|
Forum: Pentax DSLR Discussion
11-13-2013, 12:09 PM
|
| |
Interesting that you're not seeing linear metering either. For my part, I'm kind of resigned to the fact that accurate exposure metering with manual lenses is always going to be difficult on my K20D, whatever focusing screen I end up with. I just want to get it to the point where I don't actually miss any quick shots through gross exposure errors. I shoot RAW so normally if I'm there or thereabouts, I can tweak the exposure in post-processing and get the result I need. And if I have time, I can apply compensatory EV - I just need to get to know and remember where the errors are with each lens. And if I have even more time, I can simply resort to chimping!
|
|
Forum: Pentax DSLR Discussion
11-13-2013, 02:56 AM
|
| |
Yes, an impressive bit of work! I realize I have neither the dexterity nor the patience to do this, and know full well that if I attempted it, half way through I'd be wondering why on earth I didn't just pay the extra for the pre-cut one from focusingscreens.com.
I'm very interested in your use of tape there, both as a protection layer while cutting and as a means of removing dust. I have read so much of the delicate nature of these screens that I would not have been bold enough to do this, but your experience suggests there's no problem.
I'd be interested to hear how the new screen affects metering with the older lenses. My primary motivation for changing the screen was not, in fact, to do with focusing issues, but to improve exposure metering. I believe my screen (from the *ist DL) has indeed helped with this, making it much more linear than it was with the stock screen, although it's still not perfect. The curious thing is that it is now 'wrong' in the opposite way from with the stock screen, in that my M50 1.7 now underexposes slightly at f22 and overexposes slightly at f1.7.
Oh, and I believe my fears about needing to shim my screen were unfounded. The severe back-focusing problem I was seeing was, I think, user error! There is a tiny range through which I can turn the focusing screen and still have the centre point in focus, and when I tried it at the other end of this range, focus was spot on in the resultant image, so I think it's more to do with the tiny depth of field at f1.7.
|
|
Forum: Pentax DSLR Discussion
11-12-2013, 01:55 PM
|
| |
I think I may well end up going for one of the Canon S-type screens, I've been reading through some very positive old threads here about them. Whatever happens, I think I'll need to become familiar with shimming, because I've discovered that I now have a pronounced back-focusing problem with my K20D, at least with the Pentax M 50mm 1.7 lens I've been testing it with. When I focus on a bit of text at arm's length and take the shot at f 1.7, I'm finding when I look at the image that the point in sharp focus is about an inch behind where I was focusing. An inch at arm's length seems quite a lot to me. What does this suggest I need to do, in terms of shims? The amount by which it out makes me wonder whether I might have damaged something when fitting the screen.
|
|
Forum: Pentax DSLR Discussion
11-09-2013, 05:29 AM
|
| |
I'm not sure I'd want to be fiddling around with shims, to be honest. Is it tricky to do?
|
|
Forum: Pentax DSLR Discussion
11-08-2013, 10:05 AM
|
| |
Well, it's fitted and working, although I can't say it was a pleasant process. I had difficulty getting the screen out of the *ist DL, and then had difficulty getting the latch to close having put the screen in the K20D. It probably would have been a lot easier with the special tool, and I wouldn't try and do this again without one. Having said that, if the size is, in fact, wrong (it did look slightly bigger), that might explain my difficulty getting the latch to close.
I had read that the screen from the *ist helped the metering problems with manual lenses on the K10D (and, I think K20D) - I believe it was Lowell Goudge on here who mentioned it. Doing a bit more research, though, I get the impression that the screen I need is the one from the *ist DS, and the body I have is a *ist DL. It didn't strike me as a possibility that the focusing screens on the *ist DL and DS might be different sizes, but there are suggestions here - Focusing screens for Pentax DSLR's, period...: Pentax SLR Talk Forum: Digital Photography Review - that they might be.
Anyway, it hasn't proved a magic bullet. The metering appears to be better, but there is still a difference of maybe up to couple of f-stops (on my 135mm M lens) between the exposure set by the green button at f22 and f3.5, and at f8 shots are overexposed by probably half a stop or so. On the positive side, manual focusing is a little easier. I think I may try and get hold of a brand new LL-60 screen which I know fits, and see if that helps. It'll come with the requisite tool!
|
|
Forum: Pentax DSLR Discussion
11-01-2013, 09:52 AM
|
| |
Thanks, guys. I'll look around and see if I can find something around the house which could serve the function of the silicone air pipe. I'll report back.
|
|
Forum: Pentax DSLR Discussion
11-01-2013, 06:41 AM
|
| |
I have a bunch of old high quality prime lenses which I like to use with my K20D. I've been working around the well-known problems with erratic metering, but having read that the focusing screen from the *ist DL can give better results, I decided to give it a go. I've managed to pick up a non-functioning, 'for parts only' *ist DL cheap on ebay. Unfortunately I don't have one of the little tools which comes with the focusing screen when you buy one separately. Has anyone got any tips for doing the swap using some other means (like standard household tweezers, for example)? Or will I end up irreparably damaging both focusing screens, and maybe even the camera itself, which is the normal outcome when I take it upon myself to do such a job :)?
|
|
Forum: Pentax DSLR Discussion
07-14-2009, 10:41 AM
|
| |
Having done a search here, I'm rather shocked to see this problem is so widespread. In my case the problem is quite gross, immediately visible to the naked eye, so it's probably some degrees out.
