Clonk-Karl: Log for meeting on March 8.

2010-03-09T19:43:49Z

Log for meeting on March 8.

New page

== Summary ==
* Mimmo_O: Ring menu to be committed soon(?)
* ck: Investigated animation bug ([http://bugs.openclonk.org/view.php?id=142 #142]) further. Probably not an engine problem.
** Wrote a simple [http://gitorious.org/ckview OGRE mesh viewer].
* Quick slot bar for inventory, Clonk might carry up to 5 items with backpack
** Common items should be quickly accessible without having to eye-scan the inventory

== Full Log ==
<pre>19:03:37<!ck> meeting anyone?
19:04:25<!Zapper> not much time atm - i could tell that I have continued to work on the weapons and am not quite happy with them, though
19:05:11<!ck> what's the problem?
19:05:43<!Zapper> well, my aim is not to make them especially realistic or anything but fun to play - and I have not yet reached that imo
19:05:54< Randrian_> am am here too. Newton is away at the supermarket but will be back soon he said
19:05:55<!Zapper> probably because of some minor things. I just have to fiddle around with them I guess
19:06:31<!ck> so "just" playing around with parameters basically?
19:06:40<!Mimmo_O> sry for disturbance. how do i give a clonk more inventroy place?
19:06:53-!- Randrian_ [~randrian@83.171.179.euirc-7ddcc636] has quit [Connection reset by peer]
19:07:37-!- Randrian_ [~randrian@euirc-ff30a42a.212.114.236.62.ip-pool.NEFkom.net] has joined #openclonk-dev
19:08:28<!Mimmo_O> got it
19:09:31<!ck> so I did investigate that animation problem (bones at incorrect position, #142) again. Basic conclusion is that I don't think it's an engine bug.
19:09:36<!ck> Randrian_: Maybe we can discuss that after the meeting
19:11:10< Randrian_> hmm ok
19:11:27< Randrian_> sounds like i have the "Schwarzen Peter"
19:14:01<!Mimmo_O> ck:
19:14:34<!Mimmo_O> http://84.143.132.17/errors2.jpg those pictures were drawn on the circles via SetGraphics. as you can see, only objects with 3d-models are moved. they move more right, the further i zoom out
19:15:11<!ck> I have seen a similar behaviour with the HUD lately
19:15:19<!Mimmo_O> kk
19:15:25<!ck> not sure whether it's the same problem
19:15:59<!ck> Anyway can you file a bug about it?
19:16:25<!Mimmo_O> i could, but i dont have an account yet
19:16:31<!Mimmo_O> ill do it later
19:17:00<!ck> Please don't forget, or otherwise I might forget about the bug :P
19:17:07<!ck> thanks :)
19:17:13<!Mimmo_O> k
19:17:40<!Mimmo_O> by the way. i have made a ringmenu and i am about to generalize it, for other use. Newton thouht it could be useful
19:18:55<!ck> hm actually I think this is the same bug as the upperboard bug
19:19:14<!ck> Which is already filed in the bugtracker ;)
19:19:56<!ck> sure@ring menu :)
--- Log closed Mon Mar 08 19:26:55 2010
--- Log opened Mon Mar 08 19:27:22 2010
19:27:22-!- ck [~ck@euirc-8e298988.versanet.de] has joined #openclonk-dev
19:27:22-!- Irssi: #openclonk-dev: Total of 29 nicks [0 ops, 0 halfops, 1 voices, 28 normal]
19:27:22-!- mode/#openclonk-dev [+ao ck ck] by ChanServ
19:27:57-!- Irssi: Join to #openclonk-dev was synced in 35 secs
19:27:58<!Mimmo_O> do you want to test it?
19:28:15<!ck> Randrian_: Can Blender tell you the x/y/z coordinates of a bone?
19:28:47< Randrian_> hmm it can tell me some coordinates, but I don't know if they are the same as ingame.
--- Log opened Mon Mar 08 19:33:57 2010
19:33:57-!- ck [~ck@euirc-8e298988.versanet.de] has joined #openclonk-dev
19:33:57-!- Irssi: #openclonk-dev: Total of 29 nicks [0 ops, 0 halfops, 1 voices, 28 normal]
19:33:57-!- mode/#openclonk-dev [+ao ck ck] by ChanServ
19:34:25<!Mimmo_O> ck: do you want to test the ring menu?
