Skyrim Information

[jedimembrain]

It seems like item durability is also out of the game; nothing of it is shown in the menus during demos, and repairing weapons/armor hasn't been mentioned as part of the Blackmithing skill. I welcome this change. Other thoughts?

Serious? It was tedious, but I found it added a form of realism. I was hoping you'd have to find ore to repair your gear this time...Or if something was 0% you'd have to forge a new blade. So many cool ideas that will end up mods now... *cries*

 

I would gladly ditch the realism with the tediousness; it was too much for me to always carry hammers around, going through the list of weapons/armor and keeping things at maximum quality just to maintain that extra armor point/attack point. AND don't get me started on the slow, crawling grind to 50 Armorer just so I can repair my own magical items. Gyah! Apparently that's a much bigger peeve of mine from Oblivion than I thought...

 

Anyway, I actually like the idea of a weapon needing reforging at 0%. Realism, without the feeling of it being a chore! Maybe just for mundane weapons, though.

 

I wouldnt ditch repairing either but it was a little annoying .. I like the idea of an armor or weapon only loosing stat points when its completely broken though .. it would keep the realism without pissing me off when I'm not paying attention to it.

[Lowk]

I wouldnt ditch repairing either but it was a little annoying .. I like the idea of an armor or weapon only loosing stat points when its completely broken though .. it would keep the realism without pissing me off when I'm not paying attention to it.

 

Interesting...maybe instead of making the item unusable, just cut its effectiveness in half? I'd vastly prefer that.

[Ghogiel]

People have to note there will be massively more armor sets in skyrim.

 

The chest/legs were merged so the dev team could make more armor sets quicker.

 

How does merging them make it more rapid? If anything, it slows development down because you have to merge it all together. Only benefit I can see is that you have the waist seam right there to see.

It doesn't really make it quicker to make a one piece* (though it does add a very tiny amount of time. You have to draw some extra poly on there after all. It won't appear magically you have to put them there. AND you need to make another world model. I would say the time added is less than statistically negligible. like 1% increase in time ior something silly. How do I estimate that? because I have done it at least 25 times. It's still a guess.

 

So it doesn't make it slower at all. You don't have to actively combine it. you just make it from the start either with the intention it'll be 1 piece or split when you build the low poly.

 

* This is such BS anyway, the armor we have seen is obviously very modular, in other words they are almost certainly in several pieces already and in game they are using something akin to formlists and armoraddons. Meaning to achieve the variations we see, it actually is split into even more pieces than just cuirass and greaves

 

The pauldrons, the straps, the chest pieces, I have even seen at least 1 variation or the belt nearly every one is wearing, it appears in variations, one strap with that cuirass part, then again with a different cuiraas. How do you expect they achieve that, without actually making the same model twice (because that seems kinda stupid), but only making it one time making it with some extra straps?

 

IMO they did it how I would have done it. < I don't recommend it, but you could try to develope and make a set of interchangeable out fits that makes dozens of variations in game from far fewer lego pieces, do so and then lets discuss an even better way to achieve variation that is low cost to dev time and performance. :thumbsup:

 

Better idea?

Edited August 23, 2011 by Ghogiel

[jedimembrain]

People have to note there will be massively more armor sets in skyrim.

 

The chest/legs were merged so the dev team could make more armor sets quicker.

 

How does merging them make it more rapid? If anything, it slows development down because you have to merge it all together. Only benefit I can see is that you have the waist seam right there to see.

It doesn't really make it quicker to make a one piece* (though it does add a very tiny amount of time. You have to draw some extra poly on there after all. It won't appear magically you have to put them there. AND you need to make another world model. I would say the time added is less than statistically negligible. like 1% increase in time ior something silly. How do I estimate that? because I have done it at least 25 times. It's still a guess.

 

So it doesn't make it slower at all. You don't have to actively combine it. you just make it from the start either with the intention it'll be 1 piece or split when you build the low poly.

 

* This is such BS anyway, the armor we have seen is obviously very modular, in other words they are almost certainly in several pieces already and in game they are using something akin to formlists and armoraddons. Meaning to achieve the variations we see, it actually is split into even more pieces than just cuirass and greaves

 

The pauldrons, the straps, the chest pieces, I have even seen at least 1 variation or the belt nearly every one is wearing, it appears in variations, one strap with that cuirass part, then again with a different cuiraas. How do you expect they achieve that, without actually making the same model twice (because that seems kinda stupid), but only making it one time making it with some extra straps?

