 2017*permalink

User9362470513

I hired an Olympus EM1 Mark ii and shot a still life using the Hi-Res mode that produces an 80MP image using the 12-40 Pro lens. I took the same image using a Sigma DP3M.

I was surprised, but the Sigma DP3M absolutely destroyed the 80MP Hi-Res image. The Sigma image was so so sharp, even across a room in the evening light the Sigma highlighted hairs and dust on the still life which were not clearly visible with the Olympus.

Look at the studio scenes on this website, the Sigma DP2Q produces cleaner and sharper images than 90% of non-MF cameras.

Even the D850 has colour artifacts and moire in the etching that the DP2Q does not have and the Sigma is just as sharp as the best DSLR on the market!

For people who have the time and patience for MF output, the Sigma cameras are stunning. I like that no-one else has them. They are for connoisseurs.

Sep 29, 2017permalink

Naveed Akhtar

I know mate; I had SD1 and always enjoy the colours out of it; sharpness is more to do with the lenses though .. however I couldn't keep that camera for long bec of slow and limited operations.

Sep 29, 2017permalink

jwasturias

I agree, I was tempted.
I decided in the end for the Sony A7R since I already had the LE-EA4 adapter for my Minolta lenses and my canon fd85 1.2L, and a lot of my pictures are action shots of farm animals and cats, and macros using iso 800-1600 a lot.
If Sigma made a mirror less interchangeable lens version where I could adapt most of my lenses to (even if manual focus) I would have bought one.

Sep 29, 2017permalink

themcroocedvulture

Wow... there IS one already! - Sigma SD Quattro H!

Sep 29, 2017permalink

https://1x.com/photo/721714/

The SD Quattro and SD Quattro H are mirrorless, but they share the same mount as the SD1 Merrill, because Sigma didn't want to add an adapter between the lenses and the camera, and no doubt they have no intention of developing a whole new lens system.

Sep 30, 2017permalink

Paulo Ferreira

I have the SD1 Merrill. The operation is disconcerting, autofocus, redundant buttons, metering, computer processing, but, but, when I take another camera out for pictures I always think "Oh this would have looked so much better if taken with my SD1... because it does". It's like you have a tiger for a pet.
And you can make prints the size of a bus out of its files.
Something that dpreview always refused to compare were files taken with Foveon in double against the same Bayer size. It always refused to compare Bayer enlargements i.e. a 24MP bayer enlarged to 48MP against a 15MP Foveon against 48MP. A Foveon native file enlarges so much better.
https://www.flickr.com/photos/azuzarte/

Sep 29, 2017*permalink

Meuh

"Oh this would have looked so much better if taken with my SD1... because it does"

This! :D

Oct 2, 2017permalink

quintana

I owned an SD1 for some time, but I often struggled with the autofocus. The viewfinder is too small for manual focus and the accuracy of the autofocus wasn't too good.
It would have been great if it had live view with magnification of the focus point even if that would have meant carrying a lot of batteries with me.

Sep 29, 2017permalink

George_pix

There is nothing to remember. I have it and I use it.
It takes excellent pictures.

Sep 29, 2017permalink

BBnose

Will Sigma update the line with SD-2 ?

Sep 29, 2017permalink

Digiman69

The Merrill generation (SD1 and DP series) is the last real Foveon camera.
The SD/DP Quattro generation failed without the original Foveon sensor architecture, and the SD Quattro is missing a decent Oled EVF, and a new real mirrorless bayonet.
That's all I will buy tomorrow an SD Quattro with the Merrill sensor Oled EVF, up to date AF, with a real mirrorless bayonet.

Sep 29, 2017permalink

Mescalamba

Unfortunately they are Japanese, so they will pursue their way till they fail and die with honor.

10 months agopermalink

bilmenot

I always want to get one, may not be the SD1 but the SDQ, DP2Q...etc., many people said Foven is only good for low ISO, 400 is near the limit but I've never tried.

Sep 29, 2017permalink

BPA4

For anyone wondering what the SIGMA Foveon Merrill sensor can do, among other images perhaps take a look at these from photographer Mark James Ford, made with the SIGMA DP3 Merrill.

In my opinion, there is no other digital camera or sensor that creates images that appear so alive and real. I am including in this assessment the SIGMA SD1 Merrill and the SIGMA DP[1,2,3] Merrill. The SIGMA Foveon Quattro cameras are also very impressive, but the Merrill has something a bit unique.

https://1x.com/photo/1104071/

Sep 29, 2017permalink

EwanMC

Throw back?! I still want one. Anyway I do own a DP2 Merrill and its 45mm f/2.8 (equivalent) lens is sharp enough for the sensor.
Bayer sensors are pretty good these days and a Sony a6300, for example, is only a little behind in absolute resolution and superior in all other respects. The brain would say buy that instead.

Sep 29, 2017permalink

ChickenBalls

"For those who wanted to carry around something a bit more 'classy,' Sigma released a model with a burl wood veneer, which was priced at €10,000,"

ah! so this is where Hasselblad got their idea for Lunar and Stellar from. I wonder how many of that poop-colored camera Sigma managed to sell.

