[Rupert]

I am seriously considering Rupert's suggestion to carry a LED torch and use it as a manual AF assist ...

Except it probably won't work for stealth photography ...

(Recently, a man was arrested for taking lots of photos and videos up women's skirts - presumably as they were climbing stairs or using escalators - he probably wasn't using a K-5 ...)
Eastwood man held for indecent filming

A LED is a great source of light for Macros and for use when manual focusing. I have several and one that mounts on the tripod screw. I've never tried it for looking up ladies skirts....I guess it would work OK?

Regards!

[Christine Tham]

A LED is a great source of light for Macros and for use when manual focusing. I have several and one that mounts on the tripod screw.

Okay, you've sold me. I have on occasion really struggled to manual focus on a macro lens with a split prism in dark conditions (the split prism becomes totally dark and useless) - your idea would work great in those scenarios.

I'm giving you a reputation point for that idea.

[tram57]

I sent pentax some pics using the Jackson focus chart showing a front focus shift of 3cm in bright light vs low light. They suggested that they could adjust/fix the camera to rectify the issue. Has anyone tried sending in their cameras to have this adjusted? I'd be interested in knowing whether it is possible for a service tech to adjust the low light AF performance on these cameras without affecting the bright light performance, which in my case is spot on?

[Rupert]

I sent pentax some pics using the Jackson focus chart showing a front focus shift of 3cm in bright light vs low light. They suggested that they could adjust/fix the camera to rectify the issue. Has anyone tried sending in their cameras to have this adjusted? I'd be interested in knowing whether it is possible for a service tech to adjust the low light AF performance on these cameras without affecting the bright light performance, which in my case is spot on?

By all means send it in and let them try it! There is not much to lose, and plenty to gain if they succeed.


Christine, you will like what you see with the LED...and I can rest easy that you won't be looking up dresses with one! I appreciate the point, I don't think I have ever received one before? Are they good for coffee at Starbucks?

Regards

[Smeggypants]

I agree that the Assist Light is not a solution, but it would at least work for many situations that I use. From my standpoint, I am going to have to decide if I can live with the FF issues or make other choices about the future. I'm just looking at ways to stay with Pentax and not sell off all the gear I have and love.

It has been several months that this problem has been made known to Pentax, and my belief is that they are not going to address it, or we would have seen some sign, heard some word.....officially. If that is the case, many of us have to decide to move on, live with the flaws, or wait for the next model, which may be 18 months away. Each person has their own needs and will determine accordingly to those needs.

What I am saying is that I am willing to work around the FF issue if there are ways, such as the Assist Light, to do this. I have a 50-135 SDM lens, and although the first one failed, I managed to use it MF for some time and will most likely end up doing the same with the current one when it fails...assuming it does.

Not backing away from my displeasure with Pentax with the K5, but we sometimes have to live with our choices. Like Mrs Rupert, who has many fine and lovely features...but also has that addiction to plastic........this late in the game, you just try to live with it and keep that old saying close in your mind........"The Devil I know, is better than the Devil I don't know".
Best Regards

Well I'm not leaving pentax. I just want the FF problem fixed. I shall email my supplier tonight and hopefully a reply shortly. I don't want to get into a successive round of replacements if a firmware fix is on the way.

[PixleFish]

=... and the only consistent result was a shift to FF at EV 3-4 with any color light.

Im reading a lot of post that indicate a FF Problem at 3-4 "EV"

My apologies, but I have a rookie clarification question. When you say "at EV 3-4" do you actually mean "LV" (light level)? My understanding is that LV is a scale used to estimate your light level and a value of 3-4 would be something like photographing streetlights or a VERY dim lit room.

I thought LV was used to discuss how well lit your subject is where EV is a compensation setting on your camera.

I know its probably just semantics, but I just wanted to clarify for my information.

Thanks,

[Ray Pulley]

Im reading a lot of post that indicate a FF Problem at 3-4 "EV"

My apologies, but I have a rookie clarification question. When you say "at EV 3-4" do you actually mean "LV" (light level)? My understanding is that LV is a scale used to estimate your light level and a value of 3-4 would be something like photographing streetlights or a VERY dim lit room.

I thought LV was used to discuss how well lit your subject is where EV is a compensation setting on your camera.