 

IMO they did it how I would have done it. < I don't recommend it, but you could try to develope and make a set of interchangeable out fits that makes dozens of variations in game from far fewer lego pieces, do so and then lets discuss an even better way to achieve variation that is low cost to dev time and performance. :thumbsup:

 

Better idea?

 

Its MUCH faster to make a single suit than a split suit .. half the rigging .. you dont need to worry about clipping between top and bottom meshed .. half the LOD sets .. one set of textures .. one UV map for the textures .. then the topology "the layout of the wireframe of the mesh" only has to be done once too .. there is no argument about this .. no possible way its harder and its almost without a doubt quicker in every aspect .. case closed please and thank you

Edited August 23, 2011 by jedimembrain

[Alcrin]

Serious? It was tedious, but I found it added a form of realism. I was hoping you'd have to find ore to repair your gear this time...Or if something was 0% you'd have to forge a new blade. So many cool ideas that will end up mods now... *cries*

 

I would gladly ditch the realism with the tediousness; it was too much for me to always carry hammers around, going through the list of weapons/armor and keeping things at maximum quality just to maintain that extra armor point/attack point. AND don't get me started on the slow, crawling grind to 50 Armorer just so I can repair my own magical items. Gyah! Apparently that's a much bigger peeve of mine from Oblivion than I thought...

 

Anyway, I actually like the idea of a weapon needing reforging at 0%. Realism, without the feeling of it being a chore! Maybe just for mundane weapons, though.

 

You both may be interested in knowing it's confirmed in one of the new interviews (think it's the one with the redhead?) that you can "reinforce" existing items with the Smithing skill, in addition to making new ones. Scarce few details on what this actually means, of course, but hopefully it means you can have the benefits of buffing your items to 125 effectiveness without the tediousness of upkeeping it.

[Ghogiel]

People have to note there will be massively more armor sets in skyrim.

 

The chest/legs were merged so the dev team could make more armor sets quicker.

 

How does merging them make it more rapid? If anything, it slows development down because you have to merge it all together. Only benefit I can see is that you have the waist seam right there to see.

It doesn't really make it quicker to make a one piece* (though it does add a very tiny amount of time. You have to draw some extra poly on there after all. It won't appear magically you have to put them there. AND you need to make another world model. I would say the time added is less than statistically negligible. like 1% increase in time ior something silly. How do I estimate that? because I have done it at least 25 times. It's still a guess.

 

So it doesn't make it slower at all. You don't have to actively combine it. you just make it from the start either with the intention it'll be 1 piece or split when you build the low poly.

 

* This is such BS anyway, the armor we have seen is obviously very modular, in other words they are almost certainly in several pieces already and in game they are using something akin to formlists and armoraddons. Meaning to achieve the variations we see, it actually is split into even more pieces than just cuirass and greaves

 

The pauldrons, the straps, the chest pieces, I have even seen at least 1 variation or the belt nearly every one is wearing, it appears in variations, one strap with that cuirass part, then again with a different cuiraas. How do you expect they achieve that, without actually making the same model twice (because that seems kinda stupid), but only making it one time making it with some extra straps?

 

IMO they did it how I would have done it. < I don't recommend it, but you could try to develope and make a set of interchangeable out fits that makes dozens of variations in game from far fewer lego pieces, do so and then lets discuss an even better way to achieve variation that is low cost to dev time and performance. :thumbsup:

 

Better idea?

 

Its MUCH faster to make a single suit than a split suit .. half the rigging .. you dont need to worry about clipping between top and bottom meshed ..

You just split by material ID at export. I only rig the entire thing once. I'll give you the clipping though. Takes a couple mins to eyeball it. There are no problems with that on a single outfit. I wan to draw attention to this: It only becomes a problem when you are designing an entire range of interchangeable outfits, it will limit the outfit design so they have to all be somewhat compatible with each other. Which is why they combined them, forget all the other BS they have said why it was combined, all that performance gain and now they can have several more actors on screen directly because of this. It's not, it's far more likely because of combining the cuirass and greaves allows the art team to get extra funky in armor proportion and silhouette. and the outfits probably are modular anyway, that is imo how they have all those variations on the same outfit. (better idea?)

 

half the LOD sets ..

