Reimplenting crap?

A generic talk on the OpenMW project.
Locked
User avatar
lgromanowski
Site Admin
Posts: 1193
Joined: 05 Aug 2011, 22:21
Location: Wroclaw, Poland
Contact:

Reimplenting crap?

Post by lgromanowski » 21 Aug 2011, 19:48

sir_herrbatka wrote: Hi,

As we know some part of MW game rules is slightly idiotic (eg. stealth) ;)

Is the reimplenting of this crap obligatory? Maybe it would be better to develop our own formulas in some (most stupid) cases?
Zini wrote: No. That might break compatibility with some mods that rely on these features working in a specific way. With the huge number of mods out there, we can't be sure, what kind of problems we are causing by departing from the original game mechanics. Not worth the trouble. Any sub-optimal mechanics can be changed after 1.0; with a switch to turn the original behaviour back on.
sir_herrbatka wrote: I see your point here Zini. And It would be wrong to force player to use other formulas than in MW. It would be better to give a lot of options, or option to create your own formulas. If not possible - keep the current crapy rules :/
nicolay wrote: That said, this doesn't mean we can't discuss alternative rules and formulas now, or shortcomings with the old ones. We should plan ahead. I also don't think we need to have a 100% perfect rendition of everything before calling it 1.0, especially if there are details that we aren't sure about or which are a hassle to fine tune exactly.
Star-Demon wrote: I think it's inevitable that we will break mods.

The reimplementation is extending mechanics that*will* make lots of mods obsolete. The best we could do is to "turn off OPENMW.esm/esp", but that's like including a 16 bit mode for a 32 bit machine (Sorry, Assigned reading for school is getting to me).

The best thing about this project is that we can do things that were simply missing or were technically not feasible with original Morrowind due to time or technology. It goes without saying he have to preserve the structures of data that it uses so mods and file types CAN be used (this appears to be the bulk of the early work so far, we're trying very hard to make that work), but that doesn't mean we will avoid breaking or making some mods obsolete.

Some I can think of right away:

MWSE, AND all the mods that use it. All of them.
MGE.

Boom. There's a good chunk right there that will not be necessary anymore.
Are we seriously going to tiptoe around implementing OpenMW in a way that allows it to have another program run on top of it to do what we're trying to do already? That'd be silly. We have enough to do right now.

So I say we behave (a little) like the FOOKunity is looking to behave (with FO:NV; they already agreed that FOOK2 went out of control). We pick very, very carefully, preserve what we can, and don't be afraid of making certain mods obsolete or completely unnecessary

Take heart, though:
1. Many of morrowind's mods are content mods that don't necessarily care if your mechanics are different.
2. They can be rewritten, probably much easier now.
3. No more crashing and less mods to load, less conflicts.
4. We have a community to rewrite mods focused on content and not fixing mechanics like walking speed, fatigue drain, magic regen, etc. These are simple. They are easy to turn on and off, as well. THOSE we could make options for, but I think we can agree it's something that would be nice to have always.

We have a lot to learn from others, in this regard. FOOKunity is a great example to look towards, as well as the fragments of the Morrowind Community.

I myself am making my mods compatible and friendly with future releases of OpenMW and TR. (including the inevitable new meshes and collision) It's not HARD, but we're a community, we can work with each other.
raevol wrote: While there are a few things that I definitely would like to see reimplemented in OpenMW, I'd like to make the case that we should, without fail, fully implement and support the original mechanics. Altered mechanics are, in a sense, a form of cheating, and I know there is a large population of purist players out there who would be completely dis-serviced if a faithful rendition of the original mechanics wasn't available in some capacity.

Focusing on implementing the original mechanics fully for 1.0 should be the focus, in my opinion, and post-release we can re-implement things and have the new mechanics available as options.
Star-Demon wrote:
raevol wrote:While there are a few things that I definitely would like to see reimplemented in OpenMW, I'd like to make the case that we should, without fail, fully implement and support the original mechanics. Altered mechanics are, in a sense, a form of cheating, and I know there is a large population of purist players out there who would be completely dis-serviced if a faithful rendition of the original mechanics wasn't available in some capacity.

Focusing on implementing the original mechanics fully for 1.0 should be the focus, in my opinion, and post-release we can re-implement things and have the new mechanics available as options.
This is more or less what I think, but then there's the simple fact that no magic regen is crippling, some calculations are wrong, some are *missing*, and some may not be faithful to begin with.

So - I agree - the mea culpa being that these mechanics should be made to be extensible or able to be revised or improved. Hardcoding them or making an environment where anything else won't function or would break the game wouldn't be acceptable, even for a piece of OSS.

So yeah, make it WORK, then make it better.

But I want to stress, by virtue of what we're doing, a large portion of mods are probably not going to function anymore, and will need to be rewritten - I see this as an opportunity, though, not really a detriment.
Zini wrote:
But I want to stress, by virtue of what we're doing, a large portion of mods are probably not going to function anymore, and will need to be rewritten - I see this as an opportunity, though, not really a detriment.
I do not agree. The mission statement clearly says, that Morrowind and all mods should work (note that this does not necessarily covers mods made with tools like that script extender and other binary modifications of MW). If a mod does not work, we made a mistake.

