Performance KSP2 is Calculating the Physics of all Parts of all Crafts Whether They are Rendered or Not, Reducing Performance of all Scenes at all Times.

[Anth]

Reported Version: v0.1.4 (latest) | Mods: Lazy Orbits was used to create the 8001 test save then removed | Can replicate without mods? Yes 
OS: Windows 10 | CPU: i9 9900K | GPU: 3070ti | RAM: 32GB

Expected Behaviour:
That when a craft isn't rendered the game treats 1 craft as one entity (no matter how many parts it has) and that the craft is on rails to improve overall performance.
Note: This behavior might be different then a craft is using its engines.
Observed Behaviour:
That KSP2 is calculating the physics of all parts of all crafts whether they are rendered or not, causing performance issues.

3 Tests:
What the tests will prove:
Test 1 will show KSP1 maintaining maximum performance regardless of how many parts there are in the save when theres only a 1 part craft in scene.
Test 2 will show KSP2 with a massive drop in performance which is directly related to the amount of parts in a save no matter if they are rendered or not. (Used the Lazy Orbits Mod)
Test 3 will show similar to test 2 (Lazy Orbits Mod Not Used)

Test 1 + 2: KSP1+KSP2 8x1000 part crafts being deleted one at a time while focused on a 1 part craft on Vall. (Includes Videos + Save File)

Spoiler

Note that the 1000 parts are made up of 999 truss segments + 1 probe to stop it being counted debris.
I did this to attempt to stop the game from excessively using the resource system.

I did the following:

1. Put 8x1000 part crafts in orbits around different planets/moons using Lazy Orbits
2. Put a 1 part craft on the surface of Vall
3. I recorded the ms/frame (KSP2) or FPS (KSP1) for 8001 parts while having only the 1 part craft in scene on Vall.
4. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 7001 and then tested on Vall again.
5. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 6001 and then tested on Vall again.
6. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 5001 and then tested on Vall again.
7. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 4001 and then tested on Vall again.
8. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 3001 and then tested on Vall again.
9. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 2001 and then tested on Vall again.
10. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 1001 and then tested on Vall again.
11. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 1 and then tested on Vall again.
12. I recorded all of the results and put them into a graph.

Video Evidence: (Time stamped after each deletion for faster navigation)
KSP1: 1.12.5   https://youtu.be/v8Q8LkIPu_s?t=12
KSP2: 0.1.4.0  https://youtu.be/GVe6zyo9KQ0

Save File:
KSP1 8001 Save 1 Part Scene.zip (Dropbox Link)
0140 8001 Part Test.json (Dropbox Link) 

Test 3: KSP2 10x250 part crafts being deleted one at a time while focused on a 1 part craft on the Mun (Includes Video + Save File)

Spoiler

The 250 part crafts were made up of the ball tanks R-4 Dumpling, standard tanks, vector, pod and landing legs etc
I needed to actually launch it into space and the truss segments were going to be problematic.

I did the following:

1. Launched 10x250 part crafts with infinite fuel on into Kerbin low orbit
2. Launched a craft and landed on the Mun and decoupled parts so it ended up 1 pod
3. I recorded the ms/frame (KSP2) or FPS (KSP1) for 2501 parts while having only the 1 part craft in scene on the Mun
4. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 2251 and then tested on the Mun again.
5. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 2001 and then tested on the Mun again.
6. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 1751 and then tested on the Mun again.
7. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 1501 and then tested on the Mun again.
8. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 1251 and then tested on the Mun again.
9. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 1001 and then tested on the Mun again.
10. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 751 and then tested on the Mun again.
11. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 501 and then tested on the Mun again.
12. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 251 and then tested on the Mun again.
13. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 1 and then tested on the Mun again.
14. I recorded all of the results and put them into a graph.

Video Evidence: (Time stamped after each deletion for faster navigation)
KSP2: 0.1.4.0  https://youtu.be/uEFy6o8bOOg

Save File:
0140 2501 Part Test.json (Dropbox Link) 

Results from Videos on a Graph (lower is better):
Note that KSP1's FPS is converted to ms/frame.

