Link To Site, Download, Info, Etc.
ScreenShots - Fog, Lightning, Tribunal
Screenshots - First Cells, Different form of Lightning

The person behind this undertaking is tbrick who is present on these forums,


From the previous thread, this project was checked out as legal!!


This thread is for discussion and suggestions and spreading the knowledge of this project. Being an engine replacement, limitations on the modifications possible will be loosened and the OGRE engine can deliver Morrowind with much better graphics.

First Topic (Locked Due to Post Limit / Will Be)
PM Me if you want something else added / changed here.

Another idea that came to me which maybe possible with the engine change ... separate bodies of water, so we can finally have mountain lakes and the like. Just a thought (I know I have alot of dumb ones ).

ok. just a quick quistion. what exactly does the current download do?

MortX, i dont think thats dumb. now that u mention it, it sounds realy cool.

That's actually one of my biggest issues with Morrowind. In order to have any water higher than sea level it requires water meshes, but then you can't submerge under them and get the breathing/drowning/proper fog effects.

I Would LOVE to see this possible!!!!

Are there any plans to implement OpenAL support (for EAX and other environmental audio effects) in OpenMW?

Maybe the maker(s) of MWSE can add a script so that when the player is colliding with the water mesh (which is invisible, and under the 'surface mesh') (both the water meshes and surface mesh have no collision, I don't know if that would require any external programming, other than disabling them (though you can't see the surface like that)) they get a fog and limited-breath (if they don't have water breathing); having water walking causing the underwater meshes to be enabled, and gives a variable the value of '1', raising the player above it if another variable is 0 while that variable is 1, and setting that variable to 1 so that it isn't triggered... If the player does not have water walking effect but the first variable is 1, both variables are set to 0. (The second one is probably best as a global) MWE might be able to include various heights of water too. But, working from 'scrap' like OMW is, it'd be much easier to just include it in the first place .

That would be cool.

It loads an area with some static creatures and no collision and the player can move around using wasd keys. It looks very much like MW.

It's much easier to do than that. Just make it like a tree.

Take your water mesh and add a very small collision node to it, at a point near the botom of the water. Then the water will have no collision.

It will be tedious to set up correctly in the CS, as the water will have to fit into a matching depression mind.

From the previous thread:


I agree this would be great, and it certainly is very possible. It would help if the design of OpenMW tried to take future multiplayer into account, even if it is not implemented until some time after the initial release. But whatever the case, once you have some source code to work with, turning a single-player game into a multi-user game can happen. Current attempts with only the closed-source game engine are doomed in comparison.

I've thought a bit about how multiplayer could work, and it really could be quite exciting. You would even be able to support user mods, if you implemented a "sync to server" option when people connect to your MW server. So, for instance, you could install a bunch of mods, make a nice house to live in and decorate it, and then invite a friend to "come over and visit", they'd connect to your server, sync up resources, and they'd see what you see in-game in your new house.

Technically this might be possible (if difficult to code), but given the amounts of data Morrowind crunches through, it would probably require broadband or better internet, or some serious data-usage optimization to be able to run at all, much less well. And then, what do you do about dialogue, menus, and other game-pausing actions?

Edit: Not to derail, but is there a file format documentation for NIF? Something similar to the ES* documentation? I can't seem to find anything that describes the format.

Well, I have some small certification in C++, I'm taking classes in computer programming for College right now, and I know a large host of other languages. Need to read up on D - hope I can help out in some way.

Been thinking, planning, and wanting something like this for a long, long time.

tbrick, you are awesome!

I would suggest (and I'm sure it has already been suggested) getting in contact with Timeslip and adding MWSE functions into this new OpenMW engine. Or, at least making MWSE and MGE compatible with it?

Anyway, this is an ambitious project, and I stand in awe at your programming skills (I can barely work MATLab).

Lol Matlab. Everybody loves Matlab.

I know only one source besides NifTools documents and NifScope interactive help related to nif format: help for “MaxImmerse” and “NetImmerse viewers” that can be downloaded from the THANOS TOWER:

http://www.ageinc.com/Morrowind/Thanos_Tower/utilities1.html

Thanks. I'll look at that. I couldn't find any documentation for NifTools, though...

Err, a few more suggestions:

The open morrowind exe should emulate all Morrowind Enhanced and Morrowind Script Extender functions from the get go. I.E. make it so any mods requiring either ME or MWSE will work right out of the box. Not totally necessary but a good idea.

Default ai improvements. Things like the guards protecting you, random npcs running away from fights, enemies chasing you into and out of cells, and npc's wandering around a little instead of just standing still. This would be compatible or replace almost every mod out there and make the game seem much more alive at the same time.

Having looked at the OGRE forums there' are plenty of graphical effect plugins that should work automatically. Screen Space Ambient Occlusion and HDR would be nice, having depth of field activate when you're talking to an npcs would be nice too. Like I said, many of them should require little work or like SSAO just be a matter of setting up the plugin. Near instant better graphics!

I suppose that the version of OGRE is going to allow for normal maps and specular\diffuse maps? I'd happily put together a package of already improved Morrowind textures covering everything in the game if anyone would be willing to add normal and specular\diffuse values to them.

I'm sorry, I'm obsessive I know. But before I forget:

Combat and stealth enhancements (for after the entire thing is actually working). Sneaking should be light based and sound based. The heavier the stuff you have equipped the harder it is to sneak. If you sneak upon a target hitting them would be automatic and critical hit chance would be much higher.

Combat: blocking should be active! Piercing should have a much higher chance for critical hit, armor should class should only block it decently, and blocking should severely reduce the damage it does. Slashing should do high critical hit damage (slit their throats!), and should be blocked by armor and blocking normally. Bashing should have low critical hit damage and chance but be able to get through blocking and armor fairly effectively.

As for active combat concerns, the chance of an enemy blocking when you attack from the flanks or back should be much lower. Also if you attack someone from the back extra damage should be applied.

Some nice ideas Tomb, but I would say they are best left as OpenMW Modifications rather than OpenMW itself, everyone not having the same preference for play styles and all.

NO!!!
Seriously, now it starts. Inability to please everyone.
For every person who want such arcade game features, another will throw their hands up in horror.

@tombofsoldier: I believe the idea to implement all MWE and MWSE functions right into the engine is great! I mean, if you just play OpenMW without any mods (vanilla) then it won't use any of those functions anyway, so it's just like vanilla Morrowind, but it adds a HUGE amount of possibilities for scripters.

But also, I don't think all these other modifications you proposed should be implemented, too. This should be made to emulate the original Morrowind as closely as possible. Though, what I think that should be done in OpenMW, is to make all the buggy scripting functions in Morrowind actually work the way they're supposed to!

I mean, as far as I remember, walking script is broken, or something like that. That's why in Living Cities of Vvardenfell NPCs don't walk to the door and enter/exit them, but just disappear. It would be foolish to reimplement bugs like that.

Here is all I would want from the new engine if It was up to me to decide the features of a new engine,

- Dual/Quad Core Support to allow much bigger battles and things like that.
- OpenAL or something similar for a much more immersing sound experience.
- LOD
- Real Reflections on Water and on Items.
- A lightning system that is very similar to the current one but that works without seams across the meshes.
- Another lightning system that allows all objects, AI, and land mesh to cast shadows calculated upon the available light sources and time of day.
- Engine Flexibility - Smooth execution on old computers with multiple new enchantments and abilities for new hardware. (For Example: On a Gfx4, the user would have simple pixel shaded water, but on a Gfx6 the user would have reflective water, and on a Gfx7 reflective parallax water effect)
- More Options in the Options Menu. A lot more.
- Collisions should probably be done differently for more accurate physics in the not so near future.
- DOF, M-Blur, PS2 HDR, and stuff like that.

...it probably would be/is very hard to get the scripts and the AI to work.

And general engine optimisations besides that. The average Morrowind FPS is below the average Oblivion FPS on my computer - that just makes no sense.

