A900 and the future - some questions...

[bfitzgerald]

This is what I found on Dyxum

The January issue of Nippon Camera, a Japanese photog mag has an interesting interview with Mr. Tachikawa of Sony, who specifically works in product planning. He is involved with both the Alpha line and the NEX line. I won't translate the entire interview but there is one particular section that might be of interest.

Question: The A900 has gone out of production now. Will you no longer be using the reflex mirror?

Answer: We intend to listen very closely to the voice of our customers. I can't say for sure right now, but we do believe that the translucent mirror technology is capable of being used in higher end models. Nevertheless, we have no intention of dogmatically rejecting reflex mirrors or insisting that the new technology always has to be used.


IMO APS-C is just if not more important to Sony and as they don't have a current APS-C OVF Model that pretty much signals the end of that (so much for listening eh?), unless they release one next year (not very likely) As for FF well that's anyone's guess take your bets now.
I'd say they are sitting on the fence to see how SLT is selling, on that one I've no idea at all.

[Greg Beetham]

Dear photoclubalpha members,

I've now had the A900 for 3 years, and I have some questions about the future, and based on the image quality I get from the CZ 24-70 and the A900, how would you answer these fundamental questions I have:

1. If Sony no longer has a FF OVF camera, and I get eye strain looking through the A77's EVF, and prefer OVF's, how much longer (on average) will my A900 last, if I take about 2,000 to 3,000 pictures a year?

2. If Sony stops making the A900, how much longer will they service my A900 - that is, new shutter if needed, etc.

3. With the Nikon D800 possibly being 36mp and with OVF AND Sony not making a OVF comparable model, should I switch to Nikon if OVF is a key feature for me?

4. If I just love my A900 and CZ 24-70 system - should I just keep it until it breaks and Sony stops servicing it - approximately how many years would that give me - and at that point in time, should I just make all these decisions on all these questions I'm asking?

You see, Sony has put me in a spot - because with the A900, it gave me a great camera with the OVF and user interface - but now they are possibly going a different direction. The only caveat I have to all of this is that cameras like the A77 do have sweep panorama, handheld twilight, and handheld auto HDR - but only in jpegs, but assuming they can make a great camera with great jpegs - that could bring me back, as Sony does some pretty innovative things, but just don't seem to focus on basic quality of the basic fundatmental camera features, per se.

Thank you for any input - I know my CZ 24-70 should last a lifetime, but will its new Sony bodies to mount to in the future be as worthy as the A900 for me!!!

Well if I liked the A900 as much as you seem to, I’d seriously consider buying another new one now, give it a test run and then put it back in the box for later.
Greg

[artington]

Well if I liked the A900 as much as you seem to, I’d seriously consider buying another new one now, give it a test run and then put it back in the box for later.
Greg

That's what I did with the Dynax 7. Now I have two in their boxes for later

[Greg Beetham]

That's what I did with the Dynax 7. Now I have two in their boxes for later

Good thinking artington, I actually tried to get a Dynax7 at the time I bought my KM5D (as well) only to be told by the dealer here that the only new film camera available from Minolta then was a Dynax60 so I got that instead, I probably should have got one on the internet from one of the dealers down south, there were still some showing in stock in a few places as it turned out, I would have liked one in my collection even if I didn’t get around to using it much.
Greg

[Atgets_Apprentice]

Mr. Tachikawa was quoted saying the same thing in Amateur Photographer back in September, so nothing has changed. Just who Sony are "listening to", God knows, and why their thought processes are so slow, is anybodys' guess. It is just a case of smoke screens from them. If the "a99" is EVF, as I believe it will be, it will more likely be commercial suicide. As an FF "Pro-spec" camera, I do not see it taking custom away from Nikon & Canon, even with a price much less than the forthcoming D800, and it will not shift in great volumes, in these days of economic turmoil.

There may be a few a900/a850 users on here, but I have never encountered one in the real world, and I fully expect the same to be the case with the "a99"/White Elephant.

[deathvalleydave]

I tend to agree with the notion that an EVF A99 will not be the most practical marketing strategy, and if it has 24mp instead of 36 - what will be its benefits? Now it has video? Now it has auto-focus video? Now it does not have the A900's chroma noise in jpegs? These are all nice things and all nice and well, but not enough of an "upgrade" from my A900, unless the jpegs are so good that handheld sweep panorama and auto hdr are now more than just jpeg only "gimmicks".