Excel File:
KSP12TestResults.xlsx

Conclusion:

• KSP1 only calculates 1 craft as 1 part when it isn't rendered and is out of physics range regardless of its part count.
• KSP2 calculates the physics of every part of every craft in a save causing performance issues the more parts there are in a save file.

Additional Information:
Colonies has the potential to have players using a lot more than 8000 parts in a save file.
This will affect performance even more severely and also means for longer and longer load times.

Original Bug Report (0.1.1.0):

 

Edited April 24 by DibzNr

[Anth]

So this is what I was shown today:


So I kept saying physics when the problems might be everything on the right of this diagram (simulation) that is causing problems.

Edited September 2, 2023 by Anth12

[pschlik]

Surprised this isn't getting as much attention as the thread about this from previous versions did. I was getting frustrated by low FPS despite my GPU running at around 25% load and none of my CPU cores seeing sustained load and managed to gain around 10 fps (from ~35 to ~45 in space) just by removing 16 random bits of debris. If I then also removed the 19 vessels other than my own I was suddenly getting more than triple the fps (up to ~120)!

What is in essence a background task shouldn't be slowing the game down, and it really shouldn't be slowing the game down when there are more than enough system resources to handle more computation. The fact that this much of a slowdown isn't coming from a hardware bottleneck is baffling...I can't imagine what this would be like with multiple solar systems and colonies if it's this impactful in the midgame of the Kerbol system. 5 fps at 5% CPU and GPU use? We'll see...but I hope not.

[stilrare]

This is definitely a big issue, when making a space station with many parts, or there’s crafts orbiting kerbin or other planets/moons. I noticed this issue when I was launching a rocket with about 200 parts, and I had 5-8 FPS. Before patch 4, with I think 180 parts, I had 12-19 FPS when launching, even with NO crafts orbiting or anywhere expect the one that’s launching. I also tested a satellite on the launchpad (in patch 4) and I was expecting more than 25 FPS, but I had 10-14 FPS with a 20 part satellite. Still with no crafts anywhere except the satellite on the launch pad.

[TripleStaff]

1 hour ago, gluckez said:

Even that alone is a complicated task, you never know when something with a gravitational pull comes along and slingshots you into a different orbit. but ideally, yes, that's how it should be. 

It's really not, everything with gravity is on rails, so the game knows exactly when and how it will happen (Though it doesn't simulate all future orbits) due to that fixed model. 

 

1 hour ago, gluckez said:

I actually kinda like the idea of a kerbal colony where mistakes are made, it's in the nature of the game that things go wrong. not constantly obviously, but occasionally kerbals gonna kerbal.

Except it's never the Kerbals who mess up, only you, the player. Something that's fully automized and idle shouldn't need constant watching and upkeep, since that defeats the whole purpose of it. If you've done a big ksp1 campaign with lots of miners, tankers, bases, stations, etc you know the tedium of doing all that, even with help from stuff like MechJeb. (Though as I said before, some kind of lifespan limit would be fun, but not randomness)

 

2 hours ago, gluckez said:

that would be only possible when the allignment between the ship and the celestial bodies is the same. sometimes you need a bit more or less deltaV because you're landing in a crater or on a mountain.

I was assuming landing on a fixed location, like a base, that would stay the same. It would time itself to whatever target it has in orbit, so the launch, approach, docking, etc would all be the same repeating pattern.

 

2 hours ago, gluckez said:

me neither, and from what I've seen, that's definitely something that's on the roadmap. I'm just curious how it's gonna be implemented.

Changing difficulty shouldn't make your mistakes cost more, but actually make the game more challenging, something a lot of games fail at. As a veteran KSP player (I remember using winglets as landing legs...) I've done just about everything more than once, and other than the few decent contracts there's only external challenges that really test me. And as I said before, it's gotta be predictable. 

