HDRI Community

Funny, I haven't seen the "out of range message"
Does your sequence include a shot faster than 1/4000?
Or longer than 30" but you're not using a shutter cable?

The error at START is annoying. A warning sign of some kind would be better.


For your display suggestion, I would not find the bars so useful. I think they'd be hard to count if you're doing 11 frames - let alone 45! (and 45 bars wouldn't even fit I'd bet.) I'd rather see something simple at the bottom, like "1/2000 - 1/8" since the other info (Mid Exposure and Total frames) is already displayed.

Just found another little "bug." If you enable MLU on the Promote, but don't plug in the USB cable, you get twice as many shots as planned.

I set up a series of 12 shots in Time-Lapse mode. I plugged in the shutter release, but not the USB. MLU was enabled. I got 24 shots instead of 12 - I'd guess because it was sending once to lock up the mirror, and again to take the shot.

Well, this is user error, more than a bug. But it would be nice if the Promote recognized that MLU was not possible because the USB cable wasn't plugged in. A warning would be nice, as well as sending the correct number of shutter signals. (This is assuming that the lack of USB could be detected.)

Not a big deal, really.

> My suggestion would be to be to have the Promote calculate and display the shutter speeds for the brightest and darkest frames of the configured sequence

Interesting suggestion, we could make it a setup setting to choose as your preference. I'll enter that into our todo list.

> 1/2000  |  |  |  |  |  1/4

Some visualization of the chosen sequence you mean? I like that. It could only be shown in the "Mid Exp" mode, as in Start/End exp mode it wouldn't matter (user would set start/end himself).

> Funny, I haven't seen the "out of range message"
Does your sequence include a shot faster than 1/4000?
Or longer than 30" but you're not using a shutter cable?

If a shot is longer than 30" with no shutter cable, it will tell about that directly and suggest using the shutter cable. I guess we need to extend the warning about 1/4000s limit...

> Just found another little "bug." If you enable MLU on the Promote, but don't plug in the USB cable, you get twice as many shots as planned.

Not quite true... the reason for this behavior is that MLU was probably off on the camera. MLU actually is not connected to USB in any way - camera simply does not report MLU status via USB (I tried to make a warning like that but unfortunately cameras are not that chatty in this regard). But you are right, Promote sends two signals with MLU on mode, and if camera does not have MLU enabled, it will take two pics per one planned frame, just like it would if you pressed shutter release button on the camera twice for every shot.

> Has anyone tested this out on a prosumer camera model such as the Canon XSI? From my understanding, non Mark series Canons are limited at 1/200.

We tested it with the XSi, and it worked fine, with the same range of possible exposures (up to 1/4000). One thing to mention is that we don't have shutter cables for prosumer Canon DSLRs yet, but you can use any 3.5mm to 2.5mm audio cable for that purpose, and we will soon start carrying those in stock.

I spent 4 hours with the Promote the other night, shooting nothing but HDRs and filling 8 gigs of CF cards.

I ran into one "problem" that I'm going to pass on as a tip since it's happened to me twice now. On my 50D, the Promote can't set a shutter speed longer than 4 seconds if the camera is set to high- or low-speed continuous shooting mode. I got an error a few times when the frame was supposed to be 8 seconds. I finally remembered - again - to set the camera into single shooting mode.

So far, it seems like the promote works perfectly as long as I remember to set the 50D to Manual exposure, One-Shot AF mode, and Single-shooting drive mode.

I can confirm that Michael's case is likely the D3's buffer issue. I was able to reproduce it with my D3 and 27 HDR shots in a row, using NEF + JPEG S/Norm. If I shot 14 bit compressed NEF only (with no JPEGs) - the issue was solved. Using USB alone (without shutter cable) also solves the issue, but then it doesn't work as fast as with shutter cable - which is why the issue is solved. Having a shorter HDR sequence shot didn't have issues either. FWIW, I was shooting with dual Sandisk Extreme IVs.

