[DBGrip]

I've read a bit about slow shutter speeds and performance with regards to sharpness, color depth etc; but I haven''t found anything about high shutterspeeds with the K-3, or K20D for that matter.

I want to know is there a shutter speed where the sensor in the K-3 starts to lose performance just from not getting enough time viewing the scene? 1/250 th vs 1/5000th say? I'm try to ascertain if there's a point I'd rather ND to maintain aperture setting rather than speed up the shutter.

Lenses have aperture sweet spots, zooms have focal length sweet spots, just reasoning that there's a shutter sweet spot for bright exteriors.

If so, is it different for the K20 (my other body)?

Thanks,

Rob

[redcat]

I don't think there is such a problem, 1/250 vs 1/5000 is the same as long as the camera gets enough light, don't worry
You can use ND filter if you want to keep the shallow depth of field (wide aperture) while shooting at mid-day without blow out the highlight

[Rondec]

It is very light dependent. As long as you have adequate light, you won't see a deterioration in your image at 1/8000 second. The only times I really see 1/8000 second, though, are shooting wide apertures in bright light or, shooting multiple exposures, sometimes the dark ones will hit 1/8000 second.

[DBGrip]

Have you noticed differences in contrast or color vibrance say 1/500th vs 1/4000th?

Shoot outside, pretty wide open, on primes is the situation I''m thinking of. Live event so I'm not worried about getting a fast enough shutter to freeze action. I did shoot this event last year with my K20, and am just trying to make some self improvements this year. Two bodies is going to be a good start.

Thanks for your input, I'm going to go look through my shots from last year, and see if I notice anything. Was wondering if anyone had tested it, maybe I'll do it this weekend

Again, I'm just assuming there has to be a difference with the duration that the sensor "views" the scene.

[UncleVanya]

I would expect the sensor to care very little about time. Light (exposure) is more relevant. The fastest speed discussed here is achingly slow in computer terms. Sensors are computer chips and that is important mainly when talking about dropping shutters and going purely electronic.

[AquaDome]

Try AV mode. In changing light conditions, I set the aperture to f/8, ISO to 400, and let the camera pick the shutter speed. Metering back-lit or shadowed subjects can be a bit tricky, but they are even without an auto-exposure mode.

[derekkite]

I shoot to stop action with long lenses and haven't noticed a deterioration up to 1/4000. Rarely is it bright enough for me to go that high, and I tend to stay in the 1/1000 to 1/2500 range, and stop down if there is more light. I'm currently freezing action with a flash at 1/6000 ish and getting good results.

---

Last edited by derekkite; 08-10-2015 at 01:00 PM.

[stevebrot]

I've read a bit about slow shutter speeds and performance with regards to sharpness, color depth etc;

That is a new one for me.

As for shorter speeds...The sensor is engineered to record the light that strikes it. This is literally the number of photons per exposure. If it takes 1/30s @ f/22 to get x-number of photons, it is the same as if it takes 1/4000s to get that same number at f/2.0.

Your profile indicates that you own a K-3. It should be easy enough for you to test this out for yourself simply by shooting the same scene (Tv mode and fixed ISO) at both long and short exposure times.


Steve

[Bagga_Txips]

I have heard of slow shutter speeds causing problems, but the speeds in question are in dozens of seconds, or even in minutes, rather than fractions of seconds, and I thought it applied more to film than to digital. (Reciprocity failure?)

I have never heard of a shutter mechanism having a sweet spot. Unless maybe in an ancient camera suffering from metal fatigue!

[stevebrot]

I thought it applied more to film than to digital. (Reciprocity failure?)

Those are my thoughts as well. The reaction of film to light is fundamentally different than that of the digital sensor and there may be reciprocity failure with longer exposures. The film insert usually provides a guide for that case. As far as digital sensors are concerned, there may be a problem with overheating with very long exposures (many minutes), but to the best of my knowledge and experience there should be no issues within the settable range of the Pentax shutter.


Steve

[UncleVanya]

Good read if anyone's interested:
https://en.wikipedia.org/wiki/Focal-plane_shutter

[derekkite]

There are ranges where vibration occurs, and maybe a sour spot where the harmonics of the lens/body combination causes a problem. There are noise issues at long exposures. But those are negatives as opposed to sweet spots.

There are definite sweet spots where a given aperture and shutter speed on a specific lens will give you better results. I rarely have the opportunity to shoot above 1/4000 , and the occasions where it was possible I got better results stopping down the lens and dropping the Iso.

In situations where light calls for a higher Iso, say 800-1600 I tend to bias to a longer shutter speed as opposed to stopping down for nicer noise characteristics.

I don't think with any that a stop is a stop is a stop. It is a good place to start but the characteristics of the hardware give each setting certain qualities. I seem to shoot at the edge of its capabilities so the tradeoffs are real and distinct.

[K-Three]

Two thoughts:
1. Reciprocity failure, as pointed out by others, film sensitivity was (is) not linear, each emulsion had it's own characteristics. I rarely ran into trouble with short exposures on film (because I rarely used them!), it was usually the long end I had to tinker with. I would imagine the digital sensor has a similar issue at some extreme ends of it's useful range. No idea where that curve starts to bend at the ends though.

2. Actual shutter speed: I remember being very surprised that my MX was not consistently able to shoot at 1/1000 of a second. The springs just did not seem to tension the same each time. The mid-range on it was pretty good and consistent. I'm sure our modern electronic shutters are much more consistent, but I wonder how accurate they are?
Would be interesting to test shutters for actual exposure at a given setting. I bet, like the lens focusing "acceptable range" there is a variation of actual shutter speed that means a dialed in 1/4000 is not an exposed 1/4000.

[SpecialK]

I want to know is there a shutter speed where the sensor in the K-3 starts to lose performance just from not getting enough time viewing the scene?

In a word, no.

[stevebrot]

I'm sure our modern electronic shutters are much more consistent, but I wonder how accurate they are?

Very accurate...the curtains are quartz-timed.


Steve