[0.25]KSP Interstellar (Magnetic Nozzles, ISRU Revamp) Version 0.13

[JeffreyCor]

Further debug info, when generators stop producing the solar panels are also cut off from supplying power. Debug spawns numerous error entries "[Exception] InvalidOperationException: Collection was modified; enumeration operation may not execute"

Overall, there sure seems to be a lot of bugs on this release compared to 9.2

[mdapol]

Yes, ideally it'd be nice to integrate it with the stock science system such that reports get generated in the typical style. The major problem with this is that generally you're transmitting data from the active vessel back to the R&D centre, while these impactors needn't even be within physics range of an active science instrument.

It's theoretically possible that I could make it so that reports are generated and then picked up by the detectors later, and you then you send that data back. That might be a nice avenue for expansion but, that would leave you in the position of having to click back to your detectors after each impact event, record the data and send it back to Kerbin - this might make it prove to be just an irritating feature rather than a worthwhile intregration step. I'm not really sure at this stage.

In any case, certainly adding thousands of science to the game without taking any away would be a really bad idea. The Interstellar tech nodes cost thousands in themselves and it's really not right if you can advance through all of these super advanced technologies in a matter of mere moments.

You could require a vessel with the instrument to be focussed during the impact for data to be gathered, then just use the stock science system from there. Also, whatever "magic" calculates how many other accelerometers are in range could still work and just update the one vessel with the science points.

[Fractal_UK]

Further debug info, when generators stop producing the solar panels are also cut off from supplying power. Debug spawns numerous error entries "[Exception] InvalidOperationException: Collection was modified; enumeration operation may not execute"

Overall, there sure seems to be a lot of bugs on this release compared to 9.2

I don't know what you mean, when generators stop producing when exactly?

Nothing significant with generators has even changed...

[Shad0wCatcher]

Check page 611 Fractal. Second post down. His first report. Can't blame ya, this thread has gone plaid it seems with the release

[Fractal_UK]

Check page 611 Fractal. Second post down. His first report. Can't blame ya, this thread has gone plaid it seems with the release

Ah okay, thanks.

I still don't undestand though, solar panels and generators don't provide compatible power supplies, there is no connection between the two whatsoever. If the generator has no power then the megajoules bar will deplete and resource demand will go to maximum.

[mdapol]

Adding loads of new science is pointless, there is already more than enough science in the game so simply adding lots more of it is not a productive use of anyone's time. Especially when getting late technology is supposed to actually pose some kind of a challenge.

The impactor science might be a little low at the moment and need to be tweaked upward somewhat, I'm already coming to that conclusion. That said, you seem to be assuming where you once got X science, you now get zero, which is not even close to true.

That's hardly unprecendented, the stock instruments don't tell you what return you're going to get either. Ultimately, the answer is: it depends.

Why would you assume there are lots of restrictions on how long probes can run or the number of impacts? Why would you think you would have to have remote tech, a different mod? That honestly makes no sense and suggests that you haven't even run one single impact experiment (which you could do in 30 seconds on the launch pad). I'm quite happy to take feedback/criticism but it'd be kind of nice to think it was actually based on some actual experience of the new system.

The amount of science you get depends on the number of detectors you have and their position (more detectors will give you a bit more science beyond what one will give you) and the amount of science will diminish with each impact but will never reach zero. This is hardly an unprecended mechanic and it's not particularly complicated.

Ultimately, one of the biggest complaints made about the stock science system is that it doesn't involve any actual experimentation - you go somewhere, you click a few times, you fly back. The accelerometer is a good choice for alteration because the most effective way to produce seismic activity on a geologically dead world is to produce it yourself, so you know the magnitude of the impact that caused it and can actually produce meaningful data.

With this system, you can land detectors on each body, you can use multiple detectors if you like (best positioned on opposite sides of the body) and then send off special impactor missions or use discarded stages to produce science as you travel.

