Jump to content

[1.12.x] Kerbal Alarm Clock v3.13.0.0 (April 10)


TriggerAu

Recommended Posts

I have strange issue with alarm window.

I created 6 alarms but window shows only one alarm (soonest) and no scrollbar.

Alarms...txt contains 6 of them. In config alarms count before show scrollbar is set to 10.

(can't find where to attach screenshot... there is nothing to see at)

Link to comment
Share on other sites

I have strange issue with alarm window.

I created 6 alarms but window shows only one alarm (soonest) and no scrollbar.

Alarms...txt contains 6 of them. In config alarms count before show scrollbar is set to 10.

(can't find where to attach screenshot... there is nothing to see at)

Hi Kev, Simplest explanation might be that the interface is in minimized mode. Clicking the Up/Down Chevron toggles the list between normal and minimal (3rd button from the right - you can see it changing in the below shots). If thats not the problem then either post a link or PM me a link to an alarm file and I'll have a look.

AlarmList_Minimized.pngAlarmList.png

To post images you need to upload them to a hosting site and then link them in the post - imgur.com is pretty common. Similarly for logs/files you can use pastebin.com or similar

Link to comment
Share on other sites

Yep, i found it already, window was minimized.

Wanted to edit my post but it still was under inspection. :)

Anyway, plugin is really great. Very helpful in managing multiple flights, giving posibility to run tens of them.

Thanks.

Oh, by the way, sometimes alarms don't been deleted after they alarming and i choose to delete them after closing. And after switching to linked vessel i always have another pop of this alarm.

For hyperbolic orbit (when vessel just entered SOI) time to apoapsis shown earlier than time to periapsis (somewhere in the middle between vessel and periapsis). Probably apoapsis should not exists in this circumstances?

Also i noticed that AN and DN times calculated for celestial body equator (seems like that) rather than for orbit of targeted vehicle. Is that intended behavior?

Thats my observations so far. I apologize for my english. ;)

Link to comment
Share on other sites

Yep, i found it already, window was minimized.

Wanted to edit my post but it still was under inspection. :)

Anyway, plugin is really great. Very helpful in managing multiple flights, giving posibility to run tens of them.

Thanks.

Oh, by the way, sometimes alarms don't been deleted after they alarming and i choose to delete them after closing. And after switching to linked vessel i always have another pop of this alarm.

For hyperbolic orbit (when vessel just entered SOI) time to apoapsis shown earlier than time to periapsis (somewhere in the middle between vessel and periapsis). Probably apoapsis should not exists in this circumstances?

Also i noticed that AN and DN times calculated for celestial body equator (seems like that) rather than for orbit of targeted vehicle. Is that intended behavior?

Thats my observations so far. I apologize for my english. ;)

Good to hear Kev, thanks for the feedback, I'll look into these tonight - For the AN/DN these should match the nodes displayed on the flight path.

Link to comment
Share on other sites

Oh, by the way, sometimes alarms don't been deleted after they alarming and i choose to delete them after closing. And after switching to linked vessel i always have another pop of this alarm.

For hyperbolic orbit (when vessel just entered SOI) time to apoapsis shown earlier than time to periapsis (somewhere in the middle between vessel and periapsis). Probably apoapsis should not exists in this circumstances?

Also i noticed that AN and DN times calculated for celestial body equator (seems like that) rather than for orbit of targeted vehicle. Is that intended behavior?

Some feedback here:

  • Found the reason the alarms pop back up - fixed teh code and will be in next release
  • The Ap/Pe stuff looks to be working from my testing - technically its using some internal timeToAp and timeToPe variables. Can change SOI and see only the Pe on hyperbolic, and once the Pe is passed then no Ap/Pe.
  • The AN/DN stuff needs more looking - its definately not the equatorial AN/DN as it changes when you select targets (thats on my list of things to try an work out), that said in some of my tests it's not pointing at the AN/DN markers on the plan either though (on some very weird targets I might add). I'll need to review that piece of code more

If you can drop me a pic of the Ap/Pe stuff your encountering I can try and work on that some more

Thanks

Link to comment
Share on other sites

...

If you can drop me a pic of the Ap/Pe stuff your encountering I can try and work on that some more

...

Sure, there is one:

XEfcWdc.png

Now how to reproduce:

- Have a ship on mun orbit;

- Launch another ship to the mun interception and setup SOI change alarm - 5 sec;

- Switch to first ship that is orbiting mun and wait for alarm from here;

- When alarm is fired, switch to ship w/ delete alarm. After switch i have another pop of alarm, i close it w/ delete checked;

- Add alarm and i see apoapsis with some random "time to". Time to periapsis is ok.

Well, if you say its just internal vars (of KSP i assume) then bug might be a bit deeper. I noticed that apoapsis is gone in some seconds, maybe 10 or 20, but this is a bit confusing. Last time i accidentally added alarm for apoapsis instead of periapsis. Luckily i did not miss my periapsis because apo was 13 minutes earlier. :) In screenshot apo is about 20 minutes later so i could miss needed time mark this time.