We might consider adding an optional extra delay in shutter cable mode, but then it loses one of the biggest advantages of the shutter cable - speed. If you guys have any input on this, I'd love to listen.

I'd have to suspect that this would also happen to me with my Canon 50D. I'll have to give it a test tonight and see.

This may be solved by

1) Taking exposures in the opposite direction. Right now the Promote blasts away taking the fast frames first, which fills the buffer and it never catches up. By reversing and taking the slowest shots first, the buffer will always be empty until the shots are taken faster than x seconds. (With x depending on the camera itself.)

I don't see this working for every possible camera with every possible number of frames taken.

2) Have a user-configurable delay between shots when the total number of shots in the sequence exceed a user-configurable number. (So I can choose a 200ms delay when 17 shots or greater, 500ms delay when 23 shots or greater...) This is great for those of us who can figure out the correct delay for different numbers of shots in a sequence.

This is miserable for typical end-user ease-of-use issues - someone will screw up these settings, and be plenty ticked off at you (Arty) because they didn't read the manual properly. And I wouldn't blame them.

3) You (Arty) choose the delay when the sequence exceeds x shots, and solve for different values of x.

You can make this work for any situation by using excessive values - which covers your butt in any circumstance. That costs the user time, and a repeat of what you said about the benefit of the shutter cable.

4) You (Arty) *figure out* what the delay should be for each value of x, depending on the camera model that the end user configures into the Promote.

While this may seem impossible, I would bet that Rob Galbraith's CF card database of write speeds may be extremely valuable, and let you come up with reasonable delay values for each value of x.

This would require a new Setup menu item, allowing the user to select their camera model. What a pain to list each model!

5) Do the above, but simplify things. Have the user select their camera from 5 grades of camera - "entry-level, mid-tier, pro-sumer, high-end, sports model" or "slow" thru "ultra-fast".

This may be ideal, in the end, for the end user. If they find that some scenario fails (like in Michael's case) then they just need to set their camera model to 1 level down. No error should happen twice, and users should be just fine with that.

It will be easier for you (Arty) to figure out what these settings should be, too. Pick a delay, and a value of x. Use the user-configured camera "grade" selected to adjust delay and x.

As an example, a "very slow" camera gets delay is 100ms at 15 frames, and 500ms for 25 frames. A "slow" gets 100ms at 19 frames, and 500ms for 29 frames. And so on.

So you need to go into your lab and figure out what the numbers should be for 2 "grades" of cameras, like a 1d Mk II (highest grade) and a 50D (mid grade). You should then be able to figure out all the other settings from whichever two grades of camera you pick.

And then send this new feature out to beta testers who own enough cameras to cover all or most variations. Possibly allow beta testers to adjust your settings for delay and x, allowing them to give you back what might very well be exact values for each delay for each value of x frames for their specific camera.



I vote for #5 unless someone comes up with better suggestions.

Scott, I agree about the Setup setting. That's what we did in the default mode of "USB only" - it just has "milliseconds" vs "entry-mid-pro" grades, and the reason is because my pro level D3 (which needs 1000ms) is actually slower in USB only than Canon's XSi (which works fine with 500ms). So we left it at 1500ms which is what was needed by D200 - the slowest of a rather big bunch that we tested.

However, with shutter cable we never had this problem - honestly, because we didn't do a dozen long frames with all the bodies. Michael certainly picks the prize for figuring it out... thanks Michael!

Speaking of buffer vs. no buffer. I might imagine why it would happen with long vs. short exposures - the Promote Control cannot know when the camera is done shooting (at least not that reliably on 20+ different bodies with their own specialities), so it just times along what the camera is supposed to shoot. We added a safety margin to that as well (actually a few of those), but looks like it's not enough. I guess we will just go the simple route and introduce an optional setup setting that would tame it down in shutter cable mode as well. It will be a different setting from the USB only delay. I will also try to reproduce Michael's long vs. short issue on my D3 before we do any changes.