[Alex645]

Coach? Team? I don't mean to offend, but have you actually been paying attention to what I've been saying? These are scouts (i.e. boy scouts plus girls, because the Canadians - and Aussies too - play the game that way) doing their routine outdoor activities, often in fading light and inclement conditions. I have NO control over their movements, NO control over their location... it's see, shoot, respond, and the gear has to stand up to whatever weather is fit for the kids to go out in.

Sorry, if I offended you, but I'm not involved with the boy scouts, so I don't know what the leader or group is called. And yes, I did read what you posted. I was only relating as a father, photographer, ex-coach, ex-parent volunteer, that you're in a tough situation and that given the conditions, using a flash in a light drizzle or rain in the dark is problematic.
From my experience, as both an amateur and as a professional photographer and educator, my recommendation would be to consider compromising the convenience of a WR 18-135mm zoom and consider a faster prime, and thus eliminating the bulk of a flash and the zoom. If you check your meta data in a program like Bridge or Lightroom, what focal length are you mostly shooting at? I've found that most of us tend to shoot at the extremes, with something in-between on occasion. Yes, the wide angle prime is not WR, but in general primes are rarely WR because they are inherently better sealed with less moving elements and parts. If you're not doing huge printed enlargements, you can always crop a wide angle shot as well.

[Sandy Hancock]

I have had a Metz 58 AF-1 for a few years, and it has served me well. I bought a 540 FGZ II about a year ago and the Metz now only gets used when I'm using multiple flashes, or the Better Beamer (the model I have doesn't fit the Pentax).

I really like the compactness of the Pentax, and the WR is reassuring. The controls are also a bit more intuitive than the Metz. If the P-TTL delay is an issue, I shoot in Auto or Manual after a test exposure or two to get the settings which work best.

[WPRESTO]

The other thing about primes V-A-V seals and waterproofing , most SFL lenses in the near-normal range have unit focusing, which tends to "breath" less than zooming in-out so the lenses have less tendency to pull in moist air.

[Class A]

For most models, the time for preflash and calculation delay is about 0.02s.

Where do you get that figure from?

According to Imaging Resource measurements the overall time added by using flash (on the K-5) is 0.102s (see "Shutter Response (Lag Time)" and take the difference between the respective normal and "flash enabled" measurements). EDIT: Measurements by lister6520 even report 0.13-0.16s delay on the K-30 (and 0.31s-0.33s for the K-3) when using external flash.

I'd say for shooting action (moving kids), this is a considerable delay. The delay added by the K-3 (another 0.19s) leads to an overall entirely unacceptable delay.

One of the significant advantages of DSLRs over P&S cameras has been the considerable difference in shutter lag. If the OP does not need / want to capture decisive moments then I understand, but in general a noticeable shutter lag (whether caused by a pre-flash + metering or something else) is detrimental to photographing action.

The time delay involved in retaking pictures is far longer than anything the P-TTL system produces.

Sure, but with Auto-Flash (as available on the FGZ models), you don't get any delay, and with manual flash, you have an initial delay once (to figure out the right setting) but then every capture is done without further delays.

I'm not aware of any specific test comparisons between the older type 'red cross bars' AF assist beam and the new LED spotbeam technology.

I don't know of any either, but it is clear that the red cross type approach has advantages over a uniform AF light since it will create additional contrast for the AF system to latch on. With a red cross type approach, you can even focus on a white wall (or white shirt, for that matter), whereas a uniform AF light will provide no such support. I would expect the older approach to outperform the newer approach in terms of focusing speed, unless the LED is rather bright.

@pathdoc: Do you know how images using direct on-camera flash (i.e., without bouncing or diffusion) look like? If you do and don't mind the look, I'm of course fine with that. I'm of the opinion they are typically not worth taking. I'd get a light stand and a softbox (which can also protect a regular flash against some light rain) and use off-camera flash.

My theory, hmmm, hmmm, is that the AF tracking is ineffective for the same reason there is a delay in the Pttl flash; the metering sensor communication bus is slow.

AF is done through an entirely different system (using a sub-mirror behind the main mirror and a module at the bottom of the camera, whereas the metering module is attached to the prism).

