Logging run times with a camera?

insanefred

Enlightened
Joined
Sep 30, 2008
Messages
453
I assume it is accurate enough, I just need to confirm I am not crazy.
I set my camera on a tripod and set the interval timer for every 2 minutes. (1/200 sec, f/2, ISO200) on manual. With the light shining on a white wall.
Does this work?
 
Last edited:
Seems like it could work just fine. I've toyed with the idea of using the camera's meter for relative intensity measurement. I wonder if you used an aperture priority auto setting whether you could better track the dimming of the light through the shutter speed changes? Of course, the pics would all look the same.

Geoff
 
I pointed a webcam and used timershot powertoy for Xp, at my luxmeter...:)

Crenshaw
 
Seems like it could work just fine. I've toyed with the idea of using the camera's meter for relative intensity measurement. I wonder if you used an aperture priority auto setting whether you could better track the dimming of the light through the shutter speed changes? Of course, the pics would all look the same.

Geoff

I am observing the histogram. (Apparently my TK20 runs for 2hrs, 15 mins on 2600mah batteries. before 50%)
 
Last edited:
yeah that would work but that would be alot of pics at 2min increments..

Nah, last night was only 75 pictures (75 x 2 min = 2hrs 30 mins.) I usually take as many as 1200 pictures in just 3 hrs when ever I do events (i.e. Roller derby) . 8-10hrs for weddings.
 
okay, you are a professional. that explains the auto incremental shots aka canon remote shoot etc.

shoot away :twothumbs
 
Yes it is possible. It is even possible to do it in a fully automated way with logging of brightness in a single file. You don't even have to take pictures :huh:

You just need some Canon Powershot camera with installed unofficial CHDK firmware(it will add a TONS of features to your camera) and a little script written in uBasic (find a command get_bv96). And some calculations according to APEX system.

Pretty simple :D

It took me some time to figure it all out. But now I make runtime graphs(see my sig line!) with my two years old crappy compact canon camera:)
 
Yes it is possible. It is even possible to do it in a fully automated way with logging of brightness in a single file. You don't even have to take pictures :huh:

You just need some Canon Powershot camera with installed unofficial CHDK firmware(it will add a TONS of features to your camera) and a little script written in uBasic (find a command get_bv96). And some calculations according to APEX system.

Pretty simple :D

It took me some time to figure it all out. But now I make runtime graphs(see my sig line!) with my two years old crappy compact canon camera:)


I installed the CHDK pack on my g/F's camera some time ago for her to shoot raw. maybe I'll look at it again for fun.
 
Last edited:
Hey, this is an interesting thread here !


Thank you to all for sharing this info.


:twothumbs
_
 
I installed the CHDK pack on my g/F's camera some time ago for her to shoot raw. maybe I'll look at it again for fun.

This is the script I use. Sorry some comments are not in english :D

Code:
rem  Flashlight runtime measuring script by Jirik
rem  based on Divalent's "accurate intervalometer" script
rem  Cameras: all Digic II camera's, Digic III not tested
rem  CHDK version: (must have get/set prop)
rem
@title Flashlight 1,2 min Runtime logger
@param a Pocet mereni po 1 minute na zacatku
@default a 10
@param c Prumerovani z x mereni.
@default c 3

rem Seznam promennych:
rem a,c,f,g,h,n,m,p,q,t,v

rem zapneme logovani do souboru
 print_screen 1

rem nophoto sleep
 g=1500
rem Prumerovani z x mereni
 if c<1 then c=1
rem f - offset kvuli spozdeni fotaku pri mereni
 f=500+g*c

 sleep 500
rem Prvni snimek ulozime vzdy
 shoot
print "Runtime logger start"
 get_bv96 h
 q=get_tick_count
 v=get_vbatt
 print q, h, v
 

 if a<1 then goto "loop2"
rem po minute
 t=60*1000-f

for n=1 to a
 sleep t
 
 p=0
  for m=1 to c
   press "shoot_half"
   rem musime fotaku dat cas na zmereni expozice
   rem je mozne, ze pro nektere fotaky bude nutne hodnotu zvysit
   sleep g
   get_bv96 h
   release "shoot_half"
   p=p+h
  next m
 p=p/c
 q=get_tick_count
 v=get_vbatt
 print q, p, v
next n

