Critical Mounting Questions
by Gregory Dudding · in Torque Game Engine · 02/02/2005 (2:29 pm) · 20 replies
I need to mount objects to my player... several of them... I need headgear, guns, holsters, swords, capes, belts, dresses, ammo clips, goggles, helmets, etc...
So as I was getting into this, I learned of an 8 mountpoint limit in torque, which sort of put me off. I asked around in the IRC channel and received some assistance on this issue, but it was mostly a talk about the difference between mounting images and objects...
As someone who is really just starting out with this part of the engine... I need a very simple set of answers to the following questions, listed by order of importance to me:
1) Can multiple objects be mounted to the same mountpoint?
Examples:
a special glove AND a gun on the mount0 point?
A set of goggles to a mountpoint in the head, and a helmet?
A Sword and a shotgun both holstered on the player's back with different offsets, one on the left, one on the right, possibly with 2 additional objects being a holster and a scabbard?
2) What are the limitations of Torque's mounting system? Can I only have 8 Image objects as default? Can I mount my helmets and goggles and scabbards and such as Objects?
3) Where is the documentation for this stuff? I've looked over the documentation quite a bit and I don't see much regarding mountpoints in the DTS exporter stuff or the other portions... I may have overlooked it as I haven't read every shred, but, if it's out there, I'd love to see some good comments/descriptions of how mounting objects to the player works.
4) What kind of impact on the game will it have to mount several things to a player at once? Will it cause huge slowdown or can the engine take it?
I am trying to create a solid skeleton with all the mountpoints and hitboxes linked in that I can distribute to my art guys so they can create good character models in one skinning and submission. This is critical to the nature of our game as we need highly customizable characters, and visual indicators of another person's equipment (You need to see the weapon in the guy's hands and what kind of other gear he has at his disposal before you attack him... is he a miner? Is he a bodyguard? Does he have explosives strapped to his chest? Maybe I should Melee him because he doesn't have a sword on his side or back... etc)
So as I was getting into this, I learned of an 8 mountpoint limit in torque, which sort of put me off. I asked around in the IRC channel and received some assistance on this issue, but it was mostly a talk about the difference between mounting images and objects...
As someone who is really just starting out with this part of the engine... I need a very simple set of answers to the following questions, listed by order of importance to me:
1) Can multiple objects be mounted to the same mountpoint?
Examples:
a special glove AND a gun on the mount0 point?
A set of goggles to a mountpoint in the head, and a helmet?
A Sword and a shotgun both holstered on the player's back with different offsets, one on the left, one on the right, possibly with 2 additional objects being a holster and a scabbard?
2) What are the limitations of Torque's mounting system? Can I only have 8 Image objects as default? Can I mount my helmets and goggles and scabbards and such as Objects?
3) Where is the documentation for this stuff? I've looked over the documentation quite a bit and I don't see much regarding mountpoints in the DTS exporter stuff or the other portions... I may have overlooked it as I haven't read every shred, but, if it's out there, I'd love to see some good comments/descriptions of how mounting objects to the player works.
4) What kind of impact on the game will it have to mount several things to a player at once? Will it cause huge slowdown or can the engine take it?
I am trying to create a solid skeleton with all the mountpoints and hitboxes linked in that I can distribute to my art guys so they can create good character models in one skinning and submission. This is critical to the nature of our game as we need highly customizable characters, and visual indicators of another person's equipment (You need to see the weapon in the guy's hands and what kind of other gear he has at his disposal before you attack him... is he a miner? Is he a bodyguard? Does he have explosives strapped to his chest? Maybe I should Melee him because he doesn't have a sword on his side or back... etc)
#2
02/02/2005 (11:30 pm)
LOL, you only needed to wait two more minutes. But no harm done.
#3
Does anyone have a suggestion for how to do this?
I'm trying to use mountpoints to allow players to have gear on them, like shoulderpads and the things i've mentioned above...
What's the "proper" way to mount say, a helmet, to a player model in Torque?
Is a helmet, goggles, shoulderpads, 2 guns, 2 interchangeable holsters/scabbards, something on the player's back and a cloak too much?
A total of 10 things sticking to the player at a given time? Maybe less if the shoulderpads can be made into 1 mount?
02/02/2005 (11:41 pm)
Ahh, this is bad news for me... I really kinda need to be able to attach models together, and have a lot of players on the screen at once...Does anyone have a suggestion for how to do this?
I'm trying to use mountpoints to allow players to have gear on them, like shoulderpads and the things i've mentioned above...
What's the "proper" way to mount say, a helmet, to a player model in Torque?
Is a helmet, goggles, shoulderpads, 2 guns, 2 interchangeable holsters/scabbards, something on the player's back and a cloak too much?
A total of 10 things sticking to the player at a given time? Maybe less if the shoulderpads can be made into 1 mount?
#4
"Yes it can be done. No it will not be as easy as you wish."
But in your case it's still not that bad. You said you needed to have a lot of players on screen at once. This does not specify if you are running a networked game or single payer, and it does not specify real player or AI players. Having a lot of players on screen is not a problem with the mounting, but it is a problem with the networking. If you want to have 20 players around the US logged into a server and all of them doing battle with 10 mounts each, then yes, this might be pushing things a bit in TGE's current state. I've never tried such a feat so honestly I cannot say for sure if it will or will not work or how well it will work. Just by using experience and the testimony of others my gut instinct is that even with the above parameters(20 players, 10 items mounted each) that it would work, but it might be a heavy load for the net code and might cause lag and for sure some image or object lag on the mounts. Less than satisfactory performance. If this is a single player and you want to have 20 AI running around, then Net limitations will not really be a factor or a problem. Still, worse case scenario you could get this done with some tweaking to the code to make it more suitable for what functionality you need.