Sep 29, 2017permalink

merijn_s

I have very good memories about the SD1. As a matter of fact, i think it is the absolute best! Thats why i got one parked in my garage ;)

Sep 29, 2017permalink

Daft Punk

You refer to a Rover I assume?

Sep 29, 2017permalink

merijn_s

Yup sd1 vitesse owner. But ontopic: i also like this sigma camera. I always had a soft spot for the brands that tried an other way to thing. Like fuji with their super ccds.

Sep 30, 2017permalink

Tom_A

Because of the fans and detractors of foveon vs bayer below, I had a look in the new comparison between the D850 and the Sigma Quattro H, specifically examining the eye of the peacock feather at 100 iso jpeg. I was surprised to find the Sigma more detailed, but it could be due to the lens - ideally all the cameras would use the same physical lens with an adapter.

Sep 29, 2017permalink

George_pix

I think that the lens "is" the same.
Who makes it?
(I might be wrong indeed)

Sep 29, 2017permalink

I looked at the rez chart...because the oversharpening and aliasing wont help the foveon there. And as expected, the D850 destroyed the foveon on the rez chart, and other areas as well.

Moral of the story, if you shoot peacock feathers, the Sigma,is for you...for real resolution, choose the D850.

Sep 29, 2017permalink

hikerdoc

As you read the comments it seems the detractors and defenders of this camera have one thing in common, most of them have never used it.

Sep 29, 2017permalink

kodachromeguy

Well that minor oversight never affected the "experts" who make profound comments and pronouncements here on Dpreview.

Sep 29, 2017permalink

veroman

Quite agree ... but more importantly, how do we definitively identify the posting detractors and defenders who HAVE used it? It would be helpful to an extreme if they identified themselves in their posts ... as some do.

Sep 29, 2017permalink

Try looking at their profile. If they have stuff on their gear list, and a Merrill camera is listed there, it's likely they have used one.

;)

Sep 29, 2017permalink

Daryl Cheshire

Yes when I got the Canon 5D I considered the SD15 which cost the same.
I was disappointed with the pricing fiasco with the SD1 as the rumored price was more realistic and it seemed that some executive just doubled it.

Sep 29, 2017permalink

I believe the man who started Sigma, Mr. Yamaki, decided the camera should fetch prices worthy of its image quality, but alas, the customer base did not agree. If Sigma had made two cameras at the time, maybe it would have been accepted, but there was no SD15 replacement . . . just a very expensive camera, with no practical upgrade path for those who had bought into the Sigma system and bought an SD14 or SD15 years before. What Sigma SHOULD have done is introduce an SD16 (even if just announcing it) a few months after the launch of the SD1. The SD16 could have had the same pixel density, but with the 1.7x crop factor of the SD1 and slightly slower operation (3 fps) of the SD15. If Sigma had done that, there would have been very little need to drop the price of the SD1 and rebrand it to SD1 Merrill. Unfortunately Mr. Yamaki died shortly after the launch of the SD1, and then his son, Kazuto Yamaki, made the decision to drop the price of the SD1, if I'm not mistaken.

Sep 29, 2017permalink

Daryl Cheshire

Interesting reply.
That you.

Sep 29, 2017permalink

Gesture

A favorite feature, but so few of these recently describe the author's own use of the camera.

Jeff, I would have noted the initial pricing fiasco.

Sep 29, 2017permalink

taktak91

Depending on where you look, you can still get this one new.

Personally, I would like to have this one, rather than the current mirrorless.

Sep 29, 2017*permalink

FodgeandDurn

Shout-out for the dinosaurs at Crystal Palace.

Sep 28, 2017permalink

goblin

This remains to this day an ultimate tool for garment shoots, sabotaged to this day by the lack of any civilized/non-geek way of creating dcp color profiles for it.

From a happy SD1m; DP1/2/3m; DP2Q; SDQ (between others) owner.

Sep 28, 2017*permalink

George1958

a camera I still covert. I got the SD15 and no complaints but I could not have stretched to the SD1 at the time. I still have that itch but I can't work out if I should be thinking quattro?

Sep 28, 2017permalink

I suggest a used SD1 Merrill, but maybe go straight for the better focus capability of the very inexpensive and faster operating SD Quattro (though it does not shoot at 5 fps, like the SD1 Merrill can).

Sep 29, 2017permalink

EwanMC

*covet

Sep 29, 2017permalink

George1958

I am interested in the Quattro and ideally I would like to be able to use an SD1 and a Quattro for a weekend. My issue is I really like the look of the SD1 output and colour, it is closer to my SD15 and of course the DPm series.
The more recent Quattro makes a lot of sense but the output seems qualitatively different to the merril sensors ? Not sure how to express this but it's to do with colour.

