K7 AF Myth

completely disagree, never feeling slowness or imposibility with pentax. i shoot everything and in every condition. it's not the fastest but k7 is very near the other brands in many conditions.

wlachan said:

Been a Pentax user for 2 decades, and the best Pentax AF body I have used was the Z-1p (never had the chance of using the MZ-S). Though it had primitive AF sensor, its speed, torque and tracking were the best I have experience from Pentax. All Pentax DSLRs are horrible in comparsion. The repeated "truth" in the Pentax circle that Canon sacrified accuracy for speed was total BS. I have experienced it myself. When I first shot 40D with ring USM lenses, the OOF rate dropped dramatically, and that included static subjects. Tracking moving targets at close distance were mission impossible with Pentax. Another problem with all Pentax DSLRs is that they have wide area AF sensors. This seems like a fine choice for the general audience, but as soon as you try to pin point focus, or the target was not very large, or you are using wide angles, the AF would be attracted to the background. Not theory, not from white paper, but repeated experience, something I never experienced with the Z-1p. But consider how Pentax is so hooked up to their 11 point sensor being used on most models, I don't see change any time soon. On paper, most Pentax DSLRs are almost perfect, but the practical performance can be very different. Pentax should address their AF shortcoming 2 decades ago, but it seems getting worse, instead of better.
--
Alan Chan
http://www.pbase.com/wlachan

Click to expand...

--
http://www.pbase.com/jon1976

rene

Roland said:

AF speed is not only about tracking or continous shooting.

Click to expand...

The test seems limited in the info it provides for AF without tracking (note: I am not saying of no value, simply you can't come to general conclusions because it is too limited). Frst, as others have mentioned, AF performance depends on the amount of light. It would be best if that test was performed twice, in dark-ish conditions and in bright light, and maybe some levels in-between to determine at which illumination level a camera starts to focus reliably. Second. AF is sensitive to contrast: it is easier on simple target with good contrast like a black cross on a white piece of paper than it is with a "true life" target which is never as contrasty as black over white. A camera may start hunting on a real life target even in bright light, even if it passed the magazine test. Thirdly, testing with various lens would be much more informative, albeit time-consuming to set up.

Roland Mabo said:

Then in your sense all AF tests are of limited value.

Click to expand...

That one (if I understand it from the other posts) seems of very limited value. You've learned one thing: the camera focuses as well as another in very good conditions.

You haven't learned about performance in more realistic conditions. Experience with cameras show that one cannot extrapolate the results from very good to less than ideal conditions

So, how does the K7 performs in your hands and with the lens(es) you use (best test ever)?

--
Thierry

ASA_5 said:

I agree with the OP. AF is Pentax's greatest weakness. We keep fooling ourselves by comparing the k10d, k20d, k-7 to each other. The fact is they are all WAY behind the competition. Here is the proof:
http://www.youtube.com/user/55bduke

Click to expand...

Looked at the Youtube you referenced. Which is faster? I could not tell. I did notice the Pentax did a gross move and then a secondary fine-tune move. Which is more accurate consistently do you think? Do you care?

The Youtube poster must not care much about the accuracy or he/she would have tested the accuracy and shown that result along with the speed. Does the final photo matter or are we buying expensive cameras to have drag race competitions?

No BS, I actually read this on Canon's site 3 or 4 yrs ago I reckon.

Ok,

I just got my meter and metered off EV9- EV11 at ISO 1600 with Sigma 24-70 f/2.8 Non HSM. This is just my bedroom with window light.

Someone termed it very well as "micro adjustments". I still get the micro adjustments before focus. That's the "experience" I am referring to. The good will hunting or micro adjustments are still there. Is AF on the K7 "technically" faster? Perhaps, technically speaking, more torque maybe. I think it's not purposely misleading to say it's faster, but... the hunting, micro adjustments, are still there with the AF. That is the entire speed aspect to me in a nutshell. I see AF speed as how fast it could lock focus without micro adjustments. Maybe I am a bit different but to me gaining speed only means one particular thing: limit the micro adjustments. You're there. That's it. (Please ) Micro adjustments make it seem slower than it is. It's not really "slow"...

If there was an option to choose how many fine focus micro adjustments the camera takes, I'd set it at 1 and that's it. Then it would be just fine. Perhaps in the next camera, or firm... ware?

Dan Savoie said:

Here is a well known fact about Canon auto focus. The focus faster but are not accurate. This is well known!

Click to expand...

The "well known fact" that Canon's AF is fast but not accurate is as true as "the well known fact" that Pentax has a considerably slower AF than everyone else...