I know its probably just semantics, but I just wanted to clarify for my information.

Thanks,

Yes we are being a bit fast and loose with the terminology, but at ISO 100, EV and LV are the same number, but it is a confusing topic.

Here is a good quote on the subject (from http://www.earthboundlight.com/phototips/exposure-stops-here.html):


"If we are shooting at ISO 100, EV numbers are the same as LV numbers. If we set the shutter speed to 1 second and the aperture to f/1.0 (OK, just pretend for the time being that you have such a lens) with the ISO set to 100 and point the camera at something that makes the meter read dead even on zero, we are seeing a Light Value of LV 0. Switch five stops to a more reasonable f/5.6 and find something that centers the meter and we would be seeing LV 5. Since we stopped the lens down but the meter still reads zero, it must be brighter out there than it was at f/1.0."

Most have only the camera meter for judging this, and the camera's best guess at EV is encoded in the EXIF data, so that is what most are using. I have a good handheld meter, so I can read the light in various ways. Also, the AF system has a specified working range that is expressed in EV (EV-1 to 18 for the K5, I believe), which makes sense as the AF system can only work with the amount of light entering the camera. The camera has no idea how much light was in the room before it entered the camera or what percentage of that light was reflected off the subject, it simply knows how much got into the box at a given aperture (wide open on the attached lens in most cases) from whatever you pointed it at.

Ray

[jolepp]

Im reading a lot of post that indicate a FF Problem at 3-4 "EV"
...

I suppose EV has been used as measure of light intensity at the front element of a lens while it is really a convenient single number representing all the shutter speed / aperture (/ ISO) combinations that result is "correct" exposure (18% gray, I presume). LV, I understand, is an informal concept for this sort of usage (?).

One related thing I and others have been speculating about here is the effect of actual aperture used: assuming that EV can legitimately used as a measure of light intensity (close enough for the purposes of this discussion IMO) something taken at EV 3 results in different intensities at the plane of the focus sensor e.g. so that if it was taken with f=2.8 it could have been taken at EV 2 with f=2.0 and EV 1 with f=1.4 for the same intensity at the AF sensor plane. Whether the AF sensor can actually make use of that extra intensity is not clear.

Phase AF sensors are apparently automated versions of the split prism focusing aid which make use of light entering at the opposite sides of the lens surface area. Adding more of that area might actually result in making things worse if that light ends up on a different spot in the AF light detector strip; it could be that the extra light is just blocked from entering the strip maybe by an explicit countermeasure for that very thing. Would be interesting to know ...

[Ray Pulley]

I suppose EV has been used as measure of light intensity at the front element of a lens while it is really a convenient single number representing all the shutter speed / aperture (/ ISO) combinations that result is "correct" exposure (18% gray, I presume). LV, I understand, is an informal concept for this sort of usage (?).

One related thing I and others have been speculating about here is the effect of actual aperture used: assuming that EV can legitimately used as a measure of light intensity (close enough for the purposes of this discussion IMO) something taken at EV 3 results in different intensities at the plane of the focus sensor e.g. so that if it was taken with f=2.8 it could have been taken at EV 2 with f=2.0 and EV 1 with f=1.4 for the same intensity at the AF sensor plane. Whether the AF sensor can actually make use of that extra intensity is not clear.

Phase AF sensors are apparently automated versions of the split prism focusing aid which make use of light entering at the opposite sides of the lens surface area. Adding more of that area might actually result in making things worse if that light ends up on a different spot in the AF light detector strip; it could be that the extra light is just blocked from entering the strip maybe by an explicit countermeasure for that very thing. Would be interesting to know ...

I believe that the Pentax AF system has an effecitve aperture of f5.6. While it does it's thing wide open on any lens, the AF aperture is set so that the light is fed to the AF system from a circle that corresponds to an aperture of f5.6. This is done so that the slower consumer style lenses will still AF. Otherwise if the light was taken at spots corresponding to f2.8, let's say, the slower lens with an F4 or F5.6 maximum aperture would block any light from reaching the AF system.

While it does matter how much light is gathered by the lens, the base distance between the two spots, which work just like a rangefinder, does not change with a faster lens or larger aperture on the lens. If, like some Canon cameras, we had a second set of sensors at f2.8 that were used only when a faster lens was mounted, we would reap the benefit of a more accurate system by virtue of the base distance being larger.

