For the desire of minimalism

Initial requirements discussion

Moderator: seaton

For the desire of minimalism

Postby anton.tagunov » Fri Sep 21, 2007 8:48 pm

Hello gentlemen,

I'm no radio engineer. But I understand that building a good remote is hard. The primary problem AFAIK is latency. A random RC from a cheap toy is no good. A transmitter takes time to go from OFF to ON AIR.

This makes my ideal true Stobist style version 1 remote extremely spartan. It can do only one thing. But do it well. Trigger the flash. Any flash on the planet.
Other requirements are: reliable, legal, reasonably-ranged, extremely low latency, doable. Doable and reliable may imply simple.

What does it spell? To me - no microprocessor. Those who share my vision please express your opinions in this thread.
Last edited by anton.tagunov on Fri Sep 21, 2007 9:29 pm, edited 1 time in total.
anton.tagunov
 
Posts: 19
Joined: Fri Sep 21, 2007 6:52 pm

Postby brittonphotography » Fri Sep 21, 2007 9:00 pm

Would it be possible to have a microprocessor capable of higher end functions like ETTL, with a designed pass through for simple "just trigger flash as quickly as possible" function?
if we are picking a board to use, if it can do both lets get the one with the capability to upgrade, but still sync as quickly as possible with flash.
brittonphotography
 
Posts: 25
Joined: Fri Sep 21, 2007 8:52 pm

Postby anton.tagunov » Fri Sep 21, 2007 9:20 pm

brittonphotography wrote:Would it be possible to have a microprocessor capable of higher end functions ... if we are picking a board to use ... lets get the one with the capability to upgrade


Britton, everything is possible. The question is what the costs are - in money, efforts, reliability.

The post above is just my vision. I've written it solely to find out how many people agree that
upgradability is not a core requirement and can be offset by doability and simplicity

P.S. If people in the know argue that microprocessor achieves the best balance of simplicity doability and reliability I'll definitely agree that it's a good design.
anton.tagunov
 
Posts: 19
Joined: Fri Sep 21, 2007 6:52 pm

Postby brittonphotography » Fri Sep 21, 2007 9:30 pm

I would be curious about need for extra functions as well as i could see modifying the cactus/ebay design as a cheap way to make simple triggering reliable with greater range.

Is it possible to post a poll here?
would you rather a simple functional system for $35
or
a system capable of TTL/manual mode for $60

prices are guesstimates for now


My thoughts were that there were sufficient people interested in higher functions so if we could spec one system that could BOTH be used for a simple high speed tx OR complex it would be more likely to be a project that could have enough interest to be completed.
brittonphotography
 
Posts: 25
Joined: Fri Sep 21, 2007 8:52 pm

Postby JonSenior » Fri Sep 21, 2007 9:42 pm

As I posted in one of the other threads here, I believe both of these are achievable. I have proposed a system built around one of the transceivers discussed which would allow both simple triggering and the option to upgrade in the future. My take one a separate system would be to allow the possibility for a micro to take over control of the triggering and to have (if needs be) two separate radio routes. one for triggering, one for communication between devices.

Jon
JonSenior
 
Posts: 60
Joined: Fri Sep 21, 2007 7:27 pm
Location: Paris, France

Re: For the desire of minimalism

Postby mattsteg » Fri Sep 21, 2007 9:53 pm

anton.tagunov wrote:This makes my ideal true Stobist style version 1 remote extremely spartan. It can do only one thing. But do it well. Trigger the flash. Any flash on the planet.
Other requirements are: reliable, legal, reasonably-ranged, extremely low latency, doable. Doable and reliable may imply simple.

What does it spell? To me - no microprocessor. Those who share my vision please express your opinions in this thread.
Micros need not be slow. Even cheap stuff can react to an input and put a pin high in microseconds.
mattsteg
 
Posts: 26
Joined: Fri Sep 21, 2007 5:21 am

Postby mattsteg » Fri Sep 21, 2007 9:54 pm

brittonphotography wrote:Would it be possible to have a microprocessor capable of higher end functions like ETTL, with a designed pass through for simple "just trigger flash as quickly as possible" function?
if we are picking a board to use, if it can do both lets get the one with the capability to upgrade, but still sync as quickly as possible with flash.
Sure, that's extremely easy to do.
mattsteg
 
Posts: 26
Joined: Fri Sep 21, 2007 5:21 am

Postby mattsteg » Fri Sep 21, 2007 9:56 pm

anton.tagunov wrote:The post above is just my vision. I've written it solely to find out how many people agree that
upgradability is not a core requirement and can be offset by doability and simplicity

P.S. If people in the know argue that microprocessor achieves the best balance of simplicity doability and reliability I'll definitely agree that it's a good design.
Micros are really simple. They make circuit design massively easier. Look at http://snowcat.de/flashcontroller/ for example. Quite easy. Basic triggering with a micro can be implemented in hours if not minutes if the part is on-hand.
mattsteg
 
Posts: 26
Joined: Fri Sep 21, 2007 5:21 am

Postby bilmorfoto » Fri Sep 21, 2007 10:02 pm

I'm interested in an upgradeable, hackable system.