19:34:25<!ck> hnn. Why is nvidia the only vendor who is capable to create graphics drivers which do not crash all the time on Linux!?
19:34:36-!- Irssi: Join to #openclonk-dev was synced in 39 secs
19:35:08<!ck> Mimmo_O: not right now, sorry
19:35:22<!Mimmo_O> ok
19:35:30<!ck> Randrian_: I think I mean the position of the bone relative to the origin of the model
19:39:25-!- Randrian_ [~randrian@euirc-ff30a42a.212.114.236.62.ip-pool.NEFkom.net] has quit [Ping timeout]
19:49:16<!Newton> hi, i am here
19:49:21<!Newton> but much later than expected
19:49:24<!Newton> so i have to go again
19:49:37<!Mimmo_O> :<
19:49:40<!Mimmo_O> ringmenu finished..!
19:50:13<!Newton> in a few minutes
19:50:22<!Newton> yeah, read that
19:50:30<!Newton> I'll have a look at it later
19:51:01<!Mimmo_O> since its not polished yet and maybe need some changes, i wont submit it now
19:51:16<!Newton> I btw have nothing to report. Well, I finished the TGoW scenario (see blog) and renamed all the IDs plus some minor cleanup stuff
19:53:15-!- Randrian_ [~randrian@euirc-71fb6b01.83.171.155.103.ip-pool.NEFkom.net] has joined #openclonk-dev
19:53:44<!ck> 19:35:30 <!ck> Randrian_: I think I mean the position of the bone relative to the origin of the model
19:55:27<!Newton> well, cya later
19:55:37-!- Randrian_ [~randrian@euirc-71fb6b01.83.171.155.103.ip-pool.NEFkom.net] has quit [Ping timeout]
19:55:52<!Mimmo_O> cu
20:10:22-!- Randrian_ [~randrian@euirc-8b427818.83.171.176.201.ip-pool.NEFkom.net] has joined #openclonk-dev
20:11:52<!ck> 19:53:44 <!ck> 19:35:30 <!ck> Randrian_: I think I mean the position of the bone relative to the origin of the model
20:11:52<!ck> :)
20:12:34< Randrian_> did you recieve my last posts? <Randrian_> well, i have looked up the bone pos_tool1 in the CarryArms animation. The coordinates blender displays actually change during the animation, while the bone doen't move during the animation, which seems strange
20:12:34< Randrian_> <Randrian_> perhaps this is related to this bug.
20:12:56< Randrian_> seems really stange.
20:13:07< Randrian_> this could be a result of my rig.
20:13:37< Randrian_> But I don't think other games alway use just direct bone animation instead of creating them by poses.
20:14:10<!ck> what do you mean by that?
20:14:59<!ck> I wrote a simple mesh viewer using the OGRE libraries and it shows the exact same bone position as the Clonk engine does
20:15:26<!ck> I also attached the musket in OGRE in the MusketLoadArms animation, and it has the same problem
20:15:57<!ck> also, the legs of the clonk are "reverted" when (only) playing MusketLoadArms, both in OGRE and in Clonk
20:16:00< Randrian_> well you can rotate/translate the bones in blender directly. But it is ofthen easier to have a rig with constraints, so that you can say where the foot should go and blender automatically calculated how the bones have to be moved.
20:16:22-!- Mortimer [~Mortimer@euirc-0650b51d.dip.t-dialin.net] has joined #openclonk-dev
20:16:43<!ck> (see http://arbur.net/clonk/ckview.png)
20:16:46< Randrian_> well, the legs don't have keyframes, so there can't be a fixed behaviour for them I think.
20:17:05< Randrian_> oh hmm
20:17:26<!ck> I would have guessed they should just appear as in the unanimated mesh
20:17:53< Randrian_> btw. can you commit this mesh viewer somewhere? It could be usefull for making meshes.
20:17:59< Randrian_> hmm, I would have thought so too.
