[rvannatta]

Huh? Center Weighted is the metering pattern, it has nothing to do with focus at all. Do you mean you started using the center focus point?

you are right--I meant to say center weightted focus point.

[fwbigd]

Well that is a little reassuring now that theres an explanation, can those of you who have the defective (suspected) copies post your serial numbers. It would go a long way to identify the affected batches for those who are yet to walk in the store and play international ping pong with Pentax over shipping defective stock. Im hoping stockists are alerted duly by all Pentax reps if this the case. Still its better to be safe than sorry.

The first s/n was in the 90XXXXX series and the second is in the 21XXX series.

[jbcampbell]

Didn't you read my post? It does take acceptable images at f2.8. It beats the Tamron 28-75. It just is much sharper at 5.6, like most lenses.

How much are your Canon f2.8 zooms though, the IS versions to be comparable, and L glass for weather sealing to be truly equal. A hell of a lot more than $1000 USD.

Not sure there is any market for even higher end lenses for Pentax at the moment.

In my test shots I didn't get acceptable results at any focal length at F2.8. Again, I believe I got a bad copy of the lens and I hope my next one will be a better representation of the product.

Yes, my Canon L glass did cost more but it is designed to be used on a full frame sensor whereas the Pentax 16-50 is not. So I think the comparison between the 2 is legitimate, maybe I am expecting too much from the Pentax zooms? If so I guess i will just stick with Pentax primes

[Maxington]

In my test shots I didn't get acceptable results at any focal length at F2.8. Again, I believe I got a bad copy of the lens and I hope my next one will be a better representation of the product.

Yes, my Canon L glass did cost more but it is designed to be used on a full frame sensor whereas the Pentax 16-50 is not. So I think the comparison between the 2 is legitimate, maybe I am expecting too much from the Pentax zooms? If so I guess i will just stick with Pentax primes

I hope your next one is better, mine at f2.8 isn't razor sharp but its not unusable in any way. So I'm guessing a bad copy.

I don't get what you mean about full frame, as far as I know, and I could be wrong, full frame lenses have advantages over crop, since they only use the middle of the lens, and thus avoid all the problems with the edges, like softness, vignetting and so on.

[jbcampbell]

I hope your next one is better, mine at f2.8 isn't razor sharp but its not unusable in any way. So I'm guessing a bad copy.

I don't get what you mean about full frame, as far as I know, and I could be wrong, full frame lenses have advantages over crop, since they only use the middle of the lens, and thus avoid all the problems with the edges, like softness, vignetting and so on.

My point about lenses designed for 35mm film of full frame sensors is that they cost more to make because they use more glass in order to produce an image that will cover the larger sensor or film.
Just like medium format lenses are more expensive than those designed for 35mm. So if a lens is designed to only cover an APS-C sensor it should be cheaper to make than the equivalent 35mm lens.

You are correct, FF lenses on an APS-C body use the center (sweet spot) of the lens, so there is some optical advantage to that.
In my mind the Pentax 16-50 F2.8 on the K10D (APS-C, 1.5 crop factor) should give me similar results to my Canon 24-70 F2.8 on my Canon 5D (FF sensor, no crop factor).

[rvannatta]

Thanks for this post. I learned some more about my K10D. I just changed to the center weighted focusing and wowzers, no more hunting. My lens' SN is 9015nnn. I've only had the K10D for a month and I learn something new every day.

As far as getting usable images at the widest f stop at the widest focal length, slr lens designers must be having tremendous acid reflux. Back in the 60's we never expected to get good images at the widest f stop on a prime lens let alone a zoom. The designers gave us f1.4 or f1.8 so we could see to focus in low light conditions and not to give good images. The only lenses that were guaranteed to provide sharp images at the widest f stop were Leica's rangefinder lenses because it made no sense to put an unusable f stop on a RF lens. Even then the Noctilux had noticeable vignetting at f1.0! With today's technology I think it's amazing that we are getting the images we do at f2.8 on a zoom lens.

Well I spent quite a bit of time studying the lens' s behavior. My 'test pattern' was the road in front of my house, A straight road for quite a distance lined with trees on both sides. Now the dominate part of the photo was the 'hole in the middle' --- ie. down the road, that needed to be focused at infinity, but of course foliage on the sides of the road as close as the
lens could see. Slow wide angles are pretty hard to get out of focus because of the extensive hyperfocal distance but at F/2.8 something is going to be out of focus, and that causes the software to get really upset.