Regor, when you say 're-setting the focussing screen', could you perhaps say a little more about what you mean? In view of the fact that the images come out level when I have the camera sat on a known level surface, would it not suggest that the problem is not the sensor (which would be aligned correctly with the camera body) but rather somewhere in the viewfinder?
|
|
Forum: Pentax DSLR Discussion
07-14-2009, 03:42 AM
|
| |
I started noticing a while back that my photos were not level, the horizon appearing to tip down to the right. At first I put this down to user error, although it's not a problem I've usually had, but it was so consistent I decided to do a check. I established that if I rest the camera on a known level surface like a table, the shots are level - it's only when I use the viewfinder that the problem occurs. It struck me that the only rational explanation for this (other than user error) is if the view finder is somehow misleading me. I took the lens off and had a good look at the mirror, and it seems to me that it is slightly misaligned, not quite horizontal.
So, my questions are these:
1. Is this a problem anyone else has come across before?
2. Would a misaligned mirror cause the problem in the way I describe?
3. If so, is it likely to be a user-fixable problem, or will I have to send the camera away for repair?
|
|
Forum: Pentax DSLR Discussion
05-28-2009, 02:59 AM
|
| |
MTF is modulation transfer function, and it's a way of objectively measuring sharpness. There's a good explanation here: Understanding Lens Contrast. The MTF data for modern lenses is apparently stored in the lens chip and can be queried by the camera, allowing it to set the aperture to where it is sharpest, according to the MTF data (or at least, prioritise that aspect, as opposed to depth of field or shutter speed).
|
|
Forum: Pentax SLR Lens Discussion
05-27-2009, 02:59 AM
|
| |
I wouldn't be surprised if it IS the Pentax 12-24, just with a different label on it. That's certainly the case with the 18-55 kit lens - Pentax lens, badged as a Schneider on the Samsung.
|
|
Forum: Pentax DSLR Discussion
05-27-2009, 02:00 AM
|
| |
I'm sorry, I don't really get how that works. What is this chart telling me?
|
|
Forum: Pentax DSLR Discussion
05-27-2009, 01:59 AM
|
| |
That certainly seems an intelligent approach and one I should think before overriding. Thanks!
|
|
Forum: Pentax DSLR Discussion
05-26-2009, 01:59 PM
|
| |
I was a strictly aperture priority (or pure manual) film camera guy before I got my K10D, so I've never used Program mode on an SLR before, and I'm a little puzzled about it. I know on the K10D you can set the program line method to prioritise aperture, shutter speed or MTF, and I'm pretty sure I understand the effects of doing all of these. You can, though, just leave it on 'Normal', and this is the one I don't understand. How does the 'Normal' program hit upon the combination of aperture, shutter speed and ISO needed? It presumably has to prioritise something or the settings would be effectively random.
|
|
Forum: Troubleshooting and Beginner Help
05-25-2009, 01:13 PM
|
| |
Yes, I discovered this a few days ago and have posted about it elsewhere myself. The reality of the problem is more complex than merely overexposing, as at the wide open end of the range, the K10D tends to underexpose. I've been trying to find out what apertures or aperture ranges each of my M lens will meter correctly at, so I can work things out from there (e.g., if I know it's metering OK at f8, I can go 2 shutter speeds slower and use it safely at f16, ignoring what pressing the green button tells me for the 'out of range' apertures). It seems to be most accurate for my Rikenon 1:2 50mm prime at f5.6, for example. Fortunately the range where it is accurate for my Pentax 300mm seems to be quite wide, which is handy.
|
|
Forum: Troubleshooting and Beginner Help
05-25-2009, 01:02 PM
|
| |
Yes, you're quite correct, of course. I guess the off-centre sky was not quite bright enough to override the general exposure set from the rest of the scene. I'm just going through a bunch of shots I took today, mainly rather hurriedly, using my SMC-M f2.8 28mm, and trying to learn what I can from their histograms. I will check out the histogram tutorials out there on the web as well! There is a hell of a lot to learn with a DSLR like the K10D, isn't there?
|
|
Forum: Troubleshooting and Beginner Help
05-25-2009, 11:24 AM
|
| |
Well, I used an M series lens, which meant that I was inevitably using centre weighted metering, and the sky was not in the centre and was thus not metered for. All this is slowly coming together for me, and I'm very grateful for the help on here to speed the process!
|
|
Forum: Troubleshooting and Beginner Help
05-25-2009, 01:49 AM
|
| |
Peter, thanks for your lucid and helpful response.
Further digging on one image revealed that it was the blue channel which was getting blown, and looking in Raw Therapee (at which I am a complete beginner) showed highlights clipped in an area of blue sky. So could it be as simple as that ? Where I have exposed for the foreground, the sky is always likely to show clipping? In which case, would it manifest itself as a very sharp spike in the blue channel at the extreme right of the graph?