Everyone makes up its own "myths" about the other ones...
--
http://lol-photoblog.blogspot.com/

viking79 said:

Comparing the K-m to the Canon 5d I had, they felt very similar with similar lenses, and the 5d has the same focus system as the EOS 20d. The screw drive noise sometimes makes the Pentax feel slower than it is.
Eric
--
I never saw an ugly thing in my life: for let the form of an object
be what it may - light, shade, and perspective will always make it
beautiful. - John Constable (quote)

Click to expand...

All true.

The 5D with EF 35f2 feels exactly the same speedwise than the K20 with FA 35f2.

On the other hand, the 5D with 70-200f4 USM is considerably faster than K20 + 50-135 SDM and if you compare FA 100 macro with EF 100mm macro... the difference is just huge!

--
http://lol-photoblog.blogspot.com/

Always an excuse no matter how much evidence. I noticed that none of the Pentax AF defenders commented on the Popular Photography tests that show AF speed vs EV level. For both the K-7 and the k20d they show that after EV6, SAFOX slows way down. The competition doesn't do that.

I'm sure the only reply will be "you can't trust Popular Photography Tests" , or "They don't apply to the real world" etc. etc.

i show u a photo in condition far worse than ev6 and the k7 locked a fast camera in very difficult environment.

i'm not saying the af of k7 is on par with nikon d3 or canon 1 series but is good for a lot of task also sport.
k7 is not a specialized camera, excel in somethin less in another.

if u need the best af k7 is not for you, butfor me the iq of pentax is better than 7d for color and contrast for example. inbody sr is another plus. little body is a plus and so on.

ASA_5 said:

Always an excuse no matter how much evidence. I noticed that none of the Pentax AF defenders commented on the Popular Photography tests that show AF speed vs EV level. For both the K-7 and the k20d they show that after EV6, SAFOX slows way down. The competition doesn't do that.

I'm sure the only reply will be "you can't trust Popular Photography Tests" , or "They don't apply to the real world" etc. etc.

Click to expand...

--
http://www.pbase.com/jon1976

Most people people who post here - all forums - identify with their cameras and are sure the latest and greatest of whatever they have is the best.

People who don't feel that way mostly don't post here.

Many working pros or really serious image-makers of one sort or another - use whatever tool is appropriate - or whatever is available at the moment - and regard it as a tool.

I will NEVER forget my first exposure to this mindset. In an earlier life I was a serious amateur road racer (SCCA in the early days). Jaguar brought out a new successor to the C-Type (I owned a used C-type) - the D-type.

The D was a very serious improvement over the C (both were successful, winning LeMans overall more than once) - and cost a great deal of money for the day (hard to compare because the US govt deliberately devalues its money to decrease the effective debt). As I recall it was around $20,000 in that day's dollar, perhaps $200,000 in todays money.

Anyhow a big- time driver (an oil man from North Texas - that day's equivalent of a software multi-millionaire) bought the first one into the US. It was delivered to the Jaguar dealer in New Orleans, where I lived.

The dealer displayed it on his floor, isolated with posts and ribbons, with much fanfare, for a week before the oilman came to pick it up. It was absolutely gorgeous - fully streamlined, full wraparound cockpit, metal tonneau cover, long tailfin back from the driver's headrest. A marvellous shiny deep black.

I was there admiring it at the end of the week it was displayed when the owner came in to pick it up. I watched while he opened the door and tried it on for the first time. He then got out, opened the hood (the entire hood and front fenders were a single unit, aluminum of course, folding up to the front. Remarkable for the day (long ago. . .).

He then tried to get the tonneau cover off, failed to find the latch, and in jerking on it (it too was thin aluminum) - bent it. The little circle of admiring aficionados - one was me - gasped.

He looked at us kindly and smiled - "doesn't matter", he said - "we're going to use it up; it will only last one year."

It wasn't that he was rich - it was that the beautiful machine was just a tool to win races, not a thing of beauty or value in itself - to him. It was genuinely expendable, just a tool, which by its very nature had a limited life.

To some people cameras are ornaments, to some they are objects of identification, to some they are tools interesting only in that function, to some they are hobby interests (which are often absorbing).

Their is nothing either good or bad about any of these relationships to cameras - or any other artifacts, like cars. But it will help you deal with your relationships with your cameras - and with other camera owners - both on the web and in real life - if you can recognize and accept that everyone's feelings toward his camera (or car, or computer) is colored by how he, she (or you) relates to same. . . and there are lots of different ways.

And - though it seems odd - the person's perception of an artifact is deeply colored and affected by the form it takes to him - so one person's perception of the artifact may be quite literally quite different from another's perception of the exact same artifact.

