Jump to content

twostars

Administrators
  • Content count

    994
  • Joined

  • Last visited

  • Days Won

    27

Everything posted by twostars

  1. Then yes, that most definitely sounds like a crash. They tend to happen occasionally when we check in a lot of changes and don't quite check over them as well as we should before pulling them into alpha (usually because they're stuff we'd like to get tested ASAP, such as the skill changes). We'll be less update-happy for beta/official. At the moment alpha is usually always kept up-to-date with the state of the current code-base, but we'll be setting up a stable branch that beta/official will use with thoroughly tested changes. To be sure this was indeed the case (as no specific times were provided, but it really shouldn't be anything else), if this happens in future, can you please let me know on here or IRC? Thanks
  2. The short disconnect probably is due to speedhack checks. I've just disabled them. As for the timeouts, they're something that needs looking into, provided they weren't because of server crashes (we fixed a couple of crashes that came up as a result of recent changes). Can you be more specific with times? Regarding lag, what exactly are we talking about here? Everything's delayed by an extended period of time including chat? It's just teleporting you back?
  3. I assume you've already tried restarting the launcher, so perhaps try adding the client folder to your antivirus' whitelist/exclusions. Alternatively, you can download it manually from here: http://download.apexko.com/upgrade/patch2038.zip If you choose to manually patch it, be sure the files go directly inside the ApexKO client folder, and not in a subfolder like "patch2038". PS: There's really no good place for this right now, but I've moved it to the closest fit ("Game questions"). This really isn't a bug report, per se (though it is undesired behaviour we'll address later).
  4. That's entirely possible.
  5. Are you sure about that? Double-checking against 1.298, it appears to be handling it -- and we're handling both of these things the same.
  6. Firstly, I should point out that again, we always read ingame chat. In fact, all (public) chat (region/shouting) is fed straight to our dev IRC channel. If it's said in public, you don't have to worry -- we're aware of it. Having said that, I should also point out that we don't just make adjustments based purely on what people say, for obvious reasons. We always double-check what they're saying is correct before we do so. Additionally, any damage adjustments we make aren't specifically made to "increase damage by x" or whatever. We step through our damage calcs to figure out where it's diverging from official damage calcs, so the amounts produced are the same. As a result, physical damage is now fine -- we have official calcs for them, which never really changed on USKO. Magic damage however is 'iffy', because we don't have new enough official calcs to be able to step through the entire process & verify each amount it calculates (base damage we can, but when it comes to elemental damage and other new behaviour, things have changed a bit), so instead we can really only check the final damage amounts against USKO and figure out how/why they're being produced differently. A much slower process. In the state they're in right now, they appear very close to accurate -- at least for most things we've specifically looked at. Damage calcs are an ongoing, evolving process, so if there's anything specifically wrong happening you've noticed against same stats/items on USKO, please explain it so we can check it out. The one thing I do know for certain is that mages were indeed OP before, though. This was because there was a bug in our damage calcs causing it to produce insane damages which isn't correct for any version of those damage calcs we have. Aside from earlier on, this is really the only time its calc was touched. Also, people keep saying the damage of specific skills is too low (e.g. Helis, though it's a priest skill, it's still calc'd the same as mage skills). Important thing to remember is that this is all connected; so if we fix one, everything else should be improved (at least, everything else calculated the same way). So what would be really helpful at this stage is to do up some damage comparisons for ApexKO/USKO (checking amounts against resistances, different staff AP, etc), so we can get an idea where exactly things are still breaking (e.g. bonus elemental damage not being calculated right, resistances, etc).
  7. I really want to outright replace the entire patch system at some point, but I think it's because there's a limit of patches that can be downloaded at any given time (number was somewhere around 30 if I recall correctly). It probably closes to force it to start again (just in case there are more patches), though I really can't justify that behaviour over (repeatedly) just starting the update process again until we're up to date. Off-hand, I'm not sure how much work it'd be to change it, but it's possible the time would be better spent either rewriting our own implementation of it (perhaps to behave - otherwise - the same as the current launcher, for now) or even reimplementing it fully. :/
  8. At peak, on the weekends there were actually 15-20. It'll dwindle of course (especially now) as things to test dies down or general motivation to test various things drops, but we're working fairly quickly on the larger issues (as fast as we possibly can, anyway), so hopefully interest should last long enough to get these major items tested. The number of players really don't matter anyway; these people have volunteered so much time to providing us with quality feedback that I doubt you'd even be able to get from a larger population (really, more people means it's harder to 1 on 1 with everyone). As for launching alpha in this state, we could have waited and polished it more for general testing (and actually, I really didn't want to release alpha in this state), but admittedly, by doing so we've gained far more input/feedback than we would have ourselves, and things have been moving so much faster.
  9. Speedhack checks are disabled while we implement better support for latency verification (official completely removed their own means of doing so). I need to do more work with the client for 2.038, but after that's done we should be able to re-enable it on alpha for testing.
  10. Alpha stage 2 progress

