It is happening to more players Op, including me on a daily basis. However this has been going on for me since three or four days after the new server came in to action. I just put it down to my computer or just me. Hence no ticket or mention. But it become more frequent lately. Example, played for many hours with n/p, then had three ktd ( kicked to desktop ) a minute apart. Then nothing all was good again. Example today, while officer tribbs was on i was flying in sos and ktd. Logged on, then again all play was ok. No to your next question screenshots, i realize it would help. Sum up it just happens and then all is good again. my luck in todays as i said officer tribbs saw it.
Also day 2 and still all good with no eratic behaviour with the npc,s.
after some reports backups have been restored
this can be considered a failed test
Today 23/1, Server was stable. Have not had any problems what so ever. No ktd,s and no eratic beviour from npc,s. File download on launcher first one was 32 then second was 30 so again put it down to you doing something at some stage, maybe. Appreciate your work Op
yeah, making progress
everyday a small fix, then checking the logs of it works
25/1 Still stable for me. Launcher still on 30 downloads, so seems to be where it,s going to sit. No eratic behaviour, or ctd, or npc problems
Since the server is running pretty stable again and the performace did improve it is time to move on to take care of other issues.
While trying to fix the server statistics that are displayed on the CF page Crossfire 2.0 Freelancer Mod - SWAT Portal I notices some issues that forced me to completely rebuild the daily maintenance scripts as there were some incompatibilities with the new server.
The statistics got completely replaced with a new version based on a calculated player score. The rank based stats got removed but the score based stats got a sort function so individual sorting of the table is possible by clicking on the row header.
This rebuild of the PVE stats does lead to a ~85% shorter script execution and therefore less performance issues during the daily maintenance.
The PVP stats now work correctly again and some general server stats got added.
While working on the stats I noticed some problems with the backups... so that I had to rewrite all the other maintenance scripts aswell. While mostly working now there are still some quirks that I need to take care of. Especially the timing of the script execution requires a lot of fine tuning which I hopefully will be able to sort out in the next few days.
Positive side effect is that the update of the dynamic economy and dynamic market runs faster now and much more reliable than ever before. The update timeframe for this should be no longer than 5mins now, right when the server makes the announcement to restart.
Due to the lack of feedback I decided to make the new shop system dealing with clan system rent public now. You should be able to access it, in the tools section of the menu.
The next few weeks will be used to take care of the reported issues on the ticket system. I already got rid of 3 pages... but it is a step by step procedure to get all the stuff fixed or improved.
time for a small status update
The hardware failure of our webserver did cost me some time in which I could not do that much of the stuff I actually wanted to do.
For the duration of that failure stuff i needed for tests was not available to me and some other stuff just did not deliver correct results for that weekend.
So it took longer to get the fine tuning done on the new maintenance routines which greatly improve the reliability of certain features. To be precise... Thursday was the first time the new maintenance scripts run totally error free.
As reported on discord there are a few random crashes in singleplayer related to docking. First investigations showed that the issue must be related to specific docking animations. To find the issue isn't going to be easy since there are more than 40.000 lines of code controlling this stuff and finding the error in there takes some time.
Some feedback showed that about 50% of that code does not seem to cause any issues but still it is a lot of data to check.
I've decided narrow down the source of error by checking testing all the animations in a controlled test environment over and over until I can clearly say which one works... which one causes problems and therefore eventually reduce the code that needs to be checked by hand to a minimum.
Right now I simply see no alternative to this very time consuming method.
In the meantime Id like to put some focus on community discussions about some planned changes.
The first one is the rebalancing of the battleships... or more precisely the rebalancing of the cap ships.
Goal has to be to make them a viable gameplay option but not to make them overpowered.
As I proposed earlier my idea was to detach them a bit from the fighter gameplay by modifying the efficiency of shields and weapons.
Cap ships should not be able to shoot every fighter out of the sky with 1-2 shots and in the same way fighters should not be able to take out cap ships with ease.
At the same time fighters should remain lethal to other fighters and cap ships should remain lethal to other cap ships.
Balancing is going to be tough... but not impossible.
We will discuss how that can be done on a separate thread in order to keep things clean here.
The other topic to discuss will be related to how we can improve the dynamic universe.
Also some ideas were already shared.
I am just throwing in some additional stuff I have in mind here.
I want the launcher to have a news ticker showing current dynamic server events similar to the stuff that is shown on the bars. I did some investigation how to get it done... made some progress on how actually synchronize that stuff but also ran into some minor issues displaying more detailed info on the launcher. But I am sure I can find a solution for it somehow.
I also was thinking about new kinds of server events and how to visually improve the representation of these events.
e.g. jumpgate malfunctions should get get effects added showing that this gate currently is not working, stations that were attacked (a possible new event type) could show damages, burns, gas leaks etc.