All real, all valid, and many times quite different.
--
Bill Wilson

i went to canon 7d forum

first post is

http://forums.dpreview.com/forums/readflat.asp?forum=1019&thread=34195058

user can't understand why is camera cant focus proprerly with mid distance subject..

if u search in 7d forum there are tons of af inaccuracy and problems.

probably a lot of users errors but as u can see also canon has the best af for speed but accuracy is sometimes a problem.

in addition reading the thread user suggest a faster speed than normal for 7d that means pushing iso.

with my k7 and sr inbody i can use very low speed with every lens, so pushing down the iso. for example i can handhold the sigma 10-20 at half second often with sharp results.

with k7 u can dream of this.

so this is a case where k7 is better.

Well I don't care one way or another but out of my technical curiosity I just did some quick digging and found indeed the Canon AF system does sacrifice some accuracy for speed, it uses the lens DOF. Pentax system does do a double check to insure the lens is in maximum focus.

http://www.zen20934.zen.co.uk/photography/Canon%20AF%20System.htm

a quote

"Operation

When you half-press the shutter release (or the * button, if you've used the custom function to move focusing control there), the activated AF sensor "looks" at the image projected by the lens from two different directions (each line of pixels in the array looks from the opposite direction of the other) and identifies the phase difference of the light from each direction. In one "look," it calculates the distance and direction the lens must be moved to cancel the phase differences. It then commands the lens to move the appropriate distance and direction and stops. It does not "hunt" for a best focus, nor does it take a second look after the lens has moved (it is an "open loop" system).

If the starting point is so far out of focus that the sensor can't identify a phase difference, the camera racks the lens once forward and once backward to find a detectable difference. If it can't find a detectable difference during that motion, it stops.

Although the camera does not take a "second look" to see if the intended focus has been achieved, the lens does take a "second look" to ensure it has moved the direction and distance commanded by the camera (it is a "closed loop" system). This second look corrects for any slippage or backlash in the lens mechanism, and can often be detected as a small "correction" movement at the end of the longer initial movements.

So this suggests the One-Shot AF mode focus point estimation is an open loop system but the lens position control is a closed loop system with at least a second iteration.

Accuracy

When the camera determines how far and in what direction the lens must move to cancel the phase difference, it does so within a tolerance of "within the depth of focus" of lenses slower than f2.8 (down to f5.6) or "within 1/3 of the depth of focus" of lenses f2.8 and faster. The depth of focus is the range at the sensor plane within which the image of a point will be reproduced as a blur smaller than the manufacturer's designated "circle of confusion" (CoC). Canon's designated circle of confusion is 0.035mm for the 24x36mm format and 0.02mm for the APS-C format. The CoC is based on maintaining the appearance of sharpness in a 6x9 inch print at about an 10 inch viewing distance (as revealed by the Euro-Canon web site). There is no guarantee that images enlarged any greater than this will appear sharp.

As a result of this tolerance (within the depth of focus or within 1/3 of the depth of focus), the camera can place the actual plane of focus at random anywhere within the tolerance range, and not necessarily at the same place each time.

The depth of focus is the sensor side conjugate of the subject side depth of field, so neglecting pupilary magnifications issues the depth of focus and depth of field at a given distance can be considered to both have the same constraints.

Note that having said this some lenses show more AF than others, this is probably due to mechanical and servo control loop variations in lens design. In general with good light and contrast levels the AF system does seem to be fairly accurate, however it is important to remember that it will normally vary significantly from the ideal plane of sharp focus .

Also the above description does not mention light levels and contrast levels, clearly when some combination of these two factors falls below required levels, probably a signal-to-noise ratio, the AF system will stop work reliably or all together. Considering the system is phase difference driven there is probably a threshold effect. Below the threshold the system does not work at all, just hunting, just above it probably works with reduced accuracy but rapidly improves to its design specification; this is rather speculative but is deduced from a knowledge of other phase derived ranging and direction finding systems."

Do you want more evidence that the Pentax AF system does do a double check that others don't in order to achieve more AF speed? Just because an image looks sharp does not mean it could not be sharper, no?

--
jamesm007,
http://s195.photobucket.com/albums/z77/jamesm700/
WSSA member 266PX

Jorgen E said:

Basically, yes. There's one AF test I do care about, though: using the actual camera, in real life conditions. If you do that, you too will see that the K-7 and D300 are miles apart, when it comes to AF performance.