Although thread loading (dontthreadload or something like that in .ini) that makes cells load in a different thread sounds like something that might boost the performance on dual-core CPUs (Morrowind on one core and cell loading on another - sounds great), it doesn't change it at all.

Making OpenMW engine have separate threads so it benefits from multiple CPU cores is a great idea.

Also, it should be tried to make it take the load off the CPU and to the GPU, at least all that's possible to transfer to the GPU.

I am not sure whether or not that is a good idea. I mean, the current morrowind is way too inefficiently relient on the CPU... but, with a new engine, that might not be the case. If the engine is good, then the CPU probably wont be used that much / Used More Efficiently. Load balancing between the GPU and the CPU should be present but functioning both ways.

Well that's what I meant, but worded it badly. I meant the things that can be put to GPU and not the CPU should be put to GPU. Balanced load of CPU and GPU, yes.

Given how OGRE works, all by itself, a lot of your requests/problems will be fixed. Unless the developer(s) of this start doing weird stuff to it, most of the shader/lighting/render optimization/shadows/etc will be fixed.
So long as this is efficiently written, it should be much less CPU-dependent, and not too badly GPU dependent. However, the pure amount of data being processed (which will be equal or greater) will demand massive CPU and RAM resources.

Reflective water will be almost automatic, it would actually be harder to get Morrowind's water than to have OGRE's nice reflective water.
There was recently released a Paged Geometry system that's basically one of the most advanced LOD systems available. Not sure if openMW will actually use that or just a basic heightmap system, it's up to the cody dudes.
Better lighting is already in (check screenshots). Shadows are also being toyed with.
Again, it would actually be easier to have more advanced physics than Morrowinds.
DOF, HDR, and Motion blur shouldn't a problem.

Link please?

As for MW's physics, after a whole lot of crying tears of blood trying to get some reliable collision detection through the scripting engine I have come to the conclusion that MW's physics were written on the back of a napkin and implemented during a tea break one day *braces for incoming Bethsoft cannon fire*. Seriously though, I think much more could be done on them even in the vanilla conversion. Even if its just making arrows stick into things.

Paged Geometry. There's some nice videos there. There's also the proprietary Speedtree for Ogre3d, but i guess that's useless in this case.

Edit: Umm, actually i'm not sure if tombofsoldier meant Paged Terrain instead.

"- A lightning system that is very similar to the current one but that works without seams across the meshes."

This one is harder then you might think. It's possible, but it depends on the complexity. It's very difficult to, let's say, allow a normal map to have more then a few light sources and still work as it is supposed to.

Does this mean we may see havok?

It is nice, although IIRC you tend to lose a lot of the speed benefits if you have multiple meshes (as opposed to multiple entities with the same mesh). True of anything though


Sure, but not knowing the mass and inertia of any objects is a bit of a bugger isn't it?


How does havok compare to other physics libs?

Unfortunaly I haven't done anything re the ext landscape . I have been away a lot, and am trying to finish a project that I meant to do a year ago.

As I understand the nif format has at least one unused "slot(?)" or node or somesuch for which we can store those parameters. We'd need, mass, center of gravity and moment of inertia. Even if there is only space enough for one variable we can combine all three into one and parse it.



I haven't used Havok yet but I've been messing around with Bullet(freeware physics libs) and its pretty nice. I haven't seen any benchmarks but in the samples I ran it had some nice cloth effects aswell as some good ragdoll capabilities. Cloth on certain capes and things would be a really nice addition if it isn't too computationally expensive though I'm thinking if its done the same as FEA then we might be out of luck. Rigid bodies is probably all we can hope for for now.

's about a million dollars more expensive, and runs about as good. At least from what I've seen (PhysX, Havok, Newton, ODE, and a few others). I don't know how it is on the programming side, but I doubt it's all that much easier to use than the others.


Given how the modern physics work, I'd say that it might be possible to institute all the types (not just rigid body, but probably cloth and ragdoll as well). What I've seen of cloth, at least on a few physics systems, it doesn't run too poorly. Havok is actually the worst cloth dynamics I've seen yet, IIRC. Seeing as cloth is an offshoot of joints, and if you do cloth you can usually do ragdoll... Dunno.

As for physics with items, personally I'd recommend having another file (Physics/[mesh name].[something), that defines the physics properties of that object. If it has the same path and name as the mesh (except under Physics instead of Meshes, and with a different extension), it would prolly be the easiest way. Using an open slot in the NIFs would be prohibitive, since all the stock game meshes would have the be edited for that, and the new ones distributed (at least a few hundred megs, if allowed by Bethesda at all).

Havoc is free for this sort of thing


I have only played with ODE and Newton, but I was wondering why Demon xen wanted havoc in particular.

From what I recall, the API wasn't as nice as other libs, but that could be solved by a wrapper for OGRE

NiExtraData IIRC.

Doesn't matter though. If this is going to use Morrowind assets it means it must conform to the Nif 4.0.0.2 spec with Morrowind enhancements.

A lot of the maps people want just won't fly, but what will fly is true support for reflection maps, which are in that spec.
I'd settle for that as the only graphical enhancement, which would be superb. You wouldn't recognize clothing and armour, the improvement would be so huge.

This is used in several meshes.



But there is no reason as to why it can't support .mesh and the Oblivion nif version

Really? Interesting. I'll have to look at that.



The author said it wouldn't happen, although it is a good idea.


And even if it's not, all the original meshes would have to be edited. A secondary text file that can be (even for all 300 meshes or whatever) compressed down to nothing would be much easier.

Sorry to jump in here and be out of topic, but, would a NVidea GeForce 5200 FX 128mb PCI card with DX8 have trouble running this? My comp has a 3.2 ghz P4 processor and 1.5 gigs of RAM.
Thanks.

Dude, get a new video card. One that's 10x better will cost you like $50 or less. In fact get this one: http://www.newegg.com/Product/Product.aspx...N82E16814141075

That's an odd combination... Good processor, fair amount of ram, but horrible videocard.

It would run the unoptimized vanilla engine and if this would be a new engine... I have no clue, we can only guess based on current performance and requirements for OGRE engine and prediction of what will be necessary to render everything as well as scripts and AI.

Your rig reminds me of my old one before upgrades,
AMD Athlon XP 3000+
512 166Mhz DDR (2700)
NV 4 MX 64

I remember upgrading to GeForce 5200 and actually regretting it for it only ran slightly better.

I would not reccomend getting a new GPU, get a new rig... a new intel graphic card will run much better then a 5200 (espiecially if you use dual channel)

Everyone who is dreaming of Morrowind with 10.x.x.x nifs should remember that Bethesda have prohibited conversion of Morrowind assets to Oblivion, which this would require to fly.

I don't think anyone was talking of converting any assets to different mesh formats, other than to OGREs internal representation of a mesh which would be done in memory.

I was just saying that you have a mesh loader object factory based on the extension of the mesh/mesh header, which would allow easy plugging in of new loaders for whatever format anyone wanted to support.

Ooh. I just noticed this project. Very interesting

I don't know.... converting meshes and other game assets to more manageable formats beforehand might be a good idea... loading-time wise.

:/

Yup, he's very right.

which would open up the legality of the mod again.
I don't think we can touch the .esp's/esms/bsa, just use a different program to read them.

how many guys working on this project ?

I am going to do some stuff (exteriors) when I have finished my current project. Unfortunately I am away most of this week all next week and some of the week after.

So many weeks waiting. but it's still good to hear.

I think there iss two ... three with Yacoby?.

Which is fine I think, two many cooks spoil the broth as they say.

Nice idea. Wouldn't be an issue for those of who find NifSkope or Liztail's animation tool straightforward to get it working with our chosen set of mods.

Wouldn't work so well for 90% of people I suspect.
I think keeping it simple would be the best option.
Honestly, there are several capabilities of the 4.0.0.2 format that Morrowind doesn't support, much too our cost.
Axel, Vality and others have demonstrated there are several that Morrowind DOES support that Bethesda didn't use.

