LBPCentral Archive

Home

LittleBigPlanet 1 - PSP - Tearaway -Run Sackboy Run

LittleBigPlanet 1

[LBP1] Everything Else LittleBigPlanet 1 [Archive]

Problem-causing creation techniques

Archive: 30 posts

This thead is dedicated to the little things you've tried to do that just didn't turn out well, or worked inconsistently - let's tell other authors so they don't make the same mistake!

My first one is a problem I've seen break the game in several levels, and just recently it started affecting me in the latest level I'm building:

Attaching light material with a turn-bolt and hitting it with the paintinator!

Sympton:

You have a piece of light material attached to an object (such as cardboard or dissolve) using some kind of turn-bolt (wobble, motor, or sprung).

You're shooting it with the paintinator, and periodically the material just breaks off. Now, if it contains a paintinator switch and the paintinator switch does something crucial in your level, GAME OVER!

(by the way, in my extensive tests I was able to make it fail about 1 out of every 50 times - which means odds are it's going to happen to people - and they are going to rate your level lower!)

How to fix it!

A) You could use tougher material, such as metal to replace cardboard - as long as weight is not a consideration. From my testing, it takes a lot more force to break a bolted piece of metal off the object.

B) In MY case it was more complex. I had a paintinator switch on the dissolve, and it caused the dissolve to disappear and also caused additional logic. HERE'S WHAT I DID INSTEAD: I attached a magnetic switch to the object the dissolve was attached to. The dissolve has the paintinator switch AND a magnetic key. Magnetic switch set to reverse. I set the magnetic switch to activate the logic that the paintinator switch used to activate. So: Paintinator shoots dissolve, paintinator switch makes dissolve disappear, magnetic switch activates because the key is now gone, and everything works fine! So, now my crucial logic will activate if the paintinator switch is depleted OR the dissolve get's knocked off the object.

Any questions?

2009-02-16 00:44:00

Author:
CCubbage

Posts: 4430

2009-02-16 09:18:00

Author:
Wyth

Posts: 263

Hah, that's quite a good one too keep in mind. I like the thread too.

I've got one too: Consider the room the lighting elements cost in your level. I add lighting after I finish a element, but lamps need room too! At the very least you need a thin layer for a LED light, or you need to embed it in an exsisting layer... Keep it in mind.
Oh yeah, I forget about this EVERY TIME! I started adding the lighting as I'm developing the level because I forget it will take up a thin layer.

2009-02-16 12:55:00

Author:
CCubbage

Posts: 4430

Decreasing the size of a complex custom object:
It is always a problem for me. I was used to construct my custom content initially with a big size in order to increase the level of details in can put in...
However, for small details and for complex shapes, decreasing the size induce a great loss of these details and of the initial form... It is particularly true for textils matters and black and dark matter...

2009-02-17 06:45:00

Author:
Takelow

Posts: 1355

2009-02-19 13:56:00

Author:
CCubbage

Posts: 4430

2009-02-20 17:47:00

Author:
dawesbr

Posts: 3280

It's Symptom

Anyway:

Not covering your tracks

Exposed switches, blatant logic-gates and ugly dark matter structures have been seen by my loads of times. Ugly, pointless, and it's going to cause your level to be rated lower. Cover them, or move them out of the way, but please don't show them!

I actually disagree with this. This is a point of personal opinion. I consider visible switches to be instructive for new creators. The fact that people would actually rate a level based on visible switches is almost offensive, especially considering that all of the switches in the story mode levels are visible.

Anyways, I don't think this is what cubbage was getting at with his post topic.

2009-02-20 21:32:00

Author:
Thegide

Posts: 1465

^I think it depends. I wouldn't do it if your switch placement isn't symmetrical and aesthetically sensible as the rest of the level though. It would also depend on if your level is meant to be some surreal, seamless type of experience that immerses the player and doesn't feel like an LBP level at all. In that case, visible switches really have no place.

From Siberia With Love is a great example in how you can access the "off limits" area... OboroShogun has a similar secret area in Terror's Manor. As a creator, I love to see things like that and it also shows even more care and understanding of the mechanics if you can put all your switch boxes in one place without having the radius' and color coordinations conflict with each other.

