Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
06-12-2015, 01:39 AM   #31
Veteran Member
awaldram's Avatar

Join Date: May 2008
Location: Hampshire
Photos: Gallery
Posts: 732
QuoteOriginally posted by ScooterMaxi Jim Quote
I use the back button. It is entirely possible that some portion of the lag previously observed came from using the shutter button for AF as well as shutter release. However, the only fair test then would have been to shoot in MF. Otherwise you're adding an unnecessary randomness to the process. The test has to be done uniformly with prefocus.

No one should be considering pTTL for any kind of action shooting. Clearly pre-flash is a huge detriment to accurate stop action.
We know the delay to pre-flash is likely caused by an event wait that event wait could easily be AF check.

Therefore 'testing' the camera in a mode you won;t actually me using (MF) in real life for a delay that is seen when 'half press or backAF' -> full press -> delay -> preflash means nothing .

You might as well say I've tested it in MF it works fine so p-ttl is fixed

You point on none scientific testing is valid but I;d also counter your 'psuedo' scientific test is equally invalid as your reaction time and unique to your camera button hysterysis are both considerably greater than the delay and not measured during your videoing

So you have no idea form the video the start time of shutter release so explain to me again how you measure when you don;t know the start ??

I use p-ttl flash a lot on various Pentax bodies so have a good feel for the K3 delay , I also shoot two eyes open so know the time between shutter -> preflash.

The delay on the k3 is still long enough for moving targets (kids,dogs etc) to move out the frame between shutter press and actuation , On the k10d,k20d,k5,kx,q,q7 it isn't .

ON previous cameras you may get eyes shutting from pre to main but with the k3 you can get the subject missing ....... 1.21 does not alter this.

Here you go scientific - you may want to watch in slow motion

First flash is IR trigger , IR fire triggering both cameras at the same time ..... Kx fire preflash -> Kx fire main then after further delay K3 fire pre then main

So the Kx fires pre main and shutter before the k3 fire pre-flash ..... That is the delay and it still exists the same.!!

Let me expand on this for you

If any other pentax IR , P-ttl equipped camera is triggered at the same time as a K3 it will have taken the Image before the K3 has started the pre-flash
This is true for all versions of firmware and also true for the k3ii (k3ii untested)

That is a fact and the video proves it ....... !!


Last edited by awaldram; 06-12-2015 at 02:13 AM.
06-12-2015, 08:08 AM   #32
Veteran Member
ScooterMaxi Jim's Avatar

Join Date: Jul 2009
Location: Minnesota
Posts: 1,520
Original Poster
I have to admit, the video is not at all clear as to where the flashes are coming from - they are off video camera view. They seem relatively close in timing, but I'd have to be able to see frame-by-frame - I use Quicktime for that. I also have no idea at all of whether the K-3 is receiving the IR signal at the same instant as the K-X. If we know that they are, that's a good way to compare. Oddly enough, your video does prove that the K-3 is responding quicker than the original firmware. That's entirely clear because both bodies were responding quite quickly - by pTTL standards.

As for my video test, I admit that it is difficult to get the exact point of contact - but I was very consistent in my quick presses - and could tell within 1-2 frames. The fact that I was comparing from Manual to pTTL was the key for me, and the differences measured between those were pretty easy to track quite accurately.

I don't understand your point regarding AF. It is a variable to avoid in the test. If you are shooting action with pTTL flash that's your business. Its just that no one would pay me for missing so many shots due to the added delay between preflash and main flash - which is a huge factor no matter which body you're using. None of the pTTL cameras can avoid a lag of more than .2 sec, including the K-X. Manual flash is roughly half that lag - on all bodies, as measured by IR.

In any event, I also feel it is important to point out that the consistency of flash output - at least on 3rd party flashes - has improved considerably. On the K-3, I would say the situation for pTTL lag and consistency has gone from totally unusable (for paid work) to a solid middling rating.