TGE is an all around good engine, but it's built to please the masses and therefore has to strike a balance everywhere it can. Many times people like you and I will want to push that balance in some certain direction that few others(if any) have ever wished to push. That's what having the SDK is all about. You can take what's already good, and make it better for your needs. So the bottom line here is, if you want it, then by god, go get on it, and get it done. LOL
:-)
02/03/2005 (1:36 am)
The law of Torque is simple..."Yes it can be done. No it will not be as easy as you wish."
But in your case it's still not that bad. You said you needed to have a lot of players on screen at once. This does not specify if you are running a networked game or single payer, and it does not specify real player or AI players. Having a lot of players on screen is not a problem with the mounting, but it is a problem with the networking. If you want to have 20 players around the US logged into a server and all of them doing battle with 10 mounts each, then yes, this might be pushing things a bit in TGE's current state. I've never tried such a feat so honestly I cannot say for sure if it will or will not work or how well it will work. Just by using experience and the testimony of others my gut instinct is that even with the above parameters(20 players, 10 items mounted each) that it would work, but it might be a heavy load for the net code and might cause lag and for sure some image or object lag on the mounts. Less than satisfactory performance. If this is a single player and you want to have 20 AI running around, then Net limitations will not really be a factor or a problem. Still, worse case scenario you could get this done with some tweaking to the code to make it more suitable for what functionality you need.
TGE is an all around good engine, but it's built to please the masses and therefore has to strike a balance everywhere it can. Many times people like you and I will want to push that balance in some certain direction that few others(if any) have ever wished to push. That's what having the SDK is all about. You can take what's already good, and make it better for your needs. So the bottom line here is, if you want it, then by god, go get on it, and get it done. LOL
:-)
#5
I hadn't really planned to rewrite all the torque network code... Any idea how much load in k/sec per client it is as the number of mountpoints goes up? The resource that ups the number of mountable images to 32 doesn't really say anything about an impact on net performance... though it does change the number of bits I think... Is that increase too much to still have a good multiplayer game with a lot of people?
www.garagegames.com/index.php?sec=mg&mod=resource&page=view&qid=6723
That's the resource I'm talking about...
Even if there's another way to accomplish adding meshes to the player and swapping them in and out that I haven't thought of... that would help me... any ideas where I should go with this?
Quick Edit: I could solve all of my problems if mounting multiple images to one mountpoint would not mess things up for multiplayer... does each mounted image require more network overhead, etc, or is it each mountpoint with a mounted image? If I could put both helmet and goggles on the same mountpoint, disable torque's auto swap for that slot, and do that 3 or 4 times around my model, I could easily take care of everything with the 8 mounts available...
Or is mounting 2 objects to 1 mountpoint the same as mounting 2 objects to 2 mountpoints?
Thanks!
02/03/2005 (10:02 am)
Well, my plans are for a networked game...I hadn't really planned to rewrite all the torque network code... Any idea how much load in k/sec per client it is as the number of mountpoints goes up? The resource that ups the number of mountable images to 32 doesn't really say anything about an impact on net performance... though it does change the number of bits I think... Is that increase too much to still have a good multiplayer game with a lot of people?
www.garagegames.com/index.php?sec=mg&mod=resource&page=view&qid=6723
That's the resource I'm talking about...
Even if there's another way to accomplish adding meshes to the player and swapping them in and out that I haven't thought of... that would help me... any ideas where I should go with this?
Quick Edit: I could solve all of my problems if mounting multiple images to one mountpoint would not mess things up for multiplayer... does each mounted image require more network overhead, etc, or is it each mountpoint with a mounted image? If I could put both helmet and goggles on the same mountpoint, disable torque's auto swap for that slot, and do that 3 or 4 times around my model, I could easily take care of everything with the 8 mounts available...
Or is mounting 2 objects to 1 mountpoint the same as mounting 2 objects to 2 mountpoints?
Thanks!
#6
I hadn't really planned to rewrite all the torque network code... Any idea how much load in k/sec per client it is as the number of mountpoints goes up? The resource that ups the number of mountable images to 32 doesn't really say anything about an impact on net performance... though it does change the number of bits I think... Is that increase too much to still have a good multiplayer game with a lot of people?
www.garagegames.com/index.php?sec=mg&mod=resource&page=view&qid=6723
That's the resource I'm talking about...
Even if there's another way to accomplish adding meshes to the player and swapping them in and out that I haven't thought of... that would help me... any ideas where I should go with this?
02/03/2005 (3:19 pm)
Well, my plans are for a networked game...I hadn't really planned to rewrite all the torque network code... Any idea how much load in k/sec per client it is as the number of mountpoints goes up? The resource that ups the number of mountable images to 32 doesn't really say anything about an impact on net performance... though it does change the number of bits I think... Is that increase too much to still have a good multiplayer game with a lot of people?
www.garagegames.com/index.php?sec=mg&mod=resource&page=view&qid=6723
That's the resource I'm talking about...
Even if there's another way to accomplish adding meshes to the player and swapping them in and out that I haven't thought of... that would help me... any ideas where I should go with this?
