[1.3.0] Kerbal Engineer Redux 1.1.3.0 (2017-05-28)

[cybutek]

@*MajorTom*

This looks to be a known issue as I can see a heat shield in there. Please see my previous post:

On 30/03/2016 at 7:33 PM, cybutek said:

For anyone experiencing issues with no Delta-V being displayed when using heat shields and desperately require a fix, please download and replace your KerbalEngineer.dll with the one available at https://github.com/CYBUTEK/KerbalEngineer/raw/experimentals/Output/KerbalEngineer/KerbalEngineer.dll.

A workaround without downloading would be to right click on the heat shield and set it to be staged. You can then move the stage around like any decoupler to fix when KER expects everything below to be dropped.

I will be pushing out a new KER release soon that includes this fix; hopefully tomorrow.

[Blacks]

Hello
KSP 1.1 build id 01174
KerbalEngineer 1.1.0.0 + manual update KerbalEngineer.dll. from your github address
Here logs with some data what i guess it belong to your mod, no crash, no real problem but i guess more info you get is better
https://drive.google.com/file/d/0B9RK2PFOUhe3ekI4di1za1l3ZHc/view?usp=sharing
Cheer & Salute

[cybutek]

Released version 1.1.0.1 pre-release! Available to download on GitHub

Fixed: Heat shields acting as staged decouplers when not set to be staged.
Fixed: Enabling and disabling of decouplers in staging is now correctly handled.
Fixed: Issue in simulation staging that caused no deltaV above a fairing whilst landed.
Fixed: Fairing mass calculations to work using module mass only (Changed from 1.0.5).

[Jockel]

I have discovered that you can use docking ports as decouplers (there is a right click option to enable staging on docking ports)

KER doesnt notice this there is now a "decoupler" there that sperates stages. It is not so important but i wanted to let you know  Cybutek 

Edited April 1, 2016 by Jockel
Grammar issue

[*MajorTom*]

7 hours ago, cybutek said:

Released version 1.1.0.1 pre-release! Available to download on GitHub

Fixed: Heat shields acting as staged decouplers when not set to be staged.
Fixed: Enabling and disabling of decouplers in staging is now correctly handled.
Fixed: Issue in simulation staging that caused no deltaV above a fairing whilst landed.
Fixed: Fairing mass calculations to work using module mass only (Changed from 1.0.5).

Great job! 

 

[tito13kfm]

Edit: Appears to be the same issue another person has already reported and you are aware of it.

 

There seems to be a slight incompatibility between 1.1.0.1 pre-release and Fuel Tanks Plus by Necrobones or Interstellar Fuel Switch Core by FreeThinker

 

Example pictures:

 

These 2 screenshots are taken of the same craft, in the same save file, without leaving the VAB or editing the craft at all.  It just flips between the 2 values.  I THINK the issue is that it is miscalculating the mass on the higher Delta-V totals as my manual calculations show it much closer to the lower amounts.

Not sure if this is something you would be able to address, but just thought you would like to know.

Edited April 2, 2016 by tito13kfm

[harryyoung]

Hey cybutek, since I send one of them what are the plans for going over the open pull requests on GitHub?

[cybutek]

2 hours ago, harryyoung said:

Hey cybutek, since I send one of them what are the plans for going over the open pull requests on GitHub?

Your feature regarding altitude above underwater terrain has been pulled.  I've merged it into the existing 'Altitude (terrain)' instead of having a separate readout though, as I think that's more intuitive. Hope you don't mind? 

[harryyoung]

That' fine

[petersohn]

What exactly does "Intercept angle" supposed to mean? There is no help for it so I tried to figure it out from the source code, but I still don't understand what that formula represents. I put here the relevant part.

double angle = 180.0 * (1.0 - Math.Pow((originRadius + targetRadius) / (2.0 * targetRadius), 1.5));
angle = PhaseAngle - angle;

[Azimech]

There seems to be a problem with the roll rate readout.

 

I can imagine that value in rad/s, but not in deg/s.

[Padishar]

7 hours ago, Azimech said:

There seems to be a problem with the roll rate readout.

I can imagine that value in rad/s, but not in deg/s.

Well, I can't see anything obviously wrong with the code.  Could it be that the roll angle is changing so much between updates that it is reporting a aliased value (e.g. like a prop when viewed on film appears to run at various wrong speeds and even backwards)?

[Azimech]

1 minute ago, Padishar said:

Well, I can't see anything obviously wrong with the code.  Could it be that the roll angle is changing so much between updates that it is reporting a aliased value (e.g. like a prop when viewed on film appears to run at various wrong speeds and even backwards)?

 I'll compare a 1.0.5 version with V.O.I.D, which gives the readout in rad/s.

[Padishar]

Just now, Azimech said:

 I'll compare a 1.0.5 version with V.O.I.D, which gives the readout in rad/s.

Well, the calculations are done using Quaternion.eulerAngles which returns the results in degrees and then messes about with 180 and 360 so the result is definitely in degrees.

How fast is it actually spinning?  Could it be that the value just isn't being divided by the physics frame time, e.g. would 50 times that number make more sense with what you can measure the speed as?