Last edited by ScooterMaxi Jim; 06-12-2015 at 08:19 AM.
06-12-2015, 08:35 AM   #33
Veteran Member
enoeske's Avatar

Join Date: Jul 2009
Location: Surprise, Az
Photos: Gallery
Posts: 2,136
QuoteOriginally posted by ScooterMaxi Jim Quote
I have to admit, the video is not at all clear as to where the flashes are coming from - they are off video camera view. They seem relatively close in timing, but I'd have to be able to see frame-by-frame - I use Quicktime for that. I also have no idea at all of whether the K-3 is receiving the IR signal at the same instant as the K-X. If we know that they are, that's a good way to compare. Oddly enough, your video does prove that the K-3 is responding quicker than the original firmware. That's entirely clear because both bodies were responding quite quickly - by pTTL standards.
The flashes are coming from the on board flashes that are clearly in the video. Also, how could they not receive the IR signal at the same time, there is only one IR source. The video doesn't not compare FW versions on the K-3 so it does not show that its responding faster than the original FW.
06-12-2015, 09:28 AM   #34
Veteran Member
ScooterMaxi Jim's Avatar

Join Date: Jul 2009
Location: Minnesota
Posts: 1,520
Original Poster
Again, the video is a single event, no one is providing times, and not repeated. We do not know that the cameras are receiving the signal at the same time. An IR push button is not a single instant, but a prolonged signal send, AFAIK. When using IR I get delays all the time, varying, on any of the bodies - from the *istD, K-30, K-01, K20D, Q. They all use the same system, but they do not respond absolutely consistently to the signal. That's my experience. By the way, was the IR sending aimed between the cameras precisely, or directed more toward the K-X perhaps? IR senders are quite directional for good reason...

Still, no one seems to be doing actual timing. Other than IR, but that was with the original FW. Even if the K-X is the ultimate, the pTTL on that model is measured at .225 - well more than 1/5th of a second; useless for many situations. Still, better than the 1/3rd of a second that the K-3 started out at, based on the timings reported back then.

I'm not saying that the K-3 is terrific; it has lag like the rest of the bunch - and a bit more than some. Manual flash is twice as fast - roughly.

06-13-2015, 02:41 AM   #35
Veteran Member
awaldram's Avatar

Join Date: May 2008
Location: Hampshire
Photos: Gallery
Posts: 732
QuoteOriginally posted by ScooterMaxi Jim Quote
Again, the video is a single event, no one is providing times, and not repeated. We do not know that the cameras are receiving the signal at the same time. An IR push button is not a single instant, but a prolonged signal send, AFAIK. When using IR I get delays all the time, varying, on any of the bodies - from the *istD, K-30, K-01, K20D, Q. They all use the same system, but they do not respond absolutely consistently to the signal. That's my experience. By the way, was the IR sending aimed between the cameras precisely, or directed more toward the K-X perhaps? IR senders are quite directional for good reason...

Still, no one seems to be doing actual timing. Other than IR, but that was with the original FW. Even if the K-X is the ultimate, the pTTL on that model is measured at .225 - well more than 1/5th of a second; useless for many situations. Still, better than the 1/3rd of a second that the K-3 started out at, based on the timings reported back then.

I'm not saying that the K-3 is terrific; it has lag like the rest of the bunch - and a bit more than some. Manual flash is twice as fast - roughly.
Actually the IR is a single IR event launched from an Ardunio controller , its a single instruction to fire the shutter it's not repeated this is the simplified code as sent (my real code also allows Focus IR events).
It is considerably shorter in period than any human could press a shutter in !!! so I would disagree in your "prolonged"

Also what the Kx does is irrelevant I could replace it with any P-ttl K series or newer body and it would fire before the k3 pre flashed

void wait(unsigned int time){
unsigned long start = micros();
while(micros()-start<=time){
}
}

void high(unsigned int time, int freq, int pinLED){
int pause = (1000/freq/2)-4;
unsigned long start = micros();
while(micros()-start<=time){
digitalWrite(pinLED,HIGH);
delayMicroseconds(pause);
digitalWrite(pinLED,LOW);
delayMicroseconds(pause);
}
}

Pentax::Pentax(int pin)
{
pinMode(pin, OUTPUT);
_pin = pin;
_freq = 38;
}

void Pentax::shutterNow()
{
high(13000,_freq,_pin);
wait(3000);
for (int i=0;i<7;i++){
high(1000,_freq,_pin);
wait(1000);
};
}

The k3 has never been inconsistent and I haven't seem any reports as such maybe you could point them out to me ?