[cheese3660]

Doing some simple benchmarking/profiling I found, that with anths save, the update for the SpaceSimulation is taking around 50 ms per update on my machine, and that the entire fixed update chain from the game instance is taking around 60-80 ms per frame

Edited September 2, 2023 by cheese3660

[Vortygont]

It is logical if you have interstellar ship which flyes while you do other things. But countimg all crafts is ridiculous. It can have meaning if your system unstable and for example guaranted overheat, but in other case game must count before equilibrium and than simply use formula to add for example resources if something is mining

[gluckez]

On 9/2/2023 at 9:48 AM, Anth12 said:

That when a craft isn't rendered the game treats 1 craft as one entity (no matter how many parts it has) and that the craft is on rails to improve overall performance.

what if your craft that isn't rendered is on a collision course with an asteroid or another craft?
I like the effort you put into it, thanks for that, but I don't agree with the premise. There will be plenty of situations
where you aren't rendering your craft, but you still need physics. for example supply routes for colonies, where docking
occurs often off screen.

[TripleStaff]

1 hour ago, gluckez said:

what if your craft that isn't rendered is on a collision course with an asteroid or another craft?
I like the effort you put into it, thanks for that, but I don't agree with the premise. There will be plenty of situations
where you aren't rendering your craft, but you still need physics. for example supply routes for colonies, where docking
occurs often off screen.

The game would of course keep track of all objects and their possible collisions, and if one were imminent it could then load the proper physics for just that interaction. But with the game warning you when a controllable craft is on an impact trajectory, and the "traditional" way of simply deleting uncontrolled crafts/debris when the simplified on-rails model would be destroyed, you really don't need fancy simulations or physics. 

Even if we get selfcontrolled crafts that can dock and land by themselves they won't need detailed simulating, just the simple one part model. Docking doesn't need much either, just match the speed and position, add the masses together, and then consider them docked. You don't even need to add the masses if it's landing on a base. And assuming it's a repeating task you'd only need to properly simulate the active (when the craft isn't just on rails, like landing, liftoff, etc) part once and then just play it back. If that feels too deterministic just think of it as the craft adapting to different circumstances to achieve the same outcome.

A any object can be simplified as relatively few values, even when active, all the components don't need to be individually simulated until it's being rendered for the player. If the craft has something like solar panels, that can be affected by the orientation, you could just simulate it as alwats being in the ideal orientation for gathering power, so that the output is only affected by the position relative to Kerbol (though if you're using ion engines you would need more proper simulations) 

 

KSP isn't trying to be hyperrealistic, but just realistic enough so that it's challenging and fun, without the instability and randomness real space brings.

(Though I wouldn't mind the option of some more realistic difficulties, like orbits degrading and parts wearing out, giving everything a limited lifespan unless maintained, sorta like what we see with the nuclear generators)

[DancZer]

I think this is because of the planned multiplayer. That's why every object in the game is handled in the game the same way. Currently the game they act like a client server (as it is a  multiplayer game), even if you play it as a single player. At least this is my experience with developing multiplayer games. You have to design everything in mind of being multiplayer ready. It is not something you can add later easy. Probably this one one of the main reasons to start KSP 2 from scratch. It also correlates why it is fast to switch between tracking station/vessels/space station.

I'm not sure what they want/can do regarding optimization, before the multiplayer comes online. Because then "distant" would be relative.  Some simplification would definitely needed but it will have many parameters for a feature complete KSP2. 

[mattihase]

so I'm guessing al this simulationing is why orbit decay is so prominent?

 

[gluckez]

5 hours ago, TripleStaff said:

The game would of course keep track of all objects and their possible collisions, and if one were imminent it could then load the proper physics for just that interaction.

Even that alone is a complicated task, you never know when something with a gravitational pull comes along and slingshots you into a different orbit. but ideally, yes, that's how it should be. 