If you really want the stock functionality back as well, you can just delete

!MODULE[ModuleScienceExperiment]
{
}

in GameData/WarpPlugin/Science.cfg.

Just crash a ship into the ground, it doesn't need anything special on it.

Thanks, I missed a changed as I was switching everything over to the new resource draw method.

Apparently a last minute change designed to make a small tweak to the arcjet thruster caused this bug, I think I have a fix now.

I would like to echo the concerns of this poster. I am not a fan of changing the stock science system either. I just got home and haven't tried it yet, but the new Impactor instrument (if you're not going to create a new science instrument; which would be the best solution) should have the stock behavior included. So using it without an impactor experiment gives the stock Accelerometer results.

Also, I really do not like how both the new Accelerometer and the Science Lab just "Magically" update the science center, either. Keep in mind that for some players, it's not just about how many science points we get; it's _how_ we get them. If we want to limit ourselves to using RemoteTech, we would like your mod to honor that. Try to keep within the standard science system as much as possible. There's no reason that both the Science Lab and the Accelerometer could be made to work this way, and we would greatly appreciate it if it did. Consider that you added the Engine Overheating/Intercoolers to the game because "realistically" you need to cool down air intakes at hypersonic velocities. Well, RemoteTech is also trying to be "realistic" and we would like your parts to play nice.

The think with the Intercoolers also is a knock against your mod, in my opinion. While I do like the realism it adds, it also damages my enjoyment of the game because you didn't take into account non-stock intakes. Many of us use the B9 pack and I for one like the look of the Surface-Mounted hypersonic intakes. There should be a way to use those without the problem of overheating. After all, they are supposedly _designed_ for use at hypersonic velocities (it even says so in the description).

Does someone know if there is a hack to perhaps add the Intercooler function to those parts somewhere in the config file? Or some other workaround? (And to those who are about to say, "You can still use those intakes to make spaceplanes, you just have to be careful", etc. I don't _want_ to be careful, I want to use them the same way I can use the Ram Intake. I don't want to have to worry about parts overheating.

I really do appreciate your mod Fractal and I think it adds a lot to the game. Just please do not make me have to choose between your mod and the stock game.

[Shad0wCatcher]

Going to try adding the precooler module to those intakes really quickly mdapol. I'll get back with you to see if it causes a ton of weirdness. I know your pain implicitly. I ended up commenting out the heating component of the SABREs in 0.9.2 due to it. The very large rectangular intakes aren't awful since you can slap precoolers onto the rear attach node; but not so much anything else.

[JeffreyCor]

Ah okay, thanks.

I still don't undestand though, solar panels and generators don't provide compatible power supplies, there is no connection between the two whatsoever. If the generator has no power then the megajoules bar will deplete and resource demand will go to maximum.

They go provode compatible power supplies as the electric generator produces electric charge before producing megajoules. When the generators fail from power exhaustion or shutting down (as for testing), when time rate returns to normal electric charge will not rebuild despite active solar panels. This does not seem to be a problem is a generator is not present on the vessel.

Trying to put in a screen capture to illustrate the situation (a picture's worth a thousand words as they say) First time trying this so keeping fingers crossed As it shows, no power from gen, solar power supplying power, yet electric charge remains 0 with no gain.

Oh forgot to mention, there also is 123 MJ staying stored as well, not being released back into EC. It's like something in the system is frozen or looped.

Edited February 4, 2014 by JeffreyCor

[Fractal_UK]

Also, I really do not like how both the new Accelerometer and the Science Lab just "Magically" update the science center, either. Keep in mind that for some players, it's not just about how many science points we get; it's _how_ we get them. If we want to limit ourselves to using RemoteTech, we would like your mod to honor that. Try to keep within the standard science system as much as possible. There's no reason that both the Science Lab and the Accelerometer could be made to work this way, and we would greatly appreciate it if it did. Consider that you added the Engine Overheating/Intercoolers to the game because "realistically" you need to cool down air intakes at hypersonic velocities. Well, RemoteTech is also trying to be "realistic" and we would like your parts to play nice.

