Exception

> I wish my game was finished because this is awesome. :D Again, **it does not have to be finished!** So long as you are ready for players or are close to releasing it to players, I will take your servers on. :}

As I've said to a few others who PMed me, I don't do reservations, and I can see how having two reserved might be a bit misleading. Just to state, those two reserved are ones that I am using, I just wanted to indicate that those ports were being used. In retrospect, it would have been more intelligent for me to just not put them there. Apologies for any confusion!

- Port 4000: Taken (Avian Online) - Port 4001: Reserved - Port 4002: Reserved - Port 4003: Taken (Chaoscraft) - Port 4004: **Free!** - Port 4005: **Free!** - Port 4006: **Free!** - Port 4007: **Free!** - Port 4008: **Free!** **Only 5 more spots left! First come first serve!**

> Hello, I'm interested. I recently picked up Chaoscraft again, and I'm going to need the server to be 24/7 in order for development. We have custom graphics and features. I'll be using an auto updater in order to send updates out to the team constantly. If your interested in this, let me know. > > > > Thread: [http://www.eclipseorigins.com/community/index.php?/topic/134961-chaoscraft-2d-sidescrolling-skilling-orpg/](http://www.eclipseorigins.com/community/index.php?/topic/134961-chaoscraft-2d-sidescrolling-skilling-orpg/) PM me your server files (and any non-standard runtimes if needed) and I'll get you all set up.

> Oh too bad, My game is not done yet D: > Same here. :,( That's absolutely fine, if it's in a playable state and not too buggy, then that's good enough for me – I favour iterative design. :]

Hiya Eclipse, I've got a VPS, and I've not used it too much recently, so rather than let my money go to complete waste, I'm offering 24/7 hosting for your projects! I can offer 1GB of space and an FTP account, however I cannot give Remote Desktop access, so I'm trusting you to ensure your server is relatively bug free and won't crash too much. The IP Address is 85.222.228.199, and I'm willing to offer the following: - Port 4000: Taken (Avian Online) - Port 4001: Reserved - Port 4002: Reserved - Port 4003: **Free!** - Port 4004: **Free!** - Port 4005: **Free!** - Port 4006: **Free!** - Port 4007: **Free!** - Port 4008: **Free!** Backups are taken every Saturday. **Intersted?** **If you are interested, reply here with a link to the project that you would like me to host.** I only request that your project is cycle locked and is relatively active. **Refusal** I may refuse a project if I believe it is not yet ready, if it is a resource hog, or if I find that it errors too easily. If I do not see activity for a week, I will free up the spot for somebody else. My only catch is that I'm unsure of what future use I may have for it, so I reserve the right to suddenly cease running these servers (should that happen, do not worry, I will send every file) if I do have a need for it. For the next few months, that looks unlikely, so it's a non-issue at the moment. :}

I honestly think you should scale Pokémon rendering up by a factor of 1.5 to 2, simply because there's so much empty space. I reckon you'd still have enough space for two partners to each side for Triples. also, don't forget the mega evolution icon.

I find it hilarious that Eclipse is still in the loop of development > excitement > disappointment > anger > suggestion > development ad nauseum Though I suppose I can't really say much, having perpetuated one of those cycles… Anyway, from what I can figure, these things that I learned the hard way apply to Eclipse quite well: **1) Bad habits** The pattern here is that things go wrong around the excitement point, because promise of something new is either overhyped or hyped faaaaaaar too early and people become disappointed or get bored. If you don't break out of that, this will only happen again. When things start to pick up, you set yourselves up for failure or massive delays again by _**upscaling** your targets._ I saw it happen way too many times up until I left, I can still see it happening now, and it happened to me far too many times to count. It was part of the reason I left, I got fed up with game development in itself. I have 6 completed games now but tons and tons of half-finished engines and proof-of-concepts. It should be the other way around, if things go haywire, figure out why it went wrong and simplify matters. Even if you fail again, keeping to that cycle is better because at the very least, something gets _done_ at the end. **2) Too much speaking, too little action** You say you're going to do these things. Well? Do it, then. Focus on getting something functional, and by definition playable. The moment that it's a state where players can input controls to control a character…release it! Follow an iterative release pattern. Develop based on player feedback, stuff like music or pretty graphics can be done as you go along. _Don't_ start by preparing your assets, drawing up multiple plans, interface sketches, etcetc. **3) It's acceptable to quit.** From what I can figure, this is still a stigma here. If a project is draining at you, by using too much time, or draining at your health, or you're simply bored of it and it's admittedly not much fun…drop it. Learn what went wrong, and begin whatever else you want to do and get right to it. Don't even look back to the dropped project. I have probably hundreds of failed projects that failed for their own reasons whether it was out of my scope, or I got bored, or I didn't have enough time to finish it in the foreseeable future (ie. got impatient), etc, and honestly, I don't give a crap, I'm better at what I try to do now for it. tldr: do as I say not as I (mostly previously did) do. also, stop using vb6 ffs.