20:18:22<!ck> It's in a git repository. I can publish it if you want
20:18:49< loriel> the best kind of repository
20:19:15< Randrian_> well, if it has some kind of interface, so that the animations and things aren't hardcoded.
20:19:29<!ck> Yes, a command line interface :)
20:19:37-!- Ape [~ape@euirc-7fdc53a9.dhcp.inet.fi] has quit [Client exited]
20:19:37-!- Ape3000 [~ape@euirc-7fdc53a9.dhcp.inet.fi] has quit [Client exited]
20:19:55<!Isilkor> the best kind of interface
20:20:02< Randrian_> yes, this sould be enough. (at least for me)
20:20:49<!Zapper> by the way - why did we kicked out the "backpack" inventory space?
20:21:02< Randrian_> ask newton
20:21:04< Randrian_> he did it
20:22:43<!Zapper> :(
20:23:27< Randrian_> it messed all the levels up with 3 start items
20:24:01< Matthi> because it's not just an additional inventory slot, but a low level game design decision to introduce a real inventory
20:24:31<!Zapper> aaand?
20:25:43<!ck> Randrian_: So do you think it could be a problem in the mesh or the exporter?
20:25:46< Matthi> what exaclty are you missing now?
20:25:58< Matthi> the button which opens the backpack?
20:27:16<!Zapper> we _have_ a backpack?
20:27:34<!Zapper> I want "fast" access to more than two items
20:27:38< Matthi> yes
20:27:46<!Zapper> currently we dont have one slot more than in CR - we actually have less space
20:27:49< Matthi> fast as in "press a keyboard button to bring up the inventory"
20:27:52<!Zapper> considering that we outsource everything as a tool
20:28:09<!Zapper> if I have to select the item with the mouse afterwards, I wouldnt consider it fast
20:28:15< Randrian_> ck: yes I think so. Perhaps more in the armature/rig and how blender handles this.
20:28:32< Matthi> Zapper: Hm.
20:28:40<!Isilkor> I feel like I should bring up the MGS style inventory again
20:28:45<!Zapper> I mean: The idea is not bad, since nearly every solution for the 3-space-inventory was only for _3_-space inventories
20:28:49<!Zapper> mgs?
20:28:52<!Zapper> metal gear solid?
20:28:55<!Isilkor> yes
20:28:56<!Zapper> never played it
20:29:58<!Zapper> I have nothing against selecting my item from the backpack with the number keys, though
20:30:07< Matthi> Zapper: but however you turn and twist it, you'd always have some sort of limitation on how many items you can access directly without navigating through the inventory
20:30:25< Matthi> The question to answer would be "how many different items do you need on speed dial"
20:30:26<!Zapper> yepp
20:30:48<!Zapper> I would at least map the number keys to the first (few) row(s) of the backpack
20:31:04<!Isilkor> all the number keys?
20:31:05<!Zapper> in "normal" gameplay you would be able to access at least 5 items at a very high speed
20:31:07<!Zapper> sure
20:31:20< Matthi> mh, nice idea, but I wouldnt: this would require you to keep your backpack organised all the time
20:31:28<!Zapper> well
20:31:35<!Zapper> Not more organized than every other solution not using a mouse
20:31:46<!Zapper> But I actually find it ok
20:31:48<!Isilkor> also, it wouldn't really work well on Newton's SNES gamepad
20:31:51<!Isilkor> BAM killer argument
20:31:55< Matthi> I'd create 5 to 10 "quick access" slots resembling those two we have right now
20:32:09<!Zapper> without a backpack?
20:32:12< Matthi> you could then "fill" those with references to backpack-items
20:32:39<!Zapper> how would that not force the player to organize stuff? :x
20:32:47<!Zapper> you would actually have to organize even more
20:32:59< Matthi> no, not really.
20:33:01< Matthi> Here's why:
20:33:12<!Zapper> having the item at the right place in the backpack AND having the slot you want mapped to the right backpack position
20:33:22< Matthi> Imaging I set my stack of Flints on quick access button "1".