Link to comment
Share on other sites

Sure, there is one:

Now how to reproduce:

- Have a ship on mun orbit;

- Launch another ship to the mun interception and setup SOI change alarm - 5 sec;

- Switch to first ship that is orbiting mun and wait for alarm from here;

- When alarm is fired, switch to ship w/ delete alarm. After switch i have another pop of alarm, i close it w/ delete checked;

- Add alarm and i see apoapsis with some random "time to". Time to periapsis is ok.

Well, if you say its just internal vars (of KSP i assume) then bug might be a bit deeper. I noticed that apoapsis is gone in some seconds, maybe 10 or 20, but this is a bit confusing. Last time i accidentally added alarm for apoapsis instead of periapsis. Luckily i did not miss my periapsis because apo was 13 minutes earlier. :) In screenshot apo is about 20 minutes later so i could miss needed time mark this time.

Thanks Kev, Will try these steps, and see what I can do. I did notice that if you are under high warp and have the Add Ap/Pe Alarm up then it can sometimes jitter as well, so I am wondering if there is some physics thing in the background affecting those timeTo values - this sometimes can also affect ships on switches/SOI changes. I'll repeat your steps and see how I go, all my testing last night I stayed with the ship on the SOI change and monitored it.

PS. and the delete stuff is fixed in the next release

Edited by TriggerAu
added delete stuff
Link to comment
Share on other sites

Some feedback here:

  • The AN/DN stuff needs more looking - its definately not the equatorial AN/DN as it changes when you select targets (thats on my list of things to try an work out), that said in some of my tests it's not pointing at the AN/DN markers on the plan either though (on some very weird targets I might add). I'll need to review that piece of code more

Thanks

It was driving me nuts why I can't align my satellites in the same polar orbit until I noticed that your AN/DN is almost always off (according to MechJeb). I appears to change with the inclination to the orbited body (not the targeted) so the difference on equatorial orbits are only a few seconds but I got here one with 8 minutes to late. Btw it always appears to be to late:

0ehKiCf.png

I'm just throwing my thoughts about this issue in, so it may spark an idea on your side ;)

Ah and a feature request: After passing a transfer window create a new one directly (but don't delete the old one)?

Fabian

Link to comment
Share on other sites

It was driving me nuts why I can't align my satellites in the same polar orbit until I noticed that your AN/DN is almost always off (according to MechJeb). I appears to change with the inclination to the orbited body (not the targeted) so the difference on equatorial orbits are only a few seconds but I got here one with 8 minutes to late. Btw it always appears to be to late:

I'm just throwing my thoughts about this issue in, so it may spark an idea on your side ;)

Ah and a feature request: After passing a transfer window create a new one directly (but don't delete the old one)?

Fabian

Thanks for the extra info. I did notice some stuff on this yesterday when doing the same thing, I am using some of the code from Kerbal Engineer to do the AN work, so my values match those ones - but they are not always correct - and the MechJeb ones are different. There is some code in the MechJeb AN detection that swaps the Y and Z axis on the vectors, that I don't have in the functions. I'm gonna have a go at working through that when I get some time.

Link to comment
Share on other sites

Also want to ask is there way to fix wordwraping issue in the alarms window?

Changing width values at config file seems doesn't do anything...

I have that one on the list of things to work on, just a bit time poor at the moment.

Was thinking of changing it so it simply truncates the text at a point with some "..." so each alarm is always one line only

Link to comment
Share on other sites

I seams to have run into a bug with this plugin (Like it a lot otherwise) I first posted in this forum thread thinking it was a stock bug but I can only reproduce with kerbal alarm clock.

What happens, when having a satellite in a circular orbit and adding an alarm for periphrasis and hitting maximum time warp the satellites orbit has changed when the alarm triggers.

From an 1Mm by 1Mm orbit with Pe 3/4 th orbit ahead of the satellite hitting maximum timewarp changes that to an aproximatlly 1.3Mm by 0.7Mm orbit.

Steps I took to reproduce

* place 8t satelite in 1060000m by 1028000m orbit

* place an alarm for Pe that is 3/4 of an orbit ahead (I used a pause alarm)

* hit maximum timewarp by clicking at the top of the screen

* the alarm triggers and the orbit have now changed

I can provide more info once I know what is needed.

Thanks for an excellent plugin

Link to comment
Share on other sites

Hello, I also found a glitch in the closest approach implementation. When I searched for 11 orbits, it found an approach after one orbit while with 10 and 12 tested orbits it only found one after 10 orbits.

The line wraping is also strange. When the seconds of the time to closest approach or alarm where below 10 and consists of only one digit it didn't wrap them.

aj8Mihq.png

By the way is it possible to show minutes and seconds always with a leading 0?