So I fail to see just what you think 1.21 has 'fixed' regards flash
1 never inconsistent so that con't be fixed
2 delay is not altered and is still present

As I send 1 fire shutter instruction it matters not where the IR was pointing or any directional etc the cameras either recieve it and re-act (as they do in the video or they do nothing.
If they react they both received the same single instruction at the same time and hence fire synchronously.
There could be some argument that the cameras may have differing time form receiving the instruction and reacting but given the hacked coding shows much reapplication between bodies this would favour the faster CPU (hence the k3)

I suspect your issue with IR is more related to you application (controller ambient conditions etc) than the technology or implementation itself as all camera IR is very very basic leaving little area for mistakes.

---------- Post added 13-06-15 at 11:04 AM ----------

QuoteOriginally posted by ScooterMaxi Jim Quote

In any event, I also feel it is important to point out that the consistency of flash output - at least on 3rd party flashes - has improved considerably. On the K-3, I would say the situation for pTTL lag and consistency has gone from totally unusable (for paid work) to a solid middling rating.
I use cactus V6's normally for action sports and aokatec P-ttl triggers for studio presentation work the K3 has been fine like this and 1.21 makes no odds neither improving or worsening

with 2 Metz 48's 1 Metz 50 2 fgz360 1 fgz 540 and 1 45CL digital all in P-ttl I've ecperienced no inconsistencies so which 3rd parties are having consistency issue on the k3 ?

overall the k3 in P-ttl has really impressed me it is the best implementation of an Penatx D-slr but I see no changes in 1.21 and don't think your seeing anything but a standard placebo response.

Last edited by awaldram; 06-13-2015 at 02:50 AM.
06-13-2015, 08:17 AM   #36
Veteran Member
ScooterMaxi Jim's Avatar

Join Date: Jul 2009
Location: Minnesota
Posts: 1,520
Original Poster
We'll just agree to disagree.

I just did a test of my IR trigger on an IR camera in live view (IR signals show up really well in IR). When you press down on the button it stutters a signal rapidly on/off until you release the button. IR triggers are highly directional because the intent is to trigger only one device in a tight range. The code isn't clear to me, or how the controller sends the signal. You have to use an IR signal to trigger the controller, then be sure that the controller is emitting only one pulse rather than a series? A series of pulses, possibly over what duration? Still, you have no way of knowing if the two cameras are receiving the signal in the same instant.

I have timing video and compared with the much-longer original timing information from IR; you have the evidence you presented and a heavy reliance on your perceptions. Quite possibly my original set-up with shutter button-AF as opposed to current use of back-button focusing played a part in perceived improvements. I'm trying to figure out what's going on, not really concerned with pTTL - which is too slow for my needs in the majority of shooting situations I encounter in any event. You seem more concerned with winning an argument than getting to the bottom of things. You win - feel good. This is enough discussion.

Last edited by ScooterMaxi Jim; 06-13-2015 at 08:48 AM.
06-13-2015, 12:53 PM   #37
Veteran Member
awaldram's Avatar

Join Date: May 2008
Location: Hampshire
Photos: Gallery
Posts: 732
QuoteOriginally posted by ScooterMaxi Jim Quote
We'll just agree to disagree.

