SYNOPSIS: Practical Pentax Improvements (Part 1 -The Setting)...

[GordonBGood]

GordonBGood wrote:

The camera has no control over the aperture. You would be doing double duty setting the aperture ring and then re-entering the data into the camera. In the end, f/4 is two stops slower than f/2, so you are still effectively doing a sort of Ev comp with the camera.

I am under the (perhaps mistaken) impression that the camera body can control the amount of aperture for the image though how much it depresses the aperture actuation lever. I thus don't suggest that one would both set the aperture ring and the camera body setting, but rather just leave the lens set to minimum aperture (maximum aperture value) and let the camera set the aperture for the image based on the maximum aperture just as it does now for 'A' lenses.

Here is what I'm getting at. We know that right now with a typical scene, the aperture ring set wide open, and Ev comp set to zero that the camera can meter just fine with M lenses despite not knowing what the aperture is. Lets say the correct metering in this scenario is 1/1000th sec. Now the camera stops down the lens, re-meters and comes up with 1/250th sec. It still doesn't know the aperture, but it knows it looses two stops of light when it comes time to expose. If you did as you suggested and input the max aperture it could calculate the set aperture of the lens just off the exposure info. No need to fool with E-dials every time the aperture ring is changed. Then if you spin around and it meters 1/500th wide open, it knows it needs 1/125th sec for proper exposure.

Yes, your method would work as well, with the camera determining the difference between the maximum aperture as is known and the current aperture ring setting. The main problem with it is that stop down aperture metering gets less and less accurate below f/5.6 especially in low light.

and let the camera take care of the compensation calculations and the stopping down to the requested aperture, which is the behavior with 'A' lenses in that the 'A' position is equivalent to locking the camera to wide open aperture and giving the control to the camera.

It sounds like you are trying to make an M lens behave exactly like an A lens. I don't know if that can be accomplished just by the camera knowing the max f/ stop. If that were the goal, the lens would have to be set to min aperture - which is where it is on the A setting. I don't know enough about the technicals to say whether it has merit.

Yes, I suppose that is what I am suggesting as detailed further above. I don't know for sure whether it would work either, but it would seem worth investigating.

Regards, GordonBGood

 

[GordonBGood]

Limited Practical Improvements

1) Metered overexposure of high contrast scenes : I>

2) Improving the built in flash systems use of Auto ISO with flash use : When there are dark backgrounds, the Pentax flash system from even cameras before the K-7 and K-x and including them increase Auto ISO to its set maximum limit too frequently, resulting in too many flash shots taken at a higher ISO sensitivity than necessary when the focussed subject(s) is(are) much closer than so as to require use of that higher ISO sensitivity.

I agree with this point and will add another, I have found that the K7 not only tends to ramp up ISO to 1600 when using the flash (ie within 5 feet of a subject) but tends to over expose. I just took a series of shots indoors at a museum with the 12-23 DA and the 540 flash unit at average distances of 3-5 feet and it was blowing highlights at 1600.. I had to dial down the flash unit but on the second day, I forgot to do this and blew some really great shots of my son inside a train. The train was black and perfectly exposed but my sons face was blown. I lock exposure to AF point and allways use the AF indicators in the VF so I am sure I had focused on his face.....

I think that the reason Pentax cameras tend to overexpose for close subject distances is due to not properly compensating for subject (as known by focus) distance, thus my improvement as stated would also help this.

Regards, GordonBGood

 

[GordonBGood]

1) Metered overexposure of high contrast scenes : I think that the newer Pentax cameras, and in particular the K-x that lacks the new 77 element sensor of the K-7 to do more refined matrix metering, are consistently over exposing by approximately two thirds of a stop in high contrast scenes as to tendency to clip strong highlights. I know that in the past Pentax have been accused of too often underexposing too much to preserve highlights, and this may well be an overcompensation to that criticism as well as Pentax's attempt to minimize noise by strongly attempting to Expose To The Right (ETTR). In my opinion, it isn't a problem as I believe in taking enough test samples to learn my camera's exposure system so that I can predict what amount of plus or minus EV compensation I need to dial in for different scene variations, but my wife as a less experienced user feels that she shouldn't have to to that. For the sake of removing one of the somewhat major "cons" that appear in the DPR and likely some other reviews, it would likely be better if this tendency were lowered by appoximately two thirds of a stop.

Gordon - isn't your suggestion just reverting, in effect, to an exposure style that earned Pentax a lot of criticism? If we want a system that is simple for inexperienced users, surely we don't want those inexperienced users to be disappointed because their images look too dark.

