K-5 Shutter Lag: 0.107-0.125 sec

Eruditass · 03-28-2011, 01:42 PM

Originally posted by 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.

Although it's not being tested, my friend'ss 60D has a noticeable AF lag from when the AF button is pushed to when it starts finding focus, compared to my K-x.

**Ash** · 03-28-2011, 01:53 PM

Good stuff Zig. I do understand the difference.
We should definitely be concentrating on the shutter lag aspect only as AF lag/time varies with camera body, lens, lighting, starting point for focal distance and the subject itself.
So now with shutter lag, it does seem that your measurement of the K-5 shutter lag significantly exceeds that of all the other dSLRs tested on the Impulse Adventure link you provided.
Let us know how the DS and D90 fare in that regard.

Catalana · 03-28-2011, 10:31 PM

I am still somewhat confused. Shouldn't there be:

1. Shutter button press
2. Aperture set
3. Mirror up
4. Shutter curtain release

Is there a way to identify when the shutter curtain triggers?

Siegfried · 03-29-2011, 11:38 AM

Gents,
I'm sorry to report that D90 test is delayed, a friend of mine reports on no time to come to see me and give me his cam for testing.

Rat,
That was AWB but I wasn't able to find if other settings put things better or worse.

Ash,
you can refer to this post for ist DS test.

Catalana,
Ditto. But to point out for sure when shutter goes flipping I need to first record its (sound) pattern having all other parts static. So far I can not figure out how to have this done. I assume that shutter goes open right when the mirror hits the upper stop.

All,
well, as both those links above tell us that K-5 isn't any fast in terms of shutter lag comparing to mates from other vendors (at least the ones starting from C and N; I don't feel like Oly or Sigma will do it better... I mean faster). But I take it to be a feature, not a flaw - this does not prevent from taking great photos. It's like... having a stereo amplifier with only 0.01% THD+N when competitors claim to have 0.0001% - not a big deal at all. Hi-end tubes give 0.05% or 0.01% at their best having their price tags x10 more then solid-state amps with lots of leading zeros in their THD (or THD+N) specs. Sure, it's nice to have a very low shutter lag, but if it's below a certain value (say, 150 ms) then I don't really care whether it's 100 or 50 or 5ms.
For me it's just the same as when talking about lens sharpness: if it is sharp enough then it doesn't matter whether its 2800 or 3500 lph or whatever.

For me it only shows that K-5 remains being an advanced cam, but still aimed to amateurs. Yes, it's weather-sealed. Sure, it's shutter is specified for a long run. Damn, it has magnesium chassis. Indeed, it has two gyro's - but it's not designed to be pro-level, although one can surely make money on photos taken with this baby. Is it a flaw, is it smth bad? I don think so, it's just the way it is. It is a good advanced DSLR.

Btw, I bet my pinky that 645D flagship is not any faster then K-5. But none of MF was ever designed to be a super-fasting field camera, especially 645D with it's shutter spec'd for 50K actuations only.

Bottom line:
I got myself curious about this parameter and was able to retrieve some results - and I put them online since I thought that others may be curious too.

Zig

P.S.
I'll get back with D90 test right after I'm lucky to catch that friend of mine. Sorry it takes this long.

Last edited by Siegfried; 03-29-2011 at 11:47 AM.

**Ash** · 03-29-2011, 06:00 PM

Indeed Zig, the K-5 was always meant to be an advanced enthusiast's camera. Whether it is considered a pro model or not is a moot point.

Unfortunately there will always be those who in trying to justify and/or feel good about their 'big' purchase on a flagship dSLR assert how the camera should be perfect in everyway and work wonders for them. After all, they spent all that money on it. Interestingly it's quite relative as much bigger ticket items that fail, such as cars, aren't as strictly scrutinised...

We're so fickle...

Catalana · 03-29-2011, 08:40 PM

Originally posted by Siegfried

Catalana, Ditto. But to point out for sure when shutter goes flipping I need to first record its (sound) pattern having all other parts static. So far I can not figure out how to have this done. I assume that shutter goes open right when the mirror hits the upper stop.

I found some u-toobe links that might help give some insight. Obviously the mechanics differ per camera model, but from what I see, there should be some lag after the mirror hits the upper box, as there is some bounce in many cases, if not all. If recorded at a high enough fps rate, it is quite obvious (but not in the K-x video).

Maybe varying the shutter speed from slow to fast while recording the sound might let you identify an optimal spot.

You Tube

You Tube

You Tube

You Tube

Smeggypants · 03-29-2011, 09:08 PM

Originally posted by Siegfried

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

No it doesn't.

creampuff · 03-29-2011, 10:03 PM

Looking at the video, the mirror bounce of the K10D is horrid...

