Code Update - Discussion

A place for The Mana World players to discuss game-related topics outside the scope of development including guilds, player interactions, game meta and more.


User avatar
AnonDuck
TMW Adviser
TMW Adviser
Posts: 653
Joined: 02 Jan 2009, 04:19
Location: Catland

Re: Code Update - Discussion

Post by AnonDuck »

t3st3r wrote:Picking apart everything I said
Thank you for making me realize that working on TMW is a waste of time! I shall no longer be helping in any active sense, though I will keep the server running.
Head of the TMW Illuminati
User avatar
Nard
Knight
Knight
Posts: 1113
Joined: 27 Jun 2010, 12:45
Location: France, near Paris

Re: Code Update - Discussion

Post by Nard »

o11c wrote: TMW will never become great if we can't ever change anything. Particularly, it will *never*:
  1. have native guild support
  2. implement professions
  3. be rebalanced
  4. be able to have more than about 100 players online at a time without major performance problems
  5. provide the content developers with the tools necessary to make their jobs easier
  6. be able to add any serious features at all
  7. offer *any* sort of guarantee about data corruption or avoiding crashes
  • The original buggy code had some working native support of some of the features you speak about, before you removed them with no replacement and no real justification but your own opinion (sometimes). You do not even take care to backup and archive the old code.
  • TMW server had 2 years ago more than 110 connected players with lag problems, but not worse than the lag spikes we meet now with just 30 active players :/.
  • Nobody asks that no change are made, but only that they are planned by a TEAM work and released after they have been seriously proofed.
  • The first thing to do to avoid crashing and data corruption is to take care of HDs , and avoid buffer overruns, for example avoid too short save intervals.
"The language of everyday life is clogged with sentiment, and the science of human nature has not advanced so far that we can describe individual sentiment in a clear way." Lancelot Hogben, Mathematics for the Million.
“There are two motives for reading a book; one, that you enjoy it; the other, that you can boast about it.” Bertrand Russell, Conquest of Happiness.
"If you optimize everything, you will always be unhappy." Donald Knuth.
User avatar
Nard
Knight
Knight
Posts: 1113
Joined: 27 Jun 2010, 12:45
Location: France, near Paris

Re: Code Update - Discussion

Post by Nard »

o11c wrote:FWIW, the case-sensitivity made it possible to reduce the startup time of the server from about 1 minute 15 seconds to about a second, and the startup time for the char-server (no player visible change there, but same basic effect) from over 5 minutes to about 5 seconds.
LOL! If so, the case sensitivity is surely not the cause but the way it was written and compiled. BTW you could have had the idea to make a warning mailing to users to remind them their exact case sensitive login parameters. This should not be too difficult and would have shown to users that you cared about them. Another possibility would have been to warn players about the change some time and many times before the release, (news are made for that no?) instead of letting them discover it at login.
"The language of everyday life is clogged with sentiment, and the science of human nature has not advanced so far that we can describe individual sentiment in a clear way." Lancelot Hogben, Mathematics for the Million.
“There are two motives for reading a book; one, that you enjoy it; the other, that you can boast about it.” Bertrand Russell, Conquest of Happiness.
"If you optimize everything, you will always be unhappy." Donald Knuth.
User avatar
Hello=)
The Mana World
The Mana World
Posts: 708
Joined: 11 Jun 2009, 12:46

Re: Code Update - Discussion

Post by Hello=) »

MadCamel wrote:
t3st3r wrote:Picking apart everything I said
Thank you for making me realize that working on TMW is a waste of time! I shall no longer be helping in any active sense, though I will keep the server running.
Very strange conclusion IMO. You see, even best intentions often do not lead to best results. It depends on how its made. If you want to hear "thank you!" from players - maybe you have to interact and cooperate with them and get some basic understanding of how your actions would be seen by others? Just blindly throwing some "cool stuff" might fail to work and cause a lot of swearing.

For example, o11c surely had good intentions and did a remarkable job on server's code. But he haves very specific views, usually does all things on his own with minimal interactions with outside world, etc. This caused simple fact: update landed. Poorly tested and bugged. Also poorly thought in some regards. IMO, o11c failed to get idea of actual impact from his actions. Result? Collateral damage has been quite large. For example, its nice to assume everyone reads news, can remember and type login correctly, etc. However, it's overoptimistic. I wish it would be true, but its not a case. Reality is different: people having trouble to log in to server here and there. I see them on regular basis. Some new fresh alts popping up, asking "WFT?!" and swearing. On personal note, "cool new update" has actually borked my gameplay to a degree where I have trouble hunting stuff to maintain self. And it's still outstanding issue. I wish I could tell him "thank you!" but given circumstances I have trouble to do so. It's hard to thank someone for breaking my gameplay, isn't it?

