Author: Stephen

I’ve realised it has been about 4 months since the last update (and I said I would update every 2-3 months) so sorry about that!

Since the last update, I’ve done the following:

• I created the Knights app on Steam, although it hasn’t been made public yet.
• I figured out how to upload builds to Steam, so that part is working.
• The integration with Steam is progressing well, e.g. we now use Steam’s P2P networking features, and inviting friends to games works.
• Fixed some Lua build issues (this was causing the Makefile to not work properly on non-Debian Linux distributions).
• Fixed a bug where Invulnerability wasn’t working (due to a Lua error).
• Begun adding Unicode support for the in-game text rendering (although it’s not finished yet).
• Probably made a whole bunch of other minor tweaks/fixes that I have forgotten about!

What is still left to do:

• Review safety and security – e.g. I want to try to make the network code as secure and hacking-resistant as possible; also Steam themselves have some safety features (e.g. chat filtering, blocking, privacy options) that I need to make sure that Knights supports.
• Add a “Vote to End Game” button (or similar), in case players want to end a quest early.
• Maybe make some small tweaks to the UI. E.g., adding tool tips in one or two places might make sense.
• Need to add “LAN” and “Online server” modes back into the game (I temporarily removed those while getting the Steam integration working, but it would be good to put them back).
• Fix any remaining bugs, and do some testing.

Hopefully it shouldn’t take too long to do most of those. I also now have some Steam keys for the game that I can give to people, in case we want to do a multiplayer test, before the full public release of the game.

After that the only thing left would be making the artwork and videos for the Steam store page. I’ll figure that out once we get there 🙂

I guess that about covers the current state of things so I’ll wrap up for now. Hope everyone reading has a good Christmas, and I’ll report back in the new year!

Just a quick note to say what I have been working on for the last couple of months.

I’ve added a “reconnect” feature to the game. Previously, if a player left a game for any reason, they would just be removed from the game entirely — and would not be allowed to rejoin. Now, if a player leaves, their knight remains in the dungeon, but they appear as “(Disconnected)” in the player list. If they later reconnect, they can jump back in and resume from where they left off. Meanwhile, the other, non-disconnected players can continue playing, until one of them wins the quest in the normal way. (Players could also kill the disconnected knight and take their stuff, but some might consider that unchivalrous!)

Unfortunately, one side-effect of this change is that it is now quite hard to actually “abort” a quest and start a new one. (Pressing ESC-Q no longer works, because that now just makes you leave the game, with a possibility of returning later, instead of actually ending the game.) I was thinking of adding some kind of “Vote to restart” feature to address this. Basically, one player would click a button to call a vote, and if enough players agree, then the current quest would be abandoned, and the game would return to the quest selection menu. (This isn’t done yet, but hopefully I can add it soon.)

The other thing I’ve been working on is beginning the integration with Steam’s matchmaking features — or at least preparing the ground for this. I have been working on creating the concept of a “lobby” in the code. A player can create a “lobby” on their local machine, which encapsulates the state of a Knights game. Other players can join the lobby, and the lobby state can also be transferred from one machine to another, allowing for host migration. This is still work in progress, but is going reasonably well. Once I get that working, I will try hooking it up to the actual Steam APIs, so that, for example, a player could create a game and invite a friend to join it, or allow other players to join via Steam’s matchmaking system. (This will also mean creating the actual Steam app id for the game, which I think costs $100 at the moment, but that shouldn’t be too big of a deal.) This is probably still a couple of months away at least, but it will be cool to see it finally working!

Thanks for reading. I’ll try and keep posting updates here every 2-3 months or so.

About Knights

Knights is a multiplayer game in which players must explore randomly generated dungeons and race to complete various quests. For more information please visit https://www.knightsgame.org.uk/.

As mentioned in the previous post, I have begun work on bringing Knights to Steam.

Since then, I have made the following progress:

• The Lua documentation has been completed.
• I made various optimizations and code tweaks, to try to make the in-game animation smoother, reduce latency, and reduce CPU usage. For example, on my (relatively slow) computer, the game now uses about 30% of one CPU core in-game, instead of 40-50% previously.
• I’ve begun work on adding a “host migration” feature. To explain why this is needed:
  • Currently the game uses either a server model (for online games) or a host/client model (for LAN games).
  • For Steam, the plan is to move away from the server model and use a host/client model, based around the Steam P2P networking features.
  • To make this work smoothly, I want to add a “host migration” feature so that if the host player leaves the game, another player can take over as host. (The alternative would be for the game to just close if the host player leaves, but that would not be very user-friendly.)
  • For host migration to work, there needs to be a way to smoothly transfer the “game state” from one machine to another, so that the new machine can take over hosting duties. I don’t want to go into all the technical details here, but suffice to say, I have a plan for doing this, and have started working on it, but the work is not yet complete.

