[jpzk]

Well, OK, I had a tough time deciding what the title of this post should be.

Nearly everyone on this forum is talking about the K5 high ISO capabilities as one major reason for upgrading.
It is indeed one major reason for me to go for it too!

But, I thought that TWO major items were especially important to those who have pre-ordered the K5 or to those who will be upgrading/considering the new camera.
They are:

- High ISO cababilities (already a "given")
- Faster, more accurate or, at least, vastly improved AF.

The latter doesn't quite seem to get as much attention as the former and I wonder why that is.
True, there are loads of posts on the K5 "Live View" AF.
For me , that is not really what matters since I rarely use it with the K7 anyway. Will I be using it more if I get the K5? Maybe, maybe not.

What matters is:

Is the K5 overall AF "seen through the lens" system REALLY better than that of the K7?

We don't really have confirmed tests on this, do we?
Or, if there are such tests, please someone, link me to them.

To date, we have one comfirmed "item" from several photos taken with the K5 at high ISO's ... Good!
You'd presume, along with high-ISO shots, that at least some early AF tests, even with the Beta firmware, would be available by now. I mean: other than Live View AF tests.
I may be not searching properly and, if that is so, please correct me.

Which all boils down to this: it is still too early for me to consider upgrading to the K5 until such "real" tests are out to comfirm what I hope will happen ... a faster and more accurate/vastly improved AF system.

Anyone else shares those thoughts?

JP

[Adam]

Is the K5 overall AF "seen through the lens" system REALLY better than that of the K7?

Yup, it definitely will be. I'm betting on it, and I've heard reports from people using the prerelease K-5 that this is indeed the case

We don't really have confirmed tests on this, do we?

Just that one youtube video comparing its continuous performance to that of a K-7. Once I/the site gets the K-5, we'll be posting professional tests and videos.

[philbaum]

I think one reason we haven't seen lab test results YET, is that as complicated as image quality/noise comparisons are, AF performance measurements are technically more difficult, requiring specialized equipment.

Two good subjective indicators for me are:
a. the number of observations we hear from those that have handled the K5 who say that it is snappier,

b. Pentax themselves saying that AF algorithms have been "drastically" changed/improved. I don't think that Pentax would use the word drastically if they weren't confident.

be patient, we will see some actual testing eventually

[Mike.P®]

I recently sold my K-7 for the funds to buy a K-5 as the AF is the the only real let down for me on the camera. The ISO was fine for my purposes and the IQ is superb but the AF just doesn't cut it these days.

I have this nasty feeling that I am going to be disappointed though .. I think the AF itself is going to be faster but I still don't think we are going to get any decent form of tracking.

I really hope I am wrong.

[Vaikis_]

I recently sold my K-7 for the funds to buy a K-5 as the AF is the the only real let down for me on the camera. The ISO was fine for my purposes and the IQ is superb but the AF just doesn't cut it these days.

I have this nasty feeling that I am going to be disappointed though .. I think the AF itself is going to be faster but I still don't think we are going to get any decent form of tracking.

I really hope I am wrong.

and what's wrong with K-7 tracking, maybe you don't know how to use
here is one of my examples K-7 + slow DA 55-300 + AF.C mode.



Anyway I'll buy K-5

[dylansalt]

and you could go to Flikr and see several hundred equivalent and even better images from a D90/7d etc - where for the users of these cameras it's "oh hum, another day at the office" and not "ooooh look what my K10D/20D/K7 can do" - as if it's a special occasion'-)

[Rondec]

In decent light, the K7 is quite good already. Falconeye commented that he thought the K5 locked focus a little faster than his K7, but it was hard to be certain because the light was good enough at Photokina that both locked pretty fast. The question is what the autofocus does in lower light situations. I just wonder how well it will track and if SDM lenses will lock on quicker, or if they are limited by the in lens motor speed.

[Rondec]

and you could go to Flikr and see several hundred equivalent and even better images from a D90/7d etc - where for the users of these cameras it's "oh hum, another day at the office" and not "ooooh look what my K10D/20D/K7 can do" - as if it's a special occasion'-)

You can also find a lot worse photos from them. What's your point?

[Mike.P®]

and what's wrong with K-7 tracking, maybe you don't know how to use
here is one of my examples K-7 + slow DA 55-300 + AF.C mode.