 

5 hours ago, TripleStaff said:

Even if we get selfcontrolled crafts that can dock and land by themselves they won't need detailed simulating, just the simple one part model.

I actually kinda like the idea of a kerbal colony where mistakes are made, it's in the nature of the game that things go wrong. not constantly obviously, but occasionally kerbals gonna kerbal.

 

5 hours ago, TripleStaff said:

And assuming it's a repeating task you'd only need to properly simulate the active (when the craft isn't just on rails, like landing, liftoff, etc) part once and then just play it back

that would be only possible when the allignment between the ship and the celestial bodies is the same. sometimes you need a bit more or less deltaV because you're landing in a crater or on a mountain.

 

5 hours ago, TripleStaff said:

Though I wouldn't mind the option of some more realistic difficulties

me neither, and from what I've seen, that's definitely something that's on the roadmap. I'm just curious how it's gonna be implemented.

[kicka55]

On 9/3/2023 at 11:54 AM, gluckez said:

what if your craft that isn't rendered is on a collision course with an asteroid or another craft?
I like the effort you put into it, thanks for that, but I don't agree with the premise. There will be plenty of situations
where you aren't rendering your craft, but you still need physics. for example supply routes for colonies, where docking
occurs often off screen.

You could do this with treating the craft as 1 part or something. Don't need to simulate all parts when you're not actively playing it. I assume solar and heat play a role in that too but there have to be workarounds to not strain the PC too much. 

[gluckez]

12 hours ago, kicka55 said:

You could do this with treating the craft as 1 part or something. Don't need to simulate all parts when you're not actively playing it. I assume solar and heat play a role in that too but there have to be workarounds to not strain the PC too much. 

I can imagine that if that craft happens to be a space station, you wouldn't want all your kerbals killed because a piece of debris hit a solar panel. It's also not as simple as just treating it as a single part when you're actually working on the code of the game, because you're still tied to the limitations of the game engine itself. I agree there has to be a fix, not a workaround, but it's not as easy as sprinkling some "if(offscreen) Donwobble()" code over your codebase.

[cocoscacao]

On 9/3/2023 at 2:18 AM, kennyc222 said:

YEah I also experienced this

This is weird. Some people claim better performance, others the opposite.

I haven't seen any difference, but I didn't go to same lenghts like @Anth12 did. Maybe creating a poll (with hardware stuff listed) could be helpful. 

[Ruedii]

On 9/2/2023 at 3:48 AM, Anth12 said:

Reported Version: v0.1.4 (latest) | Mods: Lazy Orbits was used to create the 8001 test save then removed | Can replicate without mods? Yes 
OS: Windows 10 | CPU: i9 9900K | GPU: 3070ti | RAM: 32GB

Expected Behaviour:
That when a craft isn't rendered the game treats 1 craft as one entity (no matter how many parts it has) and that the craft is on rails to improve overall performance.
Note: This behavior might be different then a craft is using its engines.
Observed Behaviour:
That KSP2 is calculating the physics of all parts of all crafts whether they are rendered or not, causing performance issues.

3 Tests:
What the tests will prove:
Test 1 will show KSP1 maintaining maximum performance regardless of how many parts there are in the save when theres only a 1 part craft in scene.
Test 2 will show KSP2 with a massive drop in performance which is directly related to the amount of parts in a save no matter if they are rendered or not. (Used the Lazy Orbits Mod)
Test 3 will show similar to test 2 (Lazy Orbits Mod Not Used)

Test 1 + 2: KSP1+KSP2 8x1000 part crafts being deleted one at a time while focused on a 1 part craft on Vall. (Includes Videos + Save File)

  Reveal hidden contents

Note that the 1000 parts are made up of 999 truss segments + 1 probe to stop it being counted debris.
I did this to attempt to stop the game from excessively using the resource system.