20:33:27< Matthi> *Imagine
20:33:47< Matthi> So, now, I run out of flints, causing my backpack to have an empty spot
20:34:20< Matthi> I collect a loam now - YOUR system would suddenly have that loam on speed dial 1
20:34:36< Matthi> mine wouldn't ;)
20:34:43< Matthi> let's take that further
20:34:51< Matthi> after doing that, I buy new flints
20:35:11< Matthi> in your system, I'd have to check where in my backpack they went and maybe shift them around
20:35:45< Matthi> In my system, you wouldn't have to do a thing, because your speed dial is "use flint", and that works independant from its backpack-position
20:35:58<!Zapper> but when your items are not put into speed dial slots automatically you would have to select them every time (maybe in a fast fight!) when you really want to use your collected items
20:36:27<!Zapper> your system would remember the IDs? So you would have to have one slot for flints, one for superflints and so on?
20:36:37< Matthi> for example, yes
20:36:51< Matthi> hm, that argument is also valid @quick access to collected items
20:36:56< Matthi> tricky
20:37:16<!Zapper> i'd actually rather put the items on speed dial slots and maybe have to shift them around _after_ the fight than not being able to access the items in the fight :/
20:37:30<!Zapper> or _before_ the fight when you leave the base
20:38:07< Matthi> So you are saying, that the quick-dial slots don't reference the backpack items-slots, but are items-slots themself?
20:38:17< Matthi> .. works for me
20:38:28<!Zapper> not necessarily
20:38:43<!Zapper> they still can reference item slots
20:38:49<!Zapper> but I would let them reference the first few
20:38:56<!Zapper> ill be afk right now - clonkine wants to play hon :o
20:38:56< Matthi> hm
20:38:58< Matthi> okay
20:39:07< Matthi> sounds better for fast melees after all
20:39:18< Matthi> how tells newton that his controls wont work anymore?
20:39:20< Matthi> *who
20:40:05<!ck> Randrian_: Ok try "git clone git://gitorious.org/ckview/ckview.git"
20:40:41< Randrian_> ok, have to install git first
20:40:56<!ck> You'll also need ogre dev libs and ois dev libs to compile it
20:41:11< Randrian_> oh puh...
20:41:57<!ck> what distribution are you using?
20:42:42< Randrian_> ehm kubuntu
20:43:17<!ck> should be quite easy to install there. If I had to guess the package names I'd say ogre-dev and ois-dev. Or maybe libogre-dev and libois-dev ;)
20:43:53< Randrian_> what is this ois?
20:44:25<!Isilkor> an input lib
20:44:55<!Isilkor> libogre-dev and libois-dev are in universe
20:44:57<!ck> The OGRE example code was using it, so I kept it. Didn't know it before either :)
20:45:24<!Isilkor> ck: I think it was part of Ogre before, but they split it out
20:45:46<!ck> ah, makes sense
20:58:56< Randrian_> ok mesh viewer works
20:59:09-!- Luchs^away is now known as luchs
20:59:16< Randrian_> and I have seen that walk actually doesn't look very good...
20:59:33<!ck> You figured out the commandline parameters? Cool ;)
21:00:07<!ck> There's also a third one to only show a specific keyframe of an animation
21:00:11< Randrian_> well first parameter mesh second animation?
21:01:04< Randrian_> and what about attachments? or at least showing the bones?
21:01:33<!ck> I hardcoded that for testing, sorry :|
21:02:11<!ck> Also attachment is not too easy because OGRE does not allow to specify a bone for the attached mesh, just for the one to attach to I think
21:03:01-!- alexX_ [~alexX@euirc-1aa4b4f0.dip.t-dialin.net] has quit [Connection reset by peer]
21:03:24-!- alexX_ [~alexX@84.155.209.euirc-73721097] has joined #openclonk-dev
21:03:52< Randrian_> ah ok, I thought you had this already implemented, cause you said you tested it,.,
21:04:24<!ck> I did, but it's not in that public repository :)
21:04:54<!ck> I can give you the code but I didn't want to commit since it's not very ugly and hacky and whatnot
21:05:30< Randrian_> ah ok
21:05:38<!ck> er
21:05:42<!ck> it _is_ very ugly ;)
21:06:33< Randrian_> I didn't really notice you wrote "it's not" instad of _is_^^
21:07:17< Randrian_> but the viewer is already a good tool I think.
21:07:47< Randrian_> But I'll hope that I don't have to change too much with the armature of the clonk... that coule mean to break a lot of animations...