#7
It sounds like some of the meshes that you want mounted are not functional but more for looks or visibile feedback. i.e. a glove image to show that the player now has gloves. A cape to show it has a cape. My suggestion is that consider this resource for non-active mounts. Now, things like weapons and shield will likely still need to be mounted, but I'm picturing this wholly encumbered guy with 30+ mounted images walking around trying to fight. i.e. I have trouble seeing the need for a ton of mounted (active) images. The 'hiding meshes' resource looks like a possible (elegant) solution to your desire.
Note: Mounting lots of images to your shapes and then having lots of images is going to be somewhat expensive rendering wise too. If you're thinking along the lines of an RTS, you might also examine the RTS pack.
Hope some of this helps!
[HOW]EdM|EGTGE
02/03/2005 (5:33 pm)
Gregory,It sounds like some of the meshes that you want mounted are not functional but more for looks or visibile feedback. i.e. a glove image to show that the player now has gloves. A cape to show it has a cape. My suggestion is that consider this resource for non-active mounts. Now, things like weapons and shield will likely still need to be mounted, but I'm picturing this wholly encumbered guy with 30+ mounted images walking around trying to fight. i.e. I have trouble seeing the need for a ton of mounted (active) images. The 'hiding meshes' resource looks like a possible (elegant) solution to your desire.
Note: Mounting lots of images to your shapes and then having lots of images is going to be somewhat expensive rendering wise too. If you're thinking along the lines of an RTS, you might also examine the RTS pack.
Hope some of this helps!
[HOW]EdM|EGTGE
#8
Really, the full list of what we NEED according to our game design, for the characters is:
Right Hand - Weapon
Left Hand - Weapon or shield (we want John Woo gunplay for sure! Or double swords.)
Back - The player can put a weapon, maybe 2 here (we don't like the whole weapon just vanishes deal
Head - helmet, goggles, glasses, comm gear (we can deal with only having one visible at a time, or make some available combos)
Left Shoulder - For shoulder armor pad
Right Shoulder - Other shoulder armor pad
Left Hip - Holster/Scabbard for gun/sword
Right Hip - Holster/Scabbard for gun/sword
That looks like 8 to me, we could possibly forgo the shoulder pads... Torque is built to support 8 or 9 mounts... then if we do the visibility thing for the cloak, that covers our needs... The visibility trick won't really work with some of these things because we need the ability to mount different types... most of our actual character customization will come through skinning and such... but some things you just need more geometry for...
So, I mean, we don't need 30 things per se, but, a good solid 8-10 would really be sweet... and if we could mount 1-2 things per point without being penalized, that would be cool... but I don't know the effects of it for networking... like those holster mounts, if I could pin the guns to the holsters as well, that would be kosher... but would require 2 things mounted in the same spot (though the hands would no longer have anything mounted...)
I understand what you guys are saying about the non-active meshes not needing to be images, but, I can't see a feasible way for our modellers to add "hard coded" meshes... I thought about holsters, but, we kinda needed those to be of different types and sizes... I'll rethink what I can have my modellers include in the mesh... I am adding that resource just because there's no real reason not to... it adds a very useful functionality that I'm sure I can use... (maybe even use it to blow off heads :) )
Anyway, I still need more input on this... thank you for pointing me to that, Master Edward.
02/03/2005 (8:30 pm)
Interesting resource... can't segmented meshes have pieces of them hidden with a visibility animation though? In the garagegames channel I was asking around about this and someone said I could take the cloaks or flowing clothes and build them into my models as seperate meshes, and then do a visibility DSQ for the cloaks and dresses and such that just flips the visibility off for that object... I know it would still be rendered and animated, but, if I only intend to do this with a few hundred polygons per model, would this be a problem?Really, the full list of what we NEED according to our game design, for the characters is:
Right Hand - Weapon
Left Hand - Weapon or shield (we want John Woo gunplay for sure! Or double swords.)
Back - The player can put a weapon, maybe 2 here (we don't like the whole weapon just vanishes deal
Head - helmet, goggles, glasses, comm gear (we can deal with only having one visible at a time, or make some available combos)
Left Shoulder - For shoulder armor pad
Right Shoulder - Other shoulder armor pad
Left Hip - Holster/Scabbard for gun/sword
Right Hip - Holster/Scabbard for gun/sword
That looks like 8 to me, we could possibly forgo the shoulder pads... Torque is built to support 8 or 9 mounts... then if we do the visibility thing for the cloak, that covers our needs... The visibility trick won't really work with some of these things because we need the ability to mount different types... most of our actual character customization will come through skinning and such... but some things you just need more geometry for...
So, I mean, we don't need 30 things per se, but, a good solid 8-10 would really be sweet... and if we could mount 1-2 things per point without being penalized, that would be cool... but I don't know the effects of it for networking... like those holster mounts, if I could pin the guns to the holsters as well, that would be kosher... but would require 2 things mounted in the same spot (though the hands would no longer have anything mounted...)
I understand what you guys are saying about the non-active meshes not needing to be images, but, I can't see a feasible way for our modellers to add "hard coded" meshes... I thought about holsters, but, we kinda needed those to be of different types and sizes... I'll rethink what I can have my modellers include in the mesh... I am adding that resource just because there's no real reason not to... it adds a very useful functionality that I'm sure I can use... (maybe even use it to blow off heads :) )
Anyway, I still need more input on this... thank you for pointing me to that, Master Edward.
#9
Really, the full list of what we NEED according to our game design, for the characters is:
Right Hand - Weapon
Left Hand - Weapon or shield (we want John Woo gunplay for sure! Or double swords.)