Perhaps not relevant to this game, a discrete LOD system may cause issue with limited vram space on the consoles, it's why they didn't use it on heavy assets like characters in previous BGS games, it is simply faster just to keep firing more geo onto screen than it is to load more objects into mem on consoles. And the clutter never used a discrete LOD system either. They used actor and object fade to remove them from being drawn. only discrete system was for the distant buildings.

And not really. You still have to make the LOD for the model either way, it's still comprised of a cuirass and greaves whether you split or not. The added time here is negligible.

 

I think you might have worked on discrete LOD for characters? I know your pain, it's pretty damn annoying isn't it.

one set of textures .. one UV map for the textures ..

I find this actually the most time consuming part. Some artists don't find it a problem or a time sink. I think they are crazy. 1 sheet is always faster to paint than 2 imo. Sometimes by a reasonable amount.I'll concede the point. Just bare in mind technically you still only have to paint the same texture, it's just across 2 sheets.

 

then the topology "the layout of the wireframe of the mesh" only has to be done once too ..

Don't understand this. You don't need to dumb down your 3d language for me to understand. I did actually say it takes longer to make the split outfit than the one that isn't. what you seem to be arguing over is how much time it takes you to do it vs how long it takes me. I guestimate it adds a small increase in to my development time.

 

Anyway the point is, they probably have split their outfits into several pieces, and all of this they combined the greaves and cuirass because it is faster is probably not true at all.

Edited August 23, 2011 by Ghogiel

[jedimembrain]

People have to note there will be massively more armor sets in skyrim.

 

The chest/legs were merged so the dev team could make more armor sets quicker.

 

How does merging them make it more rapid? If anything, it slows development down because you have to merge it all together. Only benefit I can see is that you have the waist seam right there to see.

It doesn't really make it quicker to make a one piece* (though it does add a very tiny amount of time. You have to draw some extra poly on there after all. It won't appear magically you have to put them there. AND you need to make another world model. I would say the time added is less than statistically negligible. like 1% increase in time ior something silly. How do I estimate that? because I have done it at least 25 times. It's still a guess.

 

So it doesn't make it slower at all. You don't have to actively combine it. you just make it from the start either with the intention it'll be 1 piece or split when you build the low poly.

 

* This is such BS anyway, the armor we have seen is obviously very modular, in other words they are almost certainly in several pieces already and in game they are using something akin to formlists and armoraddons. Meaning to achieve the variations we see, it actually is split into even more pieces than just cuirass and greaves

 

The pauldrons, the straps, the chest pieces, I have even seen at least 1 variation or the belt nearly every one is wearing, it appears in variations, one strap with that cuirass part, then again with a different cuiraas. How do you expect they achieve that, without actually making the same model twice (because that seems kinda stupid), but only making it one time making it with some extra straps?

 

IMO they did it how I would have done it. < I don't recommend it, but you could try to develope and make a set of interchangeable out fits that makes dozens of variations in game from far fewer lego pieces, do so and then lets discuss an even better way to achieve variation that is low cost to dev time and performance. :thumbsup:

 

Better idea?

 

Its MUCH faster to make a single suit than a split suit .. half the rigging .. you dont need to worry about clipping between top and bottom meshed ..

You just split by material ID at export. I only rig the entire thing once. I'll give you the clipping though. Takes a couple mins to eyeball it. There are no problems with that on a single outfit. I wan to draw attention to this: It only becomes a problem when you are designing an entire range of interchangeable outfits, it will limit the outfit design so they have to all be somewhat compatible with each other. Which is why they combined them, forget all the other BS they have said why it was combined, all that performance gain and now they can have several more actors on screen directly because of this. It's not, it's far more likely because of combining the cuirass and greaves allows the art team to get extra funky in armor proportion and silhouette. and the outfits probably are modular anyway, that is imo how they have all those variations on the same outfit. (better idea?)

 

half the LOD sets ..

Perhaps not relevant to this game, a discrete LOD system may cause issue with limited vram space on the consoles, it's why they didn't use it on heavy assets like characters in previous BGS games, it is simply faster just to keep firing more geo onto screen than it is to load more objects into mem on consoles. And the clutter never used a discrete LOD system either. They used actor and object fade to remove them from being drawn. only discrete system was for the distant buildings.

And not really. You still have to make the LOD for the model either way, it's still comprised of a cuirass and greaves whether you split or not. The added time here is negligible.

 

I think you might have worked on discrete LOD for characters? I know your pain, it's pretty damn annoying isn't it.

one set of textures .. one UV map for the textures ..