If mods need to be rewritten, it will hurt the popularity of OpenMW. The mod community for MW is not as big as it was once and there are probably not enough people left to do a full conversion of a large number of mods in a reasonable time span. Also many authors of existing mods are probably not around anymore and since many mods are badly documented, rewriting them won't be easy.
nicolay wrote:
Star-Demon wrote:But I want to stress, by virtue of what we're doing, a large portion of mods are probably not going to function anymore, and will need to be rewritten - I see this as an opportunity, though, not really a detriment.
I'm not quite sure what you mean by not functioning, could you explain? I would think a large portion of mods are pure content. Ie. they either add cells/NPCs/assets/stuff (not affected at all by mechanics), or change game settings through GMST records (which we should respect, at least when using original game mechanics.) Even most script-based mods should still work, even if all the 'fixer' mods might not be needed anymore.
Star-Demon wrote:
Zini wrote:I do not agree. The mission statement clearly says, that Morrowind and all mods should work (note that this does not necessarily covers mods made with tools like that script extender and other binary modifications of MW). If a mod does not work, we made a mistake.
Okay - So: What do we do about MWSE and all the mods that use its functions?

What about mods that require the MCP? The MGE?

You see what I mean? What do we do about that?

What I'm talking about, explicitly, are the exact exceptions you mention. That's a lot of mods.

The way I look at it, as long as we preserve the data and have a good process for getting it loaded, rendered, and unloaded, the mods that need to work *will* work, right? The rest will have to be rewritten because they were written with the assumption that textures are being hooked in, dlls being injected, and the exact functions they want are going to be there, we could duplicate them, and hope it works...but wouldn't that be redundant?
If mods need to be rewritten, it will hurt the popularity of OpenMW. The mod community for MW is not as big as it was once and there are probably not enough people left to do a full conversion of a large number of mods in a reasonable time span. Also many authors of existing mods are probably not around anymore and since many mods are badly documented, rewriting them won't be easy.
I understand this, and I also alluded to it. But, I still see it as an opportunity. Keep in mind, we're talking 1.0 to subsequent milestones.

At its core, the way you describe, 1.0 just needs to load data, render it, and take input the way Morrowind would, right? The problem is that it's very likely that, as part of the capabilities of OGRE3D, myGui, and so on, that a lot of fundamental engine restrictions will be lifted, or simply different from Morrowind and we'll have to do something else that is similar.

If you start building around mods, you're protecting simply bad mods. I don't think that's good practice. (because then you have FOOK2...)
Zini wrote:
The problem is that it's very likely that, as part of the capabilities of OGRE3D, myGui, and so on, that a lot of fundamental engine restrictions will be lifted, or simply different from Morrowind and we'll have to do something else that is similar.
Not really. There are basically two areas:

- stuff that works with original MW

This should be reimplemented exactly.

- stuff that not work with original MW, e.g. size limitations for scripts; complexity limitations for expressions; certain script constructs, that crash MW; maximum number of mods

These we can change (and already have started doing so), because no legit mod can use them.
Star-Demon wrote:
nicolay wrote:
Star-Demon wrote:But I want to stress, by virtue of what we're doing, a large portion of mods are probably not going to function anymore, and will need to be rewritten - I see this as an opportunity, though, not really a detriment.
I'm not quite sure what you mean by not functioning, could you explain? I would think a large portion of mods are pure content. Ie. they either add cells/NPCs/assets/stuff (not affected at all by mechanics), or change game settings through GMST records (which we should respect, at least when using original game mechanics.) Even most script-based mods should still work, even if all the 'fixer' mods might not be needed anymore.
This is what I'm saying, but the exceptions we're discussing are mods that require things running on top of Morrowind that add functions (to be honest I don't know how mods that use MWSE functions get access to them and how they are executed by MWSE then Morrowind) that are not stock.

you're right, and I agree, though. There are mods that simply change some numbers - we don't need to include those, although I'd say they are better made into our own OPENMW_changes.esp or something like that.
Star-Demon wrote: haha! I think we're actually all in agreement but we're all concerned for different aspects. :)
sir_herrbatka wrote:
The rest will have to be rewritten because they were written with the assumption that textures are being hooked in, dlls being injected, and the exact functions they want are going to be there, we could duplicate them, and hope it works...but wouldn't that be redundant?
This is the minor part. Most of mods do not require other software.
f mods need to be rewritten, it will hurt the popularity of OpenMW. The mod community for MW is not as big as it was once and there are probably not enough people left to do a full conversion of a large number of mods in a reasonable time span.
Sad but true.

But, you know guys... I'm realy not sure if we should go into alternative formulas rather than providing tools to create your own game rules. Zini explained that it's big task (too big for now), but this would benefit in future; maybe openmw will become engine to create fpp rpg's... this would be greate.
nicolay wrote:
Star-Demon wrote:Okay - So: What do we do about MWSE and all the mods that use its functions?