Back - The player can put a weapon, maybe 2 here (we don't like the whole weapon just vanishes deal
Head - helmet, goggles, glasses, comm gear (we can deal with only having one visible at a time, or make some available combos)
Left Shoulder - For shoulder armor pad
Right Shoulder - Other shoulder armor pad
Left Hip - Holster/Scabbard for gun/sword
Right Hip - Holster/Scabbard for gun/sword
That looks like 8 to me, we could possibly forgo the shoulder pads... Torque is built to support 8 or 9 mounts... then if we do the visibility thing for the cloak, that covers our needs... The visibility trick won't really work with some of these things because we need the ability to mount different types... most of our actual character customization will come through skinning and such... but some things you just need more geometry for...
So, I mean, we don't need 30 things per se, but, a good solid 8-10 would really be sweet... and if we could mount 1-2 things per point without being penalized, that would be cool... but I don't know the effects of it for networking... like those holster mounts, if I could pin the guns to the holsters as well, that would be kosher... but would require 2 things mounted in the same spot (though the hands would no longer have anything mounted...)
I understand what you guys are saying about the non-active meshes not needing to be images, but, I can't see a feasible way for our modellers to add "hard coded" meshes... I thought about holsters, but, we kinda needed those to be of different types and sizes... I'll rethink what I can have my modellers include in the mesh... I am adding that resource just because there's no real reason not to... it adds a very useful functionality that I'm sure I can use... (maybe even use it to blow off heads :) )
Anyway, I still need more input on this... thank you for pointing me to that, Master Edward.
02/03/2005 (10:42 pm)
Interesting resource... can't segmented meshes have pieces of them hidden with a visibility animation though? In the garagegames channel I was asking around about this and someone said I could take the cloaks or flowing clothes and build them into my models as seperate meshes, and then do a visibility DSQ for the cloaks and dresses and such that just flips the visibility off for that object... I know it would still be rendered and animated, but, if I only intend to do this with a few hundred polygons per model, would this be a problem?Really, the full list of what we NEED according to our game design, for the characters is:
Right Hand - Weapon
Left Hand - Weapon or shield (we want John Woo gunplay for sure! Or double swords.)
Back - The player can put a weapon, maybe 2 here (we don't like the whole weapon just vanishes deal
Head - helmet, goggles, glasses, comm gear (we can deal with only having one visible at a time, or make some available combos)
Left Shoulder - For shoulder armor pad
Right Shoulder - Other shoulder armor pad
Left Hip - Holster/Scabbard for gun/sword
Right Hip - Holster/Scabbard for gun/sword
That looks like 8 to me, we could possibly forgo the shoulder pads... Torque is built to support 8 or 9 mounts... then if we do the visibility thing for the cloak, that covers our needs... The visibility trick won't really work with some of these things because we need the ability to mount different types... most of our actual character customization will come through skinning and such... but some things you just need more geometry for...
So, I mean, we don't need 30 things per se, but, a good solid 8-10 would really be sweet... and if we could mount 1-2 things per point without being penalized, that would be cool... but I don't know the effects of it for networking... like those holster mounts, if I could pin the guns to the holsters as well, that would be kosher... but would require 2 things mounted in the same spot (though the hands would no longer have anything mounted...)
I understand what you guys are saying about the non-active meshes not needing to be images, but, I can't see a feasible way for our modellers to add "hard coded" meshes... I thought about holsters, but, we kinda needed those to be of different types and sizes... I'll rethink what I can have my modellers include in the mesh... I am adding that resource just because there's no real reason not to... it adds a very useful functionality that I'm sure I can use... (maybe even use it to blow off heads :) )
Anyway, I still need more input on this... thank you for pointing me to that, Master Edward.
#10
@ Gregory
If you are mounting images then the net load is not going to be tremendous, but the inescapable fact is that every little thing begins to add up. I know rewritting the engine code is not appealing at first, but that's not exactly what I'm suggesting. I was thinking more along the lines of just adjusting things here and there that better suit you and possibly even taking out uneeded stuff to clear up waste.
As for some other ideas, take the gun and holsters for example. You could make the holsters with guns one image, and then when the guns are drawn you could use two images, one with the guns mounted to the hands and the holsters on the hips. Regarding the gloves, you could make a weapon without gloves and a weapon with gloves and mount the appropriate image based on inventory, etc...
02/04/2005 (6:09 pm)
Edward, I don't know if you traced back Gregory's profile to his site but it would appear that he has a full fledged team assembled and their concept art is quite good indeed. It would appear to me that he really does need to solve some of these issues before the project can really move forward. In this situation it appears that he has access to both artists and moddelers and if he cannot keep them busy then he may not be able to keep them at all. So while I would agree with your wisdom in not concentrating 100% on how good the art works, in this case it would appear that he has genuine reason to dig right into this issue and find a solution.@ Gregory
If you are mounting images then the net load is not going to be tremendous, but the inescapable fact is that every little thing begins to add up. I know rewritting the engine code is not appealing at first, but that's not exactly what I'm suggesting. I was thinking more along the lines of just adjusting things here and there that better suit you and possibly even taking out uneeded stuff to clear up waste.
As for some other ideas, take the gun and holsters for example. You could make the holsters with guns one image, and then when the guns are drawn you could use two images, one with the guns mounted to the hands and the holsters on the hips. Regarding the gloves, you could make a weapon without gloves and a weapon with gloves and mount the appropriate image based on inventory, etc...
