|
October 12th, 2008, 03:08 AM | #1 |
Join Date: Jul 2005
Posts: 145
|
The Minefield of Ring Development
The Ring is a minefield for development. In software design, this means that if you do everything right, it works. If you do anything wrong, who knows? You will probably crash and burn. Get off the path, and you are on your own, and by the way, we don't tell you where the path is.
Olepi's Moon was my first simplistic Ring scenario, and it works pretty well. Now I am trying some more advanced stuff, and I am dying in the minefield. Question to the devs, or mods, is there going to be any attempt to make the Ring solid? Test out everything? fix the bugs? Do you have anyone testing and debugging the Ring? Should I file every problem as a bug? It is pretty discouraging to talk up the unique features of the Ring in Ryzom, and then to realize that the actual code is fragile and frustrating at best. Or crap in reality. edit: just for fun, here is an early list of bugs found in the last few days: - players cannot enter the Ring - Ring scenarios dont show up on the Ring Terminal - players get zoned to some random location, instead of the Ring - mobs wont run routes after being respawned - mobs wont aggro after being respawned - simple timers cannot be re-triggered after the first time Here's a quote I got back after complaining: "Everybody who makes Ring scenarios knows exactly what you mean. So totally random and non-intuitive when a scripted event works and when it doesn't...." is this the software Ryzom is trying to sell??? Last edited by olepi : October 12th, 2008 at 03:38 AM. |
October 12th, 2008, 10:16 AM | #2 |
Join Date: Oct 2006
Posts: 355
|
Re: The Minefield of Ring Development
every time you use a basic component to do something other than its most simple objective you are entering the realm of "programing everything yourself from scratch"
Yes, you can create storage systems for items, counters for mobs killed, counters for hand ins that spawn guards, patrols that dynamically change depending on the current situation in the "world", and even random variable events that are different each time, even if a player chooses option A all the time when talking to an NPC etc. but if your going to make those things, be prepared to plan them out VERY thoroughly. But.. i cant use the ring any more, as my char has had all ring knowledge wiped (and my spawn points on atys) and i cant be arsed to go and get them all again. So meh : /
__________________
DOOMSDAY CAME =( =(
|
October 12th, 2008, 10:36 AM | #3 | |
Join Date: Sep 2004
Posts: 1,573
|
Re: The Minefield of Ring Development
Quote:
That bug has hit a few of us (me included), Allow me to tell you that, although annoying, it only takes a couple of hours to visit Atys so you get Ring knowledge again Yours Ani Last edited by iphdrunk : October 12th, 2008 at 10:59 AM. |
|
October 13th, 2008, 01:00 AM | #4 | |||
Join Date: Dec 2004
Posts: 1,331
|
Re: The Minefield of Ring Development
Quote:
Quote:
Making it more 'solid' may be too much to ask, but I think it would help a lot if we could just get more insight into the components during testing. For example, if my repeating timer doesn't re-trigger after the first time, it's extremely hard to tell what's going wrong. Could be it's not being deactivated, could be it's not being reactivated, could be it doesn't resume counting, could be the time doesn't get reset, could be it just doesn't trigger, could be it does trigger but the resulting event doesn't work (and then there's also half a dozen ways why that event doesn't work). I have to link some kind of separate visual cue to each and every one of those circumstances (and then hope those work correctly) to figure out what exactly is going wrong. It would be so much easier if selecting the timer in testing mode would just show me everything about it in realtime: whether it's active, whether it's counting, what time is currently on it, when does it trigger, etc etc. Quote:
- Can't change the weapon or spell type of NPC bandits anymore, this used to be possible. - Can't change the names of plants and primitives anymore, again this used to be possible. It's relatively minor things, but still not getting current bugs fixed is something we're sort of used to. Watching more and more functions break down over time is really discouraging. |
|||
October 13th, 2008, 04:13 AM | #5 |
