Sigma SD9 Review Posted

Phil Askey at DPreview has just posted the Sigma SD-9 review. For all those interested in how the Foveon sensor delivers - this is a must read:

http://www.dpreview.com/reviews/sigmasd9/

Cheers

James Wiseman

How many of you doubted my opinion? I told you it was the %^&&*$£'s

I can hear Nikon Canon Olympus and Fuji screaming AAAAAAAAAAGH!!!!!

with the Kodak RAM eater obsolete before it reaches the shops.

At only $1799 how much do you think the resale value is on those current digital still cameras?

One good thing about 35mm - No matter how advanced a cameras development becomes - It still uses film!!!

All we have to wait for now is the selective photochromic metering system...

Actually, like Phil I was surprised that the details of the SD9 image does match those of the D60, but the SD9 still looses on noise and dynamic range. This can be seen in Phil's comparison photos on page 20 of the review. As Phil pointed out, the SD9 image lost the color of the golden tip of the towers due to clipping which implies that image is slightly over exposed. That of course is not necessary the fault of the camera, but it does mean the exposure of that image cannot be increased. The next place to look is the shadow area that cuts across the boat's red paddle wheel down near the water. Zoom in 2 to 3x and you'll see that the SD9 image has more noise and less contrast and details. So overall the SD9 has more noise and lower dynamic range than the D60.

As far as being a RAM eater, the SD9 is putting out 10.3M pixels of data per shot vs the 6.3M of the D60 and the 14M the Kodak will have.

Given that it cost $400 less than the D60, it's a good deal. However, the economics of Silicon is dominated by chip area, and the SD9 chip is only slightly smaller than the one in the D60. Foveon sensors require special process steps not need in the D60's standard CMOS chip, so I doubt that it really cost less to produce. I think either Canon's put more $$ into the rest of the body or Sigma is just taking lower profits. It's great if they can force Canon to lower their prices, but I don't think they will wipe out Canon and Nikon.

35mm cameras are cheaper and may last forever, but film and processing cost money. On my recent two week trip to the Galapagos, Steve our trip leader came home with 60 rolls of film, and I took that many shots myself. His film and processing cost would more than pay for the depreciation of a D60 for a year.

Actually I thought the review was quite positive and the reviewer felt the noise performance was comparable to the D60. The image quality seemed superior both to me and to the reviewer. The one takeaway I got from the review was that Foveon cannot be dismissed as a technology. I'm eager to see its market adoption and its followon.

The SD9 does not produce 10.3 M pixels of data, either. It is 1/3 that.

The SD9 has 10.3M sensors at 3.4M locations. The raw image informations is 12bits x 10.3M vs 12bits x 6.3M for the D60, see the top of pg 11 in the dpreview review.

Yes. Phil seemed quite possitive. I think he is right about the resolution. I disagree with him on noise. I don't think shots at mid intensity range are a good way to measure noise. It's needs to be done at low light levels. See the review at imaging-resource

He did not comment on dynamic range.

"... As you can see noise levels are really quite similar, the most notable difference is that the SD9's noise is spread fairly equally throughout the color channels (as we would expect) where as the EOS-D60's noise is more visible in the red channel.

The other thing which isn't immediately visible is the fact that the SD9's noise has a finer 'grain' to it, that's because each individual pixel can be affected by noise without necessarily affecting its neighbour, unlike with a mosaic sensor..."

Note that ISO sensitivity is on the vertical axis of this graph. As we noted above the SD9's noise is fairly equally balanced between each color channel, unlike a mosaic sensor which tends to have noisier red or blue channels.

The EOS-D30 is still representative of the best of its sensor resolution, thus this is a good comparison of X3 versus Bayer at the same output image size. As you can see the SD9 performs much better than the EOS-D30, delivering more detail, sharper detail as well as elements which have become invisible on the D30 image. That said the D30 does an admirable job considering the considerable difference in the number of photo detectors, the SD9 has 10.3 million, the D30 just 3.1 million. The important difference here is that where the D30's clever interpolations algorithms have to guess details the X3 simply captures.

