Sony alpha A55

[Greg Beetham]

Well how does that place a camera like the D3, a FF cam with 12MP's, that puts it into the KM5D realm from a density point of view does it not? and therefore likely to produce similar P&S quality I assume. (Argorabasta you can see the thicker edges with the naked eye? or you have developed a way of recognizing that, I can't see them without magnification myself)
Greg

[agorabasta]

A 12Mp FF sensor sure is a totally oxymoronic thing. But my point was that the system res must be limited by the analog rather than digital parts of it (limited by a smooth rather than discrete function). So a 12Mp FF sensor should have a very strong LPF, and it's exactly the case with D3/D700.

I sure do see the thicker edges without any additional magnification. And you do as well, but most likely you simply do not notice. The thing to look for is e.g. the relation between the width of some thin twigs or grass blades and thickness of their edges. In your KM sample some in-focus twigs look like they consist mostly of the edges, while in the a700 sample those same twigs have real body and a very thin edge. It's all a pretty macroscopic effect, noticeable well without close peeping.

[bfitzgerald]

Not sure I'm seeing these niggles. Honestly there are a few things on the 5d that could have been better (the overexposure flash bounce at WA drives me nuts but I can work around it) IQ isn't really on that list. It's one of the few cameras I've tried that I can really say does an excellent job for skintones. Whilst the reds can cause problems at times I've yet to see any Sony DSLR have the same tonality or deliver the same skin tone quality. I'm not saying the A700 is poor or I dislike the "Sony look" it's better than some of the plastic tones I've seen from some other makers. But the colour rendition in many respects is superior to anything I've seen at any price Alpha DSLR wise.

I never found the A700 tones very appealing not bad but nothing stand out. The A100 was simply mediocre at best in this regard too. A200 is ok ish but no match for the Km in this area. Purely subjective maybe it's the film thing in me, sometimes you like one, other times it does nothing for you at all.

[Greg Beetham]

Actually after putting my reading glasses on...again I can just see some of what you are saying agorabasta, but I also noticed something else that I've never seen before from the KM5D, some blotchy artifacts here and there. I know my NVIDIA card and VISTA are currently engaged in a skirmish of unknown origin, (an update issue maybe), with the monitor profile sniping at everything in sight from the sideline...including E8 and the printer. Despite this background spat, I think I might have to dig up the KM software for the KM5D and try that, just too see if I can get rid of the artifacts.
Greg

[Greg Beetham]

I re-did and posted another copy of the KM5D photo, I went fault chasing, tell me if it's better than the first one pls.
Greg

[agorabasta]

Greg,

Your second try looks much better, quite definitely. If considered outside of comparison against the a700 sample, it shows no negative signs of insufficient resolution. Still, if a700 sample is considered along with that, you should see that thinner objects look slightly thicker in the KM sample.

The effect would be even more pronounced if you shot some brightly lit fine grass lawn in low evening sun. The brighter the edges - the more pronounced effect there is. Then if those bright edges turn into saturated highlights, the effect turns very immediately noticeable.

I should say that I noticed no red blotches in your samples. I think a 10-bit LUT really helps my monitor
You could check if the blots go away with colour management for your monitor turned off temporarily. That blotchiness is often a result of posterisation by the monitor LUT, that's a very common problem with automatic calibration. Even a 10-bit LUT with a true 10-bit panel may simply be not enough.

[Greg Beetham]

I think personally that the KM5D image is verging on being better than the A700 image quite frankly. No personal bias involved, just flipping back and forth between the two. But anyway as far as that particular image goes I think we could safely say there was no advantage in posting images on the web using a 12MP camera in favour of a 6MP cam.
Agorobasta, I went off scratching my head wondering what was wrong with the KM5D image, I deleted the monitor profile and went too the parent mfg website and found an updated (slightly) one, and a new additional calibration tool, but none of that turned out to be the real problem, (even including the trench warfare between Vista and the 8500GT card).
I traced down what I did... When I was preparing the two images originally, I realized I had forgotten the tag data, I had already gotten rid of the filenames and exif and saved them for the web, then I had too hastily re-do them from scratch and completely forgot that the KM5D's image is different too the other two cameras in ONE respect, I have to remember to set the colour space to s-RGB, something I don't have too do with the A100 or A700.
So that's what happened with the first image it 'went too air' still set too Adobe RGB....duh.
Greg

[mike2008]

I see the ex-minoltians have turned this A55 thread into a KM5D fanfest... Nice.

[Birma]

I see the ex-minoltians have turned this A55 thread into a KM5D fanfest... Nice.

Ah, but Mike, if only every gear thread resulted in someone putting the effort in to modelling some results, taking some constructive comments, reworking and coming up with something (imho) that is interesting and useful . Hard to stay strictly on topic for 17 pages I guess

[InTheSky]

The thread is that big that it can be a category by itself .

I was to order a A55 right now because of the announced aperture control with the camera, but now knowing has rumor that the Nex 5 will support aperture control, I'm little bit in waiting mode now. I probably Highly recommend people of not doing like me and ordering to fast stuff.

Talking of old camera by the way, I never found again the flexibility of my old Minolta 7D as control experience, but I'm still enjoying the Nex 5 even if it is a really not fast setting control.

On other side ... as myopic even with the GREATEST viewfinder of the A900 ... I can say that live view technology with magnification is unfortunately better for me ... so the A55 has to come in my hand soon for macro and manual lenses use.

Frank

[stevecim]

with high ISO , video with AF and live view the A33/55 would make for a good underwater camera? (+ underwater case )

[bfitzgerald]

I wonder what happens when the translucent mirror starts to accumulate dust and particles?
I've been looking at some A55/A580 images and there is def more softness to the A55 ones v the traditional DSLR model. I think it's a bit more than pixel peeping here, and I've got some doubts about the long term effect here.

It's a bit like putting a filter on all your lenses I've seen an IQ hit on that as well now you could argue it's not significant but surely a more logical design would allow photos to be taken with the mirror raised up?

[alphaomega]

Ref Barry above, I also have my reservations as well about the A55 producing as "sharp" images as the A580 because of that fixed mirror although I cannot produce any evidence. I would like whoever get their hands on an A55 to compare its A/S performance against a DSLR. I have a suspicion that if the A55 was pitted against the A580 both taking the same evenly lit picture with same ISO and Aperture, the A55 would show a lower shutter speed because of the light loss in the fixed mirror. How much? Could be 1/3 or even more of the light. Could be critical at lower light and needing a higher ISO setting to compensate.

[agorabasta]

Dust on mirror is no problem whatsoever - it's too out of focus to show in the final image, much like small scratches/blemishes on lenses. And then you just blow that dust off. There would be a real problem from tar/oil condensing from some kind of smoke like from smoking or cooking around, that would require wet cleansing.

[mike2008]

There have been lots of comparisons already showing tiny differences in IQ between the A55 and traditional DLSR's. IQ is the same. A blower is all you need to get rid of dust. At the end of the day, the people who like the EVF and compactness of the A55 will use it more than a heavier bulkier camera, and the IQ will be indistinguishable.

I've come to the conclusion that pixel-peeping is pretty much dead. All the cameras are good for IQ. The fundamental limit is the user, just like it was for film. If you make the camera fun and easy to use as well as light and small with great IQ then it's a successful camera. The fact is upsets a few people who totally fail to produce any real eveidence about IQ issues matters not a jot.