#11
@Gregory - Gonzo called my number perfectly on that. I'll take some more time to think about this properly and get back to you. I'm going to kill my reply above too so that it doesn't throw off any would be answers.
[HOW]EdM|EGTGE
02/04/2005 (6:28 pm)
@Gonzo - Ah. I am indeed properly chastised. Thanks for posting your comments. I did in fact not do the proper homework before replying to Gregory.@Gregory - Gonzo called my number perfectly on that. I'll take some more time to think about this properly and get back to you. I'm going to kill my reply above too so that it doesn't throw off any would be answers.
[HOW]EdM|EGTGE
#12
Thanks for looking into the site... It's good to see that people are willing to do a little research... the JK team isn't kidding around... we definitely mean business... (So much in fact that we have a couple programmers on backup in case we get tired of it/run down/etc... a couple 3D artists, too)... this project is fun to learn from, and we hope in the process of our artistic expression and learning... we accidentally run into something worth money to potential gamers, as well... I've actually taken your advice from that last post and simplified my requirements a little...
The gloves being visible is not necessary... we're limiting mounted things to ONLY things that are game-affecting. This means we need to be able to see the 2 guns in the player's hands, any guns holstered (the player will likely carry 2-3 ranged weapons and 1-2 melee weapons)... so, swords mounted to backs or belts, guns in the same places... for a maximum of 2 hand mounts, 2 hip mounts, 1 back mount...
We'll also need to mount helmets OR goggles, so that's a head mount as well... The combination idea will allow us to combine helmets with goggles, and so forth, so we should be able to take care of all that with 1 or 2 head mounts...
So, 6-7 mounts should take care of us... it'd be nice to have shoulder mounts because various sized shoulder attachments can look really stunning on a player since we could skin them and such, but, we can do without, with the planned 10-15 player models each with 5-6 skins, plus the various gear customizations... that allows for plenty of variety.
Do you have any tips for helping with the skinning process for character customization? Is there an "elegant" way to slap down and swap multiple textures onto the same model in game? I know it loads with many just fine, but, can you then pull off a character's hair and face skin and slap down a new one?
@Edward:
I understand where you were going... many people get consumed by the graphics part of the game and leave the gameplay to the wind. We understand gameplay to be the most important aspect of a game... the order, in my opinion, of importance, is Gameplay > Community > Graphics... However... a good blend of all 3 is required...
I've seen too many games with a heavy following that just aren't that good, and too many games with stunning graphics that are a bore to play... Never have I seen a game that had awesome gameplay but I couldn't play it because the graphics weren't good enough, but, that's just me! Everyone has opinions!
We are indeed a pretty tight team... we work from different parts of the country, but my recruiting process is quite narrowed down... I know which artists to go for based on how much drive they will exhibit for our project... simply put... if I find an artist who doesn't want to contribute their own ideas to the game... I kindly pass them up... people who want to make a mark on the game and put their work forward in the industry are the artists we want...
And of course, as for my input... I look at it sort of as my directorial debut... so I am definitely pushing this project with every spare ounce I can scrape up, and our artists are too...
The concept art you see on the page is only about half of what we have... (we're just outta space to show it is all)
@All:
Even though we are proud of our project team and our accomplishments so far, our game is still just now starting... and we always need more team members, especially people who have tackled some of these torque problems before... and we're always recruiting, so... if you're interested in lending a hand to a project as a consultant or however you want to work it out... take a look at cav.judgex.com/
Thanks for having a look... :)
02/04/2005 (7:12 pm)
@Gonzo:Thanks for looking into the site... It's good to see that people are willing to do a little research... the JK team isn't kidding around... we definitely mean business... (So much in fact that we have a couple programmers on backup in case we get tired of it/run down/etc... a couple 3D artists, too)... this project is fun to learn from, and we hope in the process of our artistic expression and learning... we accidentally run into something worth money to potential gamers, as well... I've actually taken your advice from that last post and simplified my requirements a little...
The gloves being visible is not necessary... we're limiting mounted things to ONLY things that are game-affecting. This means we need to be able to see the 2 guns in the player's hands, any guns holstered (the player will likely carry 2-3 ranged weapons and 1-2 melee weapons)... so, swords mounted to backs or belts, guns in the same places... for a maximum of 2 hand mounts, 2 hip mounts, 1 back mount...
We'll also need to mount helmets OR goggles, so that's a head mount as well... The combination idea will allow us to combine helmets with goggles, and so forth, so we should be able to take care of all that with 1 or 2 head mounts...
So, 6-7 mounts should take care of us... it'd be nice to have shoulder mounts because various sized shoulder attachments can look really stunning on a player since we could skin them and such, but, we can do without, with the planned 10-15 player models each with 5-6 skins, plus the various gear customizations... that allows for plenty of variety.
Do you have any tips for helping with the skinning process for character customization? Is there an "elegant" way to slap down and swap multiple textures onto the same model in game? I know it loads with many just fine, but, can you then pull off a character's hair and face skin and slap down a new one?
@Edward:
I understand where you were going... many people get consumed by the graphics part of the game and leave the gameplay to the wind. We understand gameplay to be the most important aspect of a game... the order, in my opinion, of importance, is Gameplay > Community > Graphics... However... a good blend of all 3 is required...
I've seen too many games with a heavy following that just aren't that good, and too many games with stunning graphics that are a bore to play... Never have I seen a game that had awesome gameplay but I couldn't play it because the graphics weren't good enough, but, that's just me! Everyone has opinions!