~~Oh look, it's everybody's favourite lazy (probably narcissistic) pessimist~~ I mean hi everybody. 242 days without VB6 and counting. In other news, since leaving, I've got the grades I wanted, got into the university I wanted, and am studying the course I wanted. Apart from that, I've not done much else; no time. :{

Byebye. I've had enough of mirage-based communities, while the forum members here have genuinely been nicer since the takeover (which I'm sure, a year ago I'd be surprised at myself for saying so), **I don't want, or need, any access or communication to anything that's a bridge between any of the linked communities**. I have no interest in posting on any of these communities any more. Thanks for the fun!

Why is this even a debate? If you want to buy it, PM as requested, if not, stop bloody derailing.

> Abhi2011, look at some different DX8 rendering in some of the custom engines… they render buttons and windows and stuff...instead of using literal controls. Where does Abhi say he wants to use GDI controls? Reading comprehension. I think Abhi is more than capable of using D3D8.

There's something here I feel obliged to point out. You, as members, may believe that Eclipse is a democracy. It isn't. I can't name any one forum that is. Carry on.

Had a go with the demo, here's my comments: * The game is FARRRRRR too small. I understand that's probably your intention, but playing on a bright 720p screen was straining enough. I can't imagine what it'll be like for people playing it on TV screens or for people who can't see very well. Different resolutions should be available. * The controls were not clear. I managed to figure that the arrow keys were for movement (although, because more people are right handed, you might want to consider WASD), and that the number keys were for the hotbar. I accidentally discovered that Z was to drop items, and then after frantically trying each other key to pick it up, I still couldn't find it. I don't know if that's possibly because I moved outside of the hitrange, because I moved a bit before I noticed. Z is also used to start conversations with NPCs, too? That's not very…good, for an action that also drops items. It seems like a recipe for accidents to happen. * Time was too fast, that fast changing text in the top-right was very distracting. Also note that time kept on ticking while in conversations with NPCs; if the time changed to an instance where the overlay would change, it would happen after finishing the conversation. The overlay is also very jarring; either make it transition so it isn't as dramatic, or tone it down, so I'm not instantly blind when it happens. * If you drop an item, move to a different region, and come back, the item is gone. Not very good for obviously important items like the hoe and watering can. * The character walks too slowly. Add running as an option, please! If you're going full on Harvest Moon style, then perhaps implement stamina. * The shop dialogue fades out your money, making it more difficult to read. The money you have should be shown in a larger font while in shops. You should also be prompted when you do not have enough money to buy anything. * The transition between the forest and the beach doesn't make sense. That's everything I could glean from about 15 minutes of playing it. Overall, it shows promise. With enough refinement, I'm sure you'll have no problem getting Greenlit. (I did vote.) :}

This sounds very pyramid scheme-y.

Oh, one more thing: Start small. Don't go too large scale, you will hate yourself for it. I mean it.

My honest candid suggestion is to simply not hang around **ANY** Mirage derivative, if you're looking to actually get anything finished. While it's fun for some, it can be an environment that inhibits the most procrastination possible, and you will have no sense of focus. It's possible to deviate, and there are obviously exceptions…but history proves that's obviously rare. Anyway. Advice. One thing I rarely see mentioned: recognise the difference between a hobby, and something you want to pursue as a career. It's quite easy to let your mental and physical health crumble worrying over a project that you later realise is silly or not worth worrying over. Ensure that you know your limits, and make sure that you sleep well, eat healthy, and exercise. Next, realise that your players don't actually give a crap about your programming prowess. It's important to optimise your code, but computers are fast becoming better and better. You don't have to waste time recoding things unless if it provides problems in active testing. Get something playable out as quickly as possible, even if it isn't finished to your standard. Just release, let others play it, evaluate, debug, release, repeat. I realise now that I have no interest in game development whatsoever (more interested in front end software development + design. I'll stick to playing games, and the occasional small game project to cure boredom), and am unlikely to ever start up a somewhat serious game project again, but if there was anything I wish I knew when I cared, it'd be the above. Ultimately, there's nothing anybody here can suggest that doesn't compare to you simply trying it out. We can advise all we like, that won't prepare you.