The next steps will be:

• Complete the “host migration” feature as described above.
• Add support for “reconnect”. Currently if a player leaves the game and rejoins, they will be added as an “observer” and won’t be able to continue playing from where they left off. This is annoying, as sometimes you get disconnected accidentally and want to come back in, but can’t currently.
• At this point, we will probably be about ready to get a dev version of the game onto Steam and start integrating with actual Steam features. This is still a little way off, but it will be a good milestone when it does happen 🙂
• There are also a couple of bugs that have crept in, that I want to look at. The most annoying is that the Invulnerability effect doesn’t seem to work any more (a Lua error message is displayed). I assume this must be a result of updating Lua to the latest version (which I did a while back) but unfortunately didn’t notice at the time. Hopefully this will be relatively straightforward to fix, though.

So, to summarize: Work is continuing on getting Knights ready to be put on Steam. The basic plan is to do a couple more months of “prep” work (e.g. improving the networking systems, fixing bugs), before beginning the integration with Steam proper once that is done.

I will continue to post updates here as things progress.

About Knights

Knights is a multiplayer game in which players must explore randomly generated dungeons and race to complete various quests. For more information please visit https://www.knightsgame.org.uk/.

I have decided to go ahead and attempt to make a Steam release of Knights!

This will be as a free game on Steam. Whilst I could conceivably try to charge money for it (and, indeed, some people have suggested that in the past), personally I think free is better, because:

• Knights has always been free.
• Keeping it free will mean that more people get to experience the game (and this is especially important for a multiplayer game, where you want as large a playerbase as possible, ideally).
• It’s not clear that a remake of a 30+ year old Amiga game would make much money anyway, so there isn’t much point trying to charge for it, in my opinion.

(If the game does somehow become popular and people start asking for a way to support the developers, then I could consider setting up a ko-fi or patreon or something, but we’ll cross that bridge when we come to it.)

The advantages of a Steam release would be:

• No need for me to host any server infrastructure myself — we can leverage Steam’s infrastructure instead. (See previous blog post for why this is a good thing.)
• Easier installation — people can just click and play, rather than having to install a separate EXE file from the internet, which they might not trust.
• Improved multiplayer features — we can leverage Steam matchmaking, “invite friends” features, etc.
• Spreading the word — people might see their friends playing Knights on Steam and decide to try it out for themselves.

I have now asked Kalle Marjola (the original Knights creator) for his permission to do this, and he has agreed, provided that I mention his name somewhere, which of course I will. (In fact, his exact words were, “Oh that sounds great!”)

So what needs to happen next?

• The first thing I want to do is finish off the Lua scripting documentation, which is still a work in progress. The point of this is to make it easier for people to make mods for Knights. (Perhaps we can even integrate with Steam Workshop to allow easier installation of mods, although I will focus on getting the base game working first!) This documentation work will take a couple of weeks at least.
• I then need to think about how the multiplayer will work (technically) on the Steam platform. My version of Knights has always been server-based, but on Steam there is no “game server”, instead just peer-to-peer connections between players. The obvious solution is to designate one player to be the “host”, and then that player acts as a “game server”. (This does give a slight “host advantage” to that player, since their ping time to the “server” will be zero, but we will have to live with that for now.) Then the technical challenges are as follows:
  • We probably need some kind of “host migration” feature — otherwise the game would just end if the host player leaves, which isn’t ideal. This is technically challenging, but I think I can see a way of doing it.
  • It would also be nice to allow new players to join a game after it has started, or existing players to reconnect after a disconnection (neither of these are possible with the current implementation), so I’ll have to think about how those could be implemented.
  • We also need to integrate with the Steam lobby and friends APIs, which means e.g. adding UIs to create a new game (either private or publicly visible), view existing games (and join them), invite a friend to a game, or respond to an invite. This will need a bit of work, but there are no major technical difficulties here — it is just a case of doing the work basically.
• There are other Steam features to consider as well:
  • Steam Community (e.g. we could set up a Steam based discussion forum for the game)
  • Rich Presence (I have only a vague idea what this is, but it might be interesting to set up, if it doesn’t take too long)
  • Steam Input (supporting different controller types, basically) — this could be good as there are all kinds of controllers that people use nowadays, not just keyboard and mouse (although I guess keyboard and mouse remains the most popular).
  • Voice Chat — may be worth researching a bit at least
  • Achievements — not sure, but maybe we could have some simple ones like “complete a destroy book with wand quest” or “kill 100 zombies” or something.
  • Ranked online games and leaderboards — might be interesting I suppose, but not an immediate priority.
  • Maybe some other stuff I haven’t thought of yet.
• The game might also need a slight UI overhaul as well, just to make it a little bit more presentable for Steam. If it’s only going to be a free game, then we don’t need to put an enormous amount of effort into this, but maybe sprucing it up a little bit would be reasonable. Can think about this once the technical side is worked out.

