[jzackery]

So I'm fairly new to the DSLR world, as I got my Pentax K-X used from craigslist maybe a month ago. I've got up to speed pretty quickly, from reading the manual, practicing with it on different settings etc, and reading this forum.

This forum is amazing, and i feel proud to own a pentax with an awesome community of people like this.

I saw a SMC F series 50mm F/1.7 (got great reviews on this forum) on ebay and decided to bid and won it at $205. It has been amazing so far, and a HUGE step up from the old kit lens that it came with. Its really interesting, the photos are ready for reviewing almost instantly after taking the photo. For some reason it was slower with the kit lens. Would anyone know why?

But anyways, my real issue seems to come mostly from shots coming in really overexposed or pretty underexposed. I take shots in any modes depending on how much I care to adjust the settings around, but I've noticed that in say Auto mode, when Im outside in bright sunlight, I have to Underexpose things pretty much as far as itll go to get a more accurate to real life picture. In daylight things seem to get blown out super easily, and lots of clipped highlights..

Any suggestions on certain settings with this prime lens to get the proper exposure, or is it up to trial and error in Manual mode or something? Would a lens hood, or filter help?

[aurele]

I had the F 50/1,7 and i never experienced problem with exposure... So i bet the problem might come from you

First, I suggest you to put the lens in "A" mode, and play with the aperture trough the body instead of the aperture ring.

Second, I suspect that at wider aperture you have more chance to clip higlights.

Third, do some test with Av mode instead of M. Av mode is pretty accurate about exposure so give a try without any compensation.

[baro-nite]

In daylight things seem to get blown out super easily, and lots of clipped highlights..

That's true with any lens. Your new lens might have slightly higher conrast than the kit lens, but not a huge difference. Are you using spot metering? That can make the issue worse, depending on what's in the center of the image. Whatever the metering mode, the meter assumes that whatever it sees should be 18% gray. I don't know about you, but it's pretty rare that I actually shoot light gray subjects. Sorry if this is too basic. To avoid clipped highlights you need to meter the highlights and/or take a test exposure and check the histogram. Note that blown highlights can be hard to spot on the histogram; it might look underexposed but for a tiny bump at the far right of the histogram.

Sanity check, are you sure the diaphragm is working properly on the new lens? With the lens off the camera, flip the actuator lever (sticking out of the back) and check to see that the diaphragm opens and closes "snappily".

Its really interesting, the photos are ready for reviewing almost instantly after taking the photo. For some reason it was slower with the kit lens. Would anyone know why?

That's a function of the noise reduction setting -- check to see that it is still on.

[audiobomber]

Its really interesting, the photos are ready for reviewing almost instantly after taking the photo. For some reason it was slower with the kit lens. Would anyone know why?

Any suggestions on certain settings with this prime lens to get the proper exposure, or is it up to trial and error in Manual mode or something? Would a lens hood, or filter help?

I suggest you start by resetting the camera to factory defaults. No telling what's been changed by now. This requires three separate menu items with the K-x, one each for Record, Playback and Set-up menus.

The fact that review is faster with the F lens than the DA leads me to believe that you have Lens Correction activated. I rarely use this feature because it bogs the camera down. I shoot raw and add corrections in post-processing where needed. Even if I weren't shooting raw I wouldn't use lens corrections because it limits high-speed drive mode to two or three frames, after which it stops for a maddeningly long time. Lens correction is unavailable with the F lens.

The K-x does tend to clip highlights. If you shoot in raw they are almost always recoverable. My K-x is normally set for center-weighted metering, because it's more predictable than matrix. I occasionally use spot metering with my K20D but almost never with the K-x, because I don't like menu diving. If you use spot metering, best to use AE-L also.

A lens hood won't affect highlight clipping, but is highly recommended to improve contrast and reduce flare. I use a hood all the time, with every lens in my bag. I have the official Pentax rectangular hood for my 50mm, but mostly I use standard rubber collapsible hoods because they let me mount a CPL easily and take up no room in the bag. The best coverage for a 50mm lens on APS-C is a metal tele hood, but they're more money and bulkier to store. If you want to use a metal hood and CPL, get a CPL to match the filter size of the hood.

---

Last edited by audiobomber; 08-03-2012 at 10:39 AM.

[audiobomber]

First, I suggest you to put the lens in "A" mode, and play with the aperture trough the body instead of the aperture ring.

Good point. Electronic exposure is more accurate than mechanical.

[baro-nite]

Good point about in-camera lens correction -- that sounds more likely than my suggestion.

[jzackery]

I checked the diaphragm and it seemed snappy.
Restored default factory settings.
Have been shooting in the aperture mode (Av) pretty much exclusively as well..

I thought what the problem initially was, was having it set to spot metering. I changed it to both center-weighted and multi-segment as well to see if it would fix it but I was primarily getting similar results with blown out images, and a strong tendency to overexpose everything.

Here's an example of what I get with the "proper" exposure as determined by the camera on the right. On the left is an image of the same place but underexposed almost as far as you can go that produced an image closer to what I actually see with my own eyes. Also, I got a better result when using spot metering, the other settings seemed to overexpose way too much in this particular situation

Data for the right image: F/2.2, 1/4 sec shutter, iso 1600, exposure bias 0 step
Data on the left: F/2.2, 1/15 sec shutter, iso 1600, exposure bias -2.3 step

Would shooting on shutter priority help, as that seemed to make the biggest difference?