As you can see the SD9 image looks sharp and detailed, it's difficult to find any part of the EOS-D60 image which is actually exhibiting more detail than the SD9 and in some places the SD9 manages to pull out detail lost on the EOS-D60. Color balance is similar although it's worth noting the SD9 loses the color at the golden tip of the towers (color clipping problem).

(The rep stated that this would be sorted before hitting the shops)

You'll also note that I'm not doing any additional processing, you could of course apply an unsharp mask to the EOS-D60 image to improve sharpness (although also introduce new artifacts and amplify noise).

Theoretical reasons why the SD9 images are sharper / exhibit better resolution:

No interpolation between surrounding pixels, each pixel has its own luminance / color detail

No low pass (anti-alias filter) to blur detail beyond the Nyquist frequency

Pixel pitch is larger therefore lens is more capable of delivering adequate resolution...

I thought it may be interesting to see how well the SD9 image enlarges. To do this the SD9 image was output from Photo Pro using the 'double size' option (creating a 4536 x 3024 output image, click here to download - 5,285 KB). This enlarged image was then loaded in Photoshop and reduced in size to match the EOS-D60 image size (3072 x 2048). Crops below were then taken from the original EOS-D60 TIFF (as above) and this newly created SD9 enlargement. Crops were then magnified 200% (nearest neighbour interpolation).

At this stage I'm thinking "this is getting interesting". As you can see the SD9 image enlarges quite well, and it shows that it's carrying at least as much resolution / detail as the EOS-D60. I for one did not expect this result but am pleasantly surprised.

What I like is that the SD9 can resolve very fine repetitive details - like small fish scales, or feathers in a featherduster. The Foveon sensor can ACTUALLY represent a 1 pixel wide line correctly - with no aliasing. Check out the samples, it's pretty cool.

On the other hand, I am very worried about the Foveon going the way of the Betamax VCR or the Netscape browser. The Foveon technology may be better, but if Sigma is the only camera company to adopt it, it'll go the way of the dodo.

The final output of the SD9 is at the ~3 megapixel size - but it resolves much more detail with those 3 million pixels. It they could pump out a 6 megapixel (final output) sensor, the other camera companies would run screaming.

My local camera shop has an SD9 - I was by there on Sunday looking for a +4 diopter. I didn't get a chance to play with it, but I will next time.

Cheers

James

It just doesn't matter. Yes, the quality of the out of camera raw image is better. This is one but factor. I don't see people flocking to this camera.

Sigma made a strategic decision to use its own mount. They will live or die by this decision. I think they will die. It will not gain appreciable market share from existing Canon or Nikon users because of the investment in glass. I have seven Nikon lenses and a Speedlight. I'm not dumping them for a camera of arguable marginally greater performance. Nikon and especially Canon has great glass across the board. Heck you won't even be able to buy a Tokina or Tamron lens for the thing.

There are also several negatives which are not problematic underwater, but would be a major issue for top side shooting.

1) ISO limited to 400. If you've ever shot movement hand held, this won't cut it.

2) Long exposures at 200 and 400 limited to one second. This can be a killer for low light situations, e.g. pre-dawn. You will end up shooting at ISO 100 with longer exposures (up to only 15 seconds)

3) No shooting JPEGS. I shoot mostly RAWs, but when I'm taking snapshots that will never be printed or printed large, or need to shoot continuous, or need to shoot lots of images, I'll shoot JPEGS.

4) Minor quibble. The LCD cover covers several buttons. What kind of design is that? I leave my D100 cover on all the time. You need to work all the buttons. I know if I take the LCD cover off it will gone forever in 60 seconds. I haven't looked it up, but Nikon probably charges $999 for a replacement cover.

I wouldn't expect a housing any time soon.

I think you may find the SD9 has a couple of housings lined up already, by Ikelite & Subal. I have 2 on order from Subal myself.

