Workaround/bug fix to stop llTargetOmega?

I'm running straight into a wall with the bug where llTargetOmega won't stop.

Setting the spinrate to 0.0 doesn't do anything (in testing, I can't change the rate, period, increasing it makes no change as well). Forcing the viewer to pay attention with llSetText/llSetAlpha doesn't do anything either. llResetScript doesn't help, although I can get it to stop by actually hitting the reset script button, albeit only sometimes. It's a child prim in an attachment, and putting the script in before linking it doesn't help. I've tried every workaround I can find, and I'm tearing my hair out, can anyone walk me through a way that will actually make it stop?

In specific- my parent prim sends out a link message to the child telling it to start or stop. The child has the receiver and the llTargetOmega commands. This is to be an attachment. Currently I can get it to start perfectly fine, but then it plugs away in that state and won't listen to any changes to it.

you sure its not just a script error?

Also you might be setting the gain to 0 with your stop command, I think the gain must be >0 for the command to be accepted (or something like that).

It isn't a script error, I can get any other information to pass through flawlessly. Gain is 1 in all calls. The client does not pay attention to the fact that llTargetOmega has been changed, and I am unable to force it to update.

Related posts, whose workarounds do not work for me:

/54/c6/185468/1.html
/54/73/79278/1.html

The prim or script might be corrupt, try recreating them.

// Stop spinning
llTargetOmega(<0,0,0>,0,0.0);

works for me.

Al isn't crazy; I've seen the same problem. However, I didn't try using a gain of zero.

heyas;

i saw the non-stopping lltargetomega bug on the jira.... and i had noticed this week that my method of writing/dropping/compiling the script in a non-linked prim prior to linking -- just doesn't do squat any more.

so i suspect somebody actually worked on that bug. and, uh, made the function more consistently not work?

i will have to check my previous examples that had actually been working (yes, starting/stopping/speed changing attachments).

al, feel free to pester me ingame with ims to remind me to do it.

Do you only see this in laggy sims?

Since target omega is a prim property (like particles/text/etc) I wonder if this is related to the build weirdness you get on laggy sims. Like when you edit something and 5 seconds later it snaps back to the pre-edit values..

Yes, llTargetOmega is seriously borked. I've never gotten any of the stated workarounds to work, and I even just built another spinny-ring thing last night to verify that it was still hosed.

If it is dropped in the root prim, you CAN stop it, but you cannot rotate the prim via script anymore (ie, no more llSetRot()).

If it is dropped into a child prim, you get random effects, but most often, you cannot stop it, llSetRot also doesn't work, and sometimes you can't even start it.

It is a well-known and oft-lamented bug in the simulator. It just fails totally to send updates to the client.

Here is the main JIRA for it: http://jira.secondlife.com/browse/SVC-54

Even though it has an internal assigned case number and Andrew Linden says he is looking into it (as of a few weeks ago, anyway), please vote to get it fixed.

allegory, and whoever wanted an updated test results on it....

i posted my latest test results to the jira entry.

i cant recite them off the top of my head, but they were slightly different than my previous results.

bump...

Have the same problem as the OP, and have voted on this issue:
http://jira.secondlife.com/browse/SVC-54

I also haven't had much joy from the 'workarounds' (they appear to 'work' randomly or intermittently ~ just when you think it's been solved, get thrown another curve!)

...so was wondering if anyone has yet discovered some solution they'd like to share...

XxX

You hit the nail on the head with your statement before your enquiry. There is no solution that works all the time or even most of the time.

Hi, the following work-around seems to work consistently, perhaps it is useful to someone..
Please give any suggestions if there's a better way to do things, i'm not really a programmer..

The end result is:

Omega rotation starts and stops consistently, and upon stopping, the prim resets itself to default rotation (pre-omega).

things to note:

- handle all llTargetOmega calls in a separate script and invoke with link messages.

- to stop omega rotation, just reset this script, since omega rotation seems linked to the script that called it in the first place

- if the script is in a child-prim, it seems necessary to reset the prim rotation also when resetting the omega script

- to set the rotation successfully after an omega call has been made it is necessary to first offset the rotation from the current rotation

- omega script must be deleted and re-copied into any freshly linked child prim, or it wont work.

in this instance, i'm reseting rotation for root prim also but dont think its necessary

<start main script>

// LINK PROTOCOL
integer ROT_START = -100;
integer ROT_STOP = -101;

integer ON = FALSE;


default
{
state_entry()
{
llSetText("",<1.0,1.0,1.0>,1.0);
}

touch_start(integer total_number)
{

if (!ON)
{
llMessageLinked(LINK_THIS,ROT_START,"",NULL_KEY);
}
else
{
llMessageLinked(LINK_THIS,ROT_STOP,"",NULL_KEY);
}
ON = !ON;
}
}

<end main script>

<start omega script>

// LINK PROTOCOL
integer ROT_START = -100;
integer ROT_STOP = -101;

// if set rot to the current rotation after stopping omega rotation, no update of rot occur
// so slightly offset first from intended rotation
setRotPostOmega(rotation rot)
{
if (llGetLinkNumber() > 1)
{
llSetLocalRot(rot + <0.0,0.0,0.1,1.0>;
llSleep(0.1);
llSetLocalRot(rot);
}
else
{
llSetRot(rot + <0.0,0.0,0.1,1.0>;
llSleep(0.1);
llSetRot(rot);
}
}

default
{
state_entry()
{
setRotPostOmega(ZERO_ROTATION);
}

link_message(integer sender_number, integer number, string message, key id)
{
if (number == ROT_START)
{
llTargetOmega(<0.0,0.0,1.0>,1.0,1.0);
}
else if (number == ROT_STOP)
{
llResetScript();
}
}
}

<end omega script>


<EDIT>

one more thing, after taking and rezzing a linked set containing these scripts, the child rotations dont seem to work anymore.. perhaps could make copies of = the omega scripts in the child prims and delete the originals via lsl upon a rez.. will look into this when i have more time