Maybe its just an issue that i'll have to be more sensitive to when im in lower light, or brighter areas. It would be nice to not have to fiddle with settings so severely when shooting casually in jpeg mode though.


THanks for all the help so far!

Attached Images

 

[Frogfish]

You have your images reversed above !

I have found that, even with the K5, it has a propensity to blow out highlights with certain lenses (which are of course never recoverable since the information has not been recorded - unlike with shadows) and almost always benefits from an EV adjustment. I usually have EV set to -1 but sometimes as much as -2.

[baro-nite]

Here's an example of what I get with the "proper" exposure as determined by the camera on the right. On the left is an image of the same place but underexposed almost as far as you can go that produced an image closer to what I actually see with my own eyes. Also, I got a better result when using spot metering, the other settings seemed to overexpose way too much in this particular situation

As Frogfish points out the images are reversed from your description, so I am assuming that the darker image is the one you feel is correct, in that it best matches what you saw. But the camera doesn't try to match how the scene appears to the eye, it tries to produce a balanced exposure. Has to, because the dynamic range of the camera is so much smaller than that of our eyes (if the eyes are allowed time to adjust to varying brightness). The brighter of your two exposures looks correct -- i.e., exactly what I would expect the camera to produce. If you meter off something dark, the image will be bright. If you meter off something bright, the image will be dark.

[audiobomber]

It looks to me like your monitor is the problem, not the camera. The darker image is underexposed and I don't think the highlights are blown in the brighter image. Also, it looks like you're using a high ISO. which seriously reduces dynamic range.

You need to understand how the camera meters:
http://www.cambridgeincolour.com/tutorials/camera-metering.htm

[audiobomber]

I have found that, even with the K5, it has a propensity to blow out highlights with certain lenses (which are of course never recoverable since the information has not been recorded - unlike with shadows)

Highlights are not recoverable in jpeg, but they certainly are if you shoot raw. I routinely recover 1 Ev with the K-x.

[Frogfish]

Highlights are not recoverable in jpeg, but they certainly are if you shoot raw. I routinely recover 1 Ev with the K-x.

I shoot RAW. And it isn't possible to recover blown highlights because the information just hasn't been recorded.

Those that you are recovering are not blown.

[audiobomber]

I shoot RAW. And it isn't possible to recover blown highlights because the information just hasn't been recorded.

Those that you are recovering are not blown.

I shoot raw and use Pentax DCU4 for conversion. I have the software set to flash on blown highlights (VIEW - Show Bright Portions - All). Often, especially with the K-x, there are highlights flashing when I open the file. If I convert to jpeg without processing, the blown highlights show no detail, i.e. they are blown, and will continue to flash. If I reduce the exposure by a stop, even a stop and a half before conversion, I can bring back detail in the same highlights. So I'm quite certain what I said is correct.

I've often heard that highlight recovery is a desirable feature of shooting raw. I believe Pentax has been criticized in reviews for only allowing about a stop of highlight recovery in raw.

[Frogfish]

I shoot raw and use Pentax DCU4 for conversion. I have the software set to flash on blown highlights (VIEW - Show Bright Portions - All). Often, especially with the K-x, there are highlights flashing when I open the file. If I convert to jpeg without processing, the blown highlights show no detail, i.e. they are blown, and will continue to flash. If I reduce the exposure by a stop, even a stop and a half before conversion, I can bring back detail in the same highlights. So I'm quite certain what I said is correct.

I've often heard that highlight recovery is a desirable feature of shooting raw. I believe Pentax has been criticized in reviews for only allowing about a stop of highlight recovery in raw.

Wikipedia :

Areas of a photo where information is lost due to extreme brightness are described as having "blown-out highlights" or "flared highlights".
In digital images this information loss is often irreversible, though small problems can be made less noticeable using photo manipulation software. Recording to RAW format can ameliorate this problem to some degree, as can using a digital camera with a better sensor.

Me :

If your highlights are 255,255,255 they are toast.

[tibbitts]

I shoot raw and use Pentax DCU4 for conversion. I have the software set to flash on blown highlights (VIEW - Show Bright Portions - All). Often, especially with the K-x, there are highlights flashing when I open the file. If I convert to jpeg without processing, the blown highlights show no detail, i.e. they are blown, and will continue to flash. If I reduce the exposure by a stop, even a stop and a half before conversion, I can bring back detail in the same highlights. So I'm quite certain what I said is correct.

I've often heard that highlight recovery is a desirable feature of shooting raw. I believe Pentax has been criticized in reviews for only allowing about a stop of highlight recovery in raw.

It seems like raw recovery should mostly depend on a combination of the sensor and the exposure. Reviews I've read for same-sensor models seem to indicate that raw recovery differences are at least mostly due to an under or over-exposure bias, or differences in firmware settings. It might be that some manufacturers' models default to "extended highlight/shadow" modes, while others don't (and some of those features apply to raw, while others don't.)

Paul