[dylansalt]

Dude, you don't understand. The way the Pentax builds a camera is the way the Photo Gods have decreed it. Features available from companies that sell oodles more cameras than Pentax? Useless fluff for unwashed ignorami (Ignore the fact that the majority of people that make there living taking pictures use those brands, trust me, these features are useless). Any point other than the center point is heresy, and then only if you're a drooling moron. Because everyone knows that autofocus is a crutch for the simpleton.

I have it on good authority that Pentax will soon introduce a true photographer's camera. 11 focusing points? How about it constantly randomly defocuses the lens, forcing you, er...allowing you the joy of constantly touching it up. And even though it's digital, it has a film wind lever, so you're not tempted to shoot more than a couple of frames per second like some crazed hooligan. And it won't take a picture until you've spot metered 23 separate parts of the scene... with an external spotmeter. Because matrix metering is for bitches. The ISO tops out at 160, because only a crazy person ever needs more than that, despite the fact that it doesn't have a hotshoe because flash is - you got it, a crutch for young whippersnappers. The only tethering option is a string that you can tie between the camera and yourself. If any lens other than a Limited is mounted, it ejects the lens, electrocutes you, and sends a request to Pentax for someone to come and retrieve it for sale to someone more worthy. It is, in fact, the perfect camera.

This really put a smile on my dial

[RBellavance]

Anyone else here hoping the extra 120g weight is from an oversized AF motor, a la the Nikon F90?

The difference might be just 20g, as we don't know if the rumoured weight is empty or with the battery (i.e. 650g vs 750g in the case of the K-7).

[zackspeed]

for sure we cann read facts about the k-5 before the 22.!
because,
photokina


world of imaging


21.9. - 26.9.2010


i think the 20. is a good day for internetwatching

[zackspeed]

The difference might be just 20g, as we don't know if the rumoured weight is empty or with the battery (i.e. 650g vs 750g in the case of the K-7).

bodyparts

[pingflood]

Unfortunately, this could only work for an isolated subject on a simple background. But for this to work reliably with background and foreground distractions there should be much more information available to that algorithm then AF system can provide.


I feel that I can safely make this claim because several years ago I was experimenting with data available in the AF system and provided by the lens in order to devise some sort of software-based focus limiter (especially for macro and telephoto lenses). I gave up eventually and one of the main reasons was because available information was not consistent between the lenses (plus several other problems). But this is what I have learned:


Each individual AF point is actually not a point, it's an AREA and there is no way for a relatively large area to reliably track any small object. Next, distance information is NOT associated with individual AF areas, there is just in/out of focus status. Overall distance scale is transmitted by lenses but it is very rough (like "closeup", "near", "medium", "far" and "infinity" if I remember well). In addition, one additional piece of information could be determined in AF-C mode on how correction is acquired, say, if reacquiring AF with new status "far" required forward correction "medium-to-far" or backward "infinity-to-far". And that's pretty much all.


Although that is enough information to track a subject moving towards or away from the camera or across the frame with plain background, it is absolutely NOT enough to track subjects moving across the frame with background or foreground distractions. The functionality as described in your post would require either pattern analysis or more accurate AF distance associated with each AF point, both not available in any AF system in consumer cameras today.

Hello,

You may theorize all day on this, but there are plenty of people out shooting that can tell you that the cameras can and do indeed track subjects this way. There is even configuration in many bodies as to how long it should attempt to track a "lost" subject (e.g. a soccer player running behind a couple of other players) before giving up and deciding the subject was actually lost entirely. It works, and it works really damned well actually. I've had series of shots of birds in flight that go behind tree branches and such and the entire series came out fine; obviously it is not actually autofocusing on the bird itself when obscured but is following the predicted path and when the bird's back out in the open it is at or very near ideal focus.

Here is Canon's brief documentation on that setting, by the way.

# C.Fn III-4-1 — Continuous AF track priority
Generally, the recommended setting if you’ve expanded the size of an AF point, and are shooting in AI Servo AF mode with moving subjects (sports, etc). If another subject should enter your active cluster of AF points, while you’re already tracking another subject, the new subject is ignored — even if it’s closer to the camera than your original moving subject.

Basically, it works like this: If something appears that would require focus to suddenly step back/forward by a certain amount, the camera figures that something temporarily got in the way of the subject or focus on the subject was lost due to something else, and keeps focusing along the predicted trajectory of the subject. Should it fail to reacquire a solid focus on something where it expects it to be within X time, it gives up and tries to focus on something else. Not ultra sophisticated but works really well.

Also, you say that the AF "areas" only indicate focus / out of focus. That is not quite correct. The camera actually gauges how far out of focus something is and tells the lens to move a certain distance based on that. Read:
LensRentals.com - How Autofocus (Often) Works.

[WMBP]

How is this announcing a rumour?

It has been been rumored that while September 9th would be the date for the official Pentax K-r announcement, we wouldn't hear anything about the K-5 until the 22nd- nearly two weeks later!

So far, here is all the credible information that we have for the K-5 specifications:
<Speculation about K-5 specs follows, indicating, among other things, 18 AF points.>

....
I find the use of "credible information" misleading.