I see a progression:

1) Trigger only
2) Remote manual adjustment of flash level
3) Full TTL automatic functionality

Actually, 1) and 2) would suit me fine.

As far as pulling numbers out of a hat goes, ~$30 is dirt cheap, ~$60 is fine, ~$100 is probably too much.

Bill
bilmorfoto
 
Posts: 5
Joined: Fri Sep 21, 2007 8:15 pm

Postby brittonphotography » Fri Sep 21, 2007 11:00 pm

as far as Bill's progression goes,

Does 2 come after 3?

Is it possible to change just the manual power of a flash from its hotshoe connection??
I do not know of that being possible.

I figured for Remote manual adjustment we would need to figure out TTL functionality and then hack that somehow to give us adjustable manual control instead of letting camera dictate ttl controls.

I saw progression as
1. Trigger Only
2. Full TTL automatic.
3. Remote manual adjustment. after hack of ttl.

so while most strobists would be fine with trigger only and manual adj. to get to manual adj we need TTL so why not add it anyways.

I see using the same setup for stage 1 and stage 2 so that we only have one product.
brittonphotography
 
Posts: 25
Joined: Fri Sep 21, 2007 8:52 pm

I fully agree with this thread

Postby Thonord » Fri Sep 21, 2007 11:18 pm

I wish I had read this before I commented to an other of Antons posts.

viewtopic.php?t=17

We need to get organised.

Tom
Ppl who agree need normally not reply, those who disagree or have questions do.
Or - just ignore me.
Thonord
 
Posts: 50
Joined: Fri Sep 21, 2007 2:32 pm
Location: Norway

Postby mattsteg » Sun Sep 23, 2007 2:27 am

brittonphotography wrote:as far as Bill's progression goes,

Does 2 come after 3?

Is it possible to change just the manual power of a flash from its hotshoe connection??
I do not know of that being possible.

I figured for Remote manual adjustment we would need to figure out TTL functionality and then hack that somehow to give us adjustable manual control instead of letting camera dictate ttl controls.

I saw progression as
1. Trigger Only
2. Full TTL automatic.
3. Remote manual adjustment. after hack of ttl.

so while most strobists would be fine with trigger only and manual adj. to get to manual adj we need TTL so why not add it anyways.

I see using the same setup for stage 1 and stage 2 so that we only have one product.
The original way of doing TTL was a signal to start and a signal to stop flashing. Trivial to implement (although you have to profile your flash). Much easier than hacking TTL.
mattsteg
 
Posts: 26
Joined: Fri Sep 21, 2007 5:21 am

Postby kcactionphoto » Sun Sep 23, 2007 6:26 am

I concur with anton.tagunov. A simple trigger just like the Cactus, but better made, reliable and bulletproof. Start simple and go from there. that's how most things are manufactured.
kcactionphoto
 
Posts: 2
Joined: Sun Sep 23, 2007 5:55 am
Location: Tulsa, OK, USA

Postby TwoLeftFeet » Sun Sep 23, 2007 6:48 am

I didnt read this thread before replying in another one about KISS... Anyway, progression looks right to me.

#1: simple trigger
#2: TTL control in stages

For #2, we may want to define an expandable "strobist standard TTL protocol" that would send the data out to the remote from the hotshoe unit. Hotshoe unit would convert from TTL to SSTTL and remote unit would convert it back to TTL.

Anyway, as long as the #1 step is 100% reliable (okay fine, 99.999%) then I think most people would be happy.
TwoLeftFeet
 
Posts: 12
Joined: Sun Sep 23, 2007 5:52 am
Location: Toronto, ON

Postby strogg » Sun Sep 23, 2007 12:56 pm

Even for our most basic model, we want to include a microprocessor. It will not affect price by any significant amount, and it will reduce development time by an order of magnitude. If you want a reliable trigger, then you want to be able to encode the signal that you send, instead of just broadcasting "HIGH" to the frequency. This is what leads to fall triggers.

If you want to encode your signal, the best way to do this is with a microcontroller.

Btw, incase people think these things are pricy, you can buy 100 atmega48's for $1.80CAD. (about $1.80USD). This is for a fairly powerful, low powered one. You can get them even cheaper if we don't need everything that an atmega48 would have to offer.
strogg
 
Posts: 19
Joined: Thu Sep 20, 2007 8:31 pm


Return to Requirements

Who is online

Users browsing this forum: No registered users and 1 guest

cron