True vanilla 4.0.0.2 support would stun people enough on it's own, but be 100% compatible with current assets.

Actually, I am gonna buy this one http://www.newegg.com/Product/Product.aspx...N82E16814143102
I think it's good

Just because it costs more, does NOT imply it is a better card. The one you have selected there is more of a business card, than gaming. GO for the 6800 recommended, it will give you MUCH better performance, for LESS money.

I spent the bucks on a HD3850 AGP card for my P4 system. I am sure that my cpu bottlenecks the hell out of the system though.....

Thanks for the advice. Just to say, I go for least price, not highest. I am looking for an AGP 4X/8X card, preferably nVidia, 512mb, 128bit,etc. Got a link for me by chance? I've checked New Egg, Ebay, Amazon, and Tiger Direct, but no luck.

Just to say, I'm upgrading this one for short term, next year, I'm buying a laptop for college. That being said, I'm not buying a new rig, this is my parent's computer anyway.

As much as I hate to say it, do not upgrade if it is not your machine. The 80 bucks you will spend might just come in handy later when you are flipping through the options list for your laptop.

I hope tbrick doesn't get too discouraged by all the high expectations and disinformation.

Well, if I was in his situation, I probably would pay little attention to our crazy ranting on how incredible his will/would be until he actually sees the light at the end of the tunnel. He seems very level-headed as well, so I doubt he'll get too swayed by anything we say unless it's something that immediately sounds more reasonable than what he originally planned.

Just dropped in to wish you good luck on achieving all the goals. Let the gods of programming guide you on this quest!

Thanks for the support guys. Don't worry about flooding me with ideas and expectations Like someone said, having a huge brain storm at this stage is a very good idea. And even if many of the ideas won't be implemented soon, we can plan the design in a way that will make them possible in the future.

The development priorities of OpenMW are pretty straight: first make an engine that can play through the original game, with mods. Then add all the extra stuff. The reason is simply that there is so much extra stuff to add (as the threads here prove!) that if we start working on all of if at once we won't get anywhere.

BTW, I've just finished an (unrelated) summer project of mine, so I'll soon have much more time to spend on OpenMW. I am setting up a dev forum / mailing list (for programmers) - Never done this before, but I'll go with google groups unless someone has a better suggestion.

@red snow: here's a list of hardware requirements for Ogre: http://www.ogre3d.org/wiki/index.php/Hardware
I don't think you will have any problems running OpenMW, but you would probably need to upgrade to DX9 to run the current exe file.

Well glad we aren't overwhelming ya to much with ideas, cause I had another for the far future.

Multiple body textures per race. In effect the body texture could be selected at character creation much like heads and hair, but allows you to choose a texture with lighter/darker tans, muscular/skinny bodies, elderly bodied, scarred bodies, tattooed bodies, etc. Perhaps there could be a way to link skin tone of face and body together ... so you don't get weird 'tan' lines above the neck.

I guess different meshes might be possible in this way? But amongst other problems would cause a lot of trouble with clothing ...

This is a long standing wet dream come true.
God bless you, tbrick. God bless you.

That sounds (from my outside of the project PoV) like one of the easier requests. At least in part. Making slightly different skin colors could be done by a saved material color, or even a file/reference for each NPC that defined which skin texture to use (and over-wrote the race-based one).

May I suggest Facegen?

I know it seems like an odd request, but if we included facegen, we could create entirely new faces for EVERY character!

Besides, this is one of my favorite things about Oblivion. This, and havok. xD

Ew no. Facegen in my opinion was one of Oblivion's worst features. Sure it had a lot of options, but you can spend hours trying to make a character and end up with nothing but ming, none of the sliders seemed to do anything positive and always seemed to work against each other.

Nothing beats a hand crafted face/head. You'd probably need to pay to get it in your game/engine anyway.

i liked the facegen thing in OB. much better than picking heads in my opinion. cause then it requires people to make faces, so other people can download them, meanwhile, everybody could make their OWN faces. i suppose u could make that feature optional, cause i think there will be alot of people who say no to facegen.

I'd have to say no to facegen too. At first I was a little miffed at Vality's dislike of facegen but...I realized she is totally right. ^^" In OB I simply could NEVER get a face to look good..or at all what I wanted it to be. In the end I just reluctantly went with a character I thought was fairly ugly and wished for Morrowind's head selector. ^^"

Being able to put in the head # and the hair # would be a nice feature during character generation though. Instead of always having to scroll through to find that particular hair or head. Input the number for the head and input the number for the hair.

I thought Oblivion's facegen was ok. I made a few faces I liked.

I don't think the idea of facegen should be thrown out, but maybe Oblivion's facegen UI might not be the best way to do it. Maybe put some grab points on facial features, so you could grab the nose and pull it out and smear it around

About Facegen SDK
Everything has cons and pros. For Morrowind-esque hand-crafted face, If you can get well-made one, you won't be disappointed, yes. But, unless you're good at 3D modelling and texturing, you can not make a decent face you want. For Facegen, you can make your own face, but you know, you can be often disappointed with the result, with older version of SDK which Oblivion used.(I learned using Facegen Modeller to make face is easier than using in-game sliders)

How would facegen data be stored?

You would have to add extra files, or modify the NIFs (bad idea, not to mention it adds NIF save code as well as just load).

Another vote AGAIST Facegen!!!!

I am not able to play OB because of the ugly plastic NPC faces (I have two OB copies: first release and gold). A couple of nice custom PC faces are not able to rescue the game: as a player I see all the time NPCs but not my PC face.
Even more, Facegen is not a free program that is not that easy to use (with exogenous face meshes) and obtain acceptable results. As a result of this problem for MW there are hundreds of custom NPC heads/faces and even more excellent textures, dozens of global head replacing mods; on the other hand, for OB there are only a couple of mods that IMHO are not much better than originals.

Moreover, there is a legal problem for the Facegen implementation in a free project.

Having said that, I think that some sort of lip-sync (not just “say” mp3 from “Vo” folder) could be reasonable. However, this is for sure NOT the number one priority.

i just thought of somthing, but could u make it so the player can pose? like actor's in charge for OB. pretty screen shots this way

that would be awesome.

-vote against facegen

I'm curious why people don't like facegen? Personally, I prefer to be able to create my own unique face as opposed to having to choose from some pre-defined set.

Edit: If it's just the UI, I see that as a separate issue. I believe the UI could be changed to do whatever works best.

I found the faces in Oblivion looked too much alike - as if all of the NPCs were related.

* Swiveller *

First off, it's not so much the UI as the entire system (or rather, it's results) people seem to hate. And changing the UI would be difficult, depending on how it's done. But I will agree with the people who have mentioned it before, every single face I've seen come out of the facegen system so far is extremely ugly. Haven't met anyone even decent looking, much less 'oh my god' like some of the Morrowind faces are. Quality wise, the facegen system is like Morrowind's vanilla faces, plus a few thousand polys (and so minus a few FPS).
Sure it adds variety. But we already have various ways of inducing vomiting that don't slow anything down.

And whether or not people like it, there's still the question of how to tie it in. It would take either saving each face as an NIF, or modifying the faces at vertex level after loading a base, and that would require some way to store the info. Neither the NIF or ES* formats have the capability, so that would necessitate another series of files, one per NPC, in their own folder. Making a mess and complicating things further.


My personal requests for new features are a better sound system and physics (sound will probably come with new software, and physics is relatively easy to integrate and easier to save/load, being new object positions).

One of the things that OpenMW will allow modders to do hopefully. To add animations to PCs like you can with NPCs.

Vote against Facegen too. Oblivion faces are too ugly.

Well that has nothing to do with facegen itself, it has to do with the way facegen was implemented in Oblivion and Oblivion's face textures.

Faces that you actually create with the Facegen program look marvellous. If the implementation is done right, you will have nearly the best faces in any game ever.