Click to expand...

fraction of a second are miles apart? ...hmm, do I miss something here.

regards,

rene

The last paragraph was not clear. Canon does not do the double check that the Pentax AF system does. This saves Canon a slight bit of time but does sacrifice some AF accuracy; the Pentax AF system may or may not be slower but it is very accurate which is something most reviews will state.
--
jamesm007,
http://s195.photobucket.com/albums/z77/jamesm700/
WSSA member 266PX

jamesm007 said:

The last paragraph was not clear. Canon does not do the double check that the Pentax AF system does. This saves Canon a slight bit of time but does sacrifice some AF accuracy; the Pentax AF system may or may not be slower but it is very accurate which is something most reviews will state.
--

Click to expand...

Again, no one seems to want to address the Popular photography tests. Even if Petnax does a "double check," the AF time differences below EV6 approach a factor of 2. that is not a "slight bit of time."

ASA_5 said:

Always an excuse no matter how much evidence. I noticed that none of the Pentax AF defenders commented on the Popular Photography tests that show AF speed vs EV level. For both the K-7 and the k20d they show that after EV6, SAFOX slows way down. The competition doesn't do that.

I'm sure the only reply will be "you can't trust Popular Photography Tests" , or "They don't apply to the real world" etc. etc.

Click to expand...

In my big living room lit by one 40w lightbulb the K-7 use 1s to focus with the 31 Limited and less than a second with the DA* 16-50/2.8. Is this too slow and a problem? I can't make it hunt unless I focus on something without contrast....

I don't understand the AF speed whining. I haven't lost a single image in god known how many years due to the lack of AF speed.

jamesm007 said:

The last paragraph was not clear. Canon does not do the double check that the Pentax AF system does. This saves Canon a slight bit of time but does sacrifice some AF accuracy; the Pentax AF system may or may not be slower but it is very accurate which is something most reviews will state.

Click to expand...

Most test states "fast and accurate" AF something that sums it up pretty well....

K7 is only faster than K20 in low light, but its much improved and very accurate below EV5. AF C is a lot faster as well, but not in the D300 or 7D league.

However some Pentax lenses are still pretty slow even on the K7 especially if they are not calibrated. As well as the AF processing, many Nikon/Canon lenses focus faster partly because they are optimised for pure AF operation and have light weight focus groups, short throw gearing and high torque ring motors. Some of their professional lenses actually focus slower because they have more glass and need to be more accurate.

Pentax, especially with DA* and FA LTD lenses which have long focus throws, can be slow but are generally easier to MF accurately as a result. The focus screen matting is especially helpful here as well. The accuracy of the off-centre points is also excellent - something I use a lot for tripod based long exposure photography. This is certainly not the case on any of the pre-7D semi pro Canons which lack cross points off centre.

Everything in life is a compromise, the trick is making the right one. Pentax make landscape cameras, and that means they spend the R&D budget on other features such as compactness and weather sealing. They also worry more about backward compatability, which is why the focus screen is usable with older MF lenses.

If I wanted to shoot sport or wildlife I would probably choose a Canon, but I dont. For travel and tripod work the Pentax's work better for me, especially as they have proper mirror lockup. Also, I find I use MF a lot in difficult conditions so its nice to have the option.

Marques Lamont said:

I just got my hands on a K7. Here is a myth I found out about it:

Faster AF - MYTH

I found the AF to be the same "experience" as K20 and K10. Granted, it might be a split faster... but it has the same hunting experience more or less, which IS the real AF "speed" problem. No exagerration. I thought the AF would be overhauled, reading through the reviews and such, and to my surprise, it's more of the same. Where are people getting this "much faster" AF from?

My Canon 30D is still faster. The AF on the 30D is pretty satisfactory, being actually 4 years old now (yikes! ) and I thought Pentax would at least get it to that level with the K7. Almost, but not really. It's along the same lines as the K20 and K10 far as hunting goes. Different techically, because it may turn the barrel faster, but it's still on the same spectrum.

"Hunt, hunt, hunt... focus, beep."

Other than that, the camera is stellar. The size is great and to my surprise, video is great, which I have absolutely no plans on seriously using. It was nice to fiddle around with it though and I could see how photogs can get caught up in it. I haven't gotten bit by the video bug yet but I can see him pretty clearly. He does have teeth and quite a bite.

Click to expand...

--
Steve

Any fool can take a picture OF something. Its much harder to take a picture ABOUT something.

jonny1976 said:

completely disagree, never feeling slowness or imposibility with pentax. i shoot everything and in every condition. it's not the fastest but k7 is very near the other brands in many conditions.

Click to expand...

May I ask which other brands, or models to be specific, you have also used? And what lenses?
--
Alan Chan
http://www.pbase.com/wlachan