Tweeking it to only look at the middle is one way, but there is also another approach.
and that is the switch around on the front -- the MF, AF.C and AF.S switch. IF you set that to theh single shot solution (AF.S) With this it will focus once and shut down which is usually not too far off, but if you have it set to continious focus, and the mission is impossible, it will
asayou say, hunt for a while and then abruptly shut down---usually about as screwed up as it can possibly be. Finallly, I have tweaked the camera to avoid taking photos 'wide open' where possible by biasing the program mode for depth of field, and increasing the ISO, and in fact enabling the ISO range function. (and I don't use the 'green mode;

I don't do all these things all the time but after a month of two of very unhappy results,
I now get along with it just fine.

It wouldn't surprise me that if one day Pentax has software smart enough to recognize the lens installed and tweak the camera settings to optimize for whatever lens is installed but until that day comes we have to do our own tweeks.

[rvannatta]

The first s/n was in the 90XXXXX series and the second is in the 21XXX series.

so far all the complaints I've seen have been about the 90xxxx series.

[Maxington]

My point about lenses designed for 35mm film of full frame sensors is that they cost more to make because they use more glass in order to produce an image that will cover the larger sensor or film.
Just like medium format lenses are more expensive than those designed for 35mm. So if a lens is designed to only cover an APS-C sensor it should be cheaper to make than the equivalent 35mm lens.

You are correct, FF lenses on an APS-C body use the center (sweet spot) of the lens, so there is some optical advantage to that.
In my mind the Pentax 16-50 F2.8 on the K10D (APS-C, 1.5 crop factor) should give me similar results to my Canon 24-70 F2.8 on my Canon 5D (FF sensor, no crop factor).

I don't quite agree, since it seems much harder to design a lens from a range of 16-50 than it is to design one from 24-70, and keep it compact, affordable and weather sealed.

Though I had a quick look at the Photozone review of the 24-70, they had to get 4 different copies to get a good one. You can't win whatever brand you go, heh.

[Canada_Rockies]

I believe mine in an early one with a 9012nnn serial number. My first experiences with it were quite negative, however, I resolved the going crazy focusing issue by simply changing a camera setting. Specifically, I went to center weighted focusing instead of screen average default method. It is my theory that with the wide angle, and the fact that the view finder doesn't show everything---- with the screen average focusing you are all too often focusing on things you can't see off the edge of the viewfinder.----particularly when you are at the 16mm setting.

Once I changed to center weighted focusing, my frequent grossly miss focused photos disappeared.

My DA* 16-50 is s/n 9013nnn, and I am having no focusing problems with it.

[roentarre]

I don't quite agree, since it seems much harder to design a lens from a range of 16-50 than it is to design one from 24-70, and keep it compact, affordable and weather sealed.

Though I had a quick look at the Photozone review of the 24-70, they had to get 4 different copies to get a good one. You can't win whatever brand you go, heh.

Da* 16-50 is an amazing lens for the price being asked for. Especially having a zoom owning such amazing contrast and colour covering this focal range is really unheard of. Plus its weather sealed feature and silent focusing, this is really a good lens to own.

This shot was shot at f3.5 16mm hand held in a dark setting where no photography and tripod was allowed. The resolution was not bad at all with the lens stuck into the gate facing up towards the ceiling.




Canon 24-70 is a lens that purple fringes more crazily than any older pentax lenses, unsharp on both ends wide open. The first zoom I sold not less than a year owning it. That zoom needs a lot of work and no second version as yet.

[Canada_Rockies]

On the grounds that images speak volumes, here are two images. The first is full frame. This is probably the worst you can do to a camera/lens combination to see how bad things can get, but here goes.

These were taken at our annual holiday bridge club party, and was held in the local Legion branch. I wanted to see what my new toys could do, and did my worst. No flash. This shot was taken at ISO 1600, 1/20 sec @ f/2.8, DA* 16-50 at 16mm focal length. What more could I do to ruin the shot? Oh, yeah: AF, AE, SR. Point and shoot. I couldn't find a shot with the focus in the corner, but this one does demonstrate to me that I have no worries about using this lens. None.

