[Nicolas06]

Sure and that is tracking, but you selected the point of focus first. What I mean is if the AF is just on auto how does the camera figure out what to focus on? There has to be something or no one with a multi-point AF system would ever get anything in focus except by luck. I get the need for lots of AF points for tracking, maybe on more sophisticated AF systems these are only used when in "tracking" mode? I guess I'm just not familiar with other systems. On Pentax I tried the multi-point, it focused on what it wanted and we agreed to disagree so I turned it off. I've used single point ever since. Maybe on newer bodies it does actually work and I'm missing out on something.

On Board Flash

If you say to the camera; "full auto choose any point in focus" this is the most brain free setting. It will just work on landscapes with a small apperture like f/8 or f/11. You say to the camera: "I don't care, manage for the best". And that good in some occasions. I think it would work on a uniform sky when shooting birds or planes. You may think "this is completely useless", but this is the "green mode" non photographer would be better of to use when they are not yet accustomed to deal with the focus yet.

Then if you want to be precise on a still subject with very shallow deph of field you use 1 point select, choose the AF point that get you the framing/composition you want and get the subject in focus and you are done.

Finally, if you are trying to focus on something that move fast you'd select a small area of focus initially so you can designate it to the camera and allow then the focus point to switch to follow the object. You'll also follow with your camera and it will greatly increase your chances to get the subject in focus rather than something else because you didn't manage to always keep the subject under the same AF point. I'am not that sure you need that many AF point to designate the subject initially but the tracking will be improved by more AF points and also a wider area allow you to have less efficiant tracking by yourself and let the camera do more work.

[Painter]

I wonder when we can expect some actual reviews and tests of the new K-3 II. I really want to see what the sensor shift can do and the improved sensor stabilization.

[Kunzite]

DPReview's, they're waiting for new Canon/Nikon cameras - so they could declare the K-3II 'outdated'. At least 6 months, and even then it might not be a full review. They will insist on the 'missing' on-board flash, and will mention 'no blinking LEDs' a dozen times.
Imaging Resources, they'll start working on it and publish updates as soon as they receive a production sample.
Pentaxforums, as soon as one of the 'usual suspects' will receive theirs?

[derekkite]

Sure and that is tracking, but you selected the point of focus first. What I mean is if the AF is just on auto how does the camera figure out what to focus on? There has to be something or no one with a multi-point AF system would ever get anything in focus except by luck. I get the need for lots of AF points for tracking, maybe on more sophisticated AF systems these are only used when in "tracking" mode? I guess I'm just not familiar with other systems. On Pentax I tried the multi-point, it focused on what it wanted and we agreed to disagree so I turned it off. I've used single point ever since. Maybe on newer bodies it does actually work and I'm missing out on something.

On Board Flash

I had the same experience. It would need some smarts and enough data to analyze, then focus on a face or eye or something like that.

[boriscleto]

DPReview's, they're waiting for new Canon/Nikon cameras - so they could declare the K-3II 'outdated'. At least 6 months, and even then it might not be a full review. They will insist on the 'missing' on-board flash, and will mention 'no blinking LEDs' a dozen times.?

The K-3 II has an older sensor than the D7200. They don't have to wait. The K-3 II doesn't even have built-in WiFi. Another thing to mention a dozen times.

[Matthew Saville]

That's a good point, It wouldn't make sense to charge $2000+ for a $1300 lens with better coatings, DC and WR

Not sure if this is a sarcastic remark or not. This is pretty much exactly what Canon / Nikon are offering, compared to lenses like the Tamron 15-30 and 24-70. Except the Tamrons are stabilized. Canikon still seem to deem their versions worth ~$2K, while the Tamrons are ~$1200...

---------- Post added 05-10-15 at 12:46 PM ----------

The K-3 II has an older sensor than the D7200. They don't have to wait. The K-3 II doesn't even have built-in WiFi. Another thing to mention a dozen times.

I'm curious to hear more about this fact / assumption. Which info in the press release etc. indicates this? (I'm actually curious, not trying to be argumentative)

It is a shame that the K-3 sensor has taken a step back in dynamic range and high ISO from what the previous 16 MP sensor had offered, but this was actually the same case in Nikon's camp for a while too. The Nikon D7000's 16 MP sensor had incredible dynamic range, and rather decent high ISO performance, but it took quite a few new generations / iterations of the 24 MP sensor before Nikon could claim improvements for both DR and high ISO.

TLDR, indeed I'm not holding my breath for the K-3 II sensor to be much different at all from the K-3 sensor, but I'd still like to think that they'll bump up both DR and ISO performance by a tiny amount, if only by improving the processing engine. But that may not even be improved, since as a Nikon convert I don't yet know what nomenclature surrounds Pentax' in-camera RAW converting / processing...