Sep 29, 2017permalink

Lukacs85

I'm big fan of Sigmas, however never owned SD1. I like my DP2m, and my SD-Q. Mainly shoot at low iso, so for me the best IQ at low iso is the most important. I can get great, somwhere similar results from PP $2500+ high resolution bayer camera. But I barely afford my SD-Q for $900. And the dp2m is a little gem, only RX-1R/2 can get close at low iso for again 3x price.
And no other camera can deliver me the same crystar clear, pop IQ of DP2m.

Sep 28, 2017permalink

iae aa eia

Under a low ISO and with the color temp well balanced, no camera matches the color depth of a Foveon sensor. No camera, not even the al-galaxy-mighty D850. It would be the ultimate advance (even more than flexible sensors, in my opinion) in sensor technology if Sigma manages to acheive better high ISO's and video from it.

Sep 28, 2017permalink

knoxphoto

The SD1-M body was great, dials/controls were well placed. Not having a top LCD was a poor decision, it also affected the battery life because you had to rely on the rear LCD screen for everything, (battery life was mediocre, I would use 3 batteries in a studio session) the rear LCD was a novelty due to the low-res Jpeg previews generated by the camera. AF points were limited but well placed, however, where the autofocus was hitting was anyone's guess and it had to be checked constantly for critical focus, even after calibrating the lenses to the body.

That being said, shooting with it tethered at ISO 100 in a studio could be magical, as long as you set a custom WB and didn't overrun the buffer, in which case you got to make small talk with the talent while you both waited. Foveon is what it is, you either love it, despite it's quirks - or it made you want to bounce the camera off of the studio's concrete floor.

Sep 28, 2017*permalink

Sezano

“the talent” - lovely euphemism

Sep 28, 2017permalink

MrImprovement

Have used the similar dp2 Quattro (yes, sensor is slightly different) and what I can say is that the image quality at ISO 100 is superb, although there are caveats, as others mention.

Re: mrgum's continual slagging of Foveon sensor: absorption of colors at different layers of silicon is well-studied, and used in many other applications, not just Foveon. Google "silicon spectral transmissivity image sensor" for instance.

You can see that the sensor does reproduce color, so exactly what is being absorbed at what depth is a false criticism: the sensor does reliably reproduce colors.

Sep 28, 2017permalink

Foveon's depth based diffusion definitely works to an extent (given enough light) but metameric errors and issues with casts are well documented. The facts remain that putting the sensels side by side instead of on top of each other just makes so much more sense:

First sensitivity is massively improved, you don't have to have a second body to shoot in low-light!

Secondly colour accuracy is improved as you can design colour filters with a particular and well defined cutoff, and you have much less overlap between the channels.

Finally resolution is also improved! Sigma themselves admit the 46MP SD1 produces images that are only as detailed as a 30MP Bayer sensor. The reason for this is that in general colour channels are highly correlated. Imagine you are photographing a banana, you will get a strong response in both the R and G channels. If the R and G channels are offset spatially instead of stacked you can utilise the overlap in response to increase your effective resolution.

Sep 29, 2017permalink

Data rates and processing requirements are also lower for Bayer sensors.

Sep 29, 2017permalink

jwasturias

@mgrum
photographing a banana is bending the rules a bit though..

Sep 29, 2017permalink

FDecker

Metameric errors go together with color resolution. If you want fine graded colors, you have to accept that metamerism can occur.

You are wrong here, real RGB sensors with very little spectral overlap are not able to distinguish colors which are in the spectral bands without overlap. Color accuracy goes down. Not that great!

You see the Foveon in a wrong way. It is a 15MP sensor which acquires all three colors at each pixel site. Therefore, it has a resolution advantage of factor 2 for G and 4 for R and B. But because the de-mosaicing algorithms for Bayer are quite good, the resolution advantage "overall" is just a factor of 2.

Anyway, back in time when the SD1 was released, a 30MP camera was at the top of what you could buy.

Sep 29, 2017permalink

But Bayer sensors DO have some overlap in response (good for resolution as well as sensitivity), but it's much less than the huge overlap of Foveon sensors (which is hard to manage) and more importantly its easy for the manufactures to tune (unlike Foveon).

I see the SD1 as a camera with 46 million photosites, that has to read out 46 million values, stores 46 million values in the RAW file. It has all of the cost and data rate implications of a 46MP camera. It's also a camera described by Sigma themselves as have 46 megapixels. The only way it's not like a 46MP Bayer is in terms of resolution...

The SD1 was first available to buy in March 2011 so they had a slight resolution advantage over the competition until the D800 was released a year later. However they failed to capitalise on this advantage due to the absurd $10,000 price tag.

Sep 29, 2017permalink

MrImprovement

Q for mgrum: Have you ever actually used a Foveon sensor camera yourself, shooting in the RAW format that it uses, and processing the images?

Sep 29, 2017permalink

@MrImprovement

No. Nor have I used a chocolate teapot but I can still tell you it's a bad idea...

Sep 29, 2017permalink

MrImprovement

@mgrum all I can tell you is, reading every issue of Playboy is hardly the same as ****ing...

Sep 30, 2017permalink

aaaand I believe we have the trifecta!