I would just spend much too much time using a facegen. I prefer to have some a hundred different faces than make my own. I would NEVER be satisfied with a facegen face. Perhaps if someone made a facegen mod for morrowind it might be a bit different?

So what if OpenMW just added in support for facegen? Meaning that a modder could utilize facegen if a modder felt inclined to do so. I'm not entirely sure how that would work, but it seems like it might appease everyone.

Oh my, I always wondered if there would be a project like this! Dream come true! Will watch.

Well, before inclusiong is considered, let this be known (unto all 'n stuff):


Last words if I've ever heard any. That usually leads to more zeros then they feel comfortable making public.

Given the the side utilities are ~$300 per, and commercial license is $1k, I think this might be a little over the top for a engine remake.

Source: http://www.facegen.com/products.htm

Example, please. Note that OB is not using original FG meshes.

http://i109.photobucket.com/albums/n45/AxelIP/Snap12.jpg

I think the original will kill me and FG authors and she will be right.


MW face in OB processed with FG:

http://i109.photobucket.com/albums/n45/AxelIP/FQ5-1.jpg

http://i109.photobucket.com/albums/n45/Axe...eenShotegt1.jpg

I do not exclude that this is only my fault.

Why use the commercial product at all is what I'm wondering? Why not start from scratch to create a new utility for making a custom face?

Can you do that? That would be great.

Sure. It's a SMOP (Small Matter Of Programming)

See, small works, until you get to programming. Then it loops around, and uses the second pass as an exponential multiplier (shortly: not small).

about facegen: well Mass Effect uses a facegen (if im understanding "facegen" correctly) and those faces look AWSOME. i played it on 360 so that might be a part of why it looked so awsome.

Theres mods for morrowind which replace every NPC with a handcrafted face, with so many choices, I don't understand the need of a oblivion-style face generator.

Something I would like to see is people being able to properly create things like fishing boats, horses and flying carpets without having to use lots of performance-sucking, save-currupting, fiddly scripts. And I look forward to seeing the smaller annoying parts fixed, such as the bug where anything too far out is cut outside the map. Or when large creatures clip through things and you can only attack their center.

full ground collision?

Physics?
Ragdoll?
much larger textures? (I'm sure Westly and Connary would go to town with this)
incorporated MGE MWSE and MGE?
The list goes on and on...


Optional facegen?

lions and tigers and bears.. OH MY!

Played it on PC, MS facegen sucks if you ask me (not many options and characters didnt look good if you ask me)
And i REALY dont like facegen idea, oblivion faces so sucked that i made new face models that dont suck...

have you seen TNR? Those faces are amazing!

Sure its an improvement but still it have that oblivion fealing...
So still pre-made faces PWNz, with stuff like Westly faces we just need that mumbo jumbo facegen

I'm not so big on full facegen because they generally end up a combination of creepy and boring, barring any huge deformities you manage to give them (not that that's any fun with perfect symmetry, ala Oblivion). Personally, I like to texture my own.

But, most of the Morrowind faces already made are mapped in just about the same manner, right? Bar all but a few Orcs and some others in my experience. It would be kind of cool to have a standard morphable mesh for races which you could apply any old premade head's texture on top of. Shift around the features to your heart's content, with no modelling know-how while still keeping a texture with personality and no odd blue 'facial hair' or any of the bizarre things you could find in Oblivion. I don't know much about how this could work and I'm sure something like this would end up with a whole lot of obvious placement issues, but hey, I'm just throwing it out there.
Really though, I think I'd cry at the simple glory of vanilla Morrowind with a stable framerate, so it wouldn't bother me terribly if this was an idea that was never touched upon.

I don't think somebody mentioned this, but will this new engine fix those horrible landscape texture seams?(as i heard those were beacause of MW's engine being incapable of seamlessly blending 3 or more textures at the same time)

I've used faegen before by itself and it produced some pretty amazing faces. I don't know why Ob's version is so bad but I know facegen can do much better.

I inquired in the last thread and it indeed seems that OpenMW will solve this problem, which alone is brill.

Ok, for those that didn't catch what was told regarding facegen:

Firstly, it's not Oblivion, nor any other games using facegen. There's a difference between an engine already implemented and one that didn't, because those games were adapted to use that function while morrowind didn't.



It would be nice having unique npcs, but there's limit on that. Since Morrowind's Nif aren't using facegen in the first place, it would be really hard to implement it and if it does, don't expect having amazing results IMO. It will also complicate things further if every npc have their own faces (how much npc in morrowind? more than 300?). It will just be like applying morrowind faces, but with hell of extra trouble.



Facegen is not free. I don't think someone will waste their money for that and find out it will not work in the end. The other thing is the license will most likely not allow using it on that sort of program anyway.

Anyway, there's plenty of amazing faces for Morrowind around the modding community. Why bother and just stay with those?



It could be an alternative, but in my opinion, it would possibly end the same way as facegen.

People can save themselves a lot of heart-ache by focusing on the possible and compatible.

An ambitious list would be replicating everything we already have perfectly, fixing all the known bugs (unarmoured anyone) and fully implementing the Nif 4.0.0.2 spec.

More than this is probably for later releases if at all.

Hello. Call me thick-skulled, but I am new to modding Morrowind, and I don't fully understand the signifigance of open mw, or what it actually does, fully. Could someone please explain to the handy-boy?

In short, it is a redone engine for morrowind with much more open possibilities than we have access to with the usual one.

I thought this would be obvious. So you can make your own unique face without modeling know-how.

Welcome to modding.

Here is somewhat of an example. With the current game engine, it is not possible to add animations to the PC (players character). With OpenMW, this may be possible.

In current morrowind, you cannot add a new skill to the skills list. With OpenMW, this may become possible as well. ^^

There are more complicated things that openMW will/could allow as well. Allow for more complex animations, fix bugs in the current morrowind, lower random game crashes, lower the chances of having a bad save (game crashes on load or something similar). The possibilities are almost limitless.

I agree with this statement. If this project can get to a level equal to what we already have it will be a massive accomplishment. I'm just as excited at the possibility as the next pleb but I feel it would be prudent to not let our enthusiasm run wild just yet.


-KWM

This. Is. Awesome.

One question, are we talking physics and collision-based combat? I'm not asking that we make it possible to turn Morrowind into Oblivion (trust me, I have a long list of things I hate about Oblivion design) but Morrowind/Tribunal/Bloodmoon + physics would be enough to keep me playing for years.

The only thing I am looking for in this is a working map. By the time TR finishes a couple more map packs, the landscape will be relatively hidden behind the map borders, which makes the notion of more land a little less appealing.


Remove the map borders!

I think FPS optimizer 2 removes map borders

http://planetelderscrolls.gamespy.com/View...etail&id=51
See the screeen shots

O.O Why was I not made aware of this?

I should go download ultimate galleon and just sail to Telvannis now.

Something that would be nice is a built in mod manager. Also a generator for list of in-game resources (like the "Lovers and Legends" list generators).

I've skipped most of the thread, but I've read most of info from homepage.
You have my support, but I have to ask: Why D?

I'm sorry if this has been asked before.

If this succeed it is possible to edit hardcoded aspects of the game like making new skills or give it a new engine.

A lot of the things I'm mentioning have mods for them that attempt to add things like sitting and horses etc. but a new engine could do this better I'm thinking so here's my two cents.

One thing that would be nice is ability to travel very far from 0,0 coordinates without any issues possible a world-space equivalent.

Another thing would be nice the ability for NPCs and creatures to interact better. MW is great but the actors are very stiff and limited.

If an alit is near an NPC the NPC ignores it and the alit ignores the NPC but it still attacks the player. Actors basically ignore each other they don't interact even implicitly. The game feels dry as a result. NPCs of different factions don't have any reaction to each other. NPCs never engage in combat with each other and so on.

Merchants occasionally wear items players sell to them.

Though there is a daily cycle of night/day no one ever sleeps and nothing is closed???
I know there are mods that add schedules but they are not very affective at properly making actors respond to time.

