Accurate USB remote sync for single camera shots?

Microfunguy

4037

Re: Accurate USB remote sync for single camera shots?

« Reply #10 on: 29 / February / 2008, 18:03:23 »

Advertisements

This definitely works on the A570, here is a gallery of sample images (enable Javascript) :-

Lake Tega (Java Script)

Many people are using this camera for precision synch.

It needs a minimum voltage of at least 3.4v.

Remove ALL files from the SD card and copy-over the new files.

Run Synch.bas and plug the USB cable into the PC for testing in case your switch has problems.

Logged

fudgey

1705
a570is

Re: Accurate USB remote sync for single camera shots?

« Reply #11 on: 02 / March / 2008, 16:34:19 »

Still can't seem to get this to work

I took an empty MMC card, put SDM from 1.40-148-a570-100e-57.zip on it, enabled synch in SDM menu and remote enable in script menu and tried Simple_Remote.bas. No change in operation, still normal non-synced remote shooting.

I loaded Synch.bas and made sure synch was enabled in script menu as well, and still no sight of falling edge remote trigger.

I tried these with both my battery powered trigger and a PC, made no difference.

Logged

crunchy_3d

25

Re: Accurate USB remote sync for single camera shots?

« Reply #12 on: 03 / March / 2008, 06:19:31 »

Quote from: fudgey on 02 / March / 2008, 16:34:19

I took an empty MMC card, put SDM from 1.40-148-a570-100e-57.zip on it, enabled synch in SDM menu and remote enable in script menu and tried Simple_Remote.bas. No change in operation, still normal non-synced remote shooting.

Have you also tried with SD card?
I cannot imagine what else could be wrong except that something happened to USB input (some overvoltage?).

« Last Edit: 03 / March / 2008, 06:26:58 by crunchy_3d »

Logged

Microfunguy

4037

Re: Accurate USB remote sync for single camera shots?

« Reply #13 on: 03 / March / 2008, 11:49:26 »

Quote from: fudgey on 02 / March / 2008, 16:34:19

Still can't seem to get this to work

If using an SD card does not solve the problem contact me directly and I will send latest unreleased version (though we know the released version works).

David

Logged

fudgey

1705
a570is

Re: Accurate USB remote sync for single camera shots?

« Reply #14 on: 10 / March / 2008, 17:41:35 »

I didn't have the time to clean my only SD card for testing this, but I don't think that would've helped anyway.
Anyway, I put SDM 1.5 on the MMC card and sync works now

I'll check the delay characteristics later when I'm closer to a digital 'scope.

Logged

fudgey

1705
a570is

Re: Accurate USB remote sync for single camera shots?

« Reply #15 on: 11 / March / 2008, 16:08:11 »

I measured the time from USB voltage release to blue led turn-off on my a570is 1.00e. In a series of 52 shots, the time varied between 60 us and 380 us, or in other words, 220 us +- 160 us (average 226 us).

Shutter lag measurement is a bit trickier so didn't do that yet.

Logged

Microfunguy

4037

Re: Accurate USB remote sync for single camera shots?

« Reply #16 on: 12 / March / 2008, 06:23:16 »

Quote from: fudgey on 10 / March / 2008, 17:41:35

Anyway, I put SDM 1.5 on the MMC card and sync works now

Memory cards seem to cause all sorts of problems.

Quote

I'll check the delay characteristics later when I'm closer to a digital 'scope.

You can if you wish but photographing a CRT monotor is easier and you need 30 or 40 readings to get representative values.

The spreadsheet provided with the sync tester calculates the average but 'median' is a more representative value.

Full details of synch-testing are on SDM website and spreadsheets of tests are in the Files section of Yahoo 'StereoDataMaker' Group.

David

Logged

fudgey

1705
a570is

Re: Accurate USB remote sync for single camera shots?

« Reply #17 on: 12 / March / 2008, 13:39:50 »

Quote from: Microfunguy on 12 / March / 2008, 06:23:16

You can if you wish but photographing a CRT monotor is easier and you need 30 or 40 readings to get representative values.

The spreadsheet provided with the sync tester calculates the average but 'median' is a more representative value.

Full details of synch-testing are on SDM website and spreadsheets of tests are in the Files section of Yahoo 'StereoDataMaker' Group.

If I'm not mistaken, the CRT sync tester is only useful for measuring the offset between two cameras. I only have one and I'm trying to determine the delay from remote release to shutter opening. The dual camera mis-sync data from the SDM archives obviously already gives a clear pointer to what kind of a deviation I should be expecting.

Logged

crunchy_3d

25

Re: Accurate USB remote sync for single camera shots?

« Reply #18 on: 06 / April / 2008, 04:41:16 »

Quote from: fudgey on 12 / March / 2008, 13:39:50

If I'm not mistaken, the CRT sync tester is only useful for measuring the offset between two cameras. I only have one and I'm trying to determine the delay from remote release to shutter opening. The dual camera mis-sync data from the SDM archives obviously already gives a clear pointer to what kind of a deviation I should be expecting.

You are not mistaken! CRT can be used for two or more cameras.
Your results are very good (up to 380us delay) and it seems that you'll be able to make good stereo or precise-timing mono photos.

Logged

plasma_physicist

4

Re: Accurate USB remote sync for single camera shots?

« Reply #19 on: 16 / May / 2008, 17:10:00 »

I am a scientist at Princeton University and was very excited to see the work by CHDK because of the possibility of using it in our research. Using a 300 $ camera to do the work of a 20K camera is very appealing in research. We want to be able to capture a plasma event that produces a very bright light that lasts 10-50 microseconds. We bought an S3 IS and confirmed with a rotating mirror and a laser that CHDK software indeed enables shutter speeds of 50 micro seconds or less. I didn't spent too much time on measuring this accurately because I also needed to know if the camera can be triggered precisely.

I tested how accurate I can trigger S3 IS. Our data system can send triggers (5 Volts square pulse ) with less than micro second accuracy. I 've read about using usb port to trigger and how its accurate to within certain milliseconds which is not nearly good enough. Anyway I didn't dig in too much so I went on and opened the camera and connected wires to the shutter button. I connected one pair of these wires which is equivalent to half pressing the shutter and the other pair was closed with a solid state relay which is equivalent to pressing the button all the way. The trigger pulse to take the shot had to be at least 30 milliseconds, the camera did not respond for shorter trigger pulses. I also triggered a photodiode separately. I could capture the photodiode lighting up within 5 miliseconds. The accuracy we want to have is 100 times more. I 'd really appreciate if anybody has a solution to this.
Is the camera hardware capable of being triggered with microsecond accuracy and how can we do it?

There are cameras that can be triggered with nano second accuracy, one billionth of a second. This type of camera goes for about 20-30 K. I know also so called streak cameras that can be triggered with picosecond accuracy one trillionth of a second which cost ~100 K. We are also planning to call and ask someone from canon, I don't know how canon feels about CHDK. A camera so cheap with microsecond imaging capabilities would be extremely valuable in worldwide research.

Logged