Siegfried · 03-30-2011, 08:29 AM

Catalana,
thanks for those youtube links, but unfortunately K10d video appears to have some sound track that is prohibited to be played here where I'm located - youtube is telling me EMI restricted my access to that video.

Nevertheless, I carefully re-examined K-5 shot track and have to say that I appear to be wrong at the decoding:

Although I wasn't able to record the shutter sound having all other body parts static, but I carefully listened to every step recorded and compared to LV and repeater sounds and have to say that at the 1st post I was wrong at the mirror sound pattern, it's 95% the shutter sound. The aperture going pre-set is too quiet and appears to be happening when mirror goes up.

I dunno if the shutter lag is the pay-off to in-body stabilization or just a safety feature or the way to increase the quietness of the K-5 when taking a photo (K-5 is a VERY quiet camera). Probably one can decrease the lag of amount of 15 ms changing its firmware - to me it looks like camera is doing nothing for 15 ms after mirror's up and aperture's pre-set before toggling the shutter. Though it can be the sensor stabilization system that's being energized (it goes energized whenever SR is on or off, see other threads on this), but to say for sure whether it's just idling or doing smth with sensor positioning I need to change my audio equipment (mostly - the sound card) since the current one is too noisy and I can not distinguish sensor stabilization noise from total noise.

What I can say for sure - K-5 does nothing for the first 27 ms after shutter button goes pressed. That's the real thing - but once again I'd like to refer my dear readers to the post above with THD+N analogy: having my K-5 shutter lag reduced by 27 ms (or even 42 ms: 27 here and 15 there) will not make me any better as a photographer.

Zig

sewebster · 03-30-2011, 09:42 AM

Whoa, the shutter goes from top to bottom. Didn't realize that.

Catalana · 03-30-2011, 09:23 PM

@ Siegfried:

Sensor stabilization occurs after "shutter button half press". This can also be seen from the exif data, which tells how long it takes for the sensor to stabilize from "shutter button half press" and indicate a stabilized shot (if SR is ON). It should occur likewise for SR OFF, but then only to pull the floating sensor into position without shake compensation. So I would not worry about that parameter too much, or you could try to record sound only while doing a "shutter button half press". If you repeat that test, you may want to wait a while between them. The sound would be very low/quiet though.

Gimbal · 03-30-2011, 11:26 PM

Not entirely right, during half press the cameras gyro sensors start collecting data about the camera shake. But the "floating" image sensor does not move into position until you fully press the shutter button.

Siegfried · 04-03-2011, 10:50 PM

Gents,
Thanks for waiting. Here come the results from D90 body compared to K-5:

K-5 on the top. D90 lag is measured at about 70-75 ms and it doesn't depend on AF/MF or any other settings.

It's clearly seen that D90 goes right away when it gets 'Action!' command from user while K-5 goes idling and waits for smth. But mechanically K-5 isn't any slower then D90, with all those gaps - between button press and mirror lift/aperture set and between mirror's up and shutter release - removed it can jump a notch ahead of D90.

Zig

Catalana · 04-04-2011, 07:29 PM

Comparing the vertical db graph, I initially thought they where at different levels, but they are the same. There is something quite "noisy" going on in the D-90, i.e., one would interpret 4-5 additional events before the exposure. Any idea...

Siegfried · 04-04-2011, 10:46 PM

Catalana,
I didn't mention it here, but D90 is much noisier camera comparing to K-5. And I was very surprised to hear the repeater (DOF preview) sound from D90 - it looks like there it's everything inside the cam that goes bouncing if you press that button! I really don't know why nikon engineers made it that way, maybe it's aperture drive hitting something in the body, I dunno.

And I won't say for sure that sound level shown for K-5, ist DS and D90 should be the same - I did recorder them on the same settings, but I could use different s/w amplification. I don't recall how much I did amplify each cam, I was amplifying as much as needed and the only limit was to avoid clipping (I used to record with -1dB level and that's what I was heading to). If my memory serves me right, I did 5dB boost for K-5 and D90 and about 3 or 4dB for ist DS.

