The key here is your assumption about data rate. I don't know the
answer. I've sent the following note to Paul Buff tech support. I'll
post the response when I get it.
----------------------------------------
In the Cybersync user's manual and in the frequently asked questions
pages on your web site there are statements that the Cybersync has a
1/4000 sec. latency and can sync as fast as 1/2500 sec. Then the user's
manual continues with a recommendation that, when using the Cybersyc,
the camera's shutter speed should be set 1/3 to 2/3 stop slower than the
camera's maximum sync speed. The specific example given is that a
camera with a max sync speed of 1/250 second be set to 1/200 or 1/160.
I don't understand "latency" here or sync as fast as 1/2500 followed by
a recommendation to use less than 1/10th that speed. I understand that
the camera in this case can't be set faster than its maximum x-sync
speed of 1/250 but why does it need to be set even slower? More
specifically, what does that mean for my Canon 5D with max sync speed of
1/200 second? Do I need to use 1/160 or slower? If the camera with
1/250 max sync speed can use 1/200 why can't I? An, once again, why
can't the 1/250 camera use 1/250 if the Cybersync can sync at up to 1/2500?
What don't I understand here? I do use my Bees outdoors for fill flash
with large groups and every 1/3 stop is important.
Thanks,
Chuck Norcutt
Endwell, NY
Ken Norton wrote:
> Chuck, I believe the latency numbers are calculated theoretical minimums.
> Kinda like how Behringer calculates the specs for their audio products. ;/
>
> Just thinking off the top of my head:
>
> 1. Transmit rate is most likely no higher than 9600 baud.
> 2. Datagram could be like four bytes or 32 bits.
> 3. The entire datagram would take 1/300 of a second to transmit
> 4. To achieve 1/4000 of a second, they could only be transmitting two
> bits--not including ANY processing time.
> 5. It is impossible to provide adequate channel assignment and protection
> from falsing with only two bits.
> 6. If one were to do this with ONE byte datagrams, at 9600 bps, it would
> take 1/1200 of a second for transmittion of the datagram.
> 7. These triggers are required to operate in the ISM (Industrial,
> Scientific, Medical) bands. As such, there is plenty of other data
> interference in the air on those same frequencies, so you'd need a larger
> datagram to coexist with other wireless devices.
>
> To do this right with adequate falsing prevention, device control and
> assignment, I can't imagine trying to do this with fewer than 8-bits.
> Figuring a 50% overhead in detection and processing, it would take 1/800 of
> a second before the trigger would fire. The next question is whether the
> trigger itself is transister-based or relay based? Either way, you will
> have a ramp-time delay.
>
> Now, consider that a strobe fires in a non-linear manner with the bulk of
> the light emitted approximately 2/3 into the pulse. Consider the other
> delays within the flash's electronics itself. I've found that using my
> Minolta A1 as a test device when I fire strobes directly attached to the
> camera that I get varying response rates with some flashes fully
> illuminating the scene at 1/640 and others as low as 1/250. Monolites seem
> to be worse than the Vivitars. At 1/250 everything works with the A1, but
> if I'm using the ebay remotes, I need to back that off to 1/200 or less.
>
> Maybe Wayne can shed a little light in here on this subject, but I suspect
> that any claims of 1/4000 response are a "bit" optimistic.
>
> AG
>
>
> ==============================================
> List usage info: http://www.zuikoholic.com
> List nannies: olympusadmin@xxxxxxxxxx
> ==============================================
>
>
> ------------------------------------------------------------------------
>
>
> No virus found in this incoming message.
> Checked by AVG - http://www.avg.com
> Version: 8.0.175 / Virus Database: 270.8.5/1757 - Release Date: 10/30/2008
> 2:35 PM
>
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|