I just did a test of my IR trigger on an IR camera in live view (IR signals show up really well in IR). When you press down on the button it stutters a signal rapidly on/off until you release the button. IR triggers are highly directional because the intent is to trigger only one device in a tight range. The code isn't clear to me, or how the controller sends the signal. You have to use an IR signal to trigger the controller, then be sure that the controller is emitting only one pulse rather than a series? A series of pulses, possibly over what duration? Still, you have no way of knowing if the two cameras are receiving the signal in the same instant.

I have timing video and compared with the much-longer original timing information from IR; you have the evidence you presented and a heavy reliance on your perceptions. Quite possibly my original set-up with shutter button-AF as opposed to current use of back-button focusing played a part in perceived improvements. I'm trying to figure out what's going on, not really concerned with pTTL - which is too slow for my needs in the majority of shooting situations I encounter in any event. You seem more concerned with winning an argument than getting to the bottom of things. You win - feel good. This is enough discussion.
It has nothing at all todo with a commercial remote the code for an ardunio it works like this

Once triggered via flash (the first flash you see in the video .

Ardunio (computer) sends 1 and only 1 fire shutter IR sequence.

Unlike a commercial IR remote which sends fire, fire , fire , fire till you remove your finger

my code sends (my choice)

Focus - Fire
or
Fire

I don't think you'll find many commercial IR that can send focus for pentax cameras.!

This is the actual libary
#include "Arduino.h"
#include "PentaxIRControl.h"

void wait(unsigned int time){
unsigned long start = micros();
while(micros()-start<=time){
}
}

void high(unsigned int time, int freq, int pinLED){
int pause = (1000/freq/2)-4;
unsigned long start = micros();
while(micros()-start<=time){
digitalWrite(pinLED,HIGH);
delayMicroseconds(pause);
digitalWrite(pinLED,LOW);
delayMicroseconds(pause);
}
}

Pentax::Pentax(int pin)
{
pinMode(pin, OUTPUT);
_pin = pin;
_freq = 38;
}

void Pentax::shutterNow()
{
high(13000,_freq,_pin);
wait(3000);
for (int i=0;i<7;i++){
high(1000,_freq,_pin);
wait(1000);
};
}

void Pentax::toggleFocus()
{
high(13000,_freq,_pin);
wait(3000);
for (int i=0;i<5;i++){
high(1000,_freq,_pin);
wait(1000);
};
high(1000,_freq,_pin);
wait(3000);
high(1000,_freq,_pin);
}

as you can see you can call it with shutternow to fire the shutter or toggle focus to focus

As to agreeing to disagree I don't think that’s pertinent I;ve presented the evidence which I believe is incontrovertible, You may argue but the video is what it is and as presented

You have so far offered no evidence, I'm also curious how you analysed frame by frame as afain mp4 if not a complete frame protocol so any frame is made up of previous frames (GOP delta compression)

Which is why I didn't attempt to define timings from the video above as it would be false as the addition of I,Pand B frames to create the 'frame' as you see in quicklime is to coarse for high accuracy timing.!

06-13-2015, 05:14 PM   #38
Veteran Member
ScooterMaxi Jim's Avatar

Join Date: Jul 2009
Location: Minnesota
Posts: 1,520
Original Poster
Wow. I understand the condescension, but you substitute your lacking command of language for some meaningless computer code. Impressive. You don't even have a handle on the pulse interval of your IR. Generally, you need some duration for successful trigger; instant is not a choice. You've obviously picked the only video proof that argues for your side of it, but I suspect you had to repeat the shooting of the video several times to get your desired result. You introduce the variable of focus-fire, but have a cap on each of the lenses! Oh, and the lenses are vastly different thereby introducing yet another uncontrolled factor.

Video frame timing indeed is used to analyse durations in all sorts of situations, even with variable rate compression of H.264; mp4 is a container and has nothing to do with it. You show a lack of even the most basic understanding of video capture. This is how editing is done, and Quicktime provides for the intervals as indicated in the EXIF of the file. No wonder the blurry, poorly aimed video you provided showed essentially nothing.