=Matt=

[Stavri]

Not sure if this is a sarcastic remark or not. This is pretty much exactly what Canon / Nikon are offering, compared to lenses like the Tamron 15-30 and 24-70. Except the Tamrons are stabilized. Canikon still seem to deem their versions worth ~$2K, while the Tamrons are ~$1200...

---------- Post added 05-10-15 at 12:46 PM ----------



I'm curious to hear more about this fact / assumption. Which info in the press release etc. indicates this? (I'm actually curious, not trying to be argumentative)

It is a shame that the K-3 sensor has taken a step back in dynamic range and high ISO from what the previous 16 MP sensor had offered, but this was actually the same case in Nikon's camp for a while too. The Nikon D7000's 16 MP sensor had incredible dynamic range, and rather decent high ISO performance, but it took quite a few new generations / iterations of the 24 MP sensor before Nikon could claim improvements for both DR and high ISO.

TLDR, indeed I'm not holding my breath for the K-3 II sensor to be much different at all from the K-3 sensor, but I'd still like to think that they'll bump up both DR and ISO performance by a tiny amount, if only by improving the processing engine. But that may not even be improved, since as a Nikon convert I don't yet know what nomenclature surrounds Pentax' in-camera RAW converting / processing...

=Matt=

I wasn't being sarcastic at all. There's a lot for work that goes into making lenses, formula, coating, polishing, gluing, aligning etc. Pentax lenses have usually higher T/stops then 3rd party products. Next time you look at the Sigma EX (high performance) line and wonder why the OOF renders the way it does (astigmatism, sub-par polishing or optical flaws) you understand why you got it for 1/2 off...

[boriscleto]

I'm curious to hear more about this fact / assumption. Which info in the press release etc. indicates this? (I'm actually curious, not trying to be argumentative)

Info about the sensor was not in the press release. Ricoh did a Q&A after the fact and said that the sensor is the same as the K-3.

[Matthew Saville]

I wasn't being sarcastic at all. There's a lot for work that goes into making lenses, formula, coating, polishing, gluing, aligning etc. Pentax lenses have usually higher T/stops then 3rd party products. Next time you look at the Sigma EX (high performance) line and wonder why the OOF renders the way it does (astigmatism, sub-par polishing or optical flaws) you understand why you got it for 1/2 off...

I think that while we do agree there is a difference, the question is, is it worth 100% more, just to get the subtle improvement in bokeh or whatnot? For most folks, no. They care about sharpness, price, and not much else. Which is why lenses like the Sigma Art primes are doing so well. They may not have the "character" of a Canon L prime or a Leica lens, but they get the job done, very sharply, and they don't cost much at all.

I guess your point was that Tokina couldn't have possibly designed a 24-70 with ALL of the optical prowess of a Canon / Nikon / Pentax lens, and in my opinion I suspect that you'd be correct about some aspects of image quality, but incorrect about others. Tokina may be a champ in the department of wide-angle sharpness, (They seem to be able to do no wrong; I've tested the 12-24, 11-16, 12-28, 11-20, and 16-28) ...they do indeed fall short sometimes, especially with their older lenses, in terms of bokeh, flare, distortion, etc.

However, lately I have indeed seen great improvements in these areas from Tokina, Tamron, and Sigma alike. They still have a ways to go, but they do seem to be eliminating many of their individual shortcomings of the past, and they're not charging $2K for it either. That was my point, I think.

Personally for astro-landscape work, I do indeed care about things such as coma, field curvature, and light falloff, which is why I often find that some third party lenses fall short. Others, however, do deliver "the goods"...

=Matt=

[Nicolas06]

I wasn't being sarcastic at all. There's a lot for work that goes into making lenses, formula, coating, polishing, gluing, aligning etc. Pentax lenses have usually higher T/stops then 3rd party products. Next time you look at the Sigma EX (high performance) line and wonder why the OOF renders the way it does (astigmatism, sub-par polishing or optical flaws) you understand why you got it for 1/2 off...

I'am not that sure the latest sigma/tamron high end offering is that bad.

[Nicolas06]

I guess your point was that Tokina couldn't have possibly designed a 24-70 with ALL of the optical prowess of a Canon / Nikon / Pentax lens

Actually Pentax and Tokina share the optical design of at least 12-24, 35 ltd macro, 16-50 and 50-135... The flows of the 16-50 & 50-135 is often said to be SDM and Tokina doesn't share it with Pentax.

I don't know why there an idea that lense maker that still make camera must for sure make better lenses in all circonstances. I have seen very good review of the sigma 35mm f/1.4 for example and I'am not sure it has the terrible bokeh stavri described.