I'll maybe post an image later on if today's experimentation doesn't resolve all this for me.
|
|
Forum: Troubleshooting and Beginner Help
05-25-2009, 01:43 AM
|
| |
The spike is cut out by the edge of the screen. It occurs on lots of images, even ones simply of foliage which was what made me wonder whether it was indicative of some camera problem. I'm doing some more experimenting today which should give me some more useful information.
|
|
Forum: Troubleshooting and Beginner Help
05-25-2009, 01:39 AM
|
| |
I think you may have hit the nail on the head here. Following the recommendations on a review I read, I had increased the JPEG saturation and contrast by +1 from their defaults. When I got around to looking at the converted RAW images side by side with the JPEGs, I found that the histograms of RAW ones were much more as I would have expected. I need to do some more experimenting today with the JPEG settings restored to normal.
|
|
Forum: Troubleshooting and Beginner Help
05-24-2009, 01:16 PM
|
| |
Yes, I'm sure Google will lead me to plenty of general information about histograms. Actually, I was wondering whether it was something I needed to be aware of with this camera specifically, whether there is something odd about the extreme spike at the right hand of the graph on what appear to be correctly exposed images.
|
|
Forum: Troubleshooting and Beginner Help
05-24-2009, 11:14 AM
|
| |
Recently got myself a K10D as my first DSLR, having been a user of Pentax film SLRs for many years, and am just getting my head around how to use it well. I'm a little puzzled about what my histograms should be telling me. I have a couple of images I took to test lens resolution, featuring a mix of tiled roofs and trees. To the naked eye the exposure looks OK, and in Photoshop when I do Image->Histogram, everything seems fine. But when I go into Adjustments->Levels, the histogram for the RGB channel shows a very sharp narrow spike at the extreme right, the full height of the graph. Is this indicative of the loss of some highlight information? What should I read into it? Sorry if this is more of a Photoshop question, I'm just not sure what belongs where at this stage of the learning process.
PS This is just with JPEG, I've not got on to playing with RAW yet.
|
|
Forum: Pentax SLR Lens Discussion
05-23-2009, 01:01 PM
|
| |
Thanks, everybody, for all the feedback. I'm interested that the superiority of the primes is something people here have actually seen with their own eyes, as so often there are assumptions made which are not tested. The kit lens will probably work pretty well for me, in fact, as (for the moment!) I have no primes below 28mm, so that 18mm is handy, and I'm likely to be happy enough with f8 and above.
I'm quite happy with the use of primes, but I have just been a little taken aback by the exposure problems with stop-down metering on M lenses, something which I had not heard about before I bought the camera. Having said that, there does seem to be an aperture, or range of apertures, for each lens where the metering is fine, and I just need to do extrapolate from that point. I did some testing with my SMC-M 300mm lens today and around f8/f11 the exposure seems fine, any problems being correctable in post-processing. I was pleased to get a good sharp shot, hand-held, of a gull in flight, taking advantage of ISO 800 setting (which allowed me a shutter speed of 1500), and shake reduction (which may not even have been necessary at that speed). It's a great camera, and I'm just beginning to get to know it.
|
|
Forum: Pentax SLR Lens Discussion
05-22-2009, 04:06 PM
|
| |
The primes I have in that range are:
1. SMC Pentax-A 1:2.8 28mm
2. SMC Pentax-M 1:2.8 35mm
3. SMC Pentax-M 1:1.7 50mm
I also have a Ricoh XR Rikenon 1:2 50mm which seems to have a glowing reputation for sharpness.
The shots I most care about are landscapes, so lens speed for low light isn't such an issue for me (although of course getting shots with a narrow depth of field can be important as well). On the whole, though, I'm more likely to be using f8 to f22 than wider apertures.
One thing which changes the game a little for me from film days is that the performance at ISO 800 seems to me to be so good that I don't need as fast a lens. I haven't really done any side by side comparisons with, say, ISO 200, but it's certainly pretty impressive.
|
|
Forum: Pentax SLR Lens Discussion
05-22-2009, 10:02 AM
|
| |
I recently got a K10D as my first digital SLR, my choice largely being driven by having a collection of good Pentax K mount lenses from film cameras. The K10D came with the 18-55mm kit lens, which I didn't have especially high hopes for. I haven't done any exhaustive tests myself yet, but various reviews I've since read suggest to me that the kit lens performs much better than one would expect, particularly away from the limits in terms of focal length and aperture - i.e., from, say 20mm up to 50mm, and from f5.6 to f16. Now, given that some of the primes I have fall into this range (namely 28mm, 35mm and 50mm), I'm wondering how much real-world difference I'm likely to note between the kit lens and my primes in this range?
The unquestioned 'truth' is that these older primes are inevitably going to be better than the kit lens, but has anyone rigorously tested this, with images to prove the point? I can't help but wonder whether the benefits of advances in computer aided lens design might offset some of the downside of cheapness of build in the kit lens.
|