---

Last edited by Class A; 11-07-2015 at 05:54 PM.

[pathdoc]

I am sure things would go a whole lot better with a softbox and off-camera flash... just not halfway through a nature walk on a drizzly dark evening through the woods on a rocky slope. If I were Reed Richards of the Fantastic Four, I might be more in a position to take your advice!

[Class A]

If I were Reed Richards of the Fantastic Four, I might be more in a position to take your advice!

So we agree, it's all your fault.

[pathdoc]

Whatever the delays with the flashes, it's got to be nowhere near as bad as mashing the shutter button repeatedly in frustration and not having it lock focus or fire. The delays being bandied about here, I can live with.

[stevebrot]

Where do you get that figure from?

I over stated. My source was the oscilloscope trace that I took for K-3 for P-TTL vs. non-P-TTL flash. It is actually for no flash vs. P-TTL. I have no idea regarding other models.

1 = K-3 No flash, 2 = K-3 P-TTL



Base latency to mirror up is 0.012s with no flash. For P-TTL, the value is increased to 0.188s with an additional 0.023s for pre-flash before mirror raise for a total lag of 0.211s (1/5s).


Taken from:

https://www.pentaxforums.com/forums/172-pentax-k-3/242556-k-3-delay-when-usin...ml#post2972999



Steve

[stevebrot]

I'd say for shooting action (moving kids), this is a considerable delay. The delay added by the K-3 (another 0.19s) leads to an overall entirely unacceptable delay.

Agreed. Thank you for the data from Imaging Resource.


Steve

[derekkite]

Class A the tracking and predictive AF uses the metering sensor. The pdaf sensors are indeed separate.

[Class A]

Whatever the delays with the flashes, it's got to be nowhere near as bad as mashing the shutter button repeatedly in frustration and not having it lock focus or fire.

Sure.

Is the AF-assist light coming on at all or is it just not strong enough?
I think for some reason, the AF-assist light is practically disabled when using AF-C (at least in combination with the rest of my settings). Switching to AF-S makes it come on for me.

I don't have any experience with using AF-assist lights from flashes so I couldn't say whether they are stronger / provide more coverage than the camera's own AF-assist light.

[Class A]

Class A the tracking and predictive AF uses the metering sensor. The pdaf sensors are indeed separate.

Ah, I missed the "tracking" part, thanks.

I still don't think that slow communication with the metering sensor is a source of problems:

1. Increasing the communication speed is easy whereas developing good predictive AF algorithms is not. If it only needed an upgrade in communication speed with the metering chip to make Pentax continuous AF competitive, I think Ricoh would have done it by now.
2. It is possible that the K-3 adds an available light metering step before taking the pre-flash in order to calculate the difference the pre-flash makes but this step shouldn't take anywhere near as long as the delay that we are seeing.

[pathdoc]

Sure.

Is the AF-assist light coming on at all or is it just not strong enough?
I think for some reason, the AF-assist light is practically disabled when using AF-C (at least in combination with the rest of my settings). Switching to AF-S makes it come on for me.

I don't have any experience with using AF-assist lights from flashes so I couldn't say whether they are stronger / provide more coverage than the camera's own AF-assist light.

Just not enough, it seems. I shot with it a couple of days ago and it was OK but those subjects had reflective jackets that lit up a charm and the focus locked beautifully.

[Class A]

Just not enough, it seems.

In that case, it may make sense to select the flash based on its AF-assist quality. They may not all be created equal in this department.

And as I said, I have my doubts about an AF-assist light that is homogeneous, i.e., not of the typical sharp cross-pattern type.
If I were you, I'd try to find FGZ II owners who also have experience with traditional AF-assist lights.

I just checked the AF-assist light of my Metz 58 AF-2. Its pattern is bigger and more contrasty (has internal structure) than the K-5 II's own AF-assist light, but it isn't brighter. On the contrary, it looks quite a bit darker. I don't know which of the two would perform better in practice, but based on the patterns, I'd say one would definitely have to perform a test to know.

[mcgregni]

I've got the older and newer AF-540FGZ models, so I'll try and run a few AF assist tests and report back.