We are indeed a pretty tight team... we work from different parts of the country, but my recruiting process is quite narrowed down... I know which artists to go for based on how much drive they will exhibit for our project... simply put... if I find an artist who doesn't want to contribute their own ideas to the game... I kindly pass them up... people who want to make a mark on the game and put their work forward in the industry are the artists we want...
And of course, as for my input... I look at it sort of as my directorial debut... so I am definitely pushing this project with every spare ounce I can scrape up, and our artists are too...
The concept art you see on the page is only about half of what we have... (we're just outta space to show it is all)
@All:
Even though we are proud of our project team and our accomplishments so far, our game is still just now starting... and we always need more team members, especially people who have tackled some of these torque problems before... and we're always recruiting, so... if you're interested in lending a hand to a project as a consultant or however you want to work it out... take a look at cav.judgex.com/
Thanks for having a look... :)
#13
Don't get me wrong, I understand completely why you deleted your post, I wasn't knocking it at all, I was just pointing out that it Gregory's determination semed to be genuine based on the observations of his site. But I wanted you to know that I thought it was really good advice and it's a shame that it's gone. To tell you the truth, I didn't even look at his site till you had given your advice. I was going to back you up on what you said but for some reason before I replied I clicked on Gregory's name and then the links to his site and ended up finding out that he had a whole team going, lol. Maybe we should build a TGE Wisdom Doc via Email. Nothing serious or intensive, just jot down some wisdom and pass it to me, I'll jot down some and pass it to you, and when we finally have a good collection, we can touch it up and make a resource out of it. If you're interested drop me a line sometime.
02/04/2005 (8:30 pm)
@ EdwardDon't get me wrong, I understand completely why you deleted your post, I wasn't knocking it at all, I was just pointing out that it Gregory's determination semed to be genuine based on the observations of his site. But I wanted you to know that I thought it was really good advice and it's a shame that it's gone. To tell you the truth, I didn't even look at his site till you had given your advice. I was going to back you up on what you said but for some reason before I replied I clicked on Gregory's name and then the links to his site and ended up finding out that he had a whole team going, lol. Maybe we should build a TGE Wisdom Doc via Email. Nothing serious or intensive, just jot down some wisdom and pass it to me, I'll jot down some and pass it to you, and when we finally have a good collection, we can touch it up and make a resource out of it. If you're interested drop me a line sometime.
#14
Since you have decided to scale back your plans a little can I offer a suggestion that might seem a little illogical? Being that you are willing to accept less than you originally started out to get, I think you should go for all you were trying to get originally while also not modifying the code any more than you need to anyway. The reason for this is simple. These ideas of mounting multiple images and objects has been batted around the forums quite a few times. And most times people either scale back like you were thinking or they went with other solutions. As far as I know, nobody has ever completed a full blown game under such specs much less actually run the tests neccesary to give good solid advice on this subject. And since the bulk of the uncertainty centers around the number of mounted images and what effect they will have and since the extra work involved is nothing more than actually placing the 8 or more nodes into the model I would recommend that you build at least one or two models with all the nodes you described originally and then start running some tests to see what happens when the image numbers start piling up. For all we REALLY know it may work just fine or work great with some tweaks. Worst case scenario is you wasted an hour or two mounting those extra nodes. But even so, you would be the first(that I'm aware of) to produce hardcore data on this subject that can be bigtime helpful to others in the future. I would be willing to be an extra player to help you test if need be, and you can email me personally if you have any questions you haven't found answers to yet. I'll make sure you make up any lost time from running the testing. But IMO, there is no such thing as lost time when you are testing, the data will be valuable for a loooong time to come. It might even lead to some changes that benefit everyone.
Also let me point out that you could make one model with all the nodes you desire, make a simple test image that is representitive of the number of poly's that you feel your images are going to be and mount 8 or 10 of those on the model and get a bunch of guys running around. Heck, make one red, one green, and one blue, and rotate each one by 90 degrees from the one before it so you could mount 2 or 3 of them to a node and run tests on them as well. The colors will help you figure out exactly what happened on each command as you mount and unmount them. If you really want to know what your limits are, you are just going to have to push them, that's all there is to it.
"Never have I seen a game that had awesome gameplay but I couldn't play it because the graphics weren't good enough,"
I agree completely. Take "Bejeweled" for example. I don't even like 2D games like those but it didn't stop that damn game from taking a couple of hundred hours of my life. I'd probably still be playing it if I didn't have the will power it took to delete that demon spawn, lol. The lesson I learned from it the most was that many a time I had gone out and paid 40-50 bucks for a console game only to get about 30-40 hours use out of it(unless it was football or something) and a free demo of a game I wouldn't normally play had consumed up to 700% more of my time. Unfortunately for them there was no way I was ever going to buy it because then I wouldn't be able to just delete it and that would mean I wouldn't be able to stay away from it.
02/04/2005 (8:30 pm)
@ GregorySince you have decided to scale back your plans a little can I offer a suggestion that might seem a little illogical? Being that you are willing to accept less than you originally started out to get, I think you should go for all you were trying to get originally while also not modifying the code any more than you need to anyway. The reason for this is simple. These ideas of mounting multiple images and objects has been batted around the forums quite a few times. And most times people either scale back like you were thinking or they went with other solutions. As far as I know, nobody has ever completed a full blown game under such specs much less actually run the tests neccesary to give good solid advice on this subject. And since the bulk of the uncertainty centers around the number of mounted images and what effect they will have and since the extra work involved is nothing more than actually placing the 8 or more nodes into the model I would recommend that you build at least one or two models with all the nodes you described originally and then start running some tests to see what happens when the image numbers start piling up. For all we REALLY know it may work just fine or work great with some tweaks. Worst case scenario is you wasted an hour or two mounting those extra nodes. But even so, you would be the first(that I'm aware of) to produce hardcore data on this subject that can be bigtime helpful to others in the future. I would be willing to be an extra player to help you test if need be, and you can email me personally if you have any questions you haven't found answers to yet. I'll make sure you make up any lost time from running the testing. But IMO, there is no such thing as lost time when you are testing, the data will be valuable for a loooong time to come. It might even lead to some changes that benefit everyone.