Fabian

Edited by xZise
direct links
Link to comment
Share on other sites

Wow, its a bug storm :wink:

I seams to have run into a bug with this plugin (Like it a lot otherwise) I first posted in this forum thread thinking it was a stock bug but I can only reproduce with kerbal alarm clock.

I'll set up a reproduction and see what I can do. I can tell you that the alarm looks ahead to see how far to the event time and then brings the warp down in stages, but if the warp is high enough and it manages to pass the time it simply stops the warp immediately and like Sal said this can cause issue like this. Will have a play today, maybe I can remove the instant changes and that will fix it.

Hello, I also found a glitch in the closest approach implementation. When I searched for 11 orbits, it found an approach after one orbit while with 10 and 12 tested orbits it only found one after 10 orbits.

The line wrapping is also strange. When the seconds of the time to closest approach or alarm where below 10 and consists of only one digit it didn't wrap them.

By the way is it possible to show minutes and seconds always with a leading 0?

  • The closest approach thing is probably rounding approximation with the period of time and the closest point actually being really close to the end/start of an orbit. Will have a look at it.
  • Yep that line wrapping is just wrong - will get that one resolved.
  • For the time, there is a time display that shows leading zero's (like 08:08:07) if you click on the clock in the main window you can cycle to it (or through the settings - called Normal Time)

Link to comment
Share on other sites

I seams to have run into a bug with this plugin (Like it a lot otherwise) I first posted in this forum thread thinking it was a stock bug but I can only reproduce with kerbal alarm clock.

OK, so got to the bottom of this one. under the hood the alarm clock is checking every 0.25 of a second to see how far off each alarm is and then reducing the warp as needed. If it passes the alarm time at speed then it simply cuts the warp immediately, and this action is the one that can cause orbit changes that you are seeing. You can cause this behaviour with any kind of alarm (e.g. set a raw alarm in 10 mins, then click on the maximum warp), this winds up the warp to 1.1 days per sec. So if the alarm is within the first interval (so 8 hours) there is a chance that this will happen. If its a 10 mins alarm then it will happen 47 times out of 48.

You don't strike this problem using the keys to adjust warp as you can't get the warp high enough, quick enough before the clock starts reducing it.

I think what I'll do is reduce the interval to 0.1 initially (is set on an interval to limit processor impact) - which makes that first interval 2.7 hours, and also find a way to make it user configurable - either in game or via the config file so people can tweak it.

Will have this change in the next release so you can try and test it.

Link to comment
Share on other sites

Hello, I also found a glitch in the closest approach implementation. When I searched for 11 orbits, it found an approach after one orbit while with 10 and 12 tested orbits it only found one after 10 orbits.

The line wraping is also strange. When the seconds of the time to closest approach or alarm where below 10 and consists of only one digit it didn't wrap them.

By the way is it possible to show minutes and seconds always with a leading 0?

Found the first one and changed the way the multi orbits works so is done, also foudn the line wrapping issue in the add window. Will be in next release.

Link to comment
Share on other sites

Firstly my apologies for the slowness of this release. I have a few other things on and have been unable to get a chunk of time on this. There are still a few other fixes in the works, but they might be a little while.

v2.4.0.1 is now on Spaceport

Lots of fixes and a few tweaks

  • Changed AN/DN code to use Mechjeb style ones to correct for my lack of math skills
  • Changed AN/DN Add so if there is no target it displays the equatorial AN/DN times
  • Adjusted Future orbits code to fix rounding issues
  • Added capacity to edit the frequency of checks - people can now tweak this for their system and helps prevent severe warp changes that can affect vesel orbits
  • Adjusted delete alarm on close code so alarm did not reappear
  • Fixed time display setting not persisting
  • Sorted out Earth alarms not updating when paused
  • Fixed Line wrapping on Add Alarm window (still need to work out the right answer for the main window)
  • Some other Graphical tweaks

To set an Equatorial AN/DN alarm, simply go to create an AN/DN alarm and have no target selected

So I added some buttons in the settings to let you set how many times a second the alarm clock will check for approaching alarms. I didn't simply increase it to a high figure as people can monitor any affect the plugin has on processing at the different levels. If you want something custom, then you add a line to the config.xml file as per below and it will be the value behind the Custom Button. This is to help with scenarios where people are close to an alarm and go directly to the highest warp level

<int name="BehaviourChecksPerSecCustom">37</int>

Link to comment
Share on other sites

Is there a way to make a specific maneuver node? I'm using KSPTOT for weird transfers, which gives me a time and prograde/radial/normal delta-v. I'd like to be able to enter it into a maneuver node at the specified time.

Not currently sorry, You could use the Maneuver Node Helper plugin (pretty sure I have used it in 0.20) http://kerbalspaceprogram.com/maneuver-node-improvement/. That will let you set those values directly, and you could maybe use the Alarm Clock to help position the node at the specific time you are after?

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...