I happen to feel that Sony's handheld jpeg features are very nice, but don't get much of a look from Canon or Nikon folks probably because they are not fundamental features of a guitar like pure image quality, noise, color, or autofocus points, etc. Therefore, they can easily be dismissed due to being able to shoot RAW with a tripod and accomplish the same in post processing. Yet, probably 9 out of 10 people don't want to take the time to use a tripod or are on the move most of the time, and the handheld jpeg features Sony offers really is something. The problem is that when I look at what I can do with the A900 and CZ 2470 and RAW and LR3 - the jpegs would have to be really, really, really good, to make me want to switch to being a jpeg shooter due to handheld hdr and sweep panorama.

Plus, if the A99 had 15 or 16 stops of dynamic range - then maybe I would switch, because they figured out ghosting. But I heard that improving dynamic range is much, much, much more difficult than increasing pixel count...

[jbtaylor]

Read David's review of the Zeiss 24mm 2.0. That will tell you why a 36 mp sensor with those Zeiss lenses may attract medium format users as well as those that bought the a900.
How many will have to be sold for it to not be a failure?

[agorabasta]

But I heard that improving dynamic range is much, much, much more difficult than increasing pixel count...

Four times more pixels give you two more stops of DR at the picture level. That's theoretically. Realistically, you should expect at least one stop of improvement over quadrupling the pixelcount.

[deathvalleydave]

With 36mp and possibly a one stop improvement of dynamic range - that would be a good thing. Also, 36mp with good CZ lenses would also be a good thing.

The only other thing that worries me is that I'll never want to buy a full body camera (with battery grip built in permanently to body). And if the rumors I heard are correct, the 36mp Sony might have the built in battery grip?

Would you think that if the A99 is still 24mp, it could also see a 1 stop improvement in dynamic range? Because I think the A77 and NEX 7 both have a stop more than the A900?

[David Kilpatrick]

JPEG dynamic range doesn't bother me - some of the best colour and tone is obtained if the range is similar to film, between 5 and 8 stops. What matters is raw clipping headroom and shadow recovery 'footroom' - but most of all, room at the highlight end for colour channel recovery.

The old Dimage A2 with its relatively small pixels did this amazingly well. It was possible to recover sunsets with the sun in pic without a hint of channel clipping. I shot the Koutoubia Mosque in Marrakesh at sunset on the last day of Ramadan in 2006, using the Nikon D70S (because the D200 was not available despite this this being a launch trip Nikon did). I also used my A2. The Nikon shots were hopelessly clipped - no chance of a realistic sunset despite correct exposure. The A2 shots have gone on to sell well and were of very high quality.

I've never had any camera since with the same capacity to shoot direct sunsets and recover channel data without clipping desaturation or peculiar saturated banding.

David

[bakubo]

Good thinking artington, I actually tried to get a Dynax7 at the time I bought my KM5D (as well) only to be told by the dealer here that the only new film camera available from Minolta then was a Dynax60 so I got that instead, I probably should have got one on the internet from one of the dealers down south, there were still some showing in stock in a few places as it turned out, I would have liked one in my collection even if I didn’t get around to using it much.

I bought a Maxxum 7 in January 2002 from B&H for $600, but I ended up also buying a Minolta D7i in May 2002 and after that I have never put another roll of film through any camera. I only put about 4-5 rolls of slide film through the 7 so it is in mint condition. I never even took it on a trip. I liked it a lot and felt it was the best Minolta AF body I have had (also had the 7000i, 7xi, 9xi, 707si) and although the D7i was lacking any many ways I was hooked on digital and haven't looked back. Before the D7i I also had a couple of digicams bought in 2000 and 2001, but the D7i was the first that was flexible enough that I found myself not using film anymore.

[deathvalleydave]

Hi David,

Could you please expand on the following - you said "5 to 8 stops similar to film" - but if film has greater dynamic range than digital, I'm confused right there, because the A900 has 12.3 stops of dynamic range, correct? So, I'm just a bit confused over your explanation with the A2 - are you indirectly saying the A2 shoots into the sun better than the A900? how about the A900 mounted to the Sigma 12-24? It's a great "into the sun" camera especially at 12mm. Only the D3X and Pentax K5 have a bit more, based on DXO Mark's measurements, knowing that your real world results of recovering highlights with the A900 renders those "measurements" rather insignificant in practice...