I certainly take this point and I am interested in looking into switching over to this approach in the future.

The think with the Intercoolers also is a knock against your mod, in my opinion. While I do like the realism it adds, it also damages my enjoyment of the game because you didn't take into account non-stock intakes. Many of us use the B9 pack and I for one like the look of the Surface-Mounted hypersonic intakes. There should be a way to use those without the problem of overheating. After all, they are supposedly _designed_ for use at hypersonic velocities (it even says so in the description).

Does someone know if there is a hack to perhaps add the Intercooler function to those parts somewhere in the config file? Or some other workaround? (And to those who are about to say, "You can still use those intakes to make spaceplanes, you just have to be careful", etc. I don't _want_ to be careful, I want to use them the same way I can use the Ram Intake. I don't want to have to worry about parts overheating.

This is not correct at all though, I did very much consider radial intakes in the design process, I considered supporting them for a while but ultimately I think their inability to be precooled is an appropriate design trade-off. There is a big difference between a hypersonic intake and an essentially orbital velocity intake. On Earth, there is a big difference between Mach 5 and orbital velocity, on Kerbin, there isn't. Having no precooling requirement up to velocities where they would be required in the real world already makes things massively easier than they should be.

The only way to remove the need for precooling in atmosphere is to have the combustion take place at supersonic velocity so you don't get all the friction heating from slowing down the airflow to subsonic velocities inside your intake. That would require a totally different engine design, no matter whether your intake is suited to hypersonic airflow or not.

Additionally, this approach actually has meaningful tradeoffs. You can either have the convenience of radial intakes and have an easy time sticking them on your craft anywhere you like, or you design around having integrated precooled intake stacks. The first approach results in you having to pack a little more rocket fuel and the second results in an engineering problem. Neither are particularly challenging but at least it adds a bit more depth to SSTO aircraft design rather than simply flying up to orbital velocity inside the atmosphere and spending a piddly few m/s of delta-v to circularise.

Adding the precoolers to radial intakes would mean it would have absolutely no impact on difficulty and might as well be removed entirely.

They go provode compatible power supplies as the electric generator produces electric charge before producing megajoules. When the generators fail from power exhaustion or shutting down (as for testing), when time rate returns to normal electric charge will not rebuild despite active solar panels. This does not seem to be a problem is a generator is not present on the vessel.

What type of reactor are you using? If it's a fusion reactor, it'll eat the electric charge trying to turn the reactor back on unless you actually turn it off.

Edited February 4, 2014 by Fractal_UK

[Termaul]

I just went ahead and removed the over heating module for now till some better intakes came out or pre coolers. I like the idea for pre-coolers just dont like how it limits my builds haha.

Edited February 4, 2014 by Termaul

[Fractal_UK]

I just went ahead and removed the over heating module for now till some better intakes came out or pre coolers. I like the idea for pre-coolers just dont like how it limits my builds haha.

To be honest, that's a good approach. It's a very easy feature to disable if it doesn't suit you.

[sidfu]

nothing wrong with chaning stock sceince. just have it in a way that people who dont like can .cfg change it out if they dont want.

also another way to do the impact science is to have when it records the impact it stores a science report. right now the only problem i see with the way the impact sceinse works is that u have to be in control of the vech that crashes into the ground. right now debris that crashes dont count.

[JeffreyCor]

Been doing more testing, it seems the solar panels are being completely bypassed at all times if there is an electric generator present on the vessel. Opening and closing panels used to cause the demand to change, I would assume because the demand is lower because of a different source of power. Now, demand remains the same regardless of panel generation. I can't recall for sure if 10.0 did this, or if this started with 10.1. Would be more than happy to try the previous .dll from the other versions to see if there is a difference if that would be beneficial.