Zig

04-03-2011, 10:50 PM	#28
Siegfried Senior Member Join Date: Apr 2010 Location: in a middle of nowhere Photos: Gallery \| Albums Posts: 165 Original Poster	K-5 and D90 compared Gents, Thanks for waiting. Here come the results from D90 body compared to K-5: K-5 on the top. D90 lag is measured at about 70-75 ms and it doesn't depend on AF/MF or any other settings. It's clearly seen that D90 goes right away when it gets 'Action!' command from user while K-5 goes idling and waits for smth. But mechanically K-5 isn't any slower then D90, with all those gaps - between button press and mirror lift/aperture set and between mirror's up and shutter release - removed it can jump a notch ahead of D90. Zig

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
K20D Shutter Lag	nanhi	Pentax DSLR Discussion	3	08-09-2010 04:20 AM
Nature Perched Ruby-Throated Male Hummingbird @ 600mm, 1/125 sec.	Marc Langille	Post Your Photos!	19	07-12-2010 07:01 AM
Shutter lag and firmware update? (K10D & 1.31)	frattofoto	Pentax DSLR Discussion	3	05-02-2010 01:46 PM
K20D slow shutter lag	RollsUp	Pentax DSLR Discussion	9	10-26-2009 04:54 AM
HELP: Camera Shutter Lag	DJey	Pentax DSLR Discussion	9	08-22-2008 06:42 PM

03-28-2011, 01:42 PM	#16
Eruditass Veteran Member Join Date: Jul 2009 Photos: Gallery Posts: 2,207	Originally posted by 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. Although it's not being tested, my friend'ss 60D has a noticeable AF lag from when the AF button is pushed to when it starts finding focus, compared to my K-x.

03-28-2011, 01:53 PM	#17
Ash Loyal Site Supporter Join Date: Mar 2007 Location: Toowoomba, Queensland Photos: Gallery \| Albums Posts: 23,920	Good stuff Zig. I do understand the difference. We should definitely be concentrating on the shutter lag aspect only as AF lag/time varies with camera body, lens, lighting, starting point for focal distance and the subject itself. So now with shutter lag, it does seem that your measurement of the K-5 shutter lag significantly exceeds that of all the other dSLRs tested on the Impulse Adventure link you provided. Let us know how the DS and D90 fare in that regard.

03-28-2011, 10:31 PM	#18
Catalana Senior Member Join Date: Feb 2010 Location: California Posts: 190	I am still somewhat confused. Shouldn't there be: 1. Shutter button press 2. Aperture set 3. Mirror up 4. Shutter curtain release Is there a way to identify when the shutter curtain triggers?

03-29-2011, 11:38 AM	#19
Siegfried Senior Member Join Date: Apr 2010 Location: in a middle of nowhere Photos: Gallery \| Albums Posts: 165 Original Poster	Gents, I'm sorry to report that D90 test is delayed, a friend of mine reports on no time to come to see me and give me his cam for testing. Rat, That was AWB but I wasn't able to find if other settings put things better or worse. Ash, you can refer to this post for ist DS test. Catalana, Ditto. But to point out for sure when shutter goes flipping I need to first record its (sound) pattern having all other parts static. So far I can not figure out how to have this done. I assume that shutter goes open right when the mirror hits the upper stop. All, well, as both those links above tell us that K-5 isn't any fast in terms of shutter lag comparing to mates from other vendors (at least the ones starting from C and N; I don't feel like Oly or Sigma will do it better... I mean faster). But I take it to be a feature, not a flaw - this does not prevent from taking great photos. It's like... having a stereo amplifier with only 0.01% THD+N when competitors claim to have 0.0001% - not a big deal at all. Hi-end tubes give 0.05% or 0.01% at their best having their price tags x10 more then solid-state amps with lots of leading zeros in their THD (or THD+N) specs. Sure, it's nice to have a very low shutter lag, but if it's below a certain value (say, 150 ms) then I don't really care whether it's 100 or 50 or 5ms. For me it's just the same as when talking about lens sharpness: if it is sharp enough then it doesn't matter whether its 2800 or 3500 lph or whatever. For me it only shows that K-5 remains being an advanced cam, but still aimed to amateurs. Yes, it's weather-sealed. Sure, it's shutter is specified for a long run. Damn, it has magnesium chassis. Indeed, it has two gyro's - but it's not designed to be pro-level, although one can surely make money on photos taken with this baby. Is it a flaw, is it smth bad? I don think so, it's just the way it is. It is a good advanced DSLR. Btw, I bet my pinky that 645D flagship is not any faster then K-5. But none of MF was ever designed to be a super-fasting field camera, especially 645D with it's shutter spec'd for 50K actuations only. Bottom line: I got myself curious about this parameter and was able to retrieve some results - and I put them online since I thought that others may be curious too. Zig P.S. I'll get back with D90 test right after I'm lucky to catch that friend of mine. Sorry it takes this long. Last edited by Siegfried; 03-29-2011 at 11:47 AM.

03-29-2011, 06:00 PM	#20
Ash Loyal Site Supporter Join Date: Mar 2007 Location: Toowoomba, Queensland Photos: Gallery \| Albums Posts: 23,920	Indeed Zig, the K-5 was always meant to be an advanced enthusiast's camera. Whether it is considered a pro model or not is a moot point. Unfortunately there will always be those who in trying to justify and/or feel good about their 'big' purchase on a flagship dSLR assert how the camera should be perfect in everyway and work wonders for them. After all, they spent all that money on it. Interestingly it's quite relative as much bigger ticket items that fail, such as cars, aren't as strictly scrutinised... We're so fickle...