21:08:05< Randrian_> I already had to redo some animatione when I introduced ellbow control constraints.
21:08:07<!ck> yeah. Can still be steadily improved. If you have a fix or so for it let me know and I'll push it
21:08:27< Randrian_> ok
21:10:02<!ck> btw I think the clonk model in the repos is not the latest so don't get confused about that
21:10:17<!ck> Just because I did get confused ;)
21:10:49<!ck> It's just that my version of OGRE (1.6.5) refused to load the current one which was created with a newer version of the XML->Binary converter
21:13:56-!- Randrian_ [~randrian@euirc-8b427818.83.171.176.201.ip-pool.NEFkom.net] has quit [Ping timeout]
21:15:18-!- Gurkenglas [~Gurkengla@euirc-5f8b9ce3.pools.arcor-ip.net] has quit [Connection reset by peer]
21:25:49<!Zapper> ] <Matthi> how tells newton that his controls wont work anymore? <- @his gamepad? well, he can still use the slow selection if he wants to :)
21:29:12-!- Randrian_ [~randrian@euirc-5d9889c4.83.171.160.200.ip-pool.NEFkom.net] has joined #openclonk-dev
21:32:42-!- B_E is now known as b_e
21:33:22< Matthi> Zapper: to get it together in a short sentence: we're going to add 3 additional quicklist-slots and a real inventory whose first 5 slots represent those in the quicklist, right?
21:34:22<!Zapper> I would sign that
21:34:26<!Zapper> but why 3 here and 5 there?
21:38:13< Matthi> err
21:38:16< Matthi> well
21:38:22< Matthi> we already have 2 in the quicklist
21:38:22< Matthi> so
21:38:29< Matthi> add three and you get 5
21:38:44< Matthi> and those five match the five in the backpack
21:39:05<!Zapper> so the hands are backpack slots, too?
21:42:37-!- luchs is now known as Luchs
21:47:52< Matthi> well, how else would you manage it?
21:49:25< Matthi> I thought that you could shift hands around in those 5 quick slots, just like before
21:49:36< Matthi> how else would it be a quick access?
21:54:17< Randrian_> I would rather swich items around. So you then have Slots: A, B, 1, 2, 3
21:54:28< Randrian_> and Slots A and B are your hands
21:55:34< Matthi> so you have 3 different slot-positions with 3 different functions: Hands, Quick-Access and Backpack
21:55:43< Matthi> I don't approve
21:56:21< Matthi> also, switching items kills the basic though behind quick-access: accessing one item by pressing a button i can remember
21:56:36< Matthi> I put flints on 3, so i expect it to be on three
21:58:35-!- JCaesar is now known as JC-weg
21:58:35< Randrian_> hmm, in a normal fight the items (and item types) may switch fast, so I don't think it is a good thing to set slots just to one object/id
21:59:31<!Mimmo_O> where is Newtooooon
22:01:52< Matthi> randrian: but switching item doesn't help with that problem
22:02:25< Matthi> If I wan't another Item than those in my hands right now, I have to press a button, there no way around that
22:02:29< Matthi> *want
22:03:30< Randrian_> yes, but if I have 3 slots beside my hands I can press buttons 1-3. But if these slots are just filled when the items are right this isn't good.
22:03:59< Randrian_> I would just have two hands and 3 extra slots no additional backpack then. this would be a bit much I think.
22:04:10< Matthi> 1) No, 2) .. I don't get it.
22:04:47< Matthi> I want 5 slots, numbered 1-5
22:05:06< Matthi> pressing 1 sets my hand to slot 1, pressing 5 sets my hand to slot 5
22:07:22< Randrian_> What I wanted is just Slot A, B, 1, 2, 3 the first two slots have the hands on it. When I press 1,2 or 3 this item changes with hand A pressing shift (or some other key) and 1, 2, or 3 I change this item with the B Slot (second hand)
22:07:37< Randrian_> having both items and hands moving around is confusing.
22:07:51< Randrian_> I think we already discussed that the hands should be static
22:08:10< Matthi> Did we?