Sep 30, 2017permalink

davev8

i dont know why sigma use the SA mount... will be better to do a nikon F and Canon EF mount like Kodak did ...all mount patents are out of date and sigma use canon protocol already for lens communication with the SD1 and the flange distance is evan the same as canon ..i think sigma will have sold many more units if they did

Sep 28, 2017permalink

danski0224

Sigma has "reverse engineered" the Canon RF lens protocol, not licensed it from Canon.

There would be less/no incentive to purchase SA mount lenses OR Sigma made lenses in EF mount, and Sigma is a family owned business that wants to be a camera maker. By their own admission, lens sales finance the camera division.

There is, most likely, a cultural taboo that is preventing Sigma from making their own camera with a Canon EF mount.

At a minimum, the Merrill generation Foveon sensor does NOT do well with adapted lenses, which would create all kinds of image quality problems. Those that have adapted and used non-Sigma lenses on Merrill and earlier sensors have had to deal with extensive post-processing to reduce or eliminate color casts. Sigma has info for their lenses in their processing software. Given the Sigma philosophy for the perfect camera, releasing something with more "issues" would go against their philosophy.

Sep 28, 2017permalink

Gesture

Yup. Even though Sigma is a lens manufacturer, they still don't get it. Pros, especially, aren't going to invest in a system where there is usually only one, in a rare time two, models in production at a given time. Would be a perfect extra platform for a Canon or Nikon pro. One has to guess that Sigma wouldn't be able to ramp up production enough if a Sigma interchangeable lens camera became more popular. Still have my DP1.

Sep 29, 2017*permalink

Real pros often have multiple systems. Some have a medium format system and a 35mm system. Some use large format, medium format, and no 35mm. For a pro who wants to try the Foveon sensor, there are the DP series cameras - 4 of them - super-wide, wide, normal, and short telephoto. Each one costs the same as a good lens. If those cameras don't convince the pro that the image quality is worth buying an SD camera with a couple of lenses, then they aren't the right customer.

As far as making a camera in a competitor's mount, does Canon do that? Does Pentax? Does Sony? No. Why should Sigma? Sigma has a great system of lenses, and they now have a low budget camera and a medium budget camera. If that's not good enough for you, then the system is not for you, so don't buy into it.

I'm hoping Sigma will soon build a more professional level camera, such as a full-frame SD Quattro F, with the same pixel pitch as their 1.5x and 1.3x crop factor cameras.

Sep 29, 2017permalink

Gesture

Why should Sigma?

No, but Sony, Fuji and to a certain extent Canon have bodies that will take a great variety of lenses with adapters.
I think this is the future of interchangeable lens photography and something this gives micro 4/3rds great traction.

Yep. But, for me, I would never buy into a system with so few bodies extant, even if I were a millionaire.

This has been a good thread and has indicated to me problems with Foveon cameras that I hadn't considered before. I believe the cameras are best left to fixed focal length versions where the lens can be matched to the sensor as great as possible.

Sep 29, 2017permalink

UnintendedKeith

Wow that wood veneer version is absolutely hideous!

Sep 28, 2017permalink

Sezano

I think of Sigma cameras as my Idiot Savants.

Sep 28, 2017permalink

Henrik Herranen

"Foveon sensors capture color in a completely different way than Bayer sensors, with each 15MP layer capturing a primary color."

I'm afraid this is incorrect. Sigma would like you to believe othewise, but the layers don't capture primary colours, but a somewhat blueish, a somewhat greenish, and a somewhat reddish colour. Examine the RAW files if you don't believe me. There's not nearly as much colour separation as there is on a Bayer sensor. This is at least part of the reason for Foveon's high-ISO colour blotches. Even though it isn't interpolating spatially, it is extrapolating chromatically.

I don't want to take anything out of Sigma's achievements. The Foveon sensor is a unique idea, and Sigma's Art lenses are wonderful (I have the 50/1.4A, and MAN!). I love the company for its innovations! But, the turth of it is that Foveon sensors for sure don't capture three primary colours as such.

Sep 28, 2017*permalink

I think it's actually more like white, dirty yellow, brown. The top layer is essentially unfiltered (save for the properties of the silicon), so has no "colour" although it will have a spectral peak who-knows-where.

Sep 28, 2017*permalink

Who cares what colors it captures? Have you seen the photos that good photographers can make with the SD1?

Sep 28, 2017permalink

Yes...I have seen superb photos. In many ways, like rez, they get close to the 36 MP Bayer sensors. In most ways, they are worse.

Sep 28, 2017permalink

@Scottelly "Who cares what colors it captures?"

I don't know people who dislike outright lies from marketing? P.S. Have you seen the photos that good photographers can make with their Rebels, Phone cameras, and film SLRs?

Sep 28, 2017permalink

Henrik Herranen

Scottelly: Anyone caring about the truth vs lies does care. As much as I respect Sigma for their very affordable super lenses (I own the 50/1.4A and it's magic), I despise their Foveon marketing, which, honestly, is based on straight lies.