Ray

[Rupert]

These are interesting conversations on the AF systems, I wish I was more technically literate where I could follow them better. I know nothing of Canon, I don't think I have ever held one, but I have used a friends Nikon D700, and the focus is very fast and accurate as well. What is in their system that seems to make them so good at fast AF?
Before anyone gets all bent out of shape, I am just asking to see if there is a reasonable explanation. If you want ten good reasons why the K5 is better than the D700 (my opinion), I can certainly give them to you, there is much more to a great camera than just fast AF, but I am curious about their AF system, and you guys seem to know a lot more about AF than I do.
Regards!

[Ray Pulley]

These are interesting conversations on the AF systems, I wish I was more technically literate where I could follow them better. I know nothing of Canon, I don't think I have ever held one, but I have used a friends Nikon D700, and the focus is very fast and accurate as well. What is in their system that seems to make them so good at fast AF?
Before anyone gets all bent out of shape, I am just asking to see if there is a reasonable explanation. If you want ten good reasons why the K5 is better than the D700 (my opinion), I can certainly give them to you, there is much more to a great camera than just fast AF, but I am curious about their AF system, and you guys seem to know a lot more about AF than I do.
Regards!

Rupert,

I have been shooting Pentax for decades and have not so much experience with other brands, but I read what I can about the technology, so I have gleaned a few nuggets along the way.

From what I have seen working some Canon Raw files from fellow shooters at events (and many would argue that this is NOT true) Canon gives up a bit of accuracy for speed. Nothing major, mind you, but sometimes speed gets an acceptable shot where my Pentax body would still be thinking about it. This is not true of all models, as the 5D series has a rep of being slow.

I think that Canon has an advantage on the higher-end gear with the ring style ultra-sonic AF lenses. This technology is a powerful and fast (and more expensive) solution that Canon has had for many years and they have had time to perfect it. Nikon has similar technology. I seem to recall that Canon has at one time or another, had some higher failure rates on the micro-motor style focusing they use in the lower end lenses, but you do not hear too much about that these days.

At the lower level, I am not sure that Canon is much better than Pentax, at least since the K7.

I think that the best Nikon machine uses a whole lot more focus points than we have available, but is pretty smart about when to use them. Most of this is selectable by the user and does require getting the setup right for a given situation or you risk having the camera make choices that might be completely wrong in a given situation. You can read some of this sort of thing happening at the other forums.

Both Nikon and Canon seem to have better programs for AF tracking than anything Pentax has made to date. You can read about how they use AF points that are not even visibly selected to start tracking the motion of the image in the frame while handing off the focus duties to the sensors in the predicted path of the subject. The lens is also running in a continuous servo mode when doing this, which means it does not stop and think and re-adjust, it simply keeps moving the focusing motor in the predicted direction.

In short, at least for AF-C, the system is intelligent and much more sophisticated than that of Pentax, at least on the higher end bodies.

I mentioned the use of higher precision f2.8 sensors on some higher-end Canon bodies which work with the pro lenses. This makes for a more accurate focusing calculation.

It also seems that the processing horsepower is higher for both Canon and Nikon, but I have no proof of this other than the cameras are generally faster in almost all regards, especially at the higher-end or at least above the low end models.

Maybe Canon and Nikon spend a bit more money on the AF optics to prevent CA or other optical aberrations from influencing the AF acccuracy? After all, with the volumes they sell, they can make better parts and probably not pay as much as Pentax does for the cheaper ones.

It could be that the best algorithms and technologies have been patented. I know that Canon ring style ultrasonic AF was. This would leave others to find some way to work around the patents and still get equivalent performance, which is not an easy task.

As for better AF, well, it isn't always true for every model. Canon has been suffering with AF issues on the 7D, and the Nikon D7000 is also having quite a few AF issues. I recall big issues with a top of the line Canon sports machine that were never really addressed, and Nikon had some issues early on with the D70, which is where the original focust test charts started showing up. The canon 10D also had AF issues as I recall.

So, we are not alone, but, for the most part, you can send your badly focusing Nikon or Canon in and they seem to get them fixed. We have no evidence of that so far on any Kr or K5 as best I can tell.