Dave

[David Kilpatrick]

Film doesn't have a greater dynamic range. It has much less, depending on the film type. Burned out highlights are clear and dense shadows are d-max - Velvia has a very limited dynamic range. The highest dynamic range is found in films like Astia for slide, or Portra 160 portrait not commercial for Kodak negs.

Typical film dynamic range is reflected in the spot metering functions of cameras and of the earlier Minolta and other spot meters. These assume 5 stops of usable dynamic range between a shadow reading ('first level of visible detail above black') and highlight ('first visible detail below pure white') with the mid-tone grey set normally 2.3 stops below highlight, 2.7 stops above shadow.

In practice, these values tended to represent worst case films like Kodachrome (very poor dynamic range unless specially processed) or conventionally developed black and white (Plus-X in D76 full srength). Later slide films definitely added one stop either way, especially self-masking films like Fujichrome 50 which could prevent highlight burnout, and complex multilayer self masking colour neg films which could give you 'five plus five' - that is, five stops dynamic range over five stops of latitude. For b/w, the use of diluted or chemically designed compensating developers added a stop either way but even the best technique could never recover as much latitude for error as colour neg films. That's one reason Agfa and Ilford developed Vario 1000 and XP1 respectively - the original claim for C41 based mono was that you did not need an exposure meter, you could use any ISO from 100 to 1600 and print a good result.

With films, whatever range was achieved was accompanied by a very distinctive contrast curve further modified by the contrast curve of printing papers.

Digital beyond 8-bit per channel has always had an excessive straight-line dynamic range - a linear response over more than 5 useful stops of final image value. An 8-bit per channel image captured with a straight line curve has in practice just five stops, like film (it actually has six but the lowest values 0-2 are not useful).

What 10-bit, 12-bit, 14-bit, 16-bit raw capture can do is to improve accuracy, extend actual values not clipped, and allow a film-like contrast curve to compress their data into the same 8-bit per channel - much the same as compensating developer or self-masking negative films.

Everyday slide films are more or less matched by 10-bit capture (which is why the Alpha 100 often seems to be a very film-like camera, responding to exposure changes much like slide film). To match the latitude of colour neg films, 16-bit capture such as the Phase One Hasselblad back is needed. Between these extremes, 12 and 14 bit capture vary a great deal depending on how the maker calibrates exposure and uses tone curves.

The dynamic range of the final image will normally be much less than the raw capture, just the same as it is much less than the film's potential. When you try to force a full-range 16-bit capture into an 8-bit output, it can look very flat and dull unless the important centre part of the contrast curve (the 'slope' in the middle of an S-curves) is kept fairly steep. If a straight line is used to map 13 stops down to a printable 5 things which look twice as bright to our eye/memory are shown just 20% brighter.

So both the highlight end and shadow end of the 16-bit capture are either compressed (an S-curve with a long heel and toe) or clipped (anything below a certain value is 0, anything above a certain value is 255).

A bit of this is necessary even with 12-bit capture like most Sony models. In practice, they exceed what slide film is capable of doing (as DRO+ which is a single raw capture can prove any day) and match what normally processed and printed negative stock can do. You can't remove all the transformation values but you can study effects in Adobe Camera Raw, Raw Photo Processor and other programs by taking a look at linear conversions to 16-bit compared to 'film like' conversions to 8-bit.

Beyond this, you get issues such as the contrast of projection screens and methods, the Callier effect where any residual or primary silver image is in the emulsion, the difference between diffuse and condensor illumination, aerial light scatter, enlargement related contrast loss, viewing conditions, computer screen brightness and contrast ratios, graphics card bit depth, CMYK conversion values for print, d-max and paper white values for print...

The sum of all this - and I think most professional users would now agree - is that digital capture tends to exceed the recording (input) range of film, but that more digital output (printing, projection, laptops, tablets) has a lower dynamic range than the printing and projection methods used for film.

Where they end up equal is on the printed magazine page, which only has about 6 bits range anyway.