But, hey, you win. Really, this is all the baloney and poor ability to form a proper sentence I can take for now. Say what ever, but never offer real proof.
06-17-2015, 11:43 PM   #39
Forum Member




Join Date: Dec 2007
Location: Moon
Posts: 87
this is nuts, you can't grasp what awaldram is saying (and it was VERY CLEAR) so you get your undies in the knot try to make pointless argument about video method used, GROW UP
06-28-2015, 02:28 PM   #40
New Member




Join Date: Apr 2015
Posts: 5
Who cares when the pre-flash goes off?

I mainly 'shoot' fast dancers, and (to me) all that matters is getting a well exposed picture when I want it...
If awaldram experiences a big lag between actuating the shutterbutton (after pre-focussing) and the actual exposure,
that's all I need to know... don't care if it 'pre-flashes' earlier, 'beeps' sooner, 'whatevers' faster in the process.

I would like acurate TTL because I like to zoom in past other couples...

Would the K3 be able to do this (in 2nd curtain sinc) in a more primitive mode (plain, non-preflash, but still TTL) ?

Thanks for your feedback

Patrick
06-28-2015, 05:08 PM - 1 Like   #41
Site Supporter
Site Supporter
Larrymc's Avatar

Join Date: Jun 2014
Location: Mississippi, USA
Posts: 5,245
Awaldram must be having problems with his K-3

QuoteOriginally posted by pavare Quote
Who cares when the pre-flash goes off?

I mainly 'shoot' fast dancers, and (to me) all that matters is getting a well exposed picture when I want it...
If awaldram experiences a big lag between actuating the shutterbutton (after pre-focussing) and the actual exposure,
that's all I need to know... don't care if it 'pre-flashes' earlier, 'beeps' sooner, 'whatevers' faster in the process.

I would like acurate TTL because I like to zoom in past other couples...

Would the K3 be able to do this (in 2nd curtain sinc) in a more primitive mode (plain, non-preflash, but still TTL) ?

Thanks for your feedback

Patrick

Hi Patrick,


I do no have any sort of appearent delay in the flash either hot shoe or pop up on my K-3 PTTL or manual. My K-3 is a late 2014 production run and I think his is a much older one. In my opinion Pentax made some un documented changes in the production of the K-3. That's the only thing I can come up with for the different flash timing. if you're buying one make sure its a later production run model and hopefully you should be fine.


Larry

Last edited by Larrymc; 06-28-2015 at 06:24 PM.
06-28-2015, 08:19 PM   #42
Veteran Member
ScooterMaxi Jim's Avatar

Join Date: Jul 2009
Location: Minnesota
Posts: 1,520
Original Poster
QuoteOriginally posted by pavare Quote
Who cares when the pre-flash goes off?

I mainly 'shoot' fast dancers, and (to me) all that matters is getting a well exposed picture when I want it...
If awaldram experiences a big lag between actuating the shutterbutton (after pre-focussing) and the actual exposure,
that's all I need to know... don't care if it 'pre-flashes' earlier, 'beeps' sooner, 'whatevers' faster in the process.

I would like acurate TTL because I like to zoom in past other couples...

Would the K3 be able to do this (in 2nd curtain sinc) in a more primitive mode (plain, non-preflash, but still TTL) ?

Thanks for your feedback

Patrick
Hello Patrick,

This isn't easily answered, unfortunately. For the most part, the news is not good.

As far as I know, all 2nd curtain flash on modern systems is tied to the use of pre-flash TTL - resulting in at least some delay. (Leaf shutters in lens is not included unless you are willing to pay the price for those high-end medium format systems.) Keep in mind that 2nd curtain flash is made necessary for proper mixing of the ambient light in the scene, so the flash allows for "trailing" in the ambient light. Otherwise, that pleasant blur is seen at the front - which makes no sense in the composition. On some of my bodies (the K-01, K-3, and presumably others as well), the 2nd curtain triggers a 1/90th maximum sync. Not ideal, of course, and a real problem for many dancing scenes where ambient light could very well call for a much higher shutter speed. You would have to live with the significant pTTL shutter lag in that situation. It isn't the end of the world because you can anticipate dancing moves if you shoot that genre often, but it remains darn tricky.