Ray

[Christine Tham]

From what I have seen working some Canon Raw files from fellow shooters at events (and many would argue that this is NOT true) Canon gives up a bit of accuracy for speed. Nothing major, mind you, but sometimes speed gets an acceptable shot where my Pentax body would still be thinking about it. This is not true of all models, as the 5D series has a rep of being slow.

Comparing the K-5 directly with the EOS 1D I would have to disagree with this statement.

A good Canon pro body and good lenses will beat the K-5 in both accuracy and speed. Especially for long lenses (Pentax doesn't really have good long lenses anyway so it's a moot point but never mind). You really have to use the Canon system to appreciate how good it is - sometimes I get the feeling the camera focuses even before I press the shutter!

But ... the Canon stuff is much more expensive than Pentax (several times more).

And I am very pleased to say the K-5 is greatly improved over the old K10D - to the point where it's behaviour is sometimes so "Canon" like it's spooky - especially in multipoint AF.C at 7fps. Single shot centre focus - Canon still has the advantage in that it will continue to use the other points to aid in focusing even though the primary AF point is the centre.

But the real point is - as good as the Canon system is, it's still not perfect. No AF system will ever be perfect, unless they have a telepathic connection to our brains.

I get misfocused shots with the Canon as easily as on Pentax.

And in low light conditions, where my K-5 starts to FF (around EV1) the Canon refuses to focus at all. So for my K-5 at least I would say it's low light focusing performance is as good as a pro Canon body.

I'm not saying at FF at EV4 is acceptable, if that's what some people are getting. Mine seems to behave itself all the way down to EV1, and I'm pretty happy with that.

[Rupert]

I appreciate the response Ray/Christine.....it seems that the different systems are not perfect either, but it also appears that money plays a large factor in performance too. I would expect that a camera 2 or 3 times the price of a Pentax would have some better performance, or no one would buy them? Even at an even price, the Pentax features I need and love are better for my needs...there are few cameras I cannot afford. A solution to the focus issues or even a good workaround would satisfy me, maybe most others too? If you have been following the shots posted from the K5, in particular the High ISO shots, it is hard to imagine how the K5 would not be more than satisfactory in most instances. I'll try out my X100 next month, it is not a replacement...it only has that 22MM lens, and by then maybe I will have to at least try the K5. Squirrels in shaded trees might demand it?
Best Regards

[Christine Tham]

Ithe Pentax features I need and love are better for my needs...

I agree completely. The EOS 1D may be okay in a studio, but it's way too bulky and heavy for me to carry with me most of the time. What's the point of having the best AF and the best lenses if I can't use them to take the shots I want in the conditions I'm in?

For me the K-5 is a nice balance between quality, ease of use and portability. And most of the time I use the DA limited lenses, and they are indeed special and give me good results most of the time.

I think sometimes it's easy to get the impression (especially from reading this thread) that there's a serious problem that needs to be fixed. For some units, probably, and maybe Pentax have some QC issues they need to address. I also suspect looking at some of the photos there may be a degree of user error - or perhaps a lack of understanding of why the shots came out the way they did. That comment is not directed at anyone in particular - it's just a possibility I am raising.

However, I have not encountered any significant issues (I don't consider a few mm of FF at EV1-2 to be a "significant issue") and I know a few other owners who are also very happy with their units - I suspect there's a silent majority of satisfied users out there.

[merengue]

I appreciate the response Ray/Christine.....it seems that the different systems are not perfect either, but it also appears that money plays a large factor in performance too. I would expect that a camera 2 or 3 times the price of a Pentax would have some better performance, or no one would buy them? Even at an even price, the Pentax features I need and love are better for my needs...there are few cameras I cannot afford. A solution to the focus issues or even a good workaround would satisfy me, maybe most others too? If you have been following the shots posted from the K5, in particular the High ISO shots, it is hard to imagine how the K5 would not be more than satisfactory in most instances. I'll try out my X100 next month, it is not a replacement...it only has that 22MM lens, and by then maybe I will have to at least try the K5. Squirrels in shaded trees might demand it?
Best Regards

With your proven track record I'm sure you can do squirrels in any environment with a "pinhole camera". Not me, I will need a K-5 in perfect working condition.

Earl