I work in audio industry, and I am an expert in audio signal processing. In my profession I have to fight against similar lies and misdirections on a constant basis. After 20 years it gets really tiresome to fight against the very same lies, perpetrated over and over again by publishers and web authors alike. That's why I really hate the Foveon lie. It is a purposeful lie, sold hook, line and sinker to even respectable publications, like DPreview, which I othewise like so much.

Sep 28, 2017*permalink

User9362470513

All these morons saying the Sigma makes worse pictures.... Just look at the studio scene on this website and compare a DP2Q against any camera - look at the feathers, look at the etching, look at the pencil drawing of the old man - the Sigma blows everything else away.

Yes the colours are whack on the portraits (which is odd, as my SIgma's always give lovely skin tones) but that can all be addressed in Post. What cannot be dealt with in post is the unbelieveable acuity and sharpness.

Sep 29, 2017permalink

Fotoguy48

So many hatin' on the SD1! I own one purchased from Sigma for 899 2 years ago. It is not my favorite camera, but I don't hate it. First, it is a very solidly built tank of a camera. Second, it is slow as molasses at xmas in processing images. But then, the images and what you can do with them! Oh my!

I also have a little DP2s bought off ebay for 200. Like its big brother it has some limitations, but produces superb images which are very amenable to processing with Sigma's proprietary software.

Sigmas are niche cameras. Always were. Alwsys will be.

Sep 28, 2017permalink

Buhl213

" slow as molasses at xmas" ?

Is that in Finland or Australia?

Sep 28, 2017permalink

User9362470513

Photographers are some of the most hate filled, bile spouting extremists this side of a Republican convention.

The Sigma DP range of cameras are stunning - tiny packages producing images that rival the best Canon and Nikon can offer. With a tripod and the right light, they are incredible and are small, light and reasonably cheap.

Sep 29, 2017permalink

worldcup1982

In resume:
Great resolution, bad DR, noise noise and noise even at iso 100, slow as hell, color shift and... At the bargain of...8000 usd... Ok, then they got real and the price went down almost 6000usd..lol
Oh, wait, I forgot, it's better than medium format according to some users, even with all the above

Sep 28, 2017permalink

In some ways it IS better than medium format. The resolution is up there with the 29 and 33 megapixels MG cameras of its day, yet it has a 1.5x crop sensor, which makes it the ultimate camera for shooting daylight wildlife even today, if image quality is what you're after. No, it won't make great photos above IS 800, but who shoots above IS 800 or even 200, when their main concern is image quality?

Sep 28, 2017permalink

worldcup1982

Pretty much all wedding, sports and many other photographers, including national geographic ones shoot at and above ISO 800.
And at ISO 100 the Merrill already has awful shadow noise, so your argument about about image quality is just no sense.
Just because is good by your standards doesn't mean is good for others.

Sep 28, 2017*permalink

rrccad

"Pretty much all wedding, sports and many other photographers, including national geographic ones shoot at and above ISO 800."

stats please. or did you pull this out of your posterior?

Sep 28, 2017permalink

The Foveon concept is sadly flawed. Instead of recording a neat red green blue sample at each pixel like the marketing implies, it really records something like white, pale yellow, dirty red. From this a lot of matrix operations are required to estimate halfway reasonable colour values. Additionally light spreads out as it penetrates deeper in the silicon necessitating heavy sharpening operations on the lower layers (this is why they abandoned the original concept when producing the quattro), this is also partially responsible for the sharpness and microcontrast (compare the SPP output to a third party converter).

All in a Foveon image requires a lot more "conversion" than a Bayer image. An oversampled Bayer sensor offers the same resolution, will less processing required, more accurate colours, increased dynamic range and orders of magnitude better sensitivity in low light. And oversampling also gets rid of the problems that Foveon sought to solve (aliasing, demosaicing artifacts).

Sep 28, 2017permalink

Have you ever used an dp Merrill or a foveon sensored camera of any kind? All that jargen you just spat out means nothing when you actually look at a photo taken with that sensor. The colour is Great, tunns and tunns of detail.

Sep 28, 2017permalink

@M W B G:

No but I've studied the images. Someone on the FM forums did a side by side comparison of the DPM2 with the D800 and replica SPP processing (deconvolution mostly). If you looked really hard you could still spot the D800 by some false colour artifacts here and there but not by sharpness or detail. I don't see that it's necessary to actually touch one to be able to talk about it (though I'm seriously considering buying one just to fend of the inevitable "you've never used one so you don't know" arguments).

All of the "jargon" that I "spat out" means something quite important: Foveon cameras = very nice detailed images, high MP Bayer sensors processed to look like Foveon = very nice detailed images. That doesn't mean what you see in the Foveon images isn't there - but it does suggest that the tradeoffs required simply aren't worth it.

Sep 28, 2017permalink

Well, I find the color and details in my 14.7 MP photos from my SD1 Merrill to be better than what I get with my 24 MP Sony A65. It seems to me that you're talking out your butt.