Edit: hmm, o11c has been impressive in fixing his bugs so mumblings about borked spells are no longer relevant.
Last edited by Hello=) on 29 Sep 2013, 16:39, edited 1 time in total.
User avatar
Hello=)
The Mana World
The Mana World
Posts: 708
Joined: 11 Jun 2009, 12:46

Re: Code Update - Discussion

Post by Hello=) »

o11c wrote:FWIW, the case-sensitivity made it possible to reduce the startup time of the server from about 1 minute 15 seconds to about a second, and the startup time for the char-server (no player visible change there, but same basic effect) from over 5 minutes to about 5 seconds
I admit it makes sense. But I can also admit restarting server is rare in normal course of events, so does it really outweights fact quite many players would fail to log in? Some maybe even abandon the game. IMO, changing anything related to logins in player visible ways should be last resort thing and avoided like a plague if possible.

p.s. and hmm, you've been fast on fixing bugs. Thanks for fixing that one with spells, this one have been quite nasty. After all I should admit that while your update has been bugged (and I foretold you this IIRC :wink:), your bugfixing still damn impressive.
Last edited by Hello=) on 29 Sep 2013, 16:55, edited 1 time in total.
User avatar
o11c
Grand Knight
Grand Knight
Posts: 2262
Joined: 20 Feb 2011, 21:09
Location: ^ ^

Re: Code Update - Discussion

Post by o11c »

t3st3r wrote:p.s. and hmm, you've been fast on fixing bugs. Thanks for fixing that one with spells, this one have been quite nasty. After all I should admit that while your update has been bugged, your bugfixing is damn impressive.
Yes, I try to fix bugs very quickly, whether they are reported during testing or on the main server. I admit I'm sometimes a bit slower if it's only on the test server though.

The magic elements bug is a great example of why we need a variety of testers. When I used magic, it seemed to work ... but either I did not test with the few mobs that have a bonus/penalty element, or I was just not familiar enough with the exact kind of numbers that magic usually produces. I spend most of my time playing as an archer.

The case-insensitive login thing is the one issue that I wish *could* have been done differently, but there was no other option since: 1. directly editing the savefiles is not something to be done lightly, 2. we have no way to contact most of the affected players.
Former programmer for the TMWA server.
User avatar
Hello=)
The Mana World
The Mana World
Posts: 708
Joined: 11 Jun 2009, 12:46

Re: Code Update - Discussion

Post by Hello=) »

o11c wrote:Yes, I try to fix bugs very quickly, whether they are reported during testing or on the main server. I admit I'm sometimes a bit slower if it's only on the test server though.
Hell yeah, it's impressive. I can imagine you had a lot of headache instead of thanks with this update and spent enough time fixing bugs. And it's not like if I like to rant instead of telling "thanks", btw. I just think it could be better for both you and players. And yeah, while I dont like many things in this particular update, I can admit you had some valid points and I cant deny you did impressive job on fixing most pressing issues in a timely manner.
The magic elements bug is a great example of why we need a variety of testers. When I used magic, it seemed to work ... but either I did not test with the few mobs that have a bonus/penalty element, or I was just not familiar enough with the exact kind of numbers that magic usually produces. I spend most of my time playing as an archer.
IMO, root cause here is the fact you've touched really many places in eA. Ideally, this requires full retest of literally everything. I'm unable to imagine who can do this reasonably well in such amounts. That's why I foretold there would be nasty bugs/issues. Experienced QA sometimes can predict things :mrgreen:. And I have feeling I had some chances to prevent at least some bugs from hitting players. But overall list of player-visible changes in this update has been demotivating enough to stay away.
The case-insensitive login thing is the one issue that I wish *could* have been done differently, but there was no other option since: 1. directly editing the savefiles is not something to be done lightly, 2. we have no way to contact most of the affected players.
The main issue here is that now some players just unable to log in. Then some quite obscure and scary error message given. And to read news on update player should be able to actually log in. Generally it's a bit like pkunzip.zip. IMO, any messing with player accounts should be an absolutely last resort thing, when skies fall on your head and you see absolutely no any chances to avoid it. Same applies to account DB purges, btw. You see, I've seen quite many returning old-time gamers who lost everything. Needless to say - [x] that suxx. I do think its really incorrect to screw up those who were part of game history. Not to mention I never seen any positive outcomes from any attempts to mess wuth players accounts for any reasons, even if it has been done with best intentions. So as for me, if you ever want to try to change something in login stuff, at very least there should be no player visible changes. Else it causes a lot of headache to players and you doomed to face backslash. Because inablilty to log in is the absolutely worst issue player can encounter at all. Needless to say players would switch to aggro mode and blame you a lot, All things related to logins/chars, player posessions, etc are really delicate.
User avatar
o11c
Grand Knight
Grand Knight
Posts: 2262
Joined: 20 Feb 2011, 21:09
Location: ^ ^

Re: Code Update - Discussion

Post by o11c »

t3st3r wrote:IMO, root cause here is the fact you've touched really many places in eA.
I'm aware of this ... but the only thing worse than changing the code is *not* changing the code.

People keep asking for features - whether they were partially implemented in eA or not - but TMWA is not extensible without a lot of prepration work.

Great example just mentioned on IRC: all the nasty corner cases that happen when you have 100 stacks in your inventory. They could be solved if 100 was a soft limit, and scripts could add beyond that.
Former programmer for the TMWA server.
Post Reply