Sitting and even mounting of creatures of course this needs animations but I think its worth mentioning.

Possibly NPC travelling from one point to another.

In the final analysis NPCs need to be more interactive and the world should not just revolve around the player character. Life in MW is very dead when the NPCs are like robots.

Read the comments - FPS OPT2 tends to break stuff.

It will be awesome if this feature is included in OpenMW, it is very handy and it would be just as well for the few users or FPS OPT2 to abandon the thing before it kills them all. ;^)

Short explanation, so you get the scope of the significance.

You know what is Quake II? Look what happened when Quake II engine was made GPL-ed (open source): clicky
(We got bunch of cool open-source (free) games based on the Quake II engine. Besides that, various customizations and improvements of the Quake II engine have been made and released so Quake II itself got a huge facelift.)

Although this project is not actually making the original Morrowind open-source, but rather making an open-source re-implementation of Morrowind engine by reverse engineering, the significance is rather the same.

the way i understand all this (even tho i dont fully understand it) is when its done we will b able to do more things, make more mods ect. like we can make graphics better than ever before, we can do so many more things. ooh i just came up with another idea/req will we b able to make more morphy like things? like Werewolves? but instead of werewolves, make a disease that (example only, not saying i want this) turns us into a fish at day (or a bird! )

edit: i quoted the wrong person xD

With open morrowind would it be possible to change the GUI? Because I sort of have a problem with how it looks. I heard that you can change the colors of it but not the design. (sorry if this was asked in a previous page)

It seems people generally don't really understand the idea of "open source" software... :/

With open MW, you can do anything you want. That is, if you are a programmer, lol... But such a thing wouldn't be a problem to implement, I reckon. So yes.

Another idea I thought I'd throw in there.

Companion Paper Dolls: Though companion share is a wonderful thing it has always irked me somewhat that you cannot have them equip specific items over others, providing the companion share window with the NPC's paper doll much like the player's inventory would be neat I think, letting you equip your companions specifically to your taste.

That kinda ruins the fourth wall, especially if your companion is supposed to have his/her own personality (like Julan, for example) and basically turns companions in mindless NPCs who just obey your commands.

And also, this isn't some modding project... I've seen many requests here that kinda fit more to a mod request, like this one. I'm pretty certain that something similar to that could be modded with current MW engine anyway, plus maybe MWE/MWSE.

That is nearly impossible, as I understand it. While many newer systems have better limits (eg, DarkPhysics works up to about 300 million units), the farther you get from origin the worse floating point positioning gets. At trillions of units away, most collision engines start to jump by 1000s.


That's already possible with mods.


Why wouldn't they wear something useful?
It might be possible to have a container checked as 'barter destination'.


That's easy to mod, and I've never had problems with actors responding.



Already possible.

There's already a mod preventing the merchants wearing armors they buy: http://planetelderscrolls.gamespy.com/View...ail&id=5518

And why it would be a problem? It's funny to see Dark Brotherhood merchants everywhere, lol XD

Define useful and I think you'll find your question is almost redundant. Weapons and armor certainly aren't useful to merchants. And they're already dressed.

Well, it's about the same thing as a player when he find something that looks useful or nice to him, lol.

Not really.
Consider that the world they live in is filled to the brim with ravenous beasts, toxic plants, powerful mages, undead, and daemons (called Daedra), armor and weapons are useful to anyone.
And if they didn't have to skills to use them, the AI won't equip it. So, they really are useful.

Could this improve the weather effects? In every TES game the rain and snow are 2D sprites placed over the player's head, even Oblivion annoyingly did this. While in games such as metroid prime, the snow floats down as actual objects which looks so much nicer. Also in Morrowind you can go under shelters and it'll still be raining, which breaks immension. I assume other games have done weather like this, but naturally im unaware of how difficult it is to implement or if it's worth the time.

From what I understand and experience that mod doesn't work very well...

and my vote FOR facegen.

Well, if OpenMW has the capability perhaps some sort of Face Gen can simply be a mod to be added or not as people desire rather than implemented directly into OpenMW? (Same for my above idea I guess).

So.....Ive just read the whole thread chokefull of great ideas, so is this like a brainstorm thread or is someone actually taking up the challenge of comming up with this?

Some one (or two or three) people are working on this ... many of the 'brainstorming' bits are more what could possibly be done with their finished product. I think most of us are throwing out suggestions for the sake of the authors knowing what to let OpenMW support .... as well as alot of general fixes to the core game and modding utilities that have been impossible till now.

So my thoughts was, we throw out some ideas and maybe the authors will go "Hey, that isn't a bad idea, I should tweak that/fix that/send flying monkeys after it".

Companion status at the hud,or a "group/party" screen when you can see an overview of your group...it must recognize all companions,julan,constance,ariela,tammy chang,laurena,mercenaries,etc

I did a very basic version of this. A bit buggy though.

To be honest, I think most of you are approaching this from the wrong direction. What would be far better is the ability to mod what you want in, without having to know any D/C++/OGRE API

Would it mean the end for overlapping alphas once and forever?...

It's not the same thing as the player at all, that's what I was trying to point out to you. The player is an adventurer; he lives and dies based on the weapons and armor he wields. Shopkeepers are shopkeepers, they don't actively seek danger and death like the character is always doing and protecting themselves 24/7 with weapons and armor is far more likely to drive away customers than to keep him safe from any bad guys. I can't imagine how a person so paranoid could even have the social skills to effectively barter with anyone his appearance didn't manage to drive away - and certainly such a person would be very rare indeed, not as common as, well, every single freaking shopkeeper in the game.


This explanation simply doesn't work. Many of the shopkeepers in the game manage small armories of weapons and armor, yet few of them wear any of their armor (for the reasons I've already pointed out and more). Given this fact, how do you explain the logic behind what you just wrote as a justification for this game mechanic?

"Yes," thinks the shopkeeper as he buys a suit of nearly-broken iron armor from an adventurer, "I could wear some of the Imperial steel or silver armor I've smithed, but this guy's stuff is so cool!"

Then, of course, at a later date you'll have this wonderful bit of brilliance to deal with: "What's that?" asks the shopkeeper when the adventurer returns, "oh, yeah, I suppose I could sell this old stuff back to you. Pity, though, I'll be fodder for the next invasion... I've got nothing else to wear in its stead."

There are a few obvious logical flaws in your explanation and how it applies to what actually happens in the game. And in any case, it's not worth anyone's effort for them to go out of their way to try to explain a faulty or broken game mechanic - if it looks like a fish, and smells like a fish, then it's probably a fish. This is a horribly illogical problem in the game and it would be nice if OpenMW included a fix, but obviously not necessary as a mod which corrects it already exists (and has been in-use in my game for many years :^).

Why wasting your time arguing with "illogical" issues if there is already something that fix it?
As far I know, OpenMW is meant to add new functions/features to the game, not to create a mod compilation.

That was a render engine issue, one I don't think OGRE has. NIF had nothing to do with it, AFAIK.

Yes, EXACTLY, thank you!

Well, no problem, lol XD

I think the 'mod' ideas are more 'I'd like to be able to create something like this' rather than 'Put this in, now! right now!'

Something mentioned in This Thread made me originally think to post this. It may be worth it for you guys, Yacoby and Tbrick, to check out that thread and see what kind of ideas people are posting - I think you could make use of some of them.

This is one thing I posted; it's rather a very wishful fantasy, but I think it would be awesome if we could get this in Morrowind (either through the new engine itself or new mod capabilities added by the new engine). Have you played Daggerfall? If so, do you think it is possible to implement some of the logic from the Daggerfall spellmaking/enchanting system into Morrowind? In Daggerfall, you could add negative effects to spells and enchantments to lower their cost to save yourself some magicka, or else to simply allow yourself to add more positive effects to the spell/enchantment to make it even better. Actually, the system was a lot more complex than that (Check This Out), but that's probably the simplest part of it and I guess the most likely to be possible. :^)

id like that too

What a nice day for an excellent idea like mentioned 2 posts above.