Also let me point out that you could make one model with all the nodes you desire, make a simple test image that is representitive of the number of poly's that you feel your images are going to be and mount 8 or 10 of those on the model and get a bunch of guys running around. Heck, make one red, one green, and one blue, and rotate each one by 90 degrees from the one before it so you could mount 2 or 3 of them to a node and run tests on them as well. The colors will help you figure out exactly what happened on each command as you mount and unmount them. If you really want to know what your limits are, you are just going to have to push them, that's all there is to it.
"Never have I seen a game that had awesome gameplay but I couldn't play it because the graphics weren't good enough,"
I agree completely. Take "Bejeweled" for example. I don't even like 2D games like those but it didn't stop that damn game from taking a couple of hundred hours of my life. I'd probably still be playing it if I didn't have the will power it took to delete that demon spawn, lol. The lesson I learned from it the most was that many a time I had gone out and paid 40-50 bucks for a console game only to get about 30-40 hours use out of it(unless it was football or something) and a free demo of a game I wouldn't normally play had consumed up to 700% more of my time. Unfortunately for them there was no way I was ever going to buy it because then I wouldn't be able to just delete it and that would mean I wouldn't be able to stay away from it.
#15
As far as what you said, I will go ahead with 8 full mountpoints rather than 6 or 7, and we'll be testing them soon. We almost have our design of the skeleton with hitboxes and mountpoints complete... and once we have that... we'll be able to start testing with our models and such...
You're right though, now is the time for stress testing... but, we simultaneously have modelling going on in one place, a guy working on levels for us in another, me coding here, and someone setting up the mountpoints here as well... so... I just kinda nail problems as they crop up and if they don't crop up I find new ones on my own :).
02/05/2005 (12:17 pm)
Lol, I love bejeweled as well... I play on yahoo all the time...As far as what you said, I will go ahead with 8 full mountpoints rather than 6 or 7, and we'll be testing them soon. We almost have our design of the skeleton with hitboxes and mountpoints complete... and once we have that... we'll be able to start testing with our models and such...
You're right though, now is the time for stress testing... but, we simultaneously have modelling going on in one place, a guy working on levels for us in another, me coding here, and someone setting up the mountpoints here as well... so... I just kinda nail problems as they crop up and if they don't crop up I find new ones on my own :).
#16
As far as what you said, I will go ahead with 8 full mountpoints rather than 6 or 7, and we'll be testing them soon. We almost have our design of the skeleton with hitboxes and mountpoints complete... and once we have that... we'll be able to start testing with our models and such...
You're right though, now is the time for stress testing... but, we simultaneously have modelling going on in one place, a guy working on levels for us in another, me coding here, and someone setting up the mountpoints here as well... so... I just kinda nail problems as they crop up and if they don't crop up I find new ones on my own :).
02/05/2005 (1:43 pm)
Lol, I love bejeweled as well... I play on yahoo all the time...As far as what you said, I will go ahead with 8 full mountpoints rather than 6 or 7, and we'll be testing them soon. We almost have our design of the skeleton with hitboxes and mountpoints complete... and once we have that... we'll be able to start testing with our models and such...
You're right though, now is the time for stress testing... but, we simultaneously have modelling going on in one place, a guy working on levels for us in another, me coding here, and someone setting up the mountpoints here as well... so... I just kinda nail problems as they crop up and if they don't crop up I find new ones on my own :).
#17
Glad you all found the two most relevant resources here, the one on mesh hiding, and the one on increasing the number of mount points.
Understand that default mounts are intended for fully functional, active objects, like guns. Mounts are designed to support heavily networked objects which require frequent updates because, again, things like guns usually need to propagate state changes very frequently to all players. This way, you can see what direction a gun is facing, and what fire state it's in, etc.
If you want mesh and texture modifiers for things that are really just for show-- like gloves-- then you'll want to use a different system. Mesh hiding can work in lots of situations. Or, you could create a simplified mounting system with less strenuous network updating. :) Even if your glove changes the stats of a character or something, you'd certainly want to use a simplified system for it.
There isn't a super-clean out of box solution here, but one isn't too difficult to implment at all. Looking at the mesh hiding resource is a good place to start, and examining the mount code in the engine and the network update calls in particular will get your head in the right place to implement a nice solution for your game. :)
Just wanted to provide the quick run-down here, I hope all this makes sense.
02/05/2005 (6:41 pm)
Just a quick summary here, for future readers who didn't wade through all the above if nothing else...Glad you all found the two most relevant resources here, the one on mesh hiding, and the one on increasing the number of mount points.
Understand that default mounts are intended for fully functional, active objects, like guns. Mounts are designed to support heavily networked objects which require frequent updates because, again, things like guns usually need to propagate state changes very frequently to all players. This way, you can see what direction a gun is facing, and what fire state it's in, etc.