[Fractal_UK]

Been doing more testing, it seems the solar panels are being completely bypassed at all times if there is an electric generator present on the vessel. Opening and closing panels used to cause the demand to change, I would assume because the demand is lower because of a different source of power. Now, demand remains the same regardless of panel generation. I can't recall for sure if 10.0 did this, or if this started with 10.1. Would be more than happy to try the previous .dll from the other versions to see if there is a difference if that would be beneficial.

Nothing on the power management display has anything to do with solar panels, solar panels will not appear there, nor will they create either supply or demand on that screen.

[Shad0wCatcher]

@Fractal - The biggest reason I can see for precooling the B9 radials is the double-hit to heat production you get when using deadly re-entry and the heating module from this. (Obviously a user-end problem at this point due to the two mods conflicting in this regard) It becomes entirely too much heat production at sub-orbital velocities causing massive unplanned disassembly. I do empathize your position entirely however in the stock game in that SSTOs are ludicrously easy to build (even prior to the RAPIER).

[Lightwarrior]

I think he was talking about DC power system usage displayed there.

It actually reduce when some source of power which produce EC directly is added, e.g. engine wich produce EC is throttled up or solar panel extended.

p.s. also i cannot reproduce this problem, solar panels work fine.

Edited February 4, 2014 by Lightwarrior

[Makeone]

Any others that have build some sort of of microwave based power system and are noticing memory leak (of sort)? Once my mods (which aren't many) have loaded, i have a stable system, but when i switch to my current career save where i have microwave network running, memory starts to fill in until...well, it crashes. For example, i can't test my ships outside sph/vab whitout a crash if i try to revert back. On some crashes i have plenty of memory left (of the 4 GB mark, that is). My save used to be very stable, even for longer periods of time, but after i started to play with mircowaves, game crashes atleast once an hour.

Also, as technically a fusion reactor is able to fit both charged particles and thermal generators, but does the plugin actually support this? I tried that on 0.10, and my mobile power unit just exploded. There is also something wrong with microwave transceivers that when activated, they seem to demand much more power from the power unit than it can possibly produce, in case of a 3.75m fusion reactor/twin generator combo it wanted 20GW out of it while its rated maxium is 13 or so GW.

[mdapol]

I certainly take this point and I am interested in looking into switching over to this approach in the future.

This is not correct at all though, I did very much consider radial intakes in the design process, I considered supporting them for a while but ultimately I think their inability to be precooled is an appropriate design trade-off. There is a big difference between a hypersonic intake and an essentially orbital velocity intake. On Earth, there is a big difference between Mach 5 and orbital velocity, on Kerbin, there isn't. Having no precooling requirement up to velocities where they would be required in the real world already makes things massively easier than they should be.

The only way to remove the need for precooling in atmosphere is to have the combustion take place at supersonic velocity so you don't get all the friction heating from slowing down the airflow to subsonic velocities inside your intake. That would require a totally different engine design, no matter whether your intake is suited to hypersonic airflow or not.

I guess what I'm after is something more like a scramjet, but since there is none at the moment I have to make do with the B9 Radial intakes to at least make it "look" like one. I see where you're coming from but I also don't want to be limited to using ugly stock intakes that require an inline design approach. The B9 parts increase the design possibilities. I look forward to Shad0wCatcher's experiments.

[Lightwarrior]

...

Also, as technically a fusion reactor is able to fit both charged particles and thermal generators, but does the plugin actually support this? I tried that on 0.10, and my mobile power unit just exploded. There is also something wrong with microwave transceivers that when activated, they seem to demand much more power from the power unit than it can possibly produce, in case of a 3.75m fusion reactor/twin generator combo it wanted 20GW out of it while its rated maxium is 13 or so GW.

It seems that now fusion reactors can produce all rated power as thermal power, or some power as charged power and some as thermal power within specified limit. There seem to be no definite values of thermal/charged power production, as it was in 0.9. So generators have maximum power of maximum reactor power for thermal and maximum charged power for direct conversion generator, which is more than 100% obviously.

