PyroSA

Reported Version: v0.1.5 (latest) | Mods: none | Can replicate without mods? Yes OS: Win 11 | CPU: AMD 5600X | GPU: NVidia 3060ti | RAM: 32GB (8x4) DDR4 (MK3 to Large Fuel tank) Fresh launch after 1.5.0 update, new game. Uninstalled mods after update, should not affect anything. There's one between the cockpit and Hydrogen, and one in hand just below the cockpit. Included Attachments: .ipsImage { width: 900px !important; }

This seems to work if I set my game to 1920x1080 -> My Screen and windows is at 2560x1440p. If I run it at more than 1920x1080 the text gets smaller every time I mouse over the next thing. If I run it at lower that 1920x1080 it actually get larger, and soon enough the game locks up. Oddly enough this does not seem to affect all items, but nothing except a game restart fixes it. Since I don't have these numbers memorized, I've pretty much stopped building anything interesting and am awaiting the next patch.

When loading with 1.5.0 it doesn't seem to add the mods menu. Pretty much anything that should show on UI does not. Content still loads however. Log contains a whole lot of: [Error :Space Warp] System.NullReferenceException: Object reference not set to an instance of an object at SpaceWarp.Patching.LoadingActions.InitializeModAction.DoAction (System.Action resolve, System.Action`1[T] reject) [0x0000d] in <fa52a487673442778cd59f22d8d81961>:0 [Error :Space Warp] System.NullReferenceException: Object reference not set to an instance of an object at SpaceWarp.Patching.LoadingActions.PostInitializeModAction.DoAction (System.Action resolve, System.Action`1[T] reject) [0x0000d] in <fa52a487673442778cd59f22d8d81961>:0 There's no longer a Mods menu item on the main screen, and I don't see any toolbars. [Info :ToolbarBackend] App bar creation event started: Flight [Info :ToolbarBackend] App bar creation event started: KSC [Info :ToolbarBackend] App bar creation event started: OAB

This man went full Soviet, UI language

What the heck is the right scale for this? Medium is 2.5 diameter right? That would mean this is roughly as big as 4 X200-64s stacked on top of one another and larger than that in diameter (2.5x3+change < 10m). So this was my quick take: More images: Imgur: The magic of the Internet First image was my first take - way overpowered and overfueled - I decoupled just before reaching the target orbit. Rest was from the final launch. Dumped more than half the fuel on the boosters, and 1/4 of the fuel in the centre. Switched to weaker engines. Came in at around 260t and reached orbit with 1500dV remaining. Launched the battery and antennas - roughly 150kg and kinda ball-shaped... No control module, so didn't take a nicer picture of it.

I honestly wish they implemented it like in DeltaV (Rings of Saturn). Those seamlessly merge the thrusters and reaction wheels (which can saturate). Ideally the reaction wheel should be used first (small movements) and then the thrusters only engage for larger movements (or to negate saturation). The SAS seems to go for max control much too often even when finer movements would suffice. It would be nice if we could somehow tweak these PID values just like you can tweak your suspension. Perhaps do it per control unit. Or allow it to be set with action groups so you can toggle between different stability profiles depending on the situation.

The SAS is not well tuned in my opinion. Obliterates RCS reserves or goes in to a rapid correction mode on aero. I don't know if it got worse in 0.1.4, as I had several issues before as well.

Pictures: https://imgur.com/gallery/0Z6oh53 - I didn't get around to editing any video. This was quite fun, the plane turned out to be super stable even at 4x, so I decided to go for that river in the distance - still don't know if it was the right one... Managed to make the entire trek over there and almost back. If my fuel tanks didn't decide to go all wobbly after the canyon, I would've had another tonne of fuel, which should have been plenty to land. Quite a few pretty sights during the flight, but the one that surprised me most was those DENSE clouds just north of the desert. Sure it shows on the satellite, but I never actually flew through them! Wasn't sure if the river actually ends under them, but managed to find something. Then after finishing the river trip, I had no idea how I'd get home. First trekked out over the ocean before deciding to gun it towards the north pole. It was pretty obvious crossing over it wouldn't be the shortest route, but as you get closer I had no idea what direction I should be flying. After the drop tank mishap the frame was a bit bent so the plane yawed to the left. Normally I'd just correct and fly toward a heading, but once the poles got closer the actual heading changed even while going straight... As the crow doesn't fly, I did manage to get home though. With tanks, approaching the canyon: Without tanks, heading home: Didn't make it... but close: A second mission was much more successful. I headed to the opposite side of the river starting from the ocean. The hairiest part was trying to navigate at night - since the flight started later. I did at least one backtrack when I took the wrong branch on the river. The map was too dark too see... Most of the flight was completed with only external tanks, and I easily made it back with plenty (I packed more) to spare. Even burned 60% throttle much of the way.

That balancing is indeed crucial. I'll post my plane shortly - it was specifically made with this in mind - the COM shifts very little between empty and full, and a little bit of trim kept it level.

I've tried building a few of those. Without the software that drives the fly-by-wire, it's really hard to get them controlled, and without a tail they slip all over. A possible option might be to pop some form of brake on the left and right that can be toggled separately, but that certainly won't give fine-grained control.

I actually managed to fly the length of the river, but due to a wobbly rockets mishap dropped my external tanks with about 1t of methane left. Most of the flight was surprisingly stable even though I was at 4x. The whole flight wasn't very well planned, but I managed to ditch it within sight of KSC even with a bit of deformation in the body that likely coincided with the emergency jettison. If I knew it would be so close, I would have throttled down earlier to save fuel. The most confusing part was flying back from the ocean west of the river end to KSC. I decided to go across the poles. Eyeballed a direction then corrected it over time. It's quite hard to fly in a 'straight' line if you have no idea what the relation between that direction and the compass heading is, so I likely could've save some more fuel by not weaving around as much. I'll go through the footage later and see if I can find something useful from it.

There's very few sci-fi ships that can pull off realistic-like dV figures. So there's just no way I'll build a ship that can do actual ground-to-orbit-to-the-mun-with-cargo-to-drop-and-back without unlimited fuel... or will I?

No idea how aerodynamic it is in practice, but it's pretty small/light for what it is. 100t of Methalox with 8t dry weight. Can use it as a type of drop tank...

That's not how things work Some calculations cannot be efficiently multi-threaded, causing a single core to do the bulk of the work -> CPU usage unlikely to hit 100% total, but likely to hit 100% on a single core. This has a knock-on effect of starving the GPU of having new frames to render -> GPU usage < 100%. There are ways of optimizing some of this, but the nature of KSP means it's not a straight forward. One way you might see higher utilization is to have multiple unconnected craft in a scene. If two ships aren't connected, the physics can be evaluated separately, so the engine could potentially use a higher % of your total CPU.

My text size seems consistent (2560x1440p - 100% scaling) - BUT - They don't always seems to display. I would have, for instance, a solar panel showing deployable, but not showing any properties, like EC generation. For parts with more complex properties, some might show on one mouseover, but flicking quickly between parts others might show while the original text dissapears.