PDA

View Full Version : How does Oly TTL flash work?


snaarman
13th October 2009, 07:52 AM
Here is what I remember from my old Nikon days. My FE2 slr had a light sensor in the base of the mirror box. When the mirror was up, it measured light from the flash reflected off the film during the exposure and sent a command to stop the flash via the hot shoe.

So, does the Oly system operate in the same way.. if so, does it really try to measure light reflected off the sensor???

And also, the wireless bit: Is the wireless link fast enough to trigger and stop the flash when the camera says its had enough??

And what if there are two flashes? Does it TTL control them individually or do all the flashes get quenched when the camera cries enough?

Answers on a postcard please, you other engineers out there...

Pete

dbutch
13th October 2009, 08:05 AM
Not sure about the current way but I guess it is similar.

The OM2 was actually the 1st camera to use this technique and had a really funky 1st shutter blind that had a pattern that reflected like an avereage image or alike - interestingly I believe that Olympus licensed the technology from Minolta.

Link to some info on the OM here: http://www.mir.com.my/rb/photography/hardwares/classics/olympusom1n2/shared/flash/index.htm

Dave

Wreckdiver
13th October 2009, 08:06 AM
And also, the wireless bit: Is the wireless link fast enough to trigger and stop the flash when the camera says its had enough??


Come on Pete you, of all people, know the speed of radio waves :rolleyes:

Steve

snaarman
13th October 2009, 09:10 AM
Come on Pete you, of all people, know the speed of radio waves :rolleyes:

Steve

Well, true I guess. I just don't have a lot of faith in wireless anything these days. Give me good old cat5 :-)

The wireless flash has to get the go/stop decision through the fog of wifi in fractions of a millisecond.
"Sorry I didn't get that packet, can you send it again?"
"Pardon?"
"Oh forget it, the shutter is closed anyway"

*chr

Adagio
13th October 2009, 11:09 AM
I don't have an oly wireless flash but was under the impression it used light from the camera flash to communicate. Am I wrong?

photo_owl
13th October 2009, 09:17 PM
Here is what I remember from my old Nikon days. My FE2 slr had a light sensor in the base of the mirror box. When the mirror was up, it measured light from the flash reflected off the film during the exposure and sent a command to stop the flash via the hot shoe.

So, does the Oly system operate in the same way.. if so, does it really try to measure light reflected off the sensor???

And also, the wireless bit: Is the wireless link fast enough to trigger and stop the flash when the camera says its had enough??

And what if there are two flashes? Does it TTL control them individually or do all the flashes get quenched when the camera cries enough?

Answers on a postcard please, you other engineers out there...

Pete

in the order of your questions

1. No it doesn't. It uses the normal exposure processing against the dedicated exposure sensor in line with the metering points selected in menu. It does all this from a preflash of known output that it instructs the flash to make 1/100th of a second before the main flash.

2. It doesn't do it that way - once it's metered the flash based on the preflash it works out the required output in advance and instructs the flash output (whether by wireless or via the hotshoe.

3. This is a $64m question. I don't have an oscilliscope to test this. My instinct is that it does each in turn against it's preset EV but I haven't tested it (even roughly which is I guess easily done by putting them at either side and different distances from a subject with low ambient using spot metering and varying the target)

shenstone
13th October 2009, 09:33 PM
Well, true I guess. I just don't have a lot of faith in wireless anything these days. Give me good old cat5 :-)

The wireless flash has to get the go/stop decision through the fog of wifi in fractions of a millisecond.
"Sorry I didn't get that packet, can you send it again?"
"Pardon?"
"Oh forget it, the shutter is closed anyway"

*chr

Cat 5... How retro ... our shiny new facttory has something like 65 miles of Cat6e... My PC at home gets to the router and other periperhals with offcuts of the same gigabit cable.... trouble is then I hit the rural telephone line and things get ... hmm.... BT!


In response to the TTL/ preflashes etc... its this that's moving me away from infrared slave units in caving... the old guns were fine, but all these new ones with preflashes are a pain and you have to get the right slave until for the device etc.

Regards
Andy

snaarman
14th October 2009, 07:28 AM
in the order of your questions

1. No it doesn't. It uses the normal exposure processing against the dedicated exposure sensor in line with the metering points selected in menu. It does all this from a preflash of known output that it instructs the flash to make 1/100th of a second before the main flash.

2. It doesn't do it that way - once it's metered the flash based on the preflash it works out the required output in advance and instructs the flash output (whether by wireless or via the hotshoe.

3. This is a $64m question. I don't have an oscilliscope to test this. My instinct is that it does each in turn against it's preset EV but I haven't tested it (even roughly which is I guess easily done by putting them at either side and different distances from a subject with low ambient using spot metering and varying the target)


Ah, I begin to understand.

I guess it is flashguns that have become more clever since lest century: Being able to output individual sequenced pwoer-specific flashes is clever. Being able to set the flash to "continuous" or FP mode is very clever.

For the moment I will stick to my cheap and cheerful setup. An old Nikon SB28 on a string and a 99p Vivitar flash with a light sensing trigger module.

:)
Pete

Ian
14th October 2009, 10:36 AM
Myself, Graham (of Rainham) and others discussed DSLR OTF (off the film) metering possibilities a while back on our DPNow forum:

http://dpnow.com/forum2/showthread.php?t=7792

The conclusion was that OTF was practically impossible to implement with a DSLR.

Ian

Graham_of_Rainham
14th October 2009, 05:58 PM
Being a bit of an optomist, I like to think that some "bright young thing" will work out a way of real time metering that will provide the same functionality of the OM OTF TTL Flash metering.

That way the "Painting with light" albeit flash or continious can be done as easily as it was with the OMs

Ok it was only used a few times a year, but the results were really good. Having said that working out the exposure and doing it manually with the DSLR is as much fun as it was before the OM2 came along, and at least we can now see the results instantly and adjust accordingly.

OR:

Fake the whole thing with Photoshop :eek:

Mike_in_Germany
15th December 2009, 11:09 AM
The Olympus radio flash uses the 433MHz band (Amateur 70cm Band), and I think there is 32 channels reserved for industrial use, of which Oly uses only 4 or 5. Most camera firms use this frequency. The time laps between trigger and flash is so small something like 0.02Sec and might even be smaller than that.

Mike

Ian
15th December 2009, 11:34 AM
The Olympus radio flash uses the 433MHz band (Amateur 70cm Band), and I think there is 32 channels reserved for industrial use, of which Oly uses only 4 or 5. Most camera firms use this frequency. The time laps between trigger and flash is so small something like 0.02Sec and might even be smaller than that.

Mike

I must admit that my knowledge of the Olympus RC flash system is pretty basic from a comms point of view, but I was under the impression that the camera communicated with the flash only via flash signalling from the on-board flash, not radio? I'm glad to be declared wrong if it means my knowledge is improved though! :)

Ian

photo_owl
15th December 2009, 03:28 PM
I must admit that my knowledge of the Olympus RC flash system is pretty basic from a comms point of view, but I was under the impression that the camera communicated with the flash only via flash signalling from the on-board flash, not radio? I'm glad to be declared wrong if it means my knowledge is improved though! :)

Ian

Don't see you being proved wrong on this - RC is remote control not radio control for the Olympus E system.

I'm facinated as to what Mike's referencing..........

Mike_in_Germany
15th December 2009, 08:00 PM
Oh dear, this is most confusing. I can remember a discussion about 3 years ago with an Olympus teach man in Hamburg, and I am sure he said that the E series emitted a 433.92 MHz trigger. Maybe he was wrong because it has nothing on the main Olympus web site. Sorry if I have confused all as well.


Mike