[Siegfried]

Gents,

This morning I got curious about K-5 shutter lag. I did some testing and came with the following results: 0.125-0.13 sec measured.

Equipment used:
- K-5 (f/w #1.03) running on original battery with Tokina AT-X PRO AF280 hooked on and Sandisk Extreme card inside (class 10)
- a microphone from Pioneer DEQ-P6600 processor (a good Hi-Fi mic, not a Fostex or Audiotechnika, but nevertheless a good one; if you don't trust its label I can get back with its acoustic response performance graph measured according to certified Phonic PAA2)
- Samsung (sorry ) R530 laptop with built-in some Realtek codec
- Sennheiser HD202 earphones
- Adobe Audition 2.0

How do things go when we take a shot? We press the shutter button, then the mirror goes up and aperture closes to pre-set value - so I started with gathering sound patterns for each of these steps. Moreover, I took two kinds of measurements of each sound: #1 having mic stuck to the K-5 body and #2 having mic in the air at ~2-3 cm (~1") from shutter release button or the mirror box.

1. Here what does it look like when we press the release button:

body sound


in the air

2. Here goes the repeater work (darn informative! )

body sound



in the air

3. And here we can see what it looks like when mirror goes up (toggling LV):

Body sound


in the air

4. Having all that stuff picked up we're ready to disassemble shot sound:

body


in the air

I don't know how come that shutter lag varies from 107 ms to 125 ms, but I carried each test (shutter button, repeater button, LV and shot) twice (or four times if we count body-air recordings) so results must be reliable: I got a couple of 107 ms lags and a pair of 123-125 ms ones.

Lemme know if you'd like to have a look at the wav files and if so then advise if I can upload them here on PF in some way - so far it appears to be restricted to images only.


UPDATE:
I forgot to say that it's measured on half-to-full shutter button release. And if one decides to carry this measurements then it's obviously best to turn to MF just in case.

Zig

P.S.
I can go on with K-7 or any other cam measurements if one gives me the output wav's, I think you've got the idea on how it should be done.

---

Last edited by Siegfried; 03-28-2011 at 06:28 AM.

[shiner]

Nice work. One wonders how other bodies and makes would fair on your same setup.

[Siegfried]

One wonders how other bodies and makes would fair on your same setup.

I'm going to try D90 nikon somewhere this week. Or you can just upload wav files recorded with the technique described and I'll parse them and post the results.

Zig

[Gimbal]

I believe Pentax is fairly slow compared to Canikon, but I guess that is due to the SR system pulling up and stabilizing the sensor from the bottom of the well.

[Siegfried]

Gimbal,
I don't see any traces of SR in the recorded shots. I mean to say that you can check the LV record to see what SR looks (sounds) like - I don't see such pattern between shutter release button click and mirror hitting the upper stop.

Nevertheless I'm going to give it another try with SR disabled. I'll get back on this later.

Zig

[Ash]

125ms seems like a long delay for shutter lag (which I'm interpreting from the time the shutter release button is fully depressed to when the camera engages the mirror). SR on or not, I don't get any appreciable lag time. I do though at times get a slight delay in AF engagement (from when I half-depress the shutter button to when AF motor engages).

[Gimbal]

Nevertheless I'm going to give it another try with SR disabled. I'll get back on this later.

Zig

There shouldn't be any difference, since even with the SR off the camera has to pick up the sensor from it's resting position and hold it absolutely still in the middle of the frame during the exposure.

[Siegfried]

Ok,
here's the update. K-5 in MF: 0.098 sec (four resettable results). It doesn't depend on whether SR is on or off. With focus switched to AF mode it appears to have no relevance to SR function either, although the result are not so consequent varying in 18 ms (105-123 ms, 8 shots).

Stay tuned and we'll get back to you with ist DS shutter lag performance shortly.

Zig

[Siegfried]

...which I'm interpreting from the time the shutter release button is fully depressed to when the camera engages the mirror...

Ash,
I measure the lag between shutter button in the half-pressed position and mirror hitting the upper stop - so I'm (or should be) eliminating focus operation. In MF mode K-5 doesn't wait for full-press and mirror goes up when the button hits half-pressed position.

Zig

[Gimbal]

In MF mode K-5 doesn't wait for full-press and mirror goes up when the button hits half-pressed position.

Zig

What? That can't be right.

[Ash]

Indeed, you may be mixing up your terminology a little Zig.
In any case an interesting study I'd like to see compared head to head with your DS.
I'm not noticing a >100ms delay myself (which is a fair amount of time) but it may just be my limited perception in the concept.

[Siegfried]

What? That can't be right.

Gimbal,
whenever it's right or wrong - it's just the way it is. I recall I saw smth on this in the manual and probably this function can be programmed, but it doesn't play much here since I'm able to record that 'click' sound from the button.

So here go the measurements from ist DS (boy, I LOVE IT'S SHUTTER SOUND! It's that CLUTZ, not FLOP from K10-20-7 or D90 or anything else from the modern market).

ist DS with the same Tokina lens, running on lithium batt and having SanDisk Extreme 4Gb (class 10) card inside. I measured 145-150 ms, refer to the shots below:

shutter button


repeater


mirror up (I used 'sensor clean-up' function from the custom menu page)


the shot

Camera was set to Manual mode with exposure time set to 1/10 sec. MF/AF mode doesn't change anything here, lag stays within 145-150 ms range at any setting (over 8 shots).

Zig

P.S.
As a preliminary announce: I'm expecting D90 to be checked tomorrow.

[Siegfried]

Ash,
I'm posting all those screenshots so everybody can perform the checks on his own methodology and terminology. But I take the shutter lag to be the time between 'motor!' command is issued and exposure starts being taken: so if a camera is set to AF mode then this is definitely the time between 'half-to-full' shutter button press and sensor is fully exposured.

Zig

[Siegfried]

Ash (and all),
I think it's worth setting up the terminology.
Link#1 - that's how I take it (moreover, I'd say that AF lag should be better called 'AF time')
Link#2 - that's probably how you take it.

In AF mode I'm set to AF.S and going release after focus is fully achieved in 'half-pressed' shutter button command. In terms of the 2nd link it's 'AF pre-focused'.

Zig

P.S.
Actually, it's the way I'm taking 99.99% of my photos: 1st go with half-press to ensuse AF and me - we're both on the same page and only then go with full-press. The only exception may be made only if I see some beauty passing by - whenever on two legs or four wheels - when I don't have time to go with all those focus checks and I need that photo taken in any way.

[Ratmagiclady]

Question, is AWB on for these things? It's not a huge difference at least, but I noticed that on an old bridge camera that really slowed things down. (I've been wondering if it affects things on like a K20d)