For non-TTL flash shooting, based on the Imaging Resource tests, it appears that the K-3 is the fastest of all Pentax cameras in terms of pre-focused triggering (assuming the K-3ii is equal - but we don't know the answer to that yet), but faster-triggering non-TTL will give you the wrong front-blur look due to the 1st curtain requirement.

My guess is that Nikon especially, and Canon higher-end cameras close the gap regarding trigger delay in preflash delay, but I haven't studied it closely. Your best bet on Pentax cameras is to use higher power non-TTL flash - with Auto thyrister yielding the best results for the situation you've described; using various bounce and multiple flash set-ups to provide a somewhat natural all-flash lighting result. This will allow you to minimize the amount of ambient blur. The Yongnuo multi-flash system and other modern non-TTL flash systems have provided improving results as they have become more sophisticated. Your only other alternative would be to shoot utilizing ambient light only, but this isn't practical unless the dance floor is especially well lit.

Last edited by ScooterMaxi Jim; 06-28-2015 at 08:24 PM.
06-29-2015, 07:11 AM   #43
Veteran Member
enoeske's Avatar

Join Date: Jul 2009
Location: Surprise, Az
Photos: Gallery
Posts: 2,136
QuoteOriginally posted by pavare Quote
Who cares when the pre-flash goes off?

I mainly 'shoot' fast dancers, and (to me) all that matters is getting a well exposed picture when I want it...
If awaldram experiences a big lag between actuating the shutterbutton (after pre-focussing) and the actual exposure,
that's all I need to know... don't care if it 'pre-flashes' earlier, 'beeps' sooner, 'whatevers' faster in the process.

I would like acurate TTL because I like to zoom in past other couples...

Would the K3 be able to do this (in 2nd curtain sinc) in a more primitive mode (plain, non-preflash, but still TTL) ?

Thanks for your feedback

Patrick
When the pre-flash goes off matters. Imaging you are trying to snap a shot of a dancer jumping in the air. You fire a shot mid jump. But due to the pre-flash delay, the dancer is already on the ground and onto the next move when your flash finally goes off. You'd have to try and memorize the exact amount of delay and shoot before the move is performed.
06-29-2015, 08:40 PM   #44
Otis Memorial Pentaxian
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 42,007
QuoteOriginally posted by enoeske Quote
You'd have to try and memorize the exact amount of delay and shoot before the move is performed.
Or shoot with available light and call the blur "art".


BTW...the dancer and everyone else on stage will thank you for not using flash during the performance.


Steve
06-29-2015, 09:22 PM   #45
Veteran Member




Join Date: Nov 2008
Location: Melbourne, Australia
Photos: Gallery | Albums
Posts: 595
Or you shoot in manual or Auto-thyristor flash mode and have no pre-flash at all.

For what it is worth I agree with ScooterMaxi Jim and believe that the latest firmware has sped up p-ttl.

Howie Be
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
blur, cameras, code, dancer, delay, dslr, event, fire, flash, ir, k-3, k-x, k3, light, move, p-ttl, pentax k-3, pre-flash, ratio, shot, shutter, stage, time
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Pentax k3II has the shutter lag problem wtih external flash ontariomark Pentax K-3 & K-3 II 24 06-09-2015 12:56 AM
FW 1.04 and longer AF lenses: any improvement? Doundounba Pentax K-01 4 09-24-2013 08:16 PM
FW 1.03 - flash modes bxf Pentax K-5 & K-5 II 6 03-11-2011 08:15 AM
Pre-flash to flash time lag? Raven Flashes, Lighting, and Studio 3 07-24-2008 06:57 AM



All times are GMT -7. The time now is 09:04 PM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top