How would we create mods for this? Will the old CS be compatiple with this because wouldn't the CS still be based off the old engine and be limited by what it can mod/create too?

Nice to see someone calling for soemthing from Daggerfall rather than yet another (implausible) Oblivion back-port!

However, as DracoNyon reminds us, anything really ought to be fully backwards compatible. (Though I'd not cry if sausage-bodies disappeared)

The CS could still be used for item placement, given that the file format won't change, and MWEdit is capable of using extended script functions (MWEditSE...), so that should take care of that.


I fully support almost anything Daggerfall, I (usually) loved the detail to which you could play that game.

Especially the character generation system was awesome, with those advantages and disadvantages you could add... But that would be even more difficult than the suggested magic idea. Much more.

Actually, that's possible scripting-wise, to a decent extent. If someone managed to institute a levelling difficulty meter (to slow levelling in alternate systems like GCD), then you could script it into CharGen with the current engine. Using MGE, you might even be able to make a visible meter.

This thread keeps coming up with great suggestions! I agree that most of these ideas are mod ideas, but I certainly don't mind getting a survey of what people want. A big brainstorm is good, and even if most of the ideas won't be put into OpenMW directly, I am planning how to make them possible through modding. It also helps to have all of them in one place - I'll be referring to these threads in the future.

Some quick news:

Version 0.4 is just out, get it at http://openmw.snaptoad.com . It should be much easier to compile on Windows since I've put all the DLLs and headers you need in one big zip file.

There's also a developers forum at http://groups.google.com/group/openmw - I just made it so it's still empty. If you have any code to contribute or questions about the code, that's the best place to post it.

Something I noticed being asked for several times is the ability to add more grades of soulgems, which is impossible due to the engine. Also the way currency is handled ought to be expanded to include silver and copper coins to alleviate some of the pressure of the economy (question is if these can be introduced as regular coinage (unlike the dwemer coins which are sellable objects) via a mod or if the changes need to be done on a more fundamental level).

But I find this project very promising, opening up for even more possibilities.

So hey that's cool. But can you post any screenshots\give details on what's been updated? I don't have the most time to say, download and install the whole thing, so if you or anyone else could do that it would be appreciated.

http://openmw.snaptoad.com/news.html

There is also a changelog in the homepage.

another new idea, (if it hasnt been said already) more bones to animate with, i understand that u cant add bones to MW's animation. if we had even 2 more bones we could make more races (or just idle anims) look better, a winged race's wings could now move and flap when u jump.

Great! It compiled from source fine. Can't wait to get into it

One more I'd don't mind not to see again is the 128 character cap on path length for nifs and textures.

You can. MW bone names are hardcoded but for human NPC skeleton tail bones (Bip01 Tail, Bip01 Tail1-3 -> 4 extra bones) as well as PC finger bones (Bip01 L/R Finger 3; Bip01 LR Finger -> 4 extra bones). So, we have 8 free bones. Moreover, this skeleton (human NPC) in original nifs does not have Bip01 L/R Toe0. Max generates these useless bones automatically. So, toe bones are present in all custom MW animations that are created in Max, probably, Blender users can find the way to get rid from them and we might have 10 extra bones.

For me the above mentioned extra bones are working very well.

But agree with that it is better to have no bone number/names restrictions as in other games. I think this is obvious.

I read about that and some of the things its being used for on the Psychodog studios forum. Very impressive stuff! I really can't wait till some of these new models get released. One in particular for me that would go great with my female character. ^^"

Imagine what we could do with no limits on bones though. Quadruped races!

The “model” is not a problem but animations (there are 205(!) required animation sequences to create nearly the scratch (there are no suitable mocaps for many of them); who has stamina/skills/tools/time/interest/animation artistic talent to accomplish such huge work?!) and clothes/armor compatibility. I think this doable/reasonable only for certain NPCs (as in the test plugin I have posted).

Quadruped races. They need hands to fight, new clothes, etc. There are many regular quadrupeds in MW – e.g. some creatures.

=S Umm... What if you replaced the combat animations with something like a bite animation for thrust, a Horizontal paw/claw swipe, and a vertical paw/claw swipe (Note: i know little to nothing about animations)? I can't think of one for spell casting. ^^"

There are some things that I'd love to see in Morrowind:
1. The ability to sit down (yes I know there's already a mod for that but it's not that good)
2. Better animations (I really dislike the choppy default ones)
3. Physics?
4. PLEASE DO NOT MAKE THE MORROWIND UI LIKE THE OBLIVION UI, THE DEFAULT IS GOOD AS IT IS!
5. Better AI ; NPC's that talk with one another, random aventurers, etc.

Did you not read the topic at all? Animations can be replaced right now but making new animations is a long and tedious process, if you want new animations by all means go ahead and make some.

I strongly agree. It made the Redguards look awful (I'm talkin' Oblivion here) and not even represent what they were supposed to look like (the men were fine, but the women? WTF! I couldn't even get the females to look the way they were supposed to without them growing facial hair. Un-friggin-believable!). Seriously, tbrick, don't do facegen.



Or why not a race with four arms to fight with (like those sand raider guys from Diablo II).
Or possibly even... eight-armed Octopus people (not your garden variety, mind you!).

Being able to add to chargen would be great. Namely the generate class by answering questions option. Would be great just to be able to add more questions to it.

As long as we don't have the "Companion Max Magicka set to zero" bug from MWE in the implementation, I'm good with it.

-E. Faust: Avatar of Metal

What confuses me is why is there stuff about that in Morrowind.ini if you can't modify it. Or can you? I never tried...

i thought if u used a tail bone on a NPC that is not classified as beast (which would make it walk beastly, no helmet, no boots.) the game would either go all wacky or just crash. i learn somthing new everyday

This is an interesting question, I have asked similar long time ago. And the answer is not that clear.

As far as I understand the situation in general CS(s) is(are) functionally tightly connected with game engine. CS is basically a higher level, mooder friendly tool for making custom changes (modding) in the original game. I totally agree with Emma that MW has that extraordinary lifetime because it is very mooder friendly and has many good tools for modders.
Hence, the new game engine is cool but would it be “modable” without a new(modified) CS? How modders (people that are not that experienced in programming) can access its numerous new features?

I am really puzzled why game industry is not able to release a game specifically for modders with completely opened simple game engine, importers/exporters for widely used software (Max, Maya (e.g. PLE), Blender, XSI…), CS, textures, models, several scenarios, etc. I would definitely buy it.

When I read this topic I fill embarrassed: one person is working, hundreds (including myself) are watching and only giving countless advices.

Okay, since you're interested in ideas, here's my personal wishlist. Please note that this is the product of a rather weird and mainly uncontrolled brainstorming, so things that are probably relatively easy to do are mentioned next to very hard tasks. Just take it as a personal quarry of ideas; pick up what you like and ignore the rest. Also I've been intentionally terse; if you'd like more details about something, just ask.

1. Bugfixing (or, to be precise, avoiding to reintroduce these bugs ):
- the local ref bug, which causes doubling, vanishing objects (esp. doors), and crashes.
- the bug which lets companions fall to death if you pass through a door while levitating
- the zero hit point bug which kills merged NPCs as soon as they get activated
- the head display bug when you're vampire and werewolf at the same time
- making quicksaves / autosaves reliable
- smarter GMST handling (e.g. use TESTool's fixing algorithm for the mod loading in-game)
- fixing buggy script functions (caveat: There *are* mods that rely on "buggy" behavior of script functions, one gold weight mod comes to my mind)

2. Limitation lifting:
- more than 255 mods
- configurable flush time (i.e. time until old data gets flushed out of the system, hardcoded to 72 hours in Morrowind)
- allow more than one exterior area (might be too drastic a change to implement, but would fix *lots* of mod incompatibilities, and might be feasible by allowing the user to page mods in or out)

3. Better diagnostics
- write more meaningful logs to make tracking crashes easier
- log which objects overwrite each other when building the game world, to better track mod incompatibilities
- supply more information for debugging in-game

4. Engine feature improvements
- include MWSE commands in the engine
- smarter AI pathfinding
- working universal companion teleport
- optional "merge load" feature, i.e. load objects with the same algorithm that TESTool uses to merge them
- store snapshot of game data with all mods loaded (and no game begun yet), afterwards on program start only load this snapshot instead of rebuilding the whole world (unless mod setup has changed, which should force rebuilding)
- scrollable main map
- player-written journal entries
- configurable text size

5. Engine performance.improvements
- increased view distance
- higher FPS rates
- quicker resolving of dialogue responses

Sooo - hope you found something you liked. Thanks for your efforts, and let me know if I can help.