There’s also stuff like setting up the store page, and whatnot, but again, we can cross that bridge when we come to it.

For those who are wondering, I’m currently working part-time, which gives me at least 1-2 days per week to work on Knights, so there is time for me to work on this stuff. Even so, it’s going to take a long time – it could be months or even years before we get to a release – so don’t expect anything soon. But I felt it would be worth announcing something here so that the Knights community (such as it is) can know where the game is heading next.

The next few months could be an exciting time, so stay tuned and I will try to keep people updated as things progress 🙂

About Knights

Knights is a multiplayer game in which players must explore randomly generated dungeons and race to complete various quests. For more information please visit https://www.knightsgame.org.uk/.

With great regret I must announce that, as of today, I have taken the official Knights server offline and made the Knights forum read-only.

The reason is the Online Safety Act, which is a piece of legislation which has recently come into effect here in the UK. The aim of the Act is to protect users of “user-to-user services” (which includes online games), and especially children, from various forms of online harm.

While this is a well-meaning goal, which I agree with in principle, the way the legislation has been implemented presents challenges for operators of small online services such as Knights.

All operators — regardless of whether they are a huge company, or a one-person operation like myself — are required to comply, and apparently the duties are quite extensive, including things like completing risk assessments and implementing safety measures to deal with any identified risks. There are also fines of up to £18 million for non-compliance.

Unfortunately, I do not currently have the time to read through the 84-page code of practice and figure out exactly what I would need to do in order to comply. Moreover, given the large potential fines, I would want to hire a lawyer to make sure that everything was being done “properly”, and this would add substantial financial costs.

Therefore, with regret, I feel that the only practical option for now is to close down those aspects of the Knights service that allow user-to-user interaction (and thus would be caught by the legislation) — this includes the posting and messaging features of the Knights forum, as well as the online Knights server itself.

However, according to my understanding, there is an exemption in the Act for blog comments (or other scenarios where users are directly commenting on “provider content”, e.g. online product reviews). (A more detailed discussion is here.) Therefore, I believe it is safe for me to leave blog comments open for now, as long as I “police” the comments section to remove anything off-topic or spammy (which is something I have always been doing anyway).

Longer term, I am wondering if the right thing to do would be to port the game to Steam. Then we could use the Steam multiplayer features and discussion forums and I would no longer have to run those features myself on this website. (Effectively, Steam would be responsible for complying with online safety rules, instead of me.) In fact, a number of people have suggested a Steam version of Knights in the past, but I have not followed up on that because (a) it would require getting back in touch with Mr. Kalle Marjola, the original creator of Knights, to get his permission (I expect he would be fine with it, but it still needs to be done); and (b) it would need quite a bit of work on my part to integrate the game with all the Steam APIs and do any other work required for a Steam release. So, up until now, I have not done that, but maybe now is the time to start considering it.

Anyway, I hope that this is not too much of a disappointment for any Knights fans out there, but I feel I have no choice at the moment, for the reasons outlined above.

Further Reading

Information about the UK’s Online Safety Act 2023

Hamster forum and local residents’ websites shut down by new internet laws – Free Speech Union

It is time to make the Online Safety Act 2023 fit for purpose – decoded.legal’s blog

Why UK Online Safety Act may not be safe for bloggers – The Register

About Knights

Knights is a multiplayer game in which players must explore randomly generated dungeons and race to complete various quests. For more information please visit https://www.knightsgame.org.uk/.

A new version of Knights has been released!

This makes some changes to potions and staffs, which were requested on the forums, and which make the gameplay closer to the original Amiga version of the game.

I have also updated the Lua and ENet versions used and made a couple of other minor changes.

The game can be downloaded from the download page as usual.

Enjoy!