That would have been a pretty good picture if it was sharper.

[Mike.P®]

Interestingly I have been doing a search lately about the K-7s lack of predictive tracking and came across a superb set of pics taken 8 months ago with the camera and Sigma 100-300mm f4. The poster was more than pleased with the test and seemed happy to show (and prove) that the K-7 does indeed have some sort of tracking.

I then noticed he had no new posts in the Pentax section and saw that a couple of months later he bought a D300s because he kept missing shots.

[opiedog]

i'd like to see the AF improvements as well, both on the k5 and kr.
this is one feature that's holding me back from jumping right in. from the k10 to k20 to k7, there have been improvements in AF, but more incremental improvements more than "big leaps".

i hope the one on the k5 is more than just incremental.

i must admire the nikon marketing engine. they really know how to flaunt their stuff.
the D7k has such great exposure that it really looks great on paper.

Sure, both k5 and d7k have features that the other doesn't have and vice versa, but ask a noob DSLR user wannabe and 9 out of 10, they'll pick the d7k.

You can't help but see what they flaunt about the D7k on the nikon web site. features, features, features, regardless if one uses it or not.

i couldn't persuade a colleague to go Pentax because it doesn't have tethering and he needs it, so please Pentax, if the k5's hardware supports it (and kr), make it a feature. enable tethering.

so far, i'm on the fence. hopefully the K5's AF is much better than the k7/kx. hopefully AFc tracking works better also, then i wouldn't care about the d7k's 39pt AF.

[JohnBee]

I recently sold my K-7 for the funds to buy a K-5 as the AF is the the only real let down for me on the camera. The ISO was fine for my purposes and the IQ is superb but the AF just doesn't cut it these days.

I have this nasty feeling that I am going to be disappointed though .. I think the AF itself is going to be faster but I still don't think we are going to get any decent form of tracking.

I really hope I am wrong.

IMO. if your looking(specifically) for tracking, then I'd say go with a Canon(7D perhaps?). Mainly because it clearly has the lead in AF performance. Either way, it's up to us(the consumers) to do our homework and prioritize our needs so as to end-up with the right product. Given the flow of reviews and end-user data going on these days, there really are no excuses for people to end-up with products that don't suite there needs.

[jpzk]

Yup, it definitely will be. I'm betting on it, and I've heard reports from people using the prerelease K-5 that this is indeed the case



Just that one youtube video comparing its continuous performance to that of a K-7. Once I/the site gets the K-5, we'll be posting professional tests and videos.

Yes Adam, I've watched this video earlier on.

It appears that the K5 AF-C is faster, but wouldn't it have been convenient to see the actual shots resulting from this rather "bland" test? Am I missing something or are there any shots available from this?
Easy enough to see the AF perfomance during LV though, although this is likely not a mode I would use that often ... perhaps I will if it is that good under certain circumstances.

I though PF would have had the K5 and the K-r by now.

Like many others, I am becoming impatient to see actual, professionally done tests for AF-C with moving subjects toward the camera.
That will be THE test, as far as I am concerned.

Hopefully soon?

Cheers.

JP

[jpzk]

I think one reason we haven't seen lab test results YET, is that as complicated as image quality/noise comparisons are, AF performance measurements are technically more difficult, requiring specialized equipment.

Two good subjective indicators for me are:
a. the number of observations we hear from those that have handled the K5 who say that it is snappier,

b. Pentax themselves saying that AF algorithms have been "drastically" changed/improved. I don't think that Pentax would use the word drastically if they weren't confident.

be patient, we will see some actual testing eventually

Patience not being one of my virtues, I nevertheless hope that PF right here will be able to provide us with comfirmation of the "hearsay".
I realize that AF-C tests require more sophisticated equipement, to be carried out properly and systematically.

JP

[jpzk]

I recently sold my K-7 for the funds to buy a K-5 as the AF is the the only real let down for me on the camera. The ISO was fine for my purposes and the IQ is superb but the AF just doesn't cut it these days.

I have this nasty feeling that I am going to be disappointed though .. I think the AF itself is going to be faster but I still don't think we are going to get any decent form of tracking.

I really hope I am wrong.

That is exactly what I was trying to convey in my opening post.

As for the K5 AF "tracking" abilities, I pretty much keep statu quo on the subject as I don't want to be disappointed, falsely hoping for such a feature.

JP