I did the following:

1. Put 8x1000 part crafts in orbits around different planets/moons using Lazy Orbits
2. Put a 1 part craft on the surface of Vall
3. I recorded the ms/frame (KSP2) or FPS (KSP1) for 8001 parts while having only the 1 part craft in scene on Vall.
4. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 7001 and then tested on Vall again.
5. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 6001 and then tested on Vall again.
6. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 5001 and then tested on Vall again.
7. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 4001 and then tested on Vall again.
8. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 3001 and then tested on Vall again.
9. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 2001 and then tested on Vall again.
10. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 1001 and then tested on Vall again.
11. Went to the tracking station and deleted 1x1000 part craft reducing the part count total to 1 and then tested on Vall again.
12. I recorded all of the results and put them into a graph.

Video Evidence: (Time stamped after each deletion for faster navigation)
KSP1: 1.12.5   https://youtu.be/v8Q8LkIPu_s?t=12
KSP2: 0.1.4.0  https://youtu.be/GVe6zyo9KQ0

Save File:
0140 8001 Part Test.json (Dropbox Link) 

Test 3: KSP2 10x250 part crafts being deleted one at a time while focused on a 1 part craft on the Mun (Includes Video + Save File)

  Reveal hidden contents

The 250 part crafts were made up of the ball tanks R-4 Dumpling, standard tanks, vector, pod and landing legs etc
I needed to actually launch it into space and the truss segments were going to be problematic.

I did the following:

1. Launched 10x250 part crafts with infinite fuel on into Kerbin low orbit
2. Launched a craft and landed on the Mun and decoupled parts so it ended up 1 pod
3. I recorded the ms/frame (KSP2) or FPS (KSP1) for 2501 parts while having only the 1 part craft in scene on the Mun
4. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 2251 and then tested on the Mun again.
5. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 2001 and then tested on the Mun again.
6. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 1751 and then tested on the Mun again.
7. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 1501 and then tested on the Mun again.
8. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 1251 and then tested on the Mun again.
9. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 1001 and then tested on the Mun again.
10. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 751 and then tested on the Mun again.
11. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 501 and then tested on the Mun again.
12. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 251 and then tested on the Mun again.
13. Went to the tracking station and deleted 1x250 part craft reducing the part count total to 1 and then tested on the Mun again.
14. I recorded all of the results and put them into a graph.

Video Evidence: (Time stamped after each deletion for faster navigation)
KSP2: 0.1.4.0  https://youtu.be/uEFy6o8bOOg

Save File:
0140 2501 Part Test.json (Dropbox Link) 

Results from Videos on a Graph (lower is better):
Note that KSP1's FPS is converted to ms/frame.

Excel File:
KSP12TestResults.xlsx

Conclusion:

• KSP1 only calculates 1 craft as 1 part when it isn't rendered and is out of physics range regardless of its part count.
• KSP2 calculates the physics of every part of every craft in a save causing performance issues the more parts there are in a save file.

Additional Information:
Colonies has the potential to have players using a lot more than 8000 parts in a save file.
This will affect performance even more severely and also means for longer and longer load times.

Original Bug Report (0.1.1.0):

 

Interesting question on a related bug, is KSP2 properly handling treating each craft and terrain layer as a separate physics layer unless they are within collision range? 

This is a very important way to reduce physics load.   Additionally turning off collisions for any body not in close contact with another body might be a good improvement for physics.  Ideally only the single quad of a surface should be active for physics on a craft that is landed, landing or taking off. 

It might be good to run a threshold algorithm on the delta of active forces on a vessel over the past few seconds to determine whether to treat it as a single ridged body or a structure made up of multiple ridged bodies.

If the forces on a vessel are within a margin of error of static, including internal forces, of course, then it be treated as a static object and thus a single ridged body mesh instead of a compound one.

[kicka55]

On 9/16/2023 at 2:37 PM, gluckez said:

I can imagine that if that craft happens to be a space station, you wouldn't want all your kerbals killed because a piece of debris hit a solar panel. It's also not as simple as just treating it as a single part when you're actually working on the code of the game, because you're still tied to the limitations of the game engine itself. I agree there has to be a fix, not a workaround, but it's not as easy as sprinkling some "if(offscreen) Donwobble()" code over your codebase.