OK, I'll try to put it plainly. Currently, the way NPCs work, the server just loops constantly, for absolutely every NPC, and just randomly moves them if no logic is required. If logic is required, then we update constantly with archaic pathfinding techniques. As you can imagine, this is absolutely wasted power, to loop through EVERY NPC on EVERY map, as you can imagine, as more players go on different maps, processing power increases more and more. When we require pathfinding, that increases even more. Eventually, we end up with a huge portion of processing onto a part that doesn't even need to be handled by the server about 90% of the time. So, what can we do? Our options range from letting the client do all of the work, but obviously, we can't do that. People will cheat. NPCs will never function properly. Peoples' gameplay experiences will differ. My solution is to generate and buffer a path for each NPC beforehand. When it's needed, (how I handled it) a queue of about 100 numbers is generated, the numbers range from 0 to 3, 0 for left, 1 for up, 2 for right, 3 for down, so we can literally send a byte array, which is -tiny-. Instead of sending when an NPC moves repetitively, we send the path to the client, and the NPCs initial position, and we let the client do all of the walking work. OK, we have a sound theory, but the problem now is, different people will walk onto maps at different times, their NPCs will act differently! Where one person might be at the 45th step, a person joining starts from 1. This is where the server comes in, alongside the path, the server also increments a single byte counter (or more, if you want varying walking speeds for NPCs). This counter keeps track of where we are in each and every NPCs step. So, we can send this out, meaning that clients that just join will be able to pick up where they were left off, and carry on with a relatively similar experience. When we hit 100 (or however long your byte array is), we roll back to 0, and we generate a new path, and send it out. This means that the only processing for idle NPCs that the server does at this point is a single counter in a simple timer, and a basic procedure to just create a path OK. Pretty good, but this doesn't work for cases where players directly influence the NPC (ie. attacking/being sighted). When this happens, I have a flag that determines whenever an NPC needs to be taken care of by the server or not. If it's being attacked, or it's staring down a player it wants to kill. At that point, I flag this true, and then the server sends out to all clients that their typical behaviour is being overridden. At this point, we process anything that needs to be processed, exactly as you have it now (well, it could be better, but more on that.). When an NPC is done, we generate a new path for it, and it carries on as normal, sending out the new path and its now current position. At this point, we require the following to be done by the server: * Increment a single variable, and loop back to 0 every now and then. When we do loop, generate a new path. * Check whether an NPC based on it's interpolated position needs to chase a player. Set an override flag if this is the case. * Check whether a player has attacked an NPC. Set an override flag if this is the case. * If the override flag is on, process pathfinding to chase players, and attack as you normally would. It's a fairly complex model, but I'm sure it can be simplified, and I'm willing to bet that Stein will either comment on whether some of these actions are necessary/improve on them, but should you choose to to go ahead, you will see a notable decrease in what's being flung back and forth by your server, and a respectable increase in your CPS (which of course, means everything. :p)

To put it plainly, quickly, and /threadly, this is generally a bad idea, for reasons behind cheating, the general convention, and the fact that it's not actually as efficient an idea as you may think it is. EDIT: That being said, I am all for making clients process NPC movement, on the basis of tiny timestamped instructions from the server. I'll explain more if you're interested.

> Is this Custom version Dead or Alive ? Judging by the 4 month gap between your post and the last post, what the bloody hell do you think? ._.

> -snip- Out of curiosity, how much would a basic tileset (so, grass, mud (with transitions), trees, a few rocks, etc) to the same quality cost?

You have two options, the simple (or lazy, depending on your outlook) option, and the technically correct option. The simple option is to simply redraw the image 4/8 times offset by 1 pixel (up/down/left/right and optionally up-left/up-right/down-left/down-right) and make them filled black. Obviously, this will impact performance. Test extensively if you try this out. Your technically correct option (but isn't necessarily always possible, depends on whether your graphics engine allows it), is to look up some shader options. WebGL certainly allows you to use shaders, dunno about anything else. I can't provide any shader code, I'm useless at creating them. :p

…is that character actually riding a dinosaur? o_o

> Honestly most all eclipse games are good because people really put effort into making a game and the stress or confusion stops them from continuing the process. Even if someone's game doesn't please yourself, it pleases someone else and that's all that matters. :P If only we were all as ~~naive~~ optimistic as you, Eclipse may still be the fun state it was back in 2005 - 2009\. :D

Bleh, fair enough. :p