03-29-2011, 09:08 PM	#22
Smeggypants Veteran Member Join Date: Sep 2010 Posts: 1,536	Originally posted by Siegfried 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 No it doesn't.

03-29-2011, 10:03 PM	#23
creampuff Veteran Member Join Date: May 2007 Location: Singapore Photos: Gallery Posts: 3,953	Looking at the video, the mirror bounce of the K10D is horrid...

03-30-2011, 08:29 AM	#24
Siegfried Senior Member Join Date: Apr 2010 Location: in a middle of nowhere Photos: Gallery \| Albums Posts: 165 Original Poster	Catalana, thanks for those youtube links, but unfortunately K10d video appears to have some sound track that is prohibited to be played here where I'm located - youtube is telling me EMI restricted my access to that video. Nevertheless, I carefully re-examined K-5 shot track and have to say that I appear to be wrong at the decoding: Although I wasn't able to record the shutter sound having all other body parts static, but I carefully listened to every step recorded and compared to LV and repeater sounds and have to say that at the 1st post I was wrong at the mirror sound pattern, it's 95% the shutter sound. The aperture going pre-set is too quiet and appears to be happening when mirror goes up. I dunno if the shutter lag is the pay-off to in-body stabilization or just a safety feature or the way to increase the quietness of the K-5 when taking a photo (K-5 is a VERY quiet camera). Probably one can decrease the lag of amount of 15 ms changing its firmware - to me it looks like camera is doing nothing for 15 ms after mirror's up and aperture's pre-set before toggling the shutter. Though it can be the sensor stabilization system that's being energized (it goes energized whenever SR is on or off, see other threads on this), but to say for sure whether it's just idling or doing smth with sensor positioning I need to change my audio equipment (mostly - the sound card) since the current one is too noisy and I can not distinguish sensor stabilization noise from total noise. What I can say for sure - K-5 does nothing for the first 27 ms after shutter button goes pressed. That's the real thing - but once again I'd like to refer my dear readers to the post above with THD+N analogy: having my K-5 shutter lag reduced by 27 ms (or even 42 ms: 27 here and 15 there) will not make me any better as a photographer. Zig

03-30-2011, 09:42 AM	#25
sewebster Veteran Member Join Date: Jun 2008 Location: Vancouver, BC Photos: Gallery Posts: 544	Whoa, the shutter goes from top to bottom. Didn't realize that.

03-30-2011, 09:23 PM	#26
Catalana Senior Member Join Date: Feb 2010 Location: California Posts: 190	@ Siegfried: Sensor stabilization occurs after "shutter button half press". This can also be seen from the exif data, which tells how long it takes for the sensor to stabilize from "shutter button half press" and indicate a stabilized shot (if SR is ON). It should occur likewise for SR OFF, but then only to pull the floating sensor into position without shake compensation. So I would not worry about that parameter too much, or you could try to record sound only while doing a "shutter button half press". If you repeat that test, you may want to wait a while between them. The sound would be very low/quiet though.

03-30-2011, 11:26 PM	#27
Gimbal Pentaxian Join Date: Apr 2007 Location: Sweden Posts: 2,106	Not entirely right, during half press the cameras gyro sensors start collecting data about the camera shake. But the "floating" image sensor does not move into position until you fully press the shutter button.

04-04-2011, 07:29 PM	#29
Catalana Senior Member Join Date: Feb 2010 Location: California Posts: 190	Comparing the vertical db graph, I initially thought they where at different levels, but they are the same. There is something quite "noisy" going on in the D-90, i.e., one would interpret 4-5 additional events before the exposure. Any idea...

04-04-2011, 10:46 PM	#30
Siegfried Senior Member Join Date: Apr 2010 Location: in a middle of nowhere Photos: Gallery \| Albums Posts: 165 Original Poster	Catalana, I didn't mention it here, but D90 is much noisier camera comparing to K-5. And I was very surprised to hear the repeater (DOF preview) sound from D90 - it looks like there it's everything inside the cam that goes bouncing if you press that button! I really don't know why nikon engineers made it that way, maybe it's aperture drive hitting something in the body, I dunno. And I won't say for sure that sound level shown for K-5, ist DS and D90 should be the same - I did recorder them on the same settings, but I could use different s/w amplification. I don't recall how much I did amplify each cam, I was amplifying as much as needed and the only limit was to avoid clipping (I used to record with -1dB level and that's what I was heading to). If my memory serves me right, I did 5dB boost for K-5 and D90 and about 3 or 4dB for ist DS. Zig