This will be the best thing to ever happen to Morrowind.

IIRC this was mentioned as already being fixed.


Should be easily possible, if some kind of loop is used to load each file in turn (which is how I would/have implement(ed) it).


Might have been mentioned, should be possible.


There will be a whole new AI system, so that might result. Or it could be worse (given tech advances, prolly better).


Can't MWSE do that?


Can already be done.


New render engine and processing engine will result in faster and more efficient all around. At least, in theory. And I know OGRE is much much better than the MW render engine.




One request (side request), a new (possibly integrated) launcher. One that isn't 3 megs because somebody decided to compile the engine into it.

I tested it earlier. You can modify them, but you can not add new ones. The ini does not show the results of the answers though. So all you can really modify is the question and the text on the buttons.

Well that's a bummer. Good thing we'll get OpenMW in a relatively near future.

So far I do not observe a significant difference from original MW on my Win PC (I have tried Dirct3D with OpenMW v.04, btw how to switch to OpenGL?). I suspect that it might be necessary to add Detail/Dark maps to all MW ex static meshes.



The project is on the very early stage to make any responsible estimations. We have no clue how very important engine components might be realized: collision, animation system, lights (MW does not use nif built-in lights), etc.

I usually get an Ogre splash screen that gives you a choice between OpenGL and Dx. Doesn't it pop up for you?

That's not the way to go. Even with MWSE Morrowind's scripting engine is hopelessly limited.
Either stick with Vanilla and complete compatibility OR
Choose another capable scripting engine. My choice would be Python but your mileage may vary.
Then, as Morrowind scripting is so primitive (even with MWSE) write a parser to convert scripts to the chosen scripting language. Bearing in mind the limited capabilities, it shouldn't be too difficult.

Well, modder *have* produced impressive mods with it. Although obviously there *are* many limits which would make modders cry for joy if they could be broken.

MWSE *does* stick with vanilla and complete compatibility, since it's an extension of vanilla. Implementing the MWSE features in OpenMW wouldn't invalidate a single vanilla mod. *Not* implementing MWSE commands would, however, render all MWSE mods unusable in OpenMW, since MWSE can't hook into the OpenMW exe. Therefore, the "complete compatibility" you mention can be achieved specifically with implementing the feature set of MWSE into the OpenMW scripting language, and not without it. Hence I don't understand your assesment?


I'd welcome Python, but tbrick has already stated that scripting will be done in his own language, Monster. Which can be a very good alternative too since this gives him a lot of freedom in how he develops the language. I'm just not sure about the details.

Anyway, my point was that I'd like the compatibility of MWSE mods to be preserved, and the scripting language improved upon. Whether this is done by first implementing the Morrowind scripting language, then the MWSE extensions, and then further extension on top of, or whether it's done by implementing a new scripting engine from scratch and converting existing scripts to it, doesn't matter for this goal.

Personally, I think it's important to either have Morrowind's scripting language as a fully compatible subset of OpenMW's scripting language, or have an automatic way of translating scripts to the new engine, preferably something which the engine could do on the fly. Anything else will probably limit the number of users and modders OpenMW will get.

Its not the loading that is a problem with the original engine, it is that only one byte is used in the save file for storing the mod index


What Psyringe said


I don't think I noticed any optimizations yet, which would have a major impact on speed. I may be wrong though.

True. I think at some point a decision will have to be made whether bidirectional save game compatibility should be maintained. Unidirectional compatibility (i.e. using Morrowind saves in OpenMW) seems to be a goal of the project (which is a good idea imho for at least three different reasons), but I don't know whether it's a goal of the project to maintain the possibility to always transfer a savegame back to the original engine. If yes, then >255 mods won't be feasible.

Probably the best course of action is to maintain bidirectional compatibility for the time being, without precluding the possibility of changing the save format (and maintaining unidirectional compatibility only) at a later date, when the engine has sufficiently matured. Especially since there are at least four different remake projects going on now. As long as they all stick to the same save format, it will be very ease to compare them, which will also make testing a lot easier.

You can also turn it on by setting "Show Ogre Config=yes" in openmw.ini, or with the -oc command line switch. By default it only pops up the first time you run the exe.


This is more or less exactly what I plan to do. The game mechanics will be written in Monster, because it's being designed from the ground up as a game scripting language and will be superior for this purpose. The Morrowind scripts can easily be compiled to the same bytecode as Monster at runtime, so the two scripting languages can run side by side in harmony. I don't see much point in NOT including the MWSE extensions with that, since they would be trivial to add in.


While this is still early on the planning stages yet, I imagine the mod and savegame system will be quite different from Morrowind's. For example, it should be possible to create several "characters" (like many other games do) and keep their savegames and mod settings separate. It should then be possible to try different mod settings easily without messing up your saves, and also to switch mods in-game without restarting the entire engine.


I've thought a lot about this, and so far I've come to the conclusion that OpenMW needs its own savegame format, with options to "import" and "export" to the original format. This is because the "game state" that OpenMW keeps is very different from that in the original engine. I think importing is easier and more useful to most people, so that will probably be done before exporting, though.

Since I didn't see this suggestion here, I just thought I'd add it. I really think it would be awesome if the Calm/Frenzy/Demoralize/Rally line of spells were able to affect players. For example, this would be a template of how Frenzy would work:

When the player is under the Frenzy effect, the player's controls are overridden with automatic actions where the player charges and attacks the person who cast the spell repeatedly (not a problem, unless you are weak in melee).

I think that should make the effects the rest would have on the player farily obvious.

While playing as a magic based character today, I realized something I would like changed in OpenMW: more spells available via the shortcutkeys, since I use a lot more than the nine spells which are available simultaniously. It could be made so you select a few spells/items for a shortcutkey like F1 and then cycle through them by pressing F1.

I just installed and tested the engine on Win2k; runs fine. Was a bit confused during installation since I didn't find any information about where to put the dll files. I simply threw them together with the exe and it seems to work. Sound is a bit choppy but that appears to be a known problem. Rendered scenes are very dark, like in the CS - at first I thought the renderer was displaying a black screen until I moved the mouse and spotted a lantern. My guess is that ambient light isn't implemented yet? Anyway, having a key which adds light (as in the CS) might be helpful.

I tried some obscure combinations of screen size and color depth with both rendering engines (OpenGL, Direct3d) in an attempt to break anything (old beta-testing habits) but didn't succeed.

Also tried to install the compiler (to check whether it works on Win2k) but got stuck early. When installing MinGW 5.1.4, the documentation (compile-win32.txt) tells me to selct gcc and g++ packages. However there is no gcc package to select, the options are "MinGW base tools", "g++ compiler", "g77 compiler", "Ada compiler", "Java compiler", "Objective C compiler", and "MinGW Make". Where is gcc, which ones do I need?

Any help on setting the project up in Code::Blocks would be very much appreciated, but bear in mind that I'm a total noob at IDEs, so I'd probably need a foolproof guide to set it up correctly.

Also just installed on Vista 32-bit. Extracted, ran, and configured just fine on a vanilla install of Morrowind.

About to start going through the source, now. So far, looking great. Eagerly awaiting the continued development.