Additionally we need to discuss what kind of new events we should go for and what it takes to somehow even control these events in a more dynamic way.
change of plans
One of my HDDs died yesterday. Most of the stuff on it was no big loss with the exception of sources for the CF installer.
I have a backup of it but that was ~2 weeks older than the final version which means that lots of the pretty important detail changes are not included.
Currently we dont need the installer but most certainly we need it at a later point. Restoring as much as possible has highest priority atm since the data integrity of what is left most likely wont get any better with time.
I believe I managed to recover the instruction set for the installer and most of its resources already but the content structure is a mess and has to be rebuild (which I probably can do with the 2 weeks old backups, dunno yet).
The stuff announced above will have to be delayed a bit.
ok, I think I managed to restore the crossfire installer and fill the gaps based on internal discussions with the testers and my bad memory.
5 days of development were lost and I really hope that the new code I wrote is good enough to cover this timeframe.
Im going to check if this restored installer version produces exactly the game installation results as the original release from 2020. If that is the case Ill provide the restored version to the same testers again in order to check if they see any differences in their specific software environments.
The malfunction of the webserver 2 weeks ago and the failure of my HDD with the installer on it did cost me a lot of time.
But if everything works now I can finally return to do some constructive work.
I did some tests on the animation scripts last weekend and at least made a bit progress with generating test results for the crash bug in singleplayer. Unfortunately I have to conduct at least 50 tests per file in order to keep the error margin low. So only 20-30 hours of testing to go
time for an general update on the topic
- Battleship rebalancing seems to be moving into the right direction. More info in this thread: Capital Ship Rebalancing
- In the meantime I have completed ~75% of the script tests to figure out why the game sometimes crashes in singleplayer while docking. Since I conduct somewhat between 50 and 100 tests per script this is a very time consuming task. Unfortunately the random behaviour of this issue require such a large amount of tests just to be sure.
- A first internal test related to the new announced dynamic economy based on fleet invasions (a very limited test actually) looks very promising but it is too soon to talk about details here.
- I was thinking about the potential crash sources in the renaissance systems related to the stations there and how to resolve this issue. The creation of large complex surface models for the stations is tricky and none of the methods is 100% error free. So I might try cut the stations into smaller pieces, generate the surfaces and then put everything back together. I think it is worth a try. It makes stuff more complicated but might resolve the crash issues related to these large objects.
- Ive read through a discussion with Huor and an idea how to make the Tour de Universe more reliable. Especially the conflicts with the Dynamic Universe could in theory be resolved this way. Depending on my time I will do a few server side tests at one of the upcoming weekends. I can not 100% tell if the stuff we came up with is going to work. Maybe it will cause unexpected side effects. We will see.
Ive read through a discussion with Huor and an idea how to make the Tour de Universe more reliable. Especially the conflicts with the Dynamic Universe could in theory be resolved this way. Depending on my time I will do a few server side tests at one of the upcoming weekends. I can not 100% tell if the stuff we came up with is going to work. Maybe it will cause unexpected side effects. We will see.
If there is anything i can doy please let me know... I still have some open changes that are not yet on the server, as we have had some troubles with the anticheat at former test times. So i havent had the guts yet to try again
do you mean the issues we had in December?
since we meanwhile know the cause....
The intended changes for the Tour de Universe were put aside.
Huor had a better idea which I applied today.
Conflicts between the Tour and the Dynamic Universe should in theory no longer happen.
I am still working on the crash issues with docking in singleplayer and the random crashes in renaisance space. This however turned out to be a bit more complicated than anticipated. The randomness makes it very hard to do reliable tests.
I have meanwhile completed the tests on the launch and landing scripts for stations to find a reported crash bug in singleplayer.
Despite having spent many weeks on a total of more 2000 tests I was not able to reproduce the reports.
I do not doubt that these crashes exist since multiple players reported such issues but like I wrote... after more than 2000 tests done on all the different station types I had not a single crash.
Maybe the issue is a combination of hardware environment and these scripts that I can not reproduce on my PC.
If you encounter such a crash somehow, somewhere I would like to get a detailed error report on the issue. Where exactly it happened and what exactly you did. Maybe this helps to narrow down the conditions under which the problem occurs.
Other than that I have no idea what to do about it other than disabling the animation scripts completely.
These long tests however, showed a few issues I would want to fix/change.
I am not so super happy with a few things on the bases.
Some of the scripts contain a few timing errors that I want to fix.
Since the Windows 10 Build 2004 update Directx causes a bug with lighting on many stations. This bug however does not seem to happen on all of the stations. There are some which still work fine. I will try to figure out if I can apply the working settings to the station which suffer from the lighting bug. I cant fix the directx bug but I can probably reduce the impact it has on the game this way.
I am also not really happy with the texturing work on some of the stations anymore. I think I can do better by masking materials like I did in other areas of the game. Next week I will check if I can improve the look of some of the station decks.
Ive spend more than a hundred hours testing these bases and this really bothers me now.