I think paint switches should always be visible unless it interferes with aesthetics or design choice, or if the paint switch is used in an environmental way (ie: shooting the broken columns in ROBO-RAMPAGE X2 by Dimo1138, or in OboroShogun's "Electric Baby Vampire Task Force"). You may have multiple paint switches connected to one object, but would only need one visible - whichever one will indicate the hitpoints or indicate that it is switched. For puzzles, and arrangements, I think it adds a great level of polish and guidance to the player, and if your mechanics are truly unique and impressive it's a nice way to show it off.

The key to making them look tasteful is symmetry, proportion and logic... the only visible switches I have in L.O.V.E. are the 5 mag switches/keys in in the small rising-platform puzzle section and it really makes it look cool, as well as adds to a sense of how to solve the problem... it also helps show what kind of engineering and thought went into it.

The truck in the level has all it's switches invisible except for the one switch box that controls the headlights and tail lights... it's on the side next to the door. It's functional, symmetrical and in line with the rest of the build... it almost makes it look like it's manually shifting gears when you throw it in reverse, and I like that.

If some of the switches weren't so randomly placed for timing purposes, and there weren't so many that are dependent on their radius and similar color I would definitely put them all into a single spot for an easter egg area... I think people would be surprised how much went into the level if they saw what's behind the scenes.

LED's and spotlights shouldn't be visible depending on your theme as well - a cave doesn't have elecricity, but a building does. No matter what, it's LBP... you're going to have checkpoints, winchs and the piston base visible (even when set to 0), two ways and three ways and big red buttons, race gates, a goofy little sackboy/girl, shiny brains, point bubbles, and a scoreboard no matter how serious or artistic it is... I think using plasma balls in a temple setting is ridiculous, but hey, what's the alternative? Cutom lethal objects don't behave in the same way that plasma does no matter what velocity you set it to.

2009-02-20 21:56:00

Author:
Unknown User

2009-02-20 23:14:00

Author:
dawesbr

Posts: 3280

Ok, so evidently no-one liked my suggestion. It just really bugs me when you see have a logic gate, or part of a switch. I don't mind presentations of complicated mechanisms, in fact I love them, but when I'm trying to get immersed in a level it's a bit of a reality check.

Anywho:

Gluing Everything To Everything

A common bug in LBP is the "doomsday" bug - everything collapses, glued parts become unglued, and everything starts rolling on the floor. This always seems to occur when Sod's law seems most appropriate - 90% of the way through creation, after a long period of pause-create. So cover your tracks. Don't glue everything to one central platform. Use discreet dark matter to hold things up, but put it out of the way! And playtest your level frequently if you are forced to glue things together. Just be warned.
This is actually a really good practice to not glue everything together. However, glue can also help to create stability.

My personal choice is to create the level in smaller sections in which things are glued together, but make sure they are not connected to other sections (carefully using the corner tool to connect sections nicely without having to glue them). This keeps issues centralized to specific areas.

Also, if you have a problem unglueing (where a ton of other things suddenly become unglued), try instead ERASING the matter with the triangle button. This allows separation of areas without needing to unglue - and doesn't have the same issue.

(and also, I'm not sure everyone dislikes your suggestions of hiding switches, it's just that MOST people don't do a good job with making them look aesthetically pleasing. However, if you follow the way MM and Thegide do it, it looks good)

2009-02-21 00:05:00

Author:
CCubbage

Posts: 4430

Sorry. I didn't mean it like that. I agree, I just think it depends on what you're going for and how it's done. 9 times out of 10, don't... occasionally it's cool, and if it's part of your design philosophy and do it right, then that's great. I only one level with visible switches, not counting paintinator switches.

editing your own objects

Say you're really careful and mindful of making a bad design decision as you build a complex machine or object... a robot or vehicle for example. A good way to ensure that you don't go too far beyond the point of no return is to capture certain pieces as you assemble it, or copy them. Say you make the legs first, and you're really happy with them, but don't know quite what you're doing with the body, arms and head yet. To avoid having to scrap the entire project if you make a stupid design choice, it's good to have a copy of the legs sitting somewhere else.