What about mods that require the MCP? The MGE?
This is a minor point in the discussion, but I don't see any big hindrance to supporting ALL of those mods in time, unless they conflict with each other. (And to be clear: this is POST-1.0 stuff I'm talking about here.) The point is, there aren't all that many of these extenders, and what they offer to mods is (as far as I can tell) pretty well defined, so adding it in later shouldn't be all that hard. Furthermore, there aren't likely to be many more extenders made for Morrowind (the game is almost a decade-old already), in other words we are dealing with a non-moving target here.
Star-Demon wrote: To be honest, what I'd really like to do, after we get the base stuff down and working, is make some sort of vehicle for a standard but very friendly set of balance changes. Maybe an actual mod, maybe a configuration file, but anything that reduces the amount of mods needed would be nice.

Oh, I just remembered one item for the mechanics wiki page...done.

Yeah, one of my complaints about all bethsoft games like this so far, is that the amount of changes made by adding items, adding mechanics, rebalancing armor, only to have more items added...it's really nuts after a while.

Fallout suffered the worst from this, although I think oblivion came in close second. For Morrowind I have quite a few to correct some serious problems for armor balance (LOLglass), but I don't even know what they are doing anymore.


There's another problem - load order and record and reference changing and resolution. I haven't gotten a close enough look at openMW to see how it works, but if the finished games had enough troubles sorting and making changes intelligible...I don't know immediately what we can do to make that less confusing and painful...not to mention more stable.
Star-Demon wrote:
nicolay wrote:This is a minor point in the discussion, but I don't see any big hindrance to supporting ALL of those mods in time, unless they conflict with each other. (And to be clear: this is POST-1.0 stuff I'm talking about here.)


Absolutely. See my "We all actually agree on this from our own perspectives" post :)
nicolay wrote:The point is, there aren't all that many of these extenders, and what they offer to mods is (as far as I can tell) pretty well defined, so adding it in later shouldn't be all that hard.
I'm not too sure about that, since we all may not understand how they work. I think we might be able to make mods that need MWSE work, but we don't know how it happens. (at least, I don't.)

While they may be few, there are many mods that use them, some may be badly made. Look at Firefox. Does firefox allow terrible code? (it doesn't...much to my frustration a few years ago! haha!)
nicolay wrote:Furthermore, there aren't likely to be many more extenders made for Morrowind (the game is almost a decade-old already), in other words we are dealing with a non-moving target here.
yeah - I think what I've been trying to say all this time is that while there are definitely not many exceptions, I do believe they exist, and I do think we shouldn't feel too bad about them. I think, looking back at the discussion, that we all agree about what should be and is being done, we like the plan, but we all have particular attention focused on different parts of the same plan. :)
silentthief wrote: Another thing to realize is, MWSE (MGE, ect) is a 3rd party program written by fans of the game to enhance morrowind because while it was a game that they loved, it was something they thought they could improve upon.

These programmers took steps above and beyond the normal user to enhance MW. I would think, then; that if they wanted to make the scripts from MWSE (that may or may not be supported by openMW, just choosing this as an example) compatible they would have the means to do so and perhaps could with ease.

ST
nicolay wrote:
Star-Demon wrote:To be honest, what I'd really like to do, after we get the base stuff down and working, is make some sort of vehicle for a standard but very friendly set of balance changes. Maybe an actual mod, maybe a configuration file, but anything that reduces the amount of mods needed would be nice.
[...]
Yeah, one of my complaints about all bethsoft games like this so far, is that the amount of changes made by adding items, adding mechanics, rebalancing armor, only to have more items added...it's really nuts after a while.
Ah, now I see what you're getting at. You're basically wishing for a better mod system, one that lets you change things without all the mess and complexity (and bug-prone-ness) of piling mods upon mods. I think we can all agree upon that as a worthy goal for the long run. I do not think such a goal has to stand at odds with supporting existing mods, and I think both solutions can happily coexist.
willrandship wrote: Isn't MGE integration kind of....pointless? the OGRE engine will already support 99%-100% of the actual graphical alterations, and the rest (animated grass, etc.) can have a much smaller, pure content mod. No need to support most of the "work" part of MGE, IMO. I'm not saying anything about MWSE, though.
Star-Demon wrote:
willrandship wrote:Isn't MGE integration kind of....pointless? the OGRE engine will already support 99%-100% of the actual graphical alterations, and the rest (animated grass, etc.) can have a much smaller, pure content mod. No need to support most of the "work" part of MGE, IMO. I'm not saying anything about MWSE, though.
I assumed that this was kinda assumed. :|
silentthief wrote: one other thing about MW that bothered me, was regarding falling and water. If, while playing MW you fall a distance that would hurt you but you fall into water it does not hurt you. So far fine. But then, it "saves" this pending hit for you if you use a waterwalk potion/spell because it then DOES hit you even though it should not.

I've even seen where it will "save" this hit if you fall into water, walk on land, then use the water walk potion, and then walk onto the water...

ST

-EDIT- for spelling error
best regards,
Lukasz

Locked