This will definitely be the MW community's most anticipated project. All of the game's known major bugs vanishing, new features never before possible in Morrowind, better FPS performance, integration with OpenGL: this is only the beginning, but I'm loving it!

not sure if this has been asked or not, but is the map going to be fixed so you can see all land?

This will be like the Morrowind Renaissance.

So, I've been playing around with OpenMW. A few areas crash... The last two that crashed that I tried were "Mournhold, Old Bazaar" and "Skaal Village, The Greathall". "Assu" of course works fine, as does "Balmora, Caius Cosades' House". Not sure what causes the crash, but thought I'd mention those areas. Been taking screenshots of what areas I set it to start in... so far, looking great. Wish I could see exterior cells, though.

I've taken courses on C++, and I know how to program in PHP and other various web-orientated languages... Working on my degree in Software Engineering. If I thought I could help more, I'd offer aid. Still, I'm browsing through your code, getting an idea where everything is.

As I said, everything is looking great. Still going through cells...

Keep up the great work - really looking forward to playing Morrowind in the new engine.

EDIT:
Actually, I have a request as well. When it comes to the HUD, it would be superb if it could be configured in XML files - sort of like Oblivion.

I posted this in the Crystal Scrolls thread and I thought it could be a welcome suggestion here, too:

A very useful idea that just appeared in my mind, about remaking the bugged scripts, but in a way that won't break the mods which actually rely on these scripts being buggy.
For example, let's say that a script named "CheckWhatever" is buggy in the original MW engine. What this new implementation should do is leave a script with the same name (CheckWhatever) buggy, so all the mods who use it and rely on its buggyness won't be broken, but instead implement a new script called, for example, "CheckWhatever_Proper" or something along the lines of that, which actually does what the original CheckWhatever was supposed to (hence the suffix proper), so any modder who wants to use that script and actually wants it to behave the way it's supposed to, can do it using that proper version of the script.

Edit: Also, another thing, which was probably mentioned already. A better scripting language should really be implemented, the original is a disaster. Something C-like if possible, and without the max lines of code and max number of variables and all those silly restrictions. If possible. I don't know if the original scripting language needs to be supported too for the already existing mods to work...

Doccolinni: Good idea, but with some disadvantages: It's prone to introduce errors in *future* mods when authors forget to add the "proper" to a function they call. In cases where the original function worls most of the time, and exhibits buggy behavior only under certain conditions, those bugs in future mods will be hard to recognize unless you know the list of updated functions by heart. Another problem is that this assumes that every use of such a function by any existing mod is intended, which might not be the case, perhaps the mod author just didn't know about the bug, and it's actually breaking his mod. With the method you're suggesting, the mod will have to be updated to the "proper" function calls to work correctly.

I think it will be better to fix the functions, but keep their names, and provide a "Compatibility mode" (like Eisenfaust suggested in the other thread), and change the names of the *buggy* functions into "(function_name)_old". Give the user the ability to specify for each mod whether or not a mod should be loaded in compatibility mode, and if yes, translate all of its relevant function calls to (function_name)_old calls. Optionally, let the user specify a "compatibility threshold time" and automatically load all mods in compatibility mode that were written before that threshold. By default, enable the threshold and set its date to the release date of the engine. (This won't be of much use to the people who know their way around with mods, since they tend to redate them anyways, but might be very convenient for users who just want to play with a couple of mods without having to spend lots of time with installation and compatibility issues).

Also, regarding the new scripting language: tbrick has already said that he will implement his own scripting language that's being developed specifically for gaming purposes, but that the engine will translate old Morrowind script into the new language, to preserve compatibilit of existing mods.

So since the graphics engine is being replaced the question has to be asked:

Will the standard by Morrowind's view fog be in, or more like Oblivion with huge view distances?

Obviously huge view distances being prettier and what most people might consider more "immersive" the option should be included. But how many people would want to play with the original, close in fog? I don't know, but if it's a lot I'm sure they appreciate having the option.

Secondly: Weather! Currently the weather effects rely on the close in view fog, and go through roofs. I haven't looked at OGRE's particle system but I can only assume you there's something that allows snow and rain to not go through static things. But this brings up the question of redoing Morrowinds weather types by hand (other than the sky textures anyway). Which I think will have to be done, I suppose they should be in keeping with the feel of the original as much as possible.

Bringing up one last thing, the skydome. It's 2d, bleh. Does OGRE support better skydomes? I'm pretty sure I saw an OGRE demo with one. It would look much better with a big 3d skybox possibly with multiple layers as opposed to Morrowinds flat 2d one, but I'm not sure the original sky textures would work. Since the goal is to recreate Morrowind I suppose if OGRE can recreate Morrowinds 2d skybox it should be kept in. But if it supports a 3d skybox it would be great if you made sure it was easy to access for modders.

Personally, I think fog suits Morrowind. Not the right-the-hell-in-there-oh-god-I-can't-see during all weather at all times of the day consistently in every region over the entire land, but I feel like most places in the game would be still pretty foggy were more capable computers available at the time.
That's me though. MGE has shown us that most people love being able to see to the edge of the world (if their pc can handle it ). So again, options are desirable.

Here's even more feature requests . Perhaps these are more appropriate for mods in the planned scripting system, but here goes.

- Stealth enhancements

Amount of lighting on the player would be taken into account when sneaking. AI for the npcs to stay a bit more alerted if they catch a player sneaking.

- Time passing while in some menus

Reading a book or having a conversation would pass time. If the player or the npc involved would get attacked, the state the conversation was in would be remembered, and after resolving the situation that conversation could be continued where it left off. The npc involved would actually initiate it if the player isn't willing. In a similar way, repairing gear or enchanting/recharching could be interrupted.

- Offensive alchemy

Oblivionesque distinction between potions and poisons, or alternatively the ability to apply any potion to a weapon. Or a way to equip potions as trowing weapons with the throwing knife animation, using them as grenades.

Hm, you're right. This is one mod that would start working properly that way:

Just curious, how long can we expect until something along the lines of a BETA, or a Relz?
Just curious.


Also:

This seems like it will be incredible, but, would MGE work with it, or would we have to use what you're thinking of implementing?

I'm guessing sometime between now and the end of the universe, if we're lucky.

Another feature idea: What about allowing more than one script per reference?

Currently, Morrowind only allows one single script for each reference. This is a major reason for mod incompatibilities. For example, take this three mods (all of which actually exist):

- Mod A adds a script to weapons to rotate them, for better placement
- Mod B adds a script to weapons to modify their effects
- Mod C adds a script to weapons to unequip them if the the character has a low level, to prevent low-level characters from becoming very powerful by using certain weapons which can be gotten early of you know how

In Morrowind, these mods are mutually exclusive, since all depend on the script they add to the weapons, and since every weapon can only have one single script. The only way make the mods compatible is to manually rewrite the scripts so that they encompass the content of all three mods, which is so difficult to do that it's hardly ever tried.

However, conceptually these three mods are totally independent. Nothing would stop them from working *if* Morrowind could deal with several scripts per reference. Allowing this would remove another major obstacle of mod compatibility.So, what do you think?

The project is in its very early stages, I think it's much too early to give an ETA or to speculate how good its chances are to actually get finished. I also suspect that any release date will be heavily dependent on the number of people who are willing to contribute to the project - everyone can help making this dream a reality.


MGE probably won't work with it, but most of the functionality that MGE provides can be programmed in OpenMW too. It might acually be easier to do it in OpenMW since OpenMW is built on OGRE, a modern rendering engine which has some features of MGE already implemented, while MGE sits on top of the original Morrowind engine and therefore keeps being bound to some of its limitations.

The only problem with that is template invalidation, the last loaded is considered correct. It would take some creative (and potentially problematic in other areas) scripting to allow multiple scripts per reference, since you would have to cache the names. If a mod replaced the script on an object, meaning for only the new one to run, that could cause conflicts.

New thread Here.