Sep 28, 2017permalink

I owned a dp2, and dp3 and I never found I put much of any work into the photos. I used sigmas software, which was slow yes, to convert to tiff only. Then with Lightroom only few tweaks. Done. I found it pretty fast. I made some large wonderful prints with those cameras.

Sep 28, 2017permalink

I've never been able to get the detail I got from them from my d810

Sep 28, 2017permalink

Well, that "sadly flawed" Foveon concept produces better quality images than my top-of-line Nikon, Canon, Pentax and Olympus dSLR's. Your comments indicate to me that your experience with Sigma is limited to reading about them rather than using them.

Sep 28, 2017permalink

@
Scottelly
M W B G
Lin Evans

You wont see the same quality in your Sony A65, D810 or "Nikon, Canon, Pentax and Olympus dSLR's" unless the images are processed in the same way as SPP, with the same amount of deconvolution applied.

If you're going to say "but I don't want to have to put my files through such processing" well fine but having zero low light ability is a large price to pay for doing so...

And no my experience is not limited to "reading about them", I have extensively studied the images they produce (which are readily available online), even though I haven't physically touched one (I'm seriously considering buying one just to avoid those comments every time...)

Sep 28, 2017*permalink

I'll stand the D810 or D850...or the Canon 5DSR against any foveon based camera. The foveon wont match the rez, and will be horrible in noise and DR comparisons. Every major test site has confirmed this for years.

Sep 28, 2017permalink

No, you're not correct. I've been shooting digital since my first dSLR (Kodak DCS-460) in 1995. I have, use and am familiar with every best deconvolution, wavelett and processing software available. Nothing can be done to get the same quality from my conventional dSLR's as from the Foveon processors in Sigma cameras. I've also been shooting Sigma since the very first one became available. There are palpable differences not only in clarity, and lack of sharpening halos but also a huge difference in enlargeability without losing details. "studying images" is not a substitute for shooting hundreds of thousands of images with both Sigma and conventional Bayer CFA cameras.

Sep 28, 2017permalink

@Lin Evans - do you have images that can demonstrate the palpable difference in clarity and enlargeability over Bayer sensors? Since you've owned so many caeras you'll have done at least one test, right? If not I'll stick with my own research.

Sep 28, 2017*permalink

I was a skeptic once, until I bought and used 2 of them. I mean to get the superb quality from the camera, its best to shoot from a tripod, but at 100%, there is no loss of fine detail and clarity at all. Not dissing the d810, it is a great camera but I could never get the fine detail I can get from the dp2. Also using sigmas proprietary software I could get plenty of shadow detail back before converting to tiff for use in Lightroom. Buy one, use it and i bet you will change your opinion. They are dirt cheap these days

Sep 28, 2017permalink

Fortunately, we have enough test images produced by review sites and people with mutliple formats that show cameras likethe D810 or D850, or 5DSR outresolve all foveon bodies. This has been beaten to death.

Sep 28, 2017permalink

Of course I have images which demonstrate the differences - hundreds of them. Many are hanging on my client's walls. We who use Sigma cameras daily get really tired of "challenges" from uninformed people with zero experience. Had you any real experience you might have seen our huge images hanging on the walls at PhotoKina, CMA and myriad other shows around the world - but it's a waste of my time and energy to try and convince every person who "thinks" they know more than people with real experience - this is obviously one of those cases.

Sep 28, 2017permalink

@Lin Evans

"I've got loads of evidence to support my case!"
"May I see it?"
"hmmm, no."

Scientists in general have way more experience that the average person, yet they always back up their claims with evidence...

Sep 28, 2017permalink

Did you not understand my English. The "evidence" has been hanging on the walls of CMA, Photokina and other shows for years. Have you ever even been to Photokina or CMA or Comdex? Sorry, but I have no more time for your nonsense.

Sep 28, 2017permalink

@mgrum
Where is your evidence? You say all that but have never used a foveon. I and many others have. I know it to be true.

Sep 28, 2017permalink

https://www.dpreview.com/galleries/9932828572/photos/3685238/sigma-vs-d810

So much for beating the D810

Sep 28, 2017*permalink

http://mattgrum.com/foveon/sppprocessing.jpg

@Lin Evans - I think understand your English perfectly - the evidence is hanging on the wall of a building that I don't have access to!

@M W B G:

Here is an example of the sharpening that SPP does (vs. irident on the right):

The D800 comparison is better, but it doesn't seem to be on FM any more - I saved a copy somewhere which I will dig out after I've been to the pub.

Sep 28, 2017permalink

That tells me nothing. Not even sure which is which on that example. My experience which I trust more than whatever that sample is, is proof enough for me. I owned both the d810 and a dp2. And I know for a fact the dp2's detail at 100% surpasses the d810. No loss of detail. The d810 is a little smudgy.

Sep 28, 2017permalink

I've only used the sigma program and Lightroom on my files. Never had a problem. You don't have to sharpen the files from the dp2 one bit. I never did and it was still sharper and had more clarity than my d810. I'm going off first hand experience, something neither of you have.

Sep 28, 2017permalink

The blurry one is Sigma. Odd comment you made....evidence tells you nothing. Interesting.