[Azimech]

22 hours ago, Padishar said:

Well, the calculations are done using Quaternion.eulerAngles which returns the results in degrees and then messes about with 180 and 360 so the result is definitely in degrees.

How fast is it actually spinning?  Could it be that the value just isn't being divided by the physics frame time, e.g. would 50 times that number make more sense with what you can measure the speed as?

Here's a screenshot with V.O.I.D. installed. As you can see, the speed is ~9 rad/s, which is accurate (and evidence of a terrible engine).

[Padishar]

10 minutes ago, Azimech said:

Here's a screenshot with V.O.I.D. installed. As you can see, the speed is ~9 rad/s, which is accurate (and evidence of a terrible engine).

So that's about 1.5 revs per second or ~90 rpm (yes, not the best revving engine ever ).  It does sound suspiciously like the change in roll angle isn't being divided by the timestep, you would need to multiply the KER value by ~26 to get the equivalent value.  I'll take another look at the code and confirm it doesn't factor the timestep in somewhere else.

[Dombi]

I'm running into horrible lag with this new version.

Pretty low specs here but version 1.0.5 was doing great. Now with 1.1 everytime I turn something on in KER I go from 60FPS to 6FPS! =|

Windows 10, Steam 64bit, intel core i7, 4GB ram.

Thanks!

[Odinz]

4 hours ago, Dombi said:

I'm running into horrible lag with this new version.

Pretty low specs here but version 1.0.5 was doing great. Now with 1.1 everytime I turn something on in KER I go from 60FPS to 6FPS! =|

Windows 10, Steam 64bit, intel core i7, 4GB ram.

Thanks!

Do you run the game on a laptop ?

 

[ZobrAA]

@cybutek First of all thank you for this amazing, essential, "musthave" mod!

One idea bothers me for a long time, so I came up here with a suggestion or feature request, if you don't mind.

So, can you add an info string into the vessel section, that indicates VESSEL_NAME of the current craft? Maybe with vessel type icon (from rename craft menu) before the name text.

My rationale for this request is that frequently happening situation when there are a lot of vessels or even wreckage on a close range around you. And you need to quickly switch around and do something with several of them. But in such situation it's hard to tell which vessel (or flag, or wreckage, or whatever) you control at the moment - you have to lose valuable seconds to zoom in or rotate camera around to figure this out...

Also it will be useful for streamers - with VESSEL_NAME info string people can understand what exactly streamer flying at the moment. Even better will be if change font size of VESSEL_NAME separately will be possible, to make this info more noticeable on the screen for audience.

I hope you'll like this idea! Cheers!

[leudaimon]

I'm not sure if I did not find this tool, or if it does not exist... Is there some way to set contract orbits (or even more interesting, a manually inserted orbit) as a target?

This would make finding the correct launch window so much easier! Also, with the manual orbit parameters, it would allow to launch into the ecliptic in RSS, which is far from trivial.

Regards!

[Padishar]

1 hour ago, ZobrAA said:

So, can you add an info string into the vessel section, that indicates VESSEL_NAME of the current craft? Maybe with vessel type icon (from rename craft menu) before the name text.

The vessel name was added as an an optional in-flight readout back in October of last year.  You will need to edit one of your windows and add the readout (it is in the Vessel category).

1 hour ago, leudaimon said:

I'm not sure if I did not find this tool, or if it does not exist... Is there some way to set contract orbits (or even more interesting, a manually inserted orbit) as a target?

I'm pretty sure it doesn't exist but it has been thought of before.  A mod that lets you effectively create a "dummy" vessel with full control over the orbit would be useful for so many things (e.g. general mission planning).  Currently, you would have to actually switch to flying it to create nodes on the orbit so it may need to be a "real" vessel consisting of  a "dummy" custom mod part to avoid issues if you manage to fly close to it (or hit it).

The current best solution is to use hyperedit to stick something (e.g. a probe core + RTG) in the orbit you want, just try not to hit it when you launch...

Edited April 5, 2016 by Padishar

[ZobrAA]

1 hour ago, Padishar said:

The vessel name was added as an an optional in-flight readout back in October of last year.  You will need to edit one of your windows and add the readout (it is in the Vessel category).

OMG! How did I missed this?!! Thx for the tip!

[petersohn]

For that matter, which parameter(s) do I need to watch to find a launch window? To be exact, while landed, I need to know how far off I am to the orbital plane of the target. Any ideas?

[leudaimon]

@Padishar Thank you for the response! Yeah, I tought there wasn't something like that. I don't think it sould be very complicated to add an option in the rendezvous selection that consists of a custom orbit, and would be great for planning, but I'm also aware that this kind of thing that looks really simple tends to be a hell to implement. I don't like the hyperedit approach because looks a little like cheating to me... but I know I'm quite crazy for that...

@petersohn What I usually do is to use the rendezvous option and look at the angle to AN and DN. When one of these is 0º, I launch into the correct inclination (Which is also available in the rendezvous category). It's usually quite accurate (In RSS I tend to get inclinations relative to the moon below 0.5º directly from launch to LEO).

[Cratzz]

The 1.1 release dosen´t work here on build 1174 of KSP. It shows up in SPH and VAB, but not at launch.