:loop2
rem po 2 minutach
 t=2*60*1000-f

:twominloop
 sleep t
 
 p=0
  for m=1 to c
   press "shoot_half"
   rem musime fotaku dat cas na zmereni expozice
   rem je mozne, ze pro nektere fotaky bude nutne hodnotu zvysit
   sleep g
   get_bv96 h
   release "shoot_half"
   p=p+h
  next m
 p=p/c
 q=get_tick_count
 v=get_vbatt
 print q, p, v
goto "twominloop"
Just save it in the file and place it in the script directory on your card. It will log runtime in file named PR_SCREEN.TXT

On the beginning it takes only one photo. Then it only measures exposure values(average from 3 measurements) but doesn't take pictures. In first 10 minutes it reads values every minute, then it takes values every 2 minutes.

the log file looks like this:

Runtime logger start
15530 722 2560
75690 719 2513
135840 718 2508
196000 717 2504
256160 717 2504
316320 716 2500
.
.
.

The first column is time in milliseconds from the start of the camera. Second column is average luminance of the scene in logarithmic scale and it is multiplied by 96. The figure for luminance in linear scale is this:
B = 0.2973*12.5*2^(measured value/96)

The last column is actual battery voltage in millivolts (just for fun:)
 
Last edited:
Just when we were getting tired of

Fenix vs. SureFire



We now return to

Canon vs. Nikon



Coming soon:

Ford vs. Chevy

_
 
Fenix vs. SureFire
So far I am liking Fenix. Does this topic really come up often?





Canon vs. Nikon
Nikon, DUH... :thumbsup:





Ford vs. Chevy
Neither, I am a Toyota guy.
 
Fenix vs. SureFire
So far I am liking Fenix. Does this topic really come up often?


:lolsign::lolsign::lolsign::lolsign::lolsign:


Only about once a week.
_
 
I assume it is accurate enough, I just need to confirm I am not crazy.
I set my camera on a tripod and set the interval timer for every 2 minutes. (1/200 sec, f/2, ISO200) on manual. With the light shining on a white wall.
Does this work?
I would imagine that 2 minute interval is a bit too long. When I was doing my setup I was using a Nikon D2h on a tripod with ijnterval of 15 sec. Worth mentioning that the picture quality do not have to be hi res. Normal low res jpeg would suffice. Furthermore displaying the photos of smaller low res pictures wouldn't be a burden on the PC/Mac. Shorter intervals gives much better representation of the performance of your light. :)
 
Yes it is possible. It is even possible to do it in a fully automated way with logging of brightness in a single file. You don't even have to take pictures :huh:

You just need some Canon Powershot camera with installed unofficial CHDK firmware(it will add a TONS of features to your camera) and a little script written in uBasic (find a command get_bv96). And some calculations according to APEX system.

Pretty simple :D

It took me some time to figure it all out. But now I make runtime graphs(see my sig line!) with my two years old crappy compact canon camera:)


jirik_cz, thanks a lot for your idea and your script!:thumbsup:

With my IXUS75, CHDK and script I was able to log the runtime of an M60 on a 6P, with almost fresh AW protected batteries. Here is the result:

runtimegraphM60_AW.jpg



Here is the graph for another M60 unit, this time in a C2, with a couple of fresh AWs. This one ran for four more minutes, and is a little brighter, too!

C2-M60-AWs.jpg



And below, an Ra Twisty 100 Tw in High mode, also running on an AW protected battery, almost fresh. After 90 minutes the values stopped making sense, as the flashlight began flashing the low batt warning. Anyway, spot on for 60 minutes runtime before starting the step down to conserve power.

RaTwisty100TwMax_AW.jpg


On a brightness direct comparison, these three lights scored 847.5 (6P+M60), 951.3 (C2+M60) and 445.7 (Twisty), respectively. These numbers are for relative comparison only.
 
Last edited:
I built my own integrating sphere (apparatice is more like it). I placed 2 photo cells one on each side of the flashlight and measuring the reflected light. I use a Radio Shack Cat# 22-812 DVM (which has a RS-232 output) to measure the ohms of the 2 photocells in series and record the info on my computer. The recording software graphs the results. The recording software can save data at 1 sec, 5 sec, 10 sec, 30 sec, and 60 sec time intervals. I almost always use the 1 sec time interval. You get a graph like crocodilo's, but you can zoom in and take extreme closeups of the data points. You can see when the flashlight's output starts to drop and how fast or even if the flashlight just go's completely out all of a sudden. The data file is in text form and you can import it to Excel and analyze things that way. You can start the test and just walk away and see what the results are at your convience. The RS-232 port can output anything that the meter can read: volts, current, ohms, frequency, etc. The DVM, cables, and software (all in the one box) are around $70. Now all I need to do is borrow a Lux meter and calibrate the setup.
 
Top