Here is the original image, exported from Lighttoom with WB and Exposures adjustments only.


And here is a crop at 100% converted to jpg of the middle left of the photo. Notice that you can read the plastic bag.

[Harald]

The first s/n was in the 90XXXXX series and the second is in the 21XXX series.

My bad copy series number is 9015xxx.

I made some comparisons more and it seems that Tamron 28-75/2.8 wins this DA* wide open at all focal lengths. At least it can be said that DA* is not better. This seems to be the case also at other apertures, but I cant be sure without further experiments. I am quite tired and disappointed to this lens and not willing to make any tests anymore.

There is just one question remaining; do I try get an good copy or get the DA 12-24, which is in my list also, instead, and forget this DA*. It will take long time before there is replacements available here because all preorders are not delivered yet....

Canada_Rockies: I do not want to be impolite, but this 100% crop isn't sharp.... but it shouldn't be because it is taken from the part of the picture which obviously is in the out of the focus area.

[roentarre]

Given that picture having various subjects at different depth of field, that 100% crop at left side corner is already not that bad in terms of sharpness.

The lens is known for a bit of backfocusing between f2.8 and f4 as Peter Zack already pointed out, the image can be soft as a result of mal focusing rather than being unsharp. This can be helped by a bit of manual focusing when you shot wide open (only a handful would do that to get bokeh anyway).

In November I posted a thread comparing corner sharpness of Da 14mm to 16mm of Da* 16-50, the difference is there but at a pixel peeping level.

16mm of Da* 16-50mm at f2.8



Da 14mm at f2.8




These two hardly have much difference. Then have a bit of pixel peeping


Centre 100% crop Da 16-50



Centre 100% crop of Da14



Da 14 appears slightly sharper but so what...


Then L lower corner crop of Da 16-50



And L lower corner crop of Da 14




Again, you might say Da 14mm f2.8 is not as sharp as Da 12-24 blah blah blah

Remember Da 14mm f2.8 is the only fantastic prime across all the competition brands. My canon EF 14mm f2.8L produces pathetic unsharp images to my naked eyes even resized to 600 by 900; My crazy friend, a Nikon D2X user also showed unusable images with his own Nikon counterpart. The shots can only be converted to abstract shots instead.

So Da* 16-50 is really not that bad as long as it is not de-centred. If I want to whine about my copy, I can write another 60 pages about it. It has a lot of issues everyone had talked about but I still can manage to love this lens...

[Harald]

Roentarre, I just wonder what should these examples to prove?

You took pictures (14mm/16mm/f:2.8) at scene including objects located very deep positioning (close and far from camera) and take 100% clips from places having very different distances from camera.... should all those clips to be sharp? at f:2.8?

If so:
At f:2.8 16mm, using 0.02mm circle of confusion, hyper focal distance is 4.54 meters... that means that if you focus to 4,54m distance all from 2.26m to infinity are (or should be) in focus.

We do not know where you have focused in these pictures, but for me it looks that you have not focused to hyper focal distance. And this maybe explains the blurriness in these pictures.

And if you have focused to hyper focal distance, those rocks in the foreground obviously shouldn't be in focus because they are closer than 2.26 meters (7.41 feet) from you, aren't they? if they are more than 2.26 meters from camera, sharpness of lenses you used is not very brilliant because then they should be tack sharp.... sorry.

But this doesn't explain why the tree in the background is also blurry in the DA* picture. It is in more acceptable level in DA 14 picture...

I tested my copy to flat surface and it didn't handle even it properly.... I used tripod, manual focusing and even with manual focus I didn't just trust my eyes but bracketed the focus until I got best possible result. If I cant get properly sharp pictures this way, the lens is a lemon, what ever others would like to say....

[roentarre]

well, the focus was on the trees in the centre. The rocks were not far away from the trees either.

The testing scenery typically involves a foreground and a background so more practical to demonstrate the points. Even the rocks are at a different plane to the trees, the sharpness are not unreasonable.

The whole point is Da* 16-50 being not that far from Da 14 in terms of sharpness. Da* 16-50 is a reasonable choice given it is flexible with the wide focal range, weather sealed and producing great imges.

Yes, there are quality control issues with this zoom but it is getting better. I found issues with this zoom but not out of proportion that bad according to the tone put forward here.

After all, I hope that one day you will find the zoom you are looking for.