For more information about Knights, please visit https://www.knightsgame.org.uk/.

Summary: In the next release of Knights I will be upgrading the versions of ENet (a networking library) and Lua that are used by Knights. I will also be removing some custom patches to Lua that were being used previously. This will have the following effects:

• Users will have to upgrade their game client (because the new ENet version is incompatible with the old one).
• The Lua code for mods might need to be updated.

Lua Change

I have updated the version of Lua used to the latest version, 5.4.6 (previously it was 5.1.4).

The previous versions of Knights also included a custom Lua patch (written by me) which added a new “&” operator to the language. This was used for merging tables, for example one could write code like the following:

i_wand_of_destruction = kts.ItemType(
  basic_wand & {
    melee_damage = 1000,
    melee_stun_time = rng_time_range(2, 3),
    melee_tile_damage = 1000
  }
)

This would make an ItemType that had all of the properties from the “basic_wand” table, together with the new properties shown.

Adding the “&” operator to Lua seemed like a good idea at the time — but it does cause problems, because it means I can’t easily upgrade to newer versions of Lua (as I would have to port my patch across to the new codebase every time).

So, from now on I am removing this patch, and the “&” table-merge operator will no longer be available.

Instead I have written a Lua function that does the same thing:

function table_merge(a, b)
    local result = {}
    for k, v in pairs(a) do
        result[k] = v
    end
    for k, v in pairs(b) do
        result[k] = v
    end
    return result
end

This function is available as “kts.table_merge”. Therefore the above “wand of destruction” code must now be written like this instead:

i_wand_of_destruction = kts.ItemType(
  kts.table_merge(
    basic_wand,
    {
      melee_damage = 1000,
      melee_stun_time = rng_time_range(2, 3),
      melee_tile_damage = 1000
    }
  }
)

This does make the code a little bit uglier I suppose, but it is not a big problem and I think this is a small price to pay, in exchange for making it much easier to do Lua upgrades in the future.

Of course, any mods that were using “&”, must now be modified (no pun intended) to use “kts.table_merge” instead, but this shouldn’t be too difficult.

ENet change

I have also taken the opportunity to upgrade ENet from version 1.2.5 to the latest available version (on Windows this will be 1.3.17, as this is the latest VCPKG package available; on Linux it will be whatever is installed on your local system, which will hopefully be reasonably up to date).

The main issue here is that ENet 1.3.x is incompatible with 1.2.x, so users running the older versions of Knights will not be able to connect to the new Knights server. They won’t even get any message telling them they need to upgrade; the connection will just fail completely.

To mitigate this, I’ve arranged for the “Connect to Server” screen to show a message like the following when players try to connect:

Hopefully this will be enough to let players know what they have to do, and it won’t be a huge issue.

A new version of Knights has been released!

This is basically just a maintenance release. I have fixed various issues with the codebase and brought the code up to date (e.g. upgrade to latest Visual Studio version, upgraded various libraries to new versions). There are no actual gameplay changes.

The release can be downloaded from: https://www.knightsgame.org.uk/download

I plan to do another release after this one with the following changes:

• Fix the bugs that Knight_Rider has posted about recently (potion changes and staff damage when using super/strength).
• Fix a bug I noticed with chat (pressing the keypad enter doesn’t seem to send a chat message, although pressing return on the main keyboard does).
• More changes to upgrade the codebase (e.g. remove deprecated C++ features, upgrade to latest Lua version).

So stay tuned for that within the next few weeks 🙂

For more information about Knights, please visit http://www.knightsgame.org.uk/.

I’m pleased to announce that I have finally moved away from SVN and moved the Knights source code into Git.

The Knights source repository is now available at: https://github.com/sdthompson1/knights

I’m also, slowly, bringing the code more up to date — e.g. updating to the latest Visual Studio version for the Windows build, moving from SDL 1 to SDL 2, removing some outdated C++ constructs.

This process will continue over the next few weeks, but hopefully fairly soon, I will be able to make a new release of the game! (The first in many years.)

I did a bit more work on the Knights Lua documentation project today. The following functions have been added to the docs:

• kts.AddMissile
• kts.AddMonster
• kts.AddMonsterGenerator
• kts.CloseDoor
• kts.Control
• kts.IsDoorOpen
• kts.LockDoor
• kts.MonsterType
• kts.OpenDoor
• kts.OpenOrCloseDoor

There is still a lot to do, but this is a good start 🙂

For more information about Knights, please visit http://www.knightsgame.org.uk/.