If you're making a vehicle... and you build the body, but haven't started cutting yet, make a copy and set it to the side before you start - that way you don't have to rewind 5000 times, or make so many additions that you can't rewind far enough to get back to square one. Same deal with making an undercarriage mechanism for the wheels... never hurts to capture or copy them before attaching or embedding anything. This also lets you have key mechanisms on hand for any future uses as well.

Now, the one drawback with this, is that you can't always edit or delete alot of things once they're captured and reused. A good fix for this, is to make whatever you're trying to change, out of dissolve and trigger it with a paint switch to effectively delete it without cutting or smearing in the event that it won't work.

I recently had a problem with aer0blue's boss rush... my boss had started malfunctioning, and it couldn't be reset to it's starting point without breaking. It was also a captured object, that I had sent him, being sent back to me in a community tagged level making it doubly protected. I couldn't delete it, to replace it with a back up of the boss... I had to simply defeat it. Activate all it's switches, and let it die and vanish... and THEN replace it with a new one.

Good tricks to keep in mind when you start getting too complex, or are meeting alot of stubborn resistance from your own creations.

thermometer control

It should be common sense that the settings on your emitters can greatly effect your thermometer... and sometimes you'll be overheating as soon as a single emitter fires, or having so many things emit that eventually nothing else will emit, and even the paintinator won't fire anymore.

You don't always have to cut down on your max emitted, frequency, or max at once, and sometimes it's crucial that you don't depending on the effect you're going for. A quick fix for this, in emitter heavy levels is to simply set all your unswitched emitters to a sensor switch that keeps it from firing or activating when the player is not around.

This will prevent the therm being battered around during play, and emitter equipped enemies and objects that die off will greatly reduce the chance of the next series of emitters overheating your level or failing. This can provide opportunity to sometimes even trick the thermometer into allowing you a fraction more space than it really should.

Also it won't be constantly firing the plasma balls that float around in your level while rewinding. These often can't be deleted for some odd reason, as they may float into the level borders and become just out of reach of the cursor, or a death to remove it... and will forever take up space in your creation.

Remember too, that if you have an emitter switched... if you disconnect it, it will fire off regardless of reconnecting it. That can take away from your max emitted count. If you had it set to 10, it will only emit 9... if it was set to 1, it won't emit anything. A fix for that, is to up it by 1 after this happens, or to replace the emitter itself with a brand new one. A copy of the same emitter will have the same negative count, so make sure it's a brand new emitter if replacing it. I'd recommend the replacement as the single emitting misfire can add a minute gain to your therm that you may or may not be able to retrieve or account for later... and in extravagant projects, that single point of therm can be all the difference between polished and ugly, or short and long.

sensor switch mishaps

Don't ever unpause while standing on the ground in create mode. Always unpause while in fly mode, and THEN land to test. If you unpause while standing on the level and in range of a sensor switch, the sensor switch will remain active even after rewinding. You can rewind twice, back past to where you were still in fly mode to correct this... but often times, you might completely forget and finish the whole level, only to playtest and find that a single sensor switch has caused a certain event to prematurely trigger and effectively ruin the course of action for the entire level. It can be a great pain to rework these and correct small mishaps like this, and it's better to just avoid it completely by never unpausing while you're feet are still on the ground.

This same principle should be applied to magnetic keys and switches as well. You will have to rewind twice if you've placed the key in range of the switch and then unpaused, i you wish to revert it back to being decatived. So don't place a key in range of a non-inverted switch while paused. Create the "test" key object while unpaused and then let it appear or move into range.

2009-02-21 00:08:00

Author:
Unknown User

Good further points on thermometer control (along with Ninja's comment above) - flatten out ALL edges that you can't see as you're building your project.

And in order to create rough terrain or caves - cut a shape using a multi-angle shape (such as a star) and then use the corner tool to carefully mold the terrain. If you do this, you will find your thermometer lasts MUCH longer.

