which lens to buy?

[agorabasta]

So the LA-EA2 has mirror up capability then agorabasta?

Sure it does! Just use a LA-EA1 instead!

[agorabasta]

I'd be inclined to add another proviso, one that covers the capability of the lens in having a sufficient LPPM/MTF for the 24MP APS-C sensor as well.

No way. The best imaginable digital system is the one that is always limited by the attached lens properties, however good the lens may be. It would simply mean that the digital part is no longer limiting the optics, as it used to be since the digital inception and still is.

[David Kilpatrick]

Re macro - I have the micro lens 25mm f/2.5 as well as the 30mm and a 70mm Sigma. I can confirm that stopping down for the required depth of field can have a drastic effect on sharpness even at 12 megapixels. Of course it's worse at 24MP.

Re landscapes - grass, trees and also fur or feathers show and problems straight away. I have noticed that Sony's NR is very good indeed at preserving sharp edges, and will sacrifice lower contrast textural detail long before you see any reduction in the quality of contrasty shapes. So it can do pretty well on a test target when a forest view will look like mush. And close-ups always look better.

What I'm concluding right now is that I have been led astray If I use exactly comparable quality lenses, and exactly comparable technique (low ambient temperature, tripod, mirror up, no long period using live view to warm up the sensor, lowest ISO, brief not extended exposure or using flash, optimum aperture, fine-tuned focus - etc) then I can get equally good results from the A77, the A900, and a Dalsa 22 megapixel 645 sensor.

That is, I'm wasting my time and money experimenting with MF digital at 22 megapixel level - both the 77 and 900 will match it. The reason they do not match it in practice is because I use the cameras differently, with convenient zoom lenses, taking loads of shots, using high ISOs and so on.

Conclusion: while the equipment has its own issues, limitations and peculiarities ultimately it's down to the photographer and their ability to fine-tune technique.

Except for that macro thing. I do not have any technique which can overcome the problems of diffraction at the very small effective apertures used for macro beyond life size. Wider apertures and focus stacking? That's how many of the best macro shooters are now overcoming the problem.

David

[agorabasta]

David, overcoming the diffraction is pretty straightforward, it has been done for years in semiconductor manufacturing using the things called 'steppers'. You can use the essentially same technique by taking a long series of slightly shifted diffraction-blurred images and stacking them in PhotoAcute or any other similar proggie, if exists. And it's better to make the camera vibrate slightly - e.g. putting a loudspeaker next to it, or using some dedicated actuator to shake your bellows/cam as needed.

[Greg Beetham]

Ha ha agorabasta you're a whacky funster, vibrate the camera with loudspeakers, what should we play Beethoven's 9th or a light sonata ha ha.
Greg

[Greg Beetham]

No way. The best imaginable digital system is the one that is always limited by the attached lens properties, however good the lens may be. It would simply mean that the digital part is no longer limiting the optics, as it used to be since the digital inception and still is.

Thing is, the diffraction dot coming from the CZ16-80 at f11 is the same whether it’s on the A55 or the A77 but the A77 is getting worse results than the A55, what has changed? Not the lens that’s for sure, the only thing left in the equation is the camera…once you subtract the lens.
Greg

[agorabasta]

... the A77 is getting worse results than the A55, what has changed?

You may blame the camera as a whole, like you do. But you should not imply that the pixelcount is the culprit, because it's not. It's the bad sensor toppings design, most likely.

And then there are very few lenses showing real diffraction. Just some rare primes come close to that. The resolution falls first from the haze caused by glass/coatings imperfections and the aperture iris edge diffuse reflections, like those 'stars' you get around an image of a bright source shot at tight apertures.

[Greg Beetham]

I'd like to see if there is in fact any lens in the Sony stable that performs better at f11 on the A77 than on the A55.
Greg

[agorabasta]

Greg, I'd guess that some Sigma 70 f/2.8 macro has a very good chance. That's the lens they use at IR for many test shots.

[agorabasta]

One more I wanted to add - the different sensor toppings are really quite capable of very strange things. I have a Samyang 85/1.4 that remains bitingly sharp at f/13 on Nex5 and C3, but it looks like mashed potatoes on my 5N (same 16Mp as the C3). Even my 14mm Samyang is a so-so lens at f/11 on 5N, being the best thing since sliced bread on all 300/700/550/55/5/C3 I own or have around.

[Greg Beetham]

So what we need to find then is someone with a Sigma 70 Macro and both the A55 and A77 to do some comparison shots to see if there is at least one lens the A77 can beat the A55 with at f11.
Greg
ps I still don't get why Sony would make things worse instead of better.

[Dusty]

ps I still don't get why Sony would make things worse instead of better.

Because they overreached by trying to get the most MPs on APS-C the fastest. If they do an A78, it will be better, because they'll have learned from their mistakes. When they said it was an Alpha77 we should have looked at the real meaning of their words. Maybe the Beta77 will be a bit better!

Dusty

[agorabasta]

ps I still don't get why Sony would make things worse instead of better.

But everystupidboby and their mother-in-law are oh-so-very happy with the 5N and even more so with its JPEG output. That's the prob. And then what Dusty just said.

[David Kilpatrick]

Greg - I would add that if you unsharp mask the full size A77 file correctly, then reduce it to the same size as the A55, the result will generally be better. The A77 can produce better images, it captures more data.

I have a specific problem, called Alamy. I submit images which are all examined at 100% pixel size. The smallest file size accepted is 24MB, or about 9 megapixel - but for various reasons there can be be big money made from very large files (over 75MB, or over 25 megapixels). Generally, if you can provide a large image size, it has more potential uses.

But - the images are examined at pixel level to check for sharpness (not all, a sample only, but if one single image is checked found to be soft out of hundreds sent the entire lot will be rejected). You can not afford to have a single image where the main focused zone is not pixel perfect.

My problem with the A77 right now is that the full size files just don't manage this all the time - the A55 has a better hit-rate, just process and they are fine. With the A77, I often have to throw away the benefit of the large file size and scale the image down to 9 megapixels or some other size smaller than 24. So I'm processing loads of data, eating card space, spending time... when a perfect shot off an Alpha 100 would be just as good!

David

[agorabasta]

... when a perfect shot off an Alpha 100 would be just as good!

Very true.
Got quite a few lenses that work best on the older 10Mp CMOS bodies, especially at ISO100.