I find this actually the most time consuming part. Some artists don't find it a problem or a time sink. I think they are crazy. 1 sheet is always faster to paint than 2 imo. Sometimes by a reasonable amount.I'll concede the point. Just bare in mind technically you still only have to paint the same texture, it's just across 2 sheets.

 

then the topology "the layout of the wireframe of the mesh" only has to be done once too ..

Don't understand this. You don't need to dumb down your 3d language for me to understand. I did actually say it takes longer to make the split outfit than the one that isn't. what you seem to be arguing over is how much time it takes you to do it vs how long it takes me. I guestimate it adds a small increase in to my development time.

 

Anyway the point is, they probably have split their outfits into several pieces, and all of this they combined the greaves and cuirass because it is faster is probably not true at all.

 

I get issues when simply splitting a mesh in two .. I work from a high res mesh to a low res one in Zbrush .. I use the retrotopology tools and rarely touch 3ds or blender unless I need to rig something .. so having broken up topology is always a pain for me and my workflow which goes "High res billion poly model with all basic features ready, Retrotopo model with converted features and a good low res topo mesh, Polypainted high res model ready to be converted to textures and normal maps which will just bake right onto the low res model, Import into the game or rig in blender" .. So spliting the waist means a new step in another program for me .. and most studios are using zbrush now too.. btw I totally agree with the UV map thing .. one UV map is always the best way to go .. but if you did that with split parts your just rendering textures that might not be on screen if the character only has one half of the torso which would mean double the textures .. minor optimization thing I guess but body textures tend to be the highest res thing on screen at any given time so double isnt good... I still say its easier over all .. less hassle making changes and you can do crazier things without having to worry about the gap in the waist.

[LeopoldCrank]

Can you guys make another topic to discuss this intensive mod stuff. I mean, I am interested in that too, but I'm trying to find information and every day I have to wade through mountain loads of discussions about whether its easier to make a mesh of two items, two items then combined or one item to represent them both...

 

I'd just like my info please... And maybe some light banter about how the dark elf chick is horrific to look at and then one random guy mentions how he really [really if you follow my drift :wink:] liked the first female Redguard screenshot and then we all go silent, half because we're disturbed and half because we were thinking the same thing...

 

That's the information thread I know and love

[Iv000]

New interview:

http://www.gametrailers.com/video/gc-11-elder-scrolls/719544

 

Info:

The beginning will be more like Morrowind, you first play for a bit and then later you create your character.

Each race has bonuses. Argonians breathe under water, Orcs are good at combat and smithing etc

There are a lot of random events. Wolves hunting a deer, guards escorting prisoners.

[Ghogiel]

 

I get issues when simply splitting a mesh in two .. I work from a high res mesh to a low res one in Zbrush .. I use the retrotopology tools and rarely touch 3ds or blender unless I need to rig something .. so having broken up topology is always a pain for me and my workflow which goes "High res billion poly model with all basic features ready, Retrotopo model with converted features and a good low res topo mesh, Polypainted high res model ready to be converted to textures and normal maps which will just bake right onto the low res model, Import into the game or rig in blender" .. So spliting the waist means a new step in another program for me .. and most studios are using zbrush now too.. btw I totally agree with the UV map thing .. one UV map is always the best way to go .. but if you did that with split parts your just rendering textures that might not be on screen if the character only has one half of the torso which would mean double the textures .. minor optimization thing I guess but body textures tend to be the highest res thing on screen at any given time so double isnt good... I still say its easier over all .. less hassle making changes and you can do crazier things without having to worry about the gap in the waist.

I do pretty much the same sort of thing. I'll use both ZB and max, for either sculpted or subdivision highpoly, depending on what is need to create the high poly, then later create the low poly. If I start in ZB and do it entirely from there, I tend to use a lot of separate sub tools, So when I go for retopo, I'll just select the subtools that comprise the lower half and retopo them as a separate object. In fact even if the thing is going to be exported as a single object, and I unwrap onto 1 UV sheet, I'll still make separate objects all the way up until I make the UV, at which point I'll combine. Like even if the mesh would in the end be one object in game, I would still have a dozen or more objects. Like straps, pauldrons, belts, boots, shoelaces are all going to be separate high poly objects/subtools. And then check this out... sometimes after UVing the low poly tom one sheet, I split off a couple pieces again just for baking :P :wacko:

 

LeopoldCrank : yeah sorry, though it is semi relevent to the current discussion: but could be continued in one of the armor ones

Edited August 23, 2011 by Ghogiel