David

[bfitzgerald]

Films vary too much to make blanket statements about dynamic range. There are significant differences between film types and emulsions. Slide film is very different to negative film and requires a different exposure approach. With negative film you expose for the shadows because underexposure is in most cases to be avoided slight overexposure is preferable, you'll land in deep trouble if you try that with slide film where slight underexposure is a better bet. But there are differences here too between makers and types of film.

B&W film is IMO simply unbeatable esp legend emulsions such as Ilford FP4 fantastic latitude on that film you can be 3 stops over exposed and not even sweat, it will also take substantial underexposure far better than most negative films would.
But the DR issue is not really the point, the real world use of negative film to me has been more "worry free" for highlight retention. With digital most of the latitude is in the shadow areas, highlights still clip even on recent digital sensors..though things have improved somewhat. We are still some way off of the highlight range we need to banish "blow outs" to fairly rare events.

The only serious attempts at this have been the Fuji Nikon mount S series..and the EXR sensors found in their compacts. Having used the EXR models they have impressed me for their ability to retail highlights and avoid channel clipping. Very impressed as it happens, it works even if you have to give up half your resolution to get there.

So the problem isn't the DR is bad on digital, it's not..but most of it involves digging in the shadows and pulling images up to get there. It's simply a lot more convenient (for daylight shooting at least) to shoot and not worry about highlights. This is why I'm far more interested in technology such as EXR, than I am boring headline resolution figures..the day they nail this down and properly is going to be far more significant to many photographers than a few more pixels ever will be.

[pakodominguez]

1. If Sony no longer has a FF OVF camera, and I get eye strain looking through the A77's EVF, and prefer OVF's, how much longer (on average) will my A900 last, if I take about 2,000 to 3,000 pictures a year?

More than 5 years if no accident happens...

2. If Sony stops making the A900, how much longer will they service my A900 - that is, new shutter if needed, etc.

Someone else already say, law in the USA says 7 years...

3. With the Nikon D800 possibly being 36mp and with OVF AND Sony not making a OVF comparable model, should I switch to Nikon if OVF is a key feature for me?

Well, do you need 36 MP next year? or you can live with 24 MP and your Zeiss glass for 3 more years? by then, the new cameras, from Sony or Nikon, will be more than 36MP...

4. If I just love my A900 and CZ 24-70 system - should I just keep it until it breaks and Sony stops servicing it - approximately how many years would that give me - and at that point in time, should I just make all these decisions on all these questions I'm asking?

At the Photo Plus Expo last October I had the chance to spend 15 minutes with Mark Weir, one of the top managers on Technology and Marketing at Sony USA for the Alpha division. I asked him about OVF and he told me that, personally, he didn't know anything about it's future in the Sony DSRL. He was not sure Sony will kill OVF but he was not able to tell me if, pressed by the demand, Sony will launch another OVF camera -FF or APS.

What I understood from this conversation is that people Sony are very enthusiastic about EVF basically for 2 reasons:
1.- the experience with the OLED EVF on the A65/A77 is better than using a camera with pentamirrow.
2.- Sony doesn't have any challenge with OVF, while even if they have now a good EVF, they have a huge marge for improving it, and they find that "exiting"...

Mr Weir counter-attack and asked me if I could picture, in 10 or 15 years, DSRL cameras from Sony or any other manufacturer, with OVF.

I don't know. I think he's right. I asked him if he tough Sony took an advance over Canon and Nikon, but he didn't dare to say yes. I bump on him again at the Nikon booth and he was amazed by all the new technologies Nikon introduced on the Nikon 1 system, specially the AF. This people are serious. They are not on the "mine is bigger" kind of discussion you can find on the forums. Sony has a plan, that's for sure...

OVF will most probably become a niche "Premium" system for people ready to pay for it. I sold my 2+ years old A850 and bought a brand new A900 that I will cherish for the next X years until Sony bring that Premium OVF FF camera, in the mean while I'm planing to convert to the NEX7 for most of my shooting, including small events and as second (or first) camera for big events, reserving the A900 for portraits and family portraits.

I think you better go through the over excitement of new cameras arriving (you can not imagine how much I'm "suffering" because I can have a NEX7 yet!) and wait a couple of years with your A900, or buy a new one as someone else suggested.

BTW, I saw an A850 at the Sony outlet in San Marcos, TX for 1600 US and I was tempted to buy it!