If you want mesh and texture modifiers for things that are really just for show-- like gloves-- then you'll want to use a different system. Mesh hiding can work in lots of situations. Or, you could create a simplified mounting system with less strenuous network updating. :) Even if your glove changes the stats of a character or something, you'd certainly want to use a simplified system for it.
There isn't a super-clean out of box solution here, but one isn't too difficult to implment at all. Looking at the mesh hiding resource is a good place to start, and examining the mount code in the engine and the network update calls in particular will get your head in the right place to implement a nice solution for your game. :)
Just wanted to provide the quick run-down here, I hope all this makes sense.
#18
Any problems spring to mind that I'd run into with this approach? I know how to export a visibility animation, but, something's telling me there's something I'm not considering.
Thank you, Josh :).
02/06/2005 (3:32 am)
Certainly does, and thanks... but what about visibility animations for some of my objects? Is that viable? I haven't heard much feedback on that, it was a trick recommended to me in the IRC channel, and I researched it a little, but am still not 100% sure if it's the way I want to go... it would knock out quite a few problems though... we could throw in a couple shoulderpad variations and cloaks and such with visibility animations for each... Any problems spring to mind that I'd run into with this approach? I know how to export a visibility animation, but, something's telling me there's something I'm not considering.
Thank you, Josh :).
#19
03/19/2015 (12:07 pm)
What do I have to do to get the weapon to stay connected to the players hand after being mounted?
#20
Id like to suggest that you start a new thread . Even though this was a rather interesting thread eh ?
anyhow , just be sure that your model has the mount , and , in the console window or a script , use this command - obj.mountimage(Xxx,Xxx,X)(assuming that you have an image) or try obj.mountobject(Xxx,Xxx,X) for mounting an object . should work . Use the Objects id numbers or names . just saying , seems that the objects image number is +1 whatever the objects number :) . dont quote me though .
example :
player id = 1000
image = 1101
mount to connect to = 0
" obj.mountimage(1000,1101,0) "
Have a nice day .
EDIT 5/2/2015
LOL , Sorry I have to laugh at my reply to Mr. McCabe but , Ill leave it for someone else to hopefully see the humor in such a reply within this thread >:D Sry Really .
Well , I did just get back on this thing after giving my eyes a 30 day break . That having been said now , (excuses) Duh
the syntax is wrong in the above reply .
Its like this , I think :)
lets say
Object = 1000 "Player01"
mounted object = 1100 "handGun01"
image = 1101 "handGun01Img"
mount = 0
so ,
1000.mountimage(1101,0)
or
player01.mountimage(handgun01Img,0)
to mount an object
object = 2000 "tank01"
mount = 5
1200.mountobject(1100,5)
tank01.mountObject(handgun01,5)
dont forget the semicolons at the end of the command .
So, yea , I think thats it :) LOL. Have a nice day .
05/01/2015 (9:48 am)
@ Mr. Darran Lane McCabe ,Id like to suggest that you start a new thread . Even though this was a rather interesting thread eh ?
anyhow , just be sure that your model has the mount , and , in the console window or a script , use this command - obj.mountimage(Xxx,Xxx,X)(assuming that you have an image) or try obj.mountobject(Xxx,Xxx,X) for mounting an object . should work . Use the Objects id numbers or names . just saying , seems that the objects image number is +1 whatever the objects number :) . dont quote me though .
example :
player id = 1000
image = 1101
mount to connect to = 0
" obj.mountimage(1000,1101,0) "
Have a nice day .
EDIT 5/2/2015
LOL , Sorry I have to laugh at my reply to Mr. McCabe but , Ill leave it for someone else to hopefully see the humor in such a reply within this thread >:D Sry Really .
Well , I did just get back on this thing after giving my eyes a 30 day break . That having been said now , (excuses) Duh
the syntax is wrong in the above reply .
Its like this , I think :)
lets say
Object = 1000 "Player01"
mounted object = 1100 "handGun01"
image = 1101 "handGun01Img"
mount = 0
so ,
1000.mountimage(1101,0)
or
player01.mountimage(handgun01Img,0)
to mount an object
object = 2000 "tank01"
mount = 5
1200.mountobject(1100,5)
tank01.mountObject(handgun01,5)
dont forget the semicolons at the end of the command .
So, yea , I think thats it :) LOL. Have a nice day .
Torque Owner Gonzo T. Clown
function Glove::onFire(%this)
{
// leave blank
}
to prevent console spam from missing function errors. You also have to take into account that when you unmount an object or image you do so by calling to the slot. So if you try to unmount the gun, the glove will go with it and will need to be added back on after the gun is removed.
2. I'm not positive, but I think the image limit had been raised to 24 or 32. Even if they haven't been increased, you can increase them. I've never experimented with such a configuration but i've read enough to know that if you start mounting to many "objects" to your player that they can lag behind when viewed on other clients. If you move to fast your mounted objects will appear to be following you, not mounted to you. It is best to mount images whenever possible. I believe it's better for engine performance to use images when you plan to mount multiple items to your player. Ben Garney could give you a better assessment of that though.
3. I don't know where the docs are(never use them) but the functionality you seek is in the engine and not the tools or exporters. Get yourself a good editor like TextPad and you can search your entire engine for "mountImage" which will pull up the 8 lines of code that you are looking for.
4. That really depends. If its a multiplayer(with lots of players) networked game, then yes, the more you mount the more load you will have to deal with and the more risk you run of having images lag on you. If you were limiting to 4-6 players, it shouldn't be a big issue and ina single player game it should be a factor at all.