2009-02-21 00:16:00

Author:
CCubbage

Posts: 4430

You know what I've found with corner editing... if you flatten on object with corner edit, it doesn't drop the thermometer as much as flatting it by cutting it with a square. In some cases I've had it INCREASE the thermometer. I think it considers pampering an object with corner editing a further calculation in the objects memory.

That does remind me though... if you can do any sectioning in your levels, do so. Building in square sections and just gluing them together or holding them together with dark matter is the best way to go. It can give you alot more freedom when editing objects and adding detail by embedding and etching. It also helps if you're trying to sticker the entire level as you can only make stickers so big, or sticker one giant object (ie: the flat background or floor) so many times. It also won't always load in at one time, or fail to show when lagging. H.A.T.E. by Voltiare is a perfect example of this design philosophy, and also a good example of where it can fail - if you're really lagging hard in the wheel with multiple players, the white sticker will glitch up and reveal the dissolve backing on the wheel. This is because it's a much wider surface area than the rest of the cubicles in the level that have less stickering and much smaller stickers on them.

2009-02-21 00:22:00

Author:
Unknown User

Sorry forgot to mention that... you definately don't want to FLATTEN using the corner editor, just to cut and create angles. A lot of the thermometer usage seems to be related to the amount of physical angles on each shape - the more angles, the more thermo usage. So using the corner editor makes it difficult to tell how many angles there are. Just flattening the edges by adding a little square material or cutting the edges seems to work better.

By the way, on Splat II I actually used the sections - I created all the crucial parts of the game up front in sections, THEN put them in order and connected afterwards. This is a GREAT way to make sure you can finish all necessary parts of the level before running out of thermo. So, I built the entire level in order of priority instead of order of story.

2009-02-21 00:44:00

Author:
CCubbage

Posts: 4430

editing your own objects

Say you're really careful and mindful of making a bad design decision as you build a complex machine or object... a robot or vehicle for example. A good way to ensure that you don't go too far beyond the point of no return is to capture certain pieces as you assemble it, or copy them. Say you make the legs first, and you're really happy with them, but don't know quite what you're doing with the body, arms and head yet. To avoid having to scrap the entire project if you make a stupid design choice, it's good to have a copy of the legs sitting somewhere else.

If you're making a vehicle... and you build the body, but haven't started cutting yet, make a copy and set it to the side before you start - that way you don't have to rewind 5000 times, or make so many additions that you can't rewind far enough to get back to square one. Same deal with making an undercarriage mechanism for the wheels... never hurts to capture or copy them before attaching or embedding anything. This also lets you have key mechanisms on hand for any future uses as well.

Now, the one drawback with this, is that you can't always edit or delete alot of things once they're captured and reused. A good fix for this, is to make whatever you're trying to change, out of dissolve and trigger it with a paint switch to effectively delete it without cutting or smearing in the event that it won't work.

Right you are... editing becomes a total pain in the butt when you do this.

I prefer not to capture along the way. Instead, I copy my bits as I work along. Using your robot example, I would build the legs first, then make at least one copy. On the copy I would add the body. Then, I would copy and paste the whole thing again and add arms.... Then copy that and add the head.

This way, I have all my basic building pieces laid out on the level floor should I wish to modify them, plus I have copies of my work at each major change. If I want to go back and modify something, it's easy because you can still detach the parts you aren't happy with and replace them. So if I wanted to edit the legs later, I'd delete them off my latest robot "version", edit the leg pieces I saved in the beginning, copy them at attach the new ones. I'm never working with captured parts, so I can always modify any part at any time.

The dissolve method works, yes, but can be incredibly tedious because you never know which key piece has to be deleted for your object to become editable again. Assembly line FTW

2009-02-21 01:23:00

Author:
Thegide

Posts: 1465

That's exactly how I built it too. The Boss Workshop level is one functioning mech, 3 mechs at various stages and a bunch of legs, arms, weapons, and stages of mechanical add ons laying around like a junk yard.

2009-02-21 02:28:00

Author:
Unknown User

2009-02-22 09