Well, I don't think I am suggesting to reverting, but rather to just having the option to apply a exposure bias to the current metering system that will optionally (as per others suggestions) bias the exposure for high contrast scenes as recognized by the matrix metering of the camera to expose not quite as strongly to the right, with perhaps the default about two thirds of a stop lower than the current default so as to prevent "cons" as per the DPR Reviews. I actually like the way the K-x exposes and just apply some -EV for high contrast scenes.

I haven't devoted a lot of thought to this so forgive me if my suggestion is naive, but isn't there more that can be done? It seems to me that the crucial phrase in your note is "high contrast scenes". Couldn't the metering system measure the DR between the bightest and darkest segment (whether 77 or fewer) and then alter the TRC depending on what it sees? Perhaps in an ideal world there might be a family of curves but at present I imagine there are only two: the "normal" one and the one used in extended D-Range.

What I'm thinking of is that when the DR is read as over (say, purely for example) 8 stops the TRC is the D-Range one that boosts shadows; otherwise it defaults to the normal curve. The camera would expose in the way you suggest to protect the highlights.

Note that I'm assuming this is only the shape of the curve, so one could still set ISO to 100 (native or not).

I actually like that the TRC is fixed as per your specified Custom Image settings including Shadow Correction and/or Highlight Correction as having this variable would introduce yet another inconsistency between images. What I am requesting is really quite simple, as the matrix metering system already knows what constitutes a high contrast scene where this proposed exposure bias would apply.

I'm not sure of the exact time difference but you are probably in the heart of your celebrations as I write. If so, they'll be done by the time you read this so I'll hope that you enjoyed them and wish you all the best for the new year.

Over and done with for this year.

Wishing you the very best in the new year!

Regards, GordonBGood

 

[GordonBGood]

To all who responded to this thread, thanks on everyone's behalf for the very positive set of suggestions and recommendations. Frankly, I was surprised to be able to expand on my short list with so many useful little tweaks that could make Pentax cameras the very best in the industry.

It was suggested further up the thread that I do a summary of what was found, and this is it, in as many parts as necessary to fit the posting limits and to divide the types of improvements into categories. I had originally intended that this thread be more restricted to those types of things that could easily be implemented with firmware changes and tweaks, but there were many other valuable suggestions which should be input to Pentax engineering for their consideration on the design directions of future Pentax models that I have included them here for ease of reference. These are as follows:

General design philosophy suggestions:

A) Make the cameara OS open source so all can contribute, which is an entirely diffent idea for any camera manufacturer to accept and unlikely to happen.

Some large general improvements that could be incorporated into future Pentax designs:

a) a Blue Tooth or other wireless flash system built both into the cameras and into future Pentax external flash units.

b) addition of a body AF clutch so that quick shift focus would be body based for all lenses, although that would only make non-SDM lenses quick shift.

c) strap lugs on the battery grip.

d) Modification of the USB port pin out so the same port supports wired remote as well as its current USB connection funtion.

e) A seal on the bayonet mount to prevent dust and moisture ingress at the joint between lens and body.

f) A built-in Acra L plate at the bottom of metal bodied cameras.

g) The PC flash sync port should be on the top or front of the camera body rather than on the side below the flash where it is on the K-7 body and the sync speed shot be at least 1/250 second to be as high as the competion.

h) It was suggested that there be an inductively coupled power connection to reduce the number of connection ports that would need to be sealed.

i) An AF assist lamp on all future Pentax DSLR bodies.

j) Various video improvements that I think will evolve on their own over the next few years, including the following:
1) full 1080p HD when the available sensor technology can support it.

2) more user adustable parameters including shutter speed within the frame rate, aperture, sensitivity gain limits, sound wind filtering, and availability of AF tracking as well as reduction of SR noise in the sound track. Also suggested as desirable would be the ability to grab a full resolution still image without causing a break in the movie stream, although this is nearly impossible unless the full sensor resolution were being read for each of the movie frames, which currently is not technically possible.

k) Implementation of HDR aligning and stacking, lens correction, and extra horizon leveling capabilities in a new hardware imaging engine (PRIME3) so that these can all be done much more quickly than using the main CPU or can be done at all.
K) more convenient access to the flash memory past the memory cover door.

l) A stiffer mode dial when there is no lock so this can not so easily be accidentally moved.

m) The status display in the viewfinder is often too dim for bright scenes.

n) GPS interface capabilities including the ability to geotag images.

To be continued in Summary 2 to cover firmware type tweaks - GordonBGood

 

[GordonBGood]

continues from Summary 1...