The word "rumored" is right there. I bolded it so you can see it. Now I suspect you are a literalist and you'll respond that what was being rumored was that the announcements would come on certain dates. But gosh, if the date of the announcements is itself a rumor, doesn't it follow, a fortiori, that anything said about what is coming in the announcements is also going to be a rumor?

Note also that "credible" doesn't mean "trustworthy" or "reliable." Credible means "believable," not "worthy of being believed."

Finally, you have some responsibility to pay attention to the context of the info you encounter. THIS IS THE INTERNET. About 87% of what you read everywhere on the 'net should be read with skepticism. No, I take that back. It's probably more like 97%.

If you want reliable info, skip the "news and rumors" forum and just wait for Pentax to make the official announcement.

Will

[janneman]

however, we must remember, the d7000 specs are rumors as well. 9/15/2010 is the expected announcement date for the d7000.
we'll see then.

I hope for Nikon that the name is a rumor as well. I mean DX000, that is their entry level number....

d7000 is RUMORED to be weather proof like the d300s.
Recap | Nikon Rumors
.

IMHO inevitable that other brands start to make cheaper sealed cameras as well, what surprise me is that Oly, never sealed anything cheaper than the E1 and E3.

AF points???
what we have now is enough for me, placed at the right spots. Ofcourse once i a while I'd like one a bit somehwre else but... more points? I shoot either cnetre or selective.... Now, that mean ofcourse that I change the selected AF spot manually... meaning, more points would mean more pressing a button to get to the spot I want thus loosing time...

[Class A]

The word "rumored" is right there.

In conjunction with the K-r announcement. Nothing to do with the K-5 "specs" at all.

But gosh, if the date of the announcements is itself a rumor, doesn't it follow, a fortiori, that anything said about what is coming in the announcements is also going to be a rumor?

No it doesn't follow. What is the connection between "Pentax K-r announcement date" rumours and "credible" K-5 specs? The two pieces of information appear in the same news article, that's the only connection.

Note also that "credible" doesn't mean "trustworthy" or "reliable."

Please have a look at this thesaurus entry for "credible". The words "trustworthy" or "reliable" are right there.

About 87% of what you read everywhere on the 'net should be read with skepticism. No, I take that back. It's probably more like 97%.

If you are happy with the fact that what is on the pentaxforums front page as news classified as "credible" and not "rumour" isn't worth more than 97% of the internet than that's you call. I expect better.

If you want reliable info, skip the "news and rumors" forum and just wait for Pentax to make the official announcement.

1. I didn't enter the "news and rumours" forum. I received an email and was directed to a news item that talked about "credible" information.
2. I would have waited for the official announcement but the news indicated that there was advance information.
3. We have some confirmed advance information already (see the K-5 camera review entry on this forum) so it wasn't completely unexpected to get some more advance info.

I think Adam would be the first to admit that the phrasing of his news item wasn't optimal (just look at how much more faithful to the unreliability of the information the next one is). I think it is fair to say that he made a small mistake but didn't repeat it and basically all would be good if you didn't try to defend an action that was clearly not worthy for a forum that wants to appear as trustworthy.

[rparmar]

The fact that they are rumors doesn't in and of itself make them false, it just makes them rumors until the 21st.

Actually, all rumours are true all the time. They are truly rumours. And the fact we call them rumours recognises that fact.

[rparmar]

When you turn on the camera, you are given a random page and paragraph number from Ansel Adams' "The Negative" and unless you recite the first sentence correctly, it won't start up. That'll sort the men from the boys!

That would certainly test for photographic memory if not any actual photographic knowledge.

[zackspeed]

till today,all af rumors are wrong......only the af processor is a safox IX+.
but no right information about focusfields and focuslogic.

[jpzk]

till today,all af rumors are wrong......only the af processor is a safox IX+.
but no right information about focusfields and focuslogic.

Take it with a grain of salt and wait until September 20ish ... !
I have become suspicious of all rumours this far with the K5.

JP

[zackspeed]

20th at 4pm berlin time

[rod_grant]

Please have a look at this thesaurus entry for "credible". The words "trustworthy" or "reliable" are right there.

Class A, while I agree with most of your reasoning here, I don't think the thesaurus is the right place to look for a definition. A thesaurus links words of similar meanings not exact definitions. Personally, I think something credible can be absolutely wrong, but be given sufficient explanation and weight to be believable. Never the less, your analysis of the structure of the sentences is, dare I say, credible.

[knoxploration]

Canon actually had eye control on some film cameras right before digital hit the market. From what I gather it worked decently well except if you wore eyeglasses. Why they have not put the tech in their DSLRs remains a mystery; myself I suspect they are keeping it on the back burner until they run out of other "upgrades" and then it will magically reappear.

Off topic: They actually had it from several years before digital took off -- their first eye-controlled AF camera was the EOS 5 in 1992. I remember trying to use it myself at a Canon dealer in Hong Kong, and being completely unable to get it to work for me. Their last camera to use eye-controlled AF was the EOS 30V in 2004.