Sep 28, 2017permalink

Who ever took that picture had no idea what they were doing.

Sep 28, 2017permalink

Who ever took that picture had no idea what they were doing.

Sep 28, 2017permalink

In their studio scene tool the dp1 is sharper than the d810 at iso 100 jpg and raw

Sep 28, 2017permalink

The Imaging Resource is a well respected test site that has bee around for many years. You however, appear to be a vacuous blowhard who doesn't accept anything that goes against what he incorrectly believes to be true. Debate with you is a waste of time because and and all evidence that shows you to be wrong, will be dismissed.

Sep 28, 2017permalink

Whatever man, I know I'm right, I've seen it with my own eyes. I'm done wasting my time. Have a lovely misguided night :)

Sep 28, 2017permalink

lauma

You sound like somebody who never made love wanting tho explain those who experienced it why according to scientific data doing sports is just as good. No offence, I just don't know how to describe it better.

The Sigma is good for one type of photography: ISO 100 and you can plan the shot. If you need ISO higher than 200, have to do lots of shots fast, can't plan your light on location then the Sigma is not the camera for the job.

It's a specialist which outperforms any other camera in that one area it is good in. Especially when it comes to food photography I simply don't know how I could get similar results with another sensor. Same when it comes to fashion photography, blues and reds. It's easy to look at another photographers pictures and judge, it's something completely different to go into a job with a certain requirement and having to fulfil it.

Sep 28, 2017permalink

mgrum, "I'm seriously considering buying one just to avoid those comments every time..."

Every time what? Every time you troll and disparage Sigma cameras? What's your deal? Do you hate the Foveon sensor for some reason?

I know there are a lot of compromises in a Sigma camera. They don't include things like wi-fi, tilt screens, GPS, or video. The SD1 Merrill unfortunately doesn't even have live view capability - one reason I want an SD Quattro H. They are slow . . . much slower than my Sony A65, not just at shooting (10 fps for my Sony vs. 5 fps for my Sigma), but they are slow to write the files out to the memory card, slow to review the photo you shot, slow and inaccurate to focus, etc. The SD Quattro and SD Quattro H are supposed to focus very well, and they are supposed to be a lot faster than the SD1 at reviewing and writing to the card. Still, their image quality is spectacular for the price . . . at low ISO, where I shoot the vast majority of my photos.

Sep 29, 2017permalink

@Scottelly

Troll: noun. Someone who posts something on the internet you disagree with.

"What's your deal? Do you hate the Foveon sensor for some reason?"

Of course not hating a piece of silicon would be absurd. I am only interested in stripping away the marketing spiel, cognitive bias and hyperbole to find out what the underlying technology does and more importantly what I am actually getting for my hard earned money.

I'm used to environments where extraordinary claims are backed up by evidence but here all you get is a lot of words (containing terms like "pop") and responses of "well you clearly haven't used one so you don't know", which is a ridiculous way to try and shut down the conversation.

You're not wrong about the list of compromises that come along with the Foveon cameras, but what if you could get the same or better results without compromise?

Sep 29, 2017permalink

@M W B G "You don't have to sharpen the files from the dp2 one bit"

Have you not read anything I've posted? That's because SPP applies a huge amount of sharpening for you without you knowing - that's what the comparison image I posted above was demonstrating.

@lauma "You sound like somebody who never made love wanting tho explain those who experienced it why according to scientific data doing sports is just as good."

But it's a camera, not a sex act. It obeys the laws of physics, you can measure it's performance, there is no magic or pixie dust involved.

Sep 29, 2017permalink

lauma

I'd agree with you if the sites that test the sensors would include all physical data. But they don't, they measure about 10-20% of what is important for the picture processing and make conclusions from that tiny little bit of information.
And they will never be able to do more as the effort would be way too huge and expensive.

That is one of the reasons why long ago I stopped reading such sites, they have absolutely nothing in common with what I experience when photographing and therefore zero value for saying which sensor is a good one or not. I always test them myself first because in the end not the data but the picture counts.

Oct 22, 2017permalink

7seven7

The SD1 is exactly the opposite of my Eos 20D and Eos 5D :
- Never shoot with a Canon Eos under ISO 400 in low light !
- Never shoot with a Sigma SDn above ISO 400 ditto !
But they all share a narrow dynamic range

Sep 28, 2017*permalink

veroman

This uninformed and unjustly brief Throwback Review begs the question, "Why was the SD1 even selected for this segment?"

The SD1, like nearly all Foveon-based cameras, draws criticism like flies to fly paper. It's so very easy to criticize, isn't it. But for low-ISO shooters who use fast, quality lenses and who have mastered the craft of taking a picture, the benefits of the Foveon sensor (even going back to the SD9) far, far outweigh what the Foveons can't do or don't do very well.

There is NO Bayer-based 35mm format camera that can deliver the image quality of the SD1 (and current Foveons) when shooting base ISO to ISO 200. Simply removing the AA filter from a Bayer camera does not a Foveon make. Yes, in less skilled hands the Foveons can be, and often are, problematic, especially if one thinks the camera should do all the work.