Join Date: Jul 2006
Location: Germany
Posts: 271
|
Re: The Minefield of Ring Development
I stopped working on advanced scenarios in the ring for one reason:
Almost all can be done, but I'm not going to spend hours to emulate missing features with other features like a stupid simple forward/backwards counter. I spend more time making up for the missing components then being creative, worst of all are the random bugs for example conditions becoming "..." out of the blue, scenarios break themselves file wise or whole paths vanishing (+ everywhere they been assigned, sigh). The Ring is a cool idea and does what it is supposed to do most of the times, but once you start to make something interesting it's sadly no fun anymore. It's the simple things that lack like status variables, counters (loops!), case selections (emulating them with dialogs and tons of conditons!) or more flexible components like item boxes or other interactive things. And even for the basic scenarios nobody thought beyound 2 acts. With the funny scenery limit counting for ALL maps you can barely fill 2 with stuff. How are you supposed to tell a story starting in the lakes at 2 settings, turning into pr with 3 settings and finally ending in pyr? Heck alone to make a burning broken crate with smoke 3 (three) scenery tokens are used. So when the time comes I have my feature and bug list ready, but I think the ring is the laaaaast thing on their list atm... sadly. |
October 13th, 2008, 10:22 AM | #6 | |
Join Date: Dec 2004
Posts: 1,331
|
Re: The Minefield of Ring Development
Quote:
I have a nice working scenario, tested and works exactly as I want it, so I save it to a file. I then make backup copies of this file just in case. After running the scenario a few times (but without having made any changes), suddenly something works differently and not the way I want it anymore. I check all the settings, but no, nothing's changed. So I figure, okay, one of those out of the blue bugs hit me and I can't find it. But I can just restore a backup of the original file, since I know that one worked perfectly and didn't have the bug yet.... Nope, the restored backup of the original file also works wrongly in the exact same way as the 'broken' file... |
|
October 13th, 2008, 04:59 PM | #7 |
Join Date: Jul 2005
Posts: 145
|
Re: The Minefield of Ring Development
I did figure out one thing about the spawned mobs not following a route. On the first invocation, the mob spawns, and follows the route into the NPC's that will kill him. Then he respawns, and does not follow the route.
The problem is that if the mob is killed before the route is finished, he won't run the route when the mob respawns. If I make the route end just in aggro range, then the mob runs the route, gets aggro, gets killed, respawns, and then runs the route again. If he gets killed before the route is finished, the mob won't run the route again when respawned. The fragility of the Ring makes me respect even more those who have made good scenarios work |
October 13th, 2008, 05:41 PM | #8 | |
Join Date: Oct 2006
Posts: 355
|
Re: The Minefield of Ring Development
Quote:
Its called simple logic Computers are stupid, you have to tell them everything just cover all bases and plan ahead
__________________
DOOMSDAY CAME =( =(
|
|
October 13th, 2008, 06:11 PM | #9 | |
Join Date: Dec 2004
Posts: 1,331
|
Re: The Minefield of Ring Development
Quote:
|
|
October 13th, 2008, 06:33 PM | #10 | |
Join Date: Sep 2004
Posts: 1,025
|
Re: The Minefield of Ring Development
Quote:
I am not completely sure, but I get the impression the mob needs to be told to start the route after spawning. (It's something I never completely figured out, but there is 'autostart' which I suspect to only autostart the first time, and a 'start after event x'. At the end I found the 'start after event x' more usefull, as scripts didnt start how and when I wanted to with 'autostart'. This might be the same thing.) In R2 language: -event: if mob spawns, then follow route x. (Or even: Event1: if mob dies; then mob eats at place y; and spawns. Event2: if mob spawns, then follow route x you might even put in a timer of 1 or 5 seconds and give a message to the GM to tell you where it is the list...) |
|
«
Previous Thread
|
Next Thread
»
Thread Tools | Search this Thread |
Display Modes | |
|
|
All times are GMT +2. The time now is 04:00 AM.