Regarding the lenses: Yes, you have a point but, Newcomers especially have the freedom and those upgrading from Nikonos or S & S systems are in the same boat and are going to opt for the Sigma. Professionals too will go to the SD9 whatever and use the older technology as a back-up - I have been using Sigma proglass for over 20 years and I defy anyone to produce glass with sharper images at under 6 x the price.

If all the arguements about competitions are anything to go by anyone entering a SD9 will, without doubt win Hands down. Especially superior will be its ability to make good use of the sun graduations in composition, and you cannot get a better UV filter than the sea!.

The majority of cons of the Sigma listed by Phil will mean squat underwater. The size doesn't matter either, afterall it's going straight into a housing.

But rest assured all of the major companies will be opting for the X3 within a year. It's a good time now to invest in Foveon stock!

It is compatible with the Microdrive - it says so in the review. As well as CF I and II.

Cheers

James

That's great news!

I can hear Nikon Canon Olympus and Fuji screaming AAAAAAAAAAGH!!!!!

with the Kodak RAM eater obsolete before it reaches the shops.

Funny, I don't think that I'd go so far as to say "obsolete". One really *small* inconvenience is that the SD9 has a focal length multiplier of 1.7 according to the review.

Personally, I find the 1.6 on my D60 to be too much on land, and even the 1.3 on a 1D would be annoying. One of the primary reasons I would move to a Canon 1Ds would be to get the full frame sensor, not just the added resolution.

Factor in wide-angle, I think it's going to be a tough sell to many underwater shooters.

The majority of cons of the Sigma listed by Phil will mean squat underwater. The size doesn't matter either, afterall it's going straight into a housing.

Uh, well, the size does matter (ok, that was pretty cliched). My back is already feeling the weight of the gear I've got, and packing space is always an issue. One of the reasons I want to house my D60 is so that I don't have to carry my Sony with housing for UW AND my D60 for shooting on land (no that has nothing to do with the size of the Sigma).

Newcomers especially have the freedom and those upgrading from Nikonos or S & S systems are in the same boat and are going to opt for the Sigma.

Maybe yes, maybe no. There are a couple of scenarios to take into account for here.

The first is a complete SLR newcomer, maybe someone coming from a film or digital point and shoot. This person's first SLR might be a film one, since the cost of entry is very low, and would most likely be either a Canon or Nikon. Factor in a couple of cheap lenses for that film body and you've got an investment in a lense mount, however small. Now, upgrade to a digital SLR, and what do you look at? Personally, I went for the Canon because I had 3 lenses, all of which were fairly reasonable in price. I would have loved to buy a Fuji after all the feedback here, but i couldn't justify replacing the lenses in my mind.

The other scenario I would consider is a Nikonos user who wants to get a dSLR. I can't speak from experience here, but probably half of these users have some investment in film SLRs, and therefore again, an investment in the lense mount.

Just my opinion, but the percentage of people who are buying a dSLR who have NO investment in any particular set of lenses is probably fairly small, less than 20%? I'd say that makes the adoption rate of the Sigma a bit lower. Now, if all the Canon and Nikon mount Sigma lenses were to work with the SD9, that might be compelling!!

I think the same adoption issue may arise with the 4/3 lense mount that Olypus and Kodak introduced. I agree with James that technology doesn't always survive because it's the best option, it needs mass acceptance.

That being said, I hope that Canon and Nikon jump onboard with the Foveon technology, because the more companies that adopt it, the better it will be! And at a minimum, hopefully they ARE scared, and it affects the street price on the top of the line dSLRs from all of them!!

Just my $0.02

Tom

Dpreview.com just posted an update to their review on the excess noise in red hues . Have a look.

To see the manifestation of this problem in a real photo, download the big jpg comparison photos on page 19 of the review and zoom in on the red paddle wheels in the shade.

Tshepherd, you almost described me to a tee. Former and current Nikonos shooter, had purchased an N90s to put in a housing some day, accumulated several lenses for Nikon, decided to go dSLR instead of housing the N90s. This was a huge selling point for the S2 Pro.