Read the reviews on Luminous Landscape and on ReidReviews if you want to know what the SD1 is truly capable of. This Throwback Review is a joke.

Sep 28, 2017permalink

Resolution is beat by the Nikon D850...as is color quality, noise, and dynamic range. Saying no Bayer camera can match the SD1 is pure rubbish.

Sep 28, 2017*permalink

I love my Nikons, but at base ISO no camera I have ever used produces files as beautiful, rich, and sharp as those from my silly old 4.7 MB Sigma DP1.

Sep 28, 2017permalink

Marty4650

I think the SD1 was selected for Throwback Thursday because it is an interesting and unique camera. I doubt that very many were sold, especially at the initial $9,700 price, but it sure makes a good story.

Sep 28, 2017permalink

What most people singing the praises of the SD1 don't realise is just how much processing goes into converting the RAW images. The sensor marketed as recording R,G,B at each photosite is really recording white,yellow,red. In order to overcome depth based dispersion, large amounts of deconvolution sharpening are applied (this is why the later Quattro models reduced the resolution of the lower layers).

If you open the Foveon RAWs up in Irident or dcraw you'll see they're a complete mess.

If you resize a D850/A7RII/5Ds or a even D800/A7R image down to 15 megapixels and apply some of the same sharpening operations and you'll see the Foveon sensor is nothing special even at ISO 100.

Sep 28, 2017*permalink

@mrkalntms

Try resizing your Nikon images to 4.7MP and sharpening them appropriately.

@Marty4650

They cut the price to $3000 two years after the launch, so they probably sold off a good deal of the produced units (albeit under the new name of SD1 Merrill)

Sep 28, 2017permalink

Raist3d

You ignore the DR gap between it and Bayer APS-C, let alone FF.

Sep 28, 2017permalink

RedFox88

Was beat before that by the canon 5Ds

Sep 28, 2017permalink

I think it's funny that you keep mentioning full-frame cameras that cost about $3,000 today, when you could have bought an SD1 Merrill for under $2,000 a few years ago. lol

Sep 28, 2017permalink

Scottelly...they started selling the camera at $9700. LOL indeed.

Sep 28, 2017permalink

@mgrum Irony!

Sep 28, 2017permalink

https://s3.amazonaws.com/masters.galleries.dpreview.com/3685238.jpg?AWSAccessKeyId=ASIAIEL32FRMEAZLCBMA&Expires=1506624192&x-amz-security-token=FQoDYXdzEJL%2F%2F%2F%2F%2F%2F%2F%2F%2F%2FwEaDEPE5fi9fOrHDmXmGiK3A%2FOesiY9I1LbjUUTFxQcGWQ3ZFvev2x76%2FJ3RX1YLA3jZWzIMFzuCbjw1EeF5wfUdsTyxIY9Q0oJWH8eiZRYjQTFnMdWWcOKttl0Ib1dOnCE3qNohGb75zOAfM93kisqxwOo9kHjIlcf4nV%2FG%2BibJ4UoMpuFPJH3hqJlAyC3wQEx%2BVRkOyyT%2FbUvfJsRQ6FtSb7XBT6pYw3tNdWcfYCqLrpM%2Fw2hy23b1K1zYEv2MMDqha6FNvj32cna7XbBDxvB%2FwD0y3n8BKgn14y7uon%2BUtEZ6kyVFwq%2B4qTmZQPVBKTEaRSyFNXQo3O8iWXKQh6J1EXc34OeJfGG%2B1KY4Uroc98Pyb3UrnbbxfSv0E%2BdDv%2ButVSZiPYRkSAgTgbZ4Od4FchvZfjif2882X4K%2FRgzfZm14jleQ4XlrBwH%2FV0p10u4yNtPZDESLlgXdDIl6TmRKnmba5KH1pRjD81rP3hPJjCLPu9jiC6o8%2FBeE4LOXTY7smRbR5zFvMHtrV1%2BYqUP%2BN7dmUkUzecRnTHyCjPpr4llopOOGA7pPPq6Jmntk4iJcSRwk%2B9OxNmrNdeZbRufmTXTmFGOAccom9G0zgU%3D&Signature=aQAj%2FowfnNPi2dwWsEkuuKk6%2FAg%3D

test between the SD1 and D810. The D850 would even be better. the less detailed shot is the SD1

Sep 28, 2017permalink

@The Davinator: You needn't feel so threatened. Praise for the Foveon isn't an attack on your manhood.

Sep 28, 2017permalink

Who said I feel attack. If someone said 2+2=5, I would post the evidence that it isn't so. Would you then ignore the evidence and discuss my manhood, or would you review the evidence instead of ignoring it? I'm beginning to see what Foveon fanboys would do.

Sep 28, 2017permalink

@The Davinator: Nobody said 2+2=5, so you can stop trying to prove to Foveon fanboys (and, by extension, reassure yourself) that Nikons take good pictures. This is a known thing.

Sep 29, 2017permalink