22:08:19< Matthi> I think that discussion was cut short in the forums
22:08:46< Matthi> also, items don't just "move" around
22:08:50<!Zapper> mh, I would keep the fixed hands and swap items around
22:08:53<!Zapper> but just swap them - not shift them
22:09:01< Matthi> ah well
22:09:05< Matthi> I guess I give up then
22:09:08<!Zapper> like "swap item in hand A with item on slot 1"
22:09:24<!Zapper> because the hand thing would not be faster
22:09:28<!Zapper> considering that you got two hands
22:09:59<+occ> Randrian * 280ebf55d623 planet (6 files in 2 dirs): new interface for projectile weapons. (only bow yet)
22:10:19< Matthi> well, I don't know if we really should keep the two hands
22:10:38<!Zapper> we got two buttons :)
22:10:51<!Zapper> i have to go for now!
22:10:54< Randrian_> well digging without it would be odd. You then have to change a lot around.
22:10:58<!Zapper> good night
22:11:01-!- Zapper [~zapper@zap.euirc.net] has quit [Quit: nacht!]
22:11:01< Randrian_> and Shield and Sword also wouldn't work
22:11:13< Matthi> Yes, "gamedesign follows control" seems like a _really_ good approach on that one
22:11:42< Randrian_> so, I have designed a new interface the bow with this commit. I hop this makes the bow a bit more robust.
22:11:58< Matthi> hm, okay @digging / sword-shield
22:12:01< Randrian_> Does anyone want to test the Windmill scen with this?
22:12:41< Matthi> but I still don't think that we need to confuse the player by switching around the items
22:12:59< Randrian_> well, switching hands confuses even more I think.
22:13:12< Randrian_> (btw. in CR also the "hand" was fixed and the items moved)
22:13:19< Randrian_> (but that's not an argument)
22:13:47< Matthi> yes, in cr, the items were not accessible otherwise
22:14:02< Matthi> you could only cycle through them or click on them
22:14:34< Matthi> now imagine you'd also have shortcuts-buttons for the slots in, lets say, knights
22:14:41< Matthi> they wouldn't do any good
22:14:52< Randrian_> what exactly do you mean?
22:15:07< Matthi> because you can't press a button without having looked at your inventory to check which item is where
22:15:15< Matthi> that takes time and is annoying
22:16:03< Randrian_> well playing "blind" won't be possible I think.
22:16:22< Matthi> thats what I want to archieve
22:16:48< Randrian_> but you have to much possible objects in clonk to have on button for every object.
22:16:52< Matthi> for example, having my magic-staff on 5 and my stack of healing items on 4
22:17:07< Matthi> why would I want to change those positions?
22:17:22< Matthi> i want to be able to cast magic on a press of 5 and a click for the direction
22:17:31< Matthi> I want to heal right away on a press of 4
22:17:40< Matthi> so why should items switch around?
22:18:26< Randrian_> well, what about having used up your heal potion and collection a new object?
22:18:31< Randrian_> where should it go?
22:18:47< Randrian_> when slot 4 is reserved to definition HealPotion
22:19:51< Matthi> When healing potions are empty, it's okay to overwrite it, I guess
22:20:28< Matthi> might be a little bit inconsistent here, but seems a good compromise in order to keep up the speed in melees
22:20:44< Randrian_> I don't want the game to consist of always configuring your slots, and which items are alloed to go where.
22:20:48< Matthi> Also, those "fixed" items would most likely be tools anyway, so they dont "disappear"
22:22:52< Matthi> configuring your slots ONCE to my own liking is much better than always having to check where my magic staff has gone to now when I need it
22:23:49-!- Icewing [~Icewing@euirc-546b0d59.pool.mediaWays.net] has quit [Client exited]
22:30:48<!Guenther> I haven't followed the discussion, but ideally, most people wouldn't need to configure anything, but could simply use the defaults - just like the default controls in CR were used by a lot of people
22:31:19< Matthi> That's not what we're discussing, but you are right anyway ;)
22:31:56<!Guenther> I realize that.
22:32:46<!Guenther> You can read is as supporting your position - configuration possibilities are not necessarily bad, if the defaults are decent.</pre>

IRCMeeting/20100308 - Revision history

Clonk-Karl: Log for meeting on March 8.