And, again, if i understand correctly, transmitter calculate power available based on sum of all generators maximum power, which in this case will lead to predictable result of requiring more than 100% of power and running out of power

[JeffreyCor]

Nothing on the power management display has anything to do with solar panels, solar panels will not appear there, nor will they create either supply or demand on that screen.

Strange, I thought I read on the Wiki that 1EC/s was 1kW, which given 4 1x6 panels exposed to sun would then produce a little over 3kW, much greater than the .01kW the power manager display goes to and would think it would be on DC Electrical System and such supply would offset the demand from the generator. At any rate, I can only tell what the symptoms are as I have no way of knowing how things actually work internally.

[Fractal_UK]

I guess what I'm after is something more like a scramjet, but since there is none at the moment I have to make do with the B9 Radial intakes to at least make it "look" like one. I see where you're coming from but I also don't want to be limited to using ugly stock intakes that require an inline design approach. The B9 parts increase the design possibilities. I look forward to Shad0wCatcher's experiments.

You can use any stock or modded stackable intakes, the b9 sabre intakes are the most appropriate.

[JeffreyCor]

I think he was talking about DC power system usage displayed there.

It actually reduce when some source of power which produce EC directly is added, e.g. engine wich produce EC is throttled up or solar panel extended.

p.s. also i cannot reproduce this problem, solar panels work fine.

Try making a vessel with both a generator and panels and get up over 800km. Extend the panels and turn off the generator. Set max time compression and wait a a week or two (game time of course ) then return to normal time. The EC should flux a few times during acceleration then fall to 0 and stay there.

Also, are you seeing the the DC system change with panel exposure? I tried it with 6 panels to give a significant amount of EC power without any change. If you're not getting these symptoms it sounds like either a corrupt download on my end or a conflict with some other mod that just hit with the .10 release.

[Lightwarrior]

Try making a vessel with both a generator and panels and get up over 800km. Extend the panels and turn off the generator. Set max time compression and wait a a week or two (game time of course ) then return to normal time. The EC should flux a few times during acceleration then fall to 0 and stay there.

Also, are you seeing the the DC system change with panel exposure? I tried it with 6 panels to give a significant amount of EC power without any change. If you're not getting these symptoms it sounds like either a corrupt download on my end or a conflict with some other mod that just hit with the .10 release.

I will try this later, for now i just selected random vessel which stayed in LKO for quite some time with reactor (but not generator) offline and tried to disable generator and retract/extend solar panels. There was no DC system change if generator was enabled, and if generator was disabled it just displayed dc system usage =0 and retracting/extending panels changed EC consumption/production correctly.

[JeffreyCor]

I will try this later, for now i just selected random vessel which stayed in LKO for quite some time with reactor (but not generator) offline and tried to disable generator and retract/extend solar panels. There was no DC system change if generator was enabled, and if generator was disabled it just displayed dc system usage =0 and retracting/extending panels changed EC consumption/production correctly.

I appreciate your assistance in testing! I am thinking the problem is linked to the time warp causing something to break. It does seem to work correctly in normal time and even high warp for relatively short durations, which is why I first noticed it when sending a probe to Duna after the update. Thanks again!

[Lightwarrior]

I've made simple test craft in sandbox mode and sent it to kerbol orbit:

...only to find that old nasty antimatter storage problem is still there, and while 2 gigantors produce enough power for antimatter containment i cannot use warp higher than 1000x because EC will instantly drop to zero and warp will be reduced to 50x with warning that antimatter storage is going to explode.

So i dumped all antimatter and tried again with only probe core consuming power. It worked fine. EC stayed full during timewarp and panels worked after it.

p.s. also antimatter storage problem seem to be caused by stock resource system (basically you need way more power stored then consumed, even if power production is higher then consumption), may be you had the same problem? The only workaround now is to either use lots of batteries or power things like antimatter storage by megajoules, not EC.

Edited February 4, 2014 by Lightwarrior