Also sigma start to offer design no other manufacturer provide at all like the 18-35 f/1.8 (this one doesn't come with a great bokeh, ok).

[monochrome]

But the thing is if you really want to do many action/sport shoots it would be a just more difficult to take great shoots with a very average AF-C camera like a K5. As new photographers don't have a clue what they want or not and may want to just try many things, this is not easy for them to choose.

My point is many if not most of dSLR consumers don't know about anything other than snapshots and taking pictures of cats and dogs and soccer games - and they don't use tracking AF for the soccer games.

But they know the difference between 65 and 27 so they buy the 'better' camera, even though they'll never use more than one AF point.

I wish there was a real study of consumer behavior around dSLR's. I really believe enthusiasts and professionals would make rational feature / benefit / cost decisions about how these feature mixes contribute to final output - but camera manufacturers are forced by market economies to make cameras for consumers, who respond to false claims of utility..

[Rondec]

Good point. I rarely shoot less than f/11, sometimes f/8 if I'm daring so not really an issue for me.

Yes, focus and re-compose is what I am used to. Fiddling with the AF points seems to me to be just that, fiddly. I have certainly used it, when I remember, but mostly it seems to be one more thing that gets in the way of me taking pictures.

Yes, I can understand that.

It seems the reason that I don't understand the need for more points is that the only situations where that is an advantage are things I never, or rarely, do. Good enough, thanks for the information.

---------- Post added 05-10-15 at 08:05 AM ----------

Ok, after thinking about this some more one thing still puzzles me.

I shoot with single center point when not on the tripod and when on the tripod single select and move that point to where I want the focus to be. So 1 AF point = 1 area of focus. I have a 100% chance of getting what I want in focus.

But if I have a camera with 65 AF points in auto or whatever it is called and I take a picture the camera will select one of those 65 points to be the focus point. So I now have a 1 in 65 chance that what I want in focus is going to be the focus area.

Now, I am sure I am missing something because otherwise people would be clamoring for less points not more so what am I missing? Maybe I am old, but I just cannot figure out how having 65 AF points does not cause chaos.

You choose an auto focus point based on the composition you want. This is not going to be so useful for landscape photography, but say you want a portrait with the eyes in focus in the right upper third of the photo, you can choose that auto focus point and shoot away. Recomposing is not a great option in these situations, as small movements of the camera or subject can be enough to throw your focus on the ears or the nose instead. Choosing auto focus points is really not that hard. Hitting the OK button centers the auto focus point and then a couple of taps and you have the point where you want it.

The biggest problem in the past hasn't been the number of auto focus points, but their size. The K5 II would lock on any subject, you just couldn't always tell where it would lock and with a lens like the FA 77 shooting at f2, you could easily have the focus point on someone's shirt and not on their eyes.

[jatrax]

You choose an auto focus point based on the composition you want. This is not going to be so useful for landscape photography, but say you want a portrait with the eyes in focus in the right upper third of the photo, you can choose that auto focus point and shoot away. Recomposing is not a great option in these situations, as small movements of the camera or subject can be enough to throw your focus on the ears or the nose instead. Choosing auto focus points is really not that hard. Hitting the OK button centers the auto focus point and then a couple of taps and you have the point where you want it.

So the only use for more AF points is when you actually choose an AF point, or when using the tracking? Tracking, OK I can see more is better assuming the camera can use those to hand off the target from one to another. But for portraits I'm not convinced 65 is better than 27 or 11. Eye in the upper right third is fine, but that does not need 65 AF points to do.

---------- Post added 05-10-15 at 01:51 PM ----------

My point is many if not most of dSLR consumers don't know about anything other than snapshots and taking pictures of cats and dogs and soccer games - and they don't use tracking AF for the soccer games. But they know the difference between 65 and 27 so they buy the 'better' camera, even though they'll never use more than one AF point.

Seems like AF points are the new "megapixel wars". More is better, even if they are never used. So on the k-3 let's say I want to set the focus point up and right as far as possible. That is: press select button to move from drive mode function to AF select, press OK to center, then up, up, right, right. Total of 6 button presses to get where I want. Way too fiddly for me, but OK it's doable. Now if we have 65 AF points how many button presses is that? I've no idea.

I'm not opposed to more AF points if there is a benefit to some. I don't think it will ever benefit me, but as long they stay out of my way I don't care. I just do not see any advantage of 65 over 27 for anything except maybe tracking. Except that the marketing department can brag about it. Or am I missing something? Not trying to argue with anyone, just trying to learn something new.

[Stavri]

I'am not that sure the latest sigma/tamron high end offering is that bad.

I own the Art 18-35/f1.8 the frogegg bokeh under strong specular lights is the proof