SGP Beta v2.6.0.1 / 2 Center & Rotate bug


#1

When using the manual rotator and doing a slew to target, then a center and rotate the value reported is way, way off. In my case the camera is at 270 and SGP reports Zero (0).

If you do a slew, center (only), then run the center and rotate things are fine.


#2

Please see here and we’ll take a look:


#3

Ken,

There’s nothing in the log files other than me cancelling the routine after being presented the following dialog box.


#4

you just gotta post the logs… what looks like nothing to you might be evidence of a problem for ken & jared.


#5

The failure is happening because the initial value for the rotator is invalid as it has never been synced we don’t have a value for it. Which is why when it runs with an initial center that it works fine. Also if a Solve and Sync had been performed at some point it would also have worked.

I’ll have to think about how to reconcile this. But clicking “Ok” should allow the verification to run, which will Solve and Sync and populate a valid angle. From that point things should work as expected.

Thanks,
Jared


#6

Jared,

Thanks for that! I usually do a Slew to Target, then Center on Target as part of my routine probably 99% of the time. Once centered, I then proceed with a Center Target & Rotate Camera.

[quote=“Jared, post:1, topic:4689, full:true”]
Which is why when it runs with an initial center that it works fine.[/quote]

Maybe I’m missing something here, but the code classes for Slew, Center and Rotate would seem to be components that could be called independently and/or sequentially.

Would it be wrong to assume the user who just clicked Center Target & Rotate Camera would expect that the core Center function/class would be executed and thus provide the Rotate function/class the required value?


#7

If you recall in the release notes, we removed “step 2” of center and rotate. This step forced sync of both mount and rotator allowing step 3 (slew and rotate) to perform as expected. We also indicated that the removal of this step places more onus on the user to ensure that your equipment is in sync with reality before these actions are run. In your case, the rotator has never been synced.

We will add a message that indicates if a rotation is requested without the rotator ever having been synced. Hopefully this will help transition folks to the new method.


#8

Actually… giving this a little thought, it may just be better to move the old auto centering “step 2” to the front of a sequence (since SGPro only needs this data once to get started)


#9

I guess I misinterpreted, and/or under estimated, what the impact of that statement in the Release Notes for 2.6.0.0 would turn out to be. :disappointed:


#10

It’s really not all that dire. It essentially means that when using the manual rotator that you need to sync before hand OR click “OK” on the rotation dialog without doing anything and let it fall through to the verification step which will solve and sync for you.

Since “step 2” has been removed we no longer do the initial sync for the Manual Rotator that would have been done here. However this is a pretty small price to pay since you just have to do the initial sync once or let it fall through to verification. This should end up saving you about a minute per Auto Center/Center Here.

Thanks,
Jared


#11

@rjbokleman

If you are using the manual rotator, this should be fairly transparent to you now (meaning you won’t need to click “OK” to arbitrarily force a solve, or worse, actually attempt to rotate to an erroneous angle then be very wrong after validation). If you run a solve and sync prior to sequence start, SGPro will be fine. If you don’t SGPro will automatically perform the steps to acquire missing information (one time at sequence start) and then be good from there. This is essentially no different that centering has been in the past except you only need to incur the initial “pre-sync” one time instead of every time.


#12

Got it. Thanks, Ken!


www.mainsequencesoftware.com