As per my original list of things that could mostly be done with firmware tweaks, most seem to feel that my ideas were quite positive improvments, with the following added tweaks:

1) To correcting the tendency of overexposure for high contrast scenes, it was suggested that there by a (Record Menu?) option to tune this exposure bias in stops as fine as one sixth stops, although I think even one third stop increments would be adequate.

2) To improving the flash system as to incorporating focus distance in the setting of the maximum Auto ISO, most seemed to find this a good idea although there were many other flash improvements suggested, which are summarized other places.

3) To allowing Auto ISO in Manual Mode, most found this to be a good idea for the K-x and were ambivilent about it's use with the K-7 due to that camera already having TAv mode. It was suggested this this capability be able to be enabled as a menu option, with current status of disabled the default, especially for the K-7, and I think that would be a good idea.

4) No one objected to to idea of being able to save a raw file from the recorded raw data which is in memory until the next shot of the camera is powered off, even for JPEG shooters.

5) There were several suggestions to increase the limits of things that seem to be needlessly limited, among them the limit of only 99 shots for the intervalometer mode of the K-7, + - EV Exposure bracketing increased to + -5 EV for all bracketing modes and for EV overrides for all cameras, the forcing of Dark Frame Subtraction (DFS) Long Exposure NR ("Noise Reduction") for time exposures over 30 seconds even when the control is set to off, etc.

6) Continued improvement to the Auto Focus (AF) system, especially as to predictive continuous tracking at least up to the standards of the Nikon D5000/D90, whether this can be done in current cameras or in the next generation. Related to this, be sure that Continuous AF does not slow the maximum continuous frame rate. This would include improving the contrast detect AF time.

7) Improvements to the user experience when using manual lenses with the camera in using the input focal length used for SR in helping choose a shutter speed, and also allowing the input of a maximum aperture so that when combined with the input of current desired focal length, manual lens metering could work as for automatic aperture lenses and the usability of Av mode would not necessarily be limited to use of manual lenses wide open. Related to this, it was suggested that the Aperture Ring Enable setting in the Custom Menu is unnecessary, as most of us who use manual lenses leave this enabled at all times, but the point could be made that it prevents those who might use lenses with an 'A' setting from trying to use the lens when they have accidentally moved the Aperture Ring away from the 'A' position.

8) It is pretty commonly accepted that all users would like to have the ability to adjust focus for a range of lenses as the K-7 has but over a suffiently wide range to cover the needs of all lense, and I see no reason that that capability should not be available for all Pentax models, with models distinguished by other features.

9) A good suggestion is that when one hits the Menu button, the menu open at the last selected item the last time the Menu system was closed for convenience in adjusting one specific parameter.

10) Another added convenience would be the addition of multiple Custom White Balance setting, although perhaps this is not possible for current models in that it would require some redesign of the MakerNotes format in order to record the extra settings.

11) A common theme for many was the request for a programmable Fn button, which is not likely possible for such cameras as the K-x in that it doesn't have many extra buttons and the green button is already programmable, but for cameras such as the K-7 there are a couple of canditates while in shooting mode in the RAW button and the Delete/Trash button. Possible functions that could be programmed into these would be things like the missing direct access to Auto Exposure Bracketing, Direct setting of Flash mode, Drive mode, etc. There may be some improvement that could be made as to setting WB preset that would not require that it have an optional direct access button.

to be concluded in Summary Conclusion - GordonBGood

 

[GordonBGood]

Continued from Summary 2...

12) Use of the e-dial where ever possible, which should include selecting WB presets for all cameras and selecting Auto ISO upper limit and manually selected ISO for the K-x range of single e-dial cameras. This would actually help alleviate the requirement for a programmable direct button for WB presets.

13) Improvements of Multiple Exposure and HDR "state" as to user interface so that one can review the settings and results to the current "stack" without leaving the mode and having to re-enter it.

14) A Custom Mode that allows forcing SDM lenses to revert to screw type AF for those lenses that support it for the case where SDM fails but the screw drive still works.

15) It is the general feeling that all cameras should support tethered shooting via the USB port. If there were a standard PTP interface that supported this, the software that supports it would not have to be supplied by Pentax.

16) Display of ISO sensitivity for every Standard Info display screen where aperture value and shutter speed are displayed.

17) Option of displaying a raw data histogram rather than a JPEG histogram even though the raw histogram is shown with a non-linear horizontal scale to be more useful.

18) More attention to the AF point multi select mode to avoid changing the selected AF point inadvertantly.

19) An addition to a Custom Menu so as to be able to change the wireless flash channel without having to mount the flash to the camera and/or autosetting the channels by attempted communication on all four channels. Other flash improvments would be better integration with flash Auto modes. Also, have the aperture stop down before the pre-flash for manual lenses rather than just for the actual exposure as now.