    Just a note, these damages were fixed yesterday.
  11. All chars wiped

    It was (is) alpha though, wipes/rollbacks/insane breakage should have been expected regardless. We needed to wipe for a second stage as everyone had injected themselves with +30 items and what-not, which isn't really conducive to testing damages etc.
  12. This is implemented for 2.038, though we'll probably have to add server-side checks (and preferably client-side checks similar to attempting to exit) to prevent it from being abused in PVP and such. May or may not also have issues with effects not being reset (noticeable issue for GMs, anyway -- not sure if it's likely to be an issue with players), so that may be something we'll have to look out for as well.
  13. Right now we have notices in chat & scrolling at the top whenever a nation clears a room, i.e. "The <nation> nation has cleared room X / 3." It has been suggested, however, to remove this because it'd add unnecessary pressure to the other team: We're personally in favour of these because they provide some indication of what the other team is doing (the event is intended to be/described as a race to the finish), but we'd like to get an idea what everyone else's opinion is on it before doing anything with it. Thanks!
  14. Well, I'm convinced. As does the poll (100% for yes!). Small compromise, but I'll have the notice sent to that nation only (i.e. "You have cleared room x / y."). Edit: This is done.
  15. Sure, I have no objections to new quests/culling out useless quests. Chain quests being broken right now though is just something we'll be fixing, regardless of whether or not the existing quests themselves actually get used.
  16. They'll be fixed eventually, even if it means manually doing so. As all quest scripts are being generated by a tool we've written, we want to avoid as much manual intervention with quest scripts as we can -- automating script generation for those types of quests (well, all types of quests...) is preferred to writing them up manually. This means if they add new quests, or things generally change, it's just a matter of re-running the tool on the new data and voila -- new, updated quests in mere seconds... as opposed to hours of tedious back and forthing over data manually. Whether they actually stay is up for debate, I suppose.
  17. Server's currently in closed alpha at the moment. You can still apply for testing via the website (register -> "Become an alpha tester" -> fill out the form and submit). Also, please refrain from spamming your question in every topic in these forums. You only need to post it once.
  18. I think that for right now, broad testing like this is exactly what it needs; there's clearly a lot of problems out there, so the more people testing whatever they currently feel like at the time, the better. So far some pretty crazy things have come up that I'm not even sure we'd have thought to specifically test. As such, I think it'd be a better idea to wait until we have most of the serious bugs under control, and then attempt more targeted testing. Not sure if a full wipe is really needed for that (be ideal, though), but if so we'd increase the rates to compensate.
  19. ReOpen?

    Shortly, sorry -- just took a bit longer than usual to address a bug before restarting. Server's compiling right now, so it'll be up in a few minutes. Edit: It's up now.
  20. Level range for BDW

    This is done, though the client will still show the old level bracket until we next update it.
  21. Alpha Schedule

    Good question. Actually, they've always been saved both server-side and client-side. I assume it'll only use the server's skillbar data if the client doesn't already have data for your character. On older clients this was stored in the registry, so permission issues caused it to act up -- but with newer clients like this, they're stored in files in the client's folder. I've not had any problems with my skillbar data since.
  22. Alpha Schedule

    Really, I think having an alpha testing stage on a project like this is more than justified. Most other servers just use the official binaries, which means they really only need content testing. However, we are not using the official binaries; we've implemented every (server-side) feature ourselves, which means there's a great deal which needs extensive testing before we can even really start to think about content issues.
  23. Alpha Schedule

    Not officially -- not for a while yet. We're currently at the alpha stage, which essentially means we're testing functionality, and getting feedback & suggestions on things before it's polished. We still need to go through: - Closed alpha (what the alpha applications & testing schedule are for), where we'll test the existing functionality within a controlled environment. Alpha is extremely raw, so really we're making sure everything's working before proceeding with polishing & setting things up properly, using the community's feedback on various things. - Open alpha, where we'll open up testing for more widespread testing/feedback (not to mention stress testing). Again, this is raw so not much should be expected. (after further development/polishing from alpha input) - Closed beta, where we'll be doing final tests of polished functionality (and custom content) in a controlled environment identical to production (i.e. the one we'll be using for our official launch). - Open beta, where we'll open up these final tests for everyone to access. Further stress testing, etc. - Official! (finally) It may seem a bit overkill, but we really, really want to be sure things run as smoothly as they can when we do release. The server is pretty complete, though, so we don't think the alpha testing stage should last terribly long -- just long enough to run over all the functionality to verify it's all working correctly, and get our remaining tasks [including our custom content] finished before beta. Right now we're still looking over alpha applications, but you can still apply -- just register on the website (http://alpha.apexko.com) and submit an application ("Become an alpha tester"). We really appreciate everybody who's interested in helping us improve the server!
×