Nothing ever is easy but they already have this function for timewarp. In timewarp the craft changes from physics simulated to non physical. Even with engines on now in KSP2. In KSP1 they only managed that with physical timewarp - which does not exist in KSP2 anymore in space (only in atmosphere).

[gluckez]

10 hours ago, kicka55 said:

Nothing ever is easy but they already have this function for timewarp. In timewarp the craft changes from physics simulated to non physical. Even with engines on now in KSP2.

then the question becomes: is that the desired behaviour? because if it's a non physical craft, how is it accelerating with the engines on? and with that implementation, parts of the ship will simply phase through objects during timewarp. that would make it very easy to avoid collisions when playing multiplayer, just turn on timewarp and wait until you passed through entirely. and being able to provide thrust during timewarp would also mean someone could mod the game and add a 1x timewarp that does the same, so you would never experience any wobble or collisions during takeoff.

[kicka55]

On 9/25/2023 at 9:50 AM, gluckez said:

then the question becomes: is that the desired behaviour? because if it's a non physical craft, how is it accelerating with the engines on? and with that implementation, parts of the ship will simply phase through objects during timewarp. that would make it very easy to avoid collisions when playing multiplayer, just turn on timewarp and wait until you passed through entirely. and being able to provide thrust during timewarp would also mean someone could mod the game and add a 1x timewarp that does the same, so you would never experience any wobble or collisions during takeoff.

That's exactly what happens in timewarp. Vehicles pass through each other and even through planets sometimes. If someone mods anything it's considered a cheat so people can cheat mod anything anyways. They can simply turn all rigidity up to max in the config for that matter to have 0 wobble. 

[QuiescentRabbitt]

Reported Version: v0.2.0 (latest) | Mods: none | Can replicate without mods? Yes 
OS: Windows 10 | CPU: AMD Ryzen 7 3700x | GPU: Nvidia RTX 3080 Ti | RAM: DDR4 16GB 3200Mhz

 

Despite a fix being implemented into 0.2 intended to improve background vessel performance, background vessels still continue to impact game performance in a very significant way that needs further work. This can be seen in the chart I provided where a save file with 40 satellites around Kerbin nearly halves the frame-rate both in orbit and around KSC. This result is nearly unchanged from 0.1.5 which experienced nearly the same performance hit.

As things are now, save files that accumulate vessels from continual play will experience severe performance slow-down.

 

Included Attachments:

Mk1PodonLaunchpad40Mini.json

Edited April 24 by DibzNr

[Selvek]

I have five probes in the Jool system and am down to about 4 seconds per frame.   Ragequitting again after attempting to play on 0.2.1 for an hour.

[Jaypeg]

On 2/1/2024 at 4:41 AM, Selvek said:

I have five probes in the Jool system and am down to about 4 seconds per frame.   Ragequitting again after attempting to play on 0.2.1 for an hour.

get rid of unused crafts for now

[TFA3393]

This report isn't getting a lot of upvotes but people are asking about this all over reddit.  I'm sure its more of a long term issues but it is still very annoying.  How am I supposed to put colonies on every single celestial body 5 times over with performance like this!?

[Spicat]

This post is specifically about worse performance on background vessel compared to before 0.2.

If you just want to talk (and report) about bad performance related to background vessel in general, there is already a report about it:

This has (and will be) addressed by devs.

[DibzNr]

@QuiescentRabbitt We initially kept your report thread separate from this bug report due to wanting to catalogue how this bug may have changed post-0.2.0.0, however we've seen no significant change in it, as such we've merged it into this report thread

[kennyc222]

YEah I also experienced this! BEfore 0.1.4.0, my high-ended PC handled the KSP2 so well with satisfactory performance, but after the upgrade, it kept lagging every second in all time! no matter the campaign has a ship or not! it can't handle a single small ship!