20) A trap focus mode that works for all lenses and which when armed stays armed until the image is shot or explictly ended. This mode might even include an automatic re-arm for bird feeder shots.

21) The ability to specify a SR focal length multiplier for TC's that aren't recognized by the camera.

22) That the mirror should be locked up over the entire span of HDR images.

23) Optional use of the LV button to enter movie mode through some button sequence for convenience over having to use the mode dial.

And that concludes the list of this type of suggestions that I was able to glean from the thread. If I've missed or glossed over any that you feel are important, please accept my apologies, but these are the things that I feel are possible within the terms of this thread. Again, thanks for your participation. I will use the remaining few posts to try to reply to any further comments, but there are just a very few posts left before the thread is full so may be limited in my ability to do so. I think that most discussion has been done and see no need to open another follow-on thread, at least on the same subject.

So, to wrap things up: To Pentax cameras and World Peace.

Best regards, GordonBGood

 

[GordonBGood]

Ryan, I'm not sure what "cheat the sync" means, but I've added the others to the master list.

Regards, GordonBGood

 

[GordonBGood]

Forget about MF zooms, they rare anyway.

Agreed, and even at that many had the same maximum aperture for all focal lengths over the zoom range.

Regards, GordonBGood

 

[GordonBGood]

Well, deleted most of the response as it would be a hijack of this thread, but do you know whether the D300 and D5000 share the same jpeg engine ? (ell let me check the specs and see for myself...so, never mind...)

If you look at IR photos, it's clear that D300s does more smearing than D5000, at least the default NR does. As a result D5000 seems to have more detail than D300s at ISO 6400. Why would Nikon do that to its more expensive model? I don't know, but maybe sensors are not identical? Or maybe jpeg engines are not identical? Whatever it is, the main reason probably is fps. Nikon probably had to change something for D300s to perform 8 fps.

This is really OT to the thread but I think that Nikon just modified its thinking about JPEG Noise Reduction (NR) between models, and has even modified their thinking on raw NR between even more similar models in the D90, which has extra raw NR at high ISO's and the D5000 which does not. NR as applied by the current Nikon cameras is performed by the hardware imaging engine during the workflow streaming and thus doesn't really take any real processing time that would affect continuous shooting rates.

Regards, GordonBGood

 

[GordonBGood]

A common API... yes, so if Pentax won't we will. But I thought about my 26' long USB cable I use, its a 20' USB cable with a booster amp (from RadioShack) to allow long USB runs, I plug in a regular 6' cable and have about 26' of USB cable for a cost of $45. I can tether it to my notebook and just about completely control my K20D from inside the house while the K20D sits outside on a tripod, I love it. However I thought of a major improvement Bluetooth I just bought a bluetooth adapter for my PC, so now I can download pics from my phones to any PC or notebook I plug it into. If a dSLR had Bluetooth I could transfer pics (slowly) and tether it with no cables, allowing more freedom of placement and saving the hassle of the hook up; plus the Pentax dSLR can remain shut tight to allow you to use it tethered in any weather . One problem Bluetooth has low transfer speed.

So maybe both a wired tether option and Bluetooth option. The Bluetooth could just send a compressed JPEG to the PC during operation so you can see the pics as fast as the Pentax dSLR takes them. Also design the dSLR to use the tripod mount as an extension for the Bluetooth antenna (if possible). When the user is finished they could transfer the data off the SD card in the normal manner. This gives IMO a cheap but powerful alternative to tethering in short (30') range situations that has many benefits over wire, and should have just enough range for most people. Wire has it benefits of speed but really requires special equipment for long range (over 20') operation and its cumbersome. Also this allows wireless communication with printers and... I believe some cameras do have Bluetooth but honestly don't know. I just know we don't.

Whoops kinda late in the night when I was thinking of this and I missed what Pentax should really use in my excitement over my new little Bluetooth adapter Wi-Fi that ends the slow speed problem.

So..

1. Bring back remote operation of Pentax dSLRs

2. Improve upon the USB cable method which was cumbersome by using some wireless transmission such as Bluetooth or better Wi-Fi which allows many other benefits like downloading to your PC without touching the SD card.

I'm sorry, James, but I see nothing new here on which to comment.

Please make a new post Gordon when we hit 150

As I say in my summary, I am not inclined to start another thread on this subject, as I feel that everything has been pretty well covered here, at least as to my stated purposes in starting the thread in the first place.

I just want to get back to my Thai whiskey, girly bars, and a few mathematical analyses thrown in between, maybe even very occasionally taking an actual real image

Best regards, GordonBGood