You finally mentioned the 1.7x crop. I'm still trying to figure how to cope with the 1.5 on my Fuji.

Another important thing, the lack of a built in strobe. I actually use mine a lot in places I prefer not to take an external strobe.

I have a Canon D60 but am not married to it; I'd get a new camera system if there were some reason to.

I saw nothing in the Sigma SD9 that would make me want to buy one, even if I had no camera at all right now.

I don't see the Foveon being used by any of the major manufacturers, it doesn't appear to offer any significant advantages and does come with serious downsides, and Canon in particular has a big investment in their own technology.

I'm a bit amazed that people actually care what the sensor is; what matters more to me is image quality, camera feature set, and lenses. The SD9 has a long list of areas were it is, IMO, not nearly as nice a camera as either a D60 or a Nikon D100. The sensor used is not on this list.

The majority of cons of the Sigma listed by Phil will mean squat underwater. The size doesn't matter either, afterall it's going straight into a housing.

I agree, but the success of the camera, any land camera, will have little to do with it successful use underwater. Size doesn't matter much. An Ikelite MD case is huge to begin with. It's the same size case regardless of the camera inside.

I don't just shoot underwater, so some of those cons matter in making a purchase decision.

Regarding the lenses: Yes, you have a point but, Newcomers especially have the freedom and those upgrading from Nikonos or S & S systems are in the same boat and are going to opt for the Sigma.

This alludes to the marketing decision by Sigma with regards to the mount. As Tom mentioned, that decision targets users without the glass investment, upgraders to SLR. This decision made a lot more sense when it was anounced, although I still question it. Now that the camera has arrived late to the party, there is a world of people with Canon and Nikon d-SLR's who aren't going to be switching. Plus as "prosumer" digicams creep up in performance (e.g. Olympus E20, Minolta Dimage 7Hi, Sony DSC717), those who might have upgraded into the SLR world shrinks.

Professionals too will go to the SD9 whatever and use the older technology as a back-up - I have been using Sigma proglass for over 20 years and I defy anyone to produce glass with sharper images at under 6 x the price.

First, most professionals already have land based SLRs, even if they shoot Nikonos. Most all will have Canon or Nikon. They will already have an investment in Canon or Nikon glass.

Second, if I concede that Sigma has the best glass at under 6x the price of Canon or Nikon, so what. I don't buy the best glass under xxx dollars. Also, the best Sigma glass isn't 1/6th the price of Nikon glass. Cheaper, but not 1/6th. Example: Nikon 18-35mm $500. Sigma 17-35mm $436. Prices from B&H

Third, all those Sigma lenses you've been using for over 20 years...what mount are they? Are you willing to buy them all again? Even the majority of Sigma lens fans own Sigma lenses with another mount. They will have to start all over again.

But rest assured all of the major companies will be opting for the X3 within a year. It's a good time now to invest in Foveon stock!

Foveon is not a public company.

The 1.7 crop is another major reason this camera may be shunned, especially by the pros. Wide angle definitely isn't wide enough. The 20mm lens favored by most pros now becomes a 34mm lens. This isn't going to cut it.

It is my opinion that there are problems/tradeoffs in lens performance and price performance when you go below 20mm in a rectilinear lens. Sigma produces a 14mm lens for $800 that would give you 24mm at 1.7. That isn't going to appeal to many pros. Add to that issues of flaring and chromatic aberation common to that wide a lens and I don't think it is all that appealing.

Another thing that concerns me with the Foveon technology is product life cycle. When Foveon anounced their new sensor, it was on the leading edge of pixel technology. Since that time we have seen the advent of the 6mp sensor, now the 11mp and 14mp full frame sensors. It has taken Foveon/Sigma a long time to get this camera to market. I'm not big on pixel counts, but I do see more interest in the full frame sensor. The existing sensor makers continue to innovate, but Foveon is still on the same generation as anounced over a year ago.

I once worked for a company within a family of companies. One of our sister companies had developed a new technology for manufacturing disk drive platters. It was far superior to the metal oxide sputtering technique being used at the time. Years passed by as this new plating technology was being perfected. It had the promise of being able to exceed the 40mb limits in existence (yes, that's megabytes not gigabytes) In the mean time, the old technology continually improved to the point that the new technology was of no benefit. The factory was shut down.

As I would expect most of the negativity would come from people who have already invested in technology preceding the X3.

It has been pointed out this older tecnology has now advanced to 6, 11 & 14 mp, but at what cost - even if the performance could match what the x3's can at under 4 mp?

Regarding lens investment and costs: I was talking "Glass" and not the plastic equivelent developed for less strain on the autofocus motors.

Yes, these have unacceptable aborations.

Working in a variety of formats my lens investment has been considerable, but staying in 35mm I use have predominantly used Minolta cameras since the inception of the 9000 in 83 to my present day Dinax 9's. However, I have Canon Nikon and a Hasselblad too.

The net investment on exta lenses for use underwater is minimal anyway - maybe 3 + a few diopters and an extension tube. Most newcomers would probably opt for just 1 zoom coming in at $250 - $450.

To draw on a suitable analogy - Spanners in a toolbox. There was probably a similar arguement following the invention of the hexagonal concept.

Professionals for a long time to come will still be shooting film for the majority of their work, but they will, and are using digital systems in areas it excells like Press photography where speed to hq is important. We are also using the digital technology as a replacement for pre-shot Polaroids - here it's saving a fortune.

While we are on the subject of film: You should also check out the advancements in this technology too before blindly insulting its future potential or its very place in future photography.

A 16mm lens on the SD9 will still offer 105 degrees without adapters and the abilty to be just mm away from a subject.

Why I, and most professionals want digital underwater is to shoot just macro & close-up subjects, but with the added bonus of the X3 wide angle has potential now too. However this will still depend on a photo editors decision to accept digital files which will have to be in CMYK.

Foveon is not a public company yet, but I reckon it will be soon and this superior product is not limited to just Sigma. Foveon are looking to improve on their baby and will be looking for major investment, and now proven it will arrive soon.

Yes, it's "proven" not arguably: Proven

I see no one has argued about its competition potential yet, and the probable arguement about "If they are good enough as photographers!" Check out submittions by 'Slippery', she has ordered an SD9 too after speaking to Phil on Monday!

I don't think many are arguing about the results from the SD 9. I think the discussions are leaning toward their packaging of the new technology. I'm not a marketting guru, IMO this seems to be a bad decision. Any time you sole source a product, you narrow the market.

I must say I weighed up all of the pro's and con's regarding this camera after reading 6 reviews by experts in the field and consulting with several of the writers of these reviews. I did this because buying a camera system is a major investment for me and I wanted quality but value for money too.

After all of this I contacted Jessops in UK and ordered one, together with 2 zoom lenses. The Jessops head office in Leicester told me that I was on a waiting list of 1641 (paid up) people, but every one would be delivered ASAP when their bulk stock arrives during December.

On to the continuous debate about marketing, please guys I am the women here! Look at it in simple English.

SD9 with new technology £1299 - $ 1799

Canon 1Ds £6999 - $10500

Nikon D100 £1699 - $ 2500

The Kodak 14 Megpixel release, according to Jessops, is now looking doubtful.

I can buy 3 SD9's a housing and and a load of lenses for the same as just the Canon 1Ds body What is any sane person going to do given the choice.

Whine all you like but these are real unbias facts. A whole host of new lenses will be released shortly too obliterating the 1: whatever problem.

This was the quibble over APS and 35mm some time ago.

Price and quality dictates its marketing value - trust me I am in the business.

P.

SWW, nobody is whining, merely performing constructive criticism as this forum was meant. I'm sure you will be as excited to get your new camera as I was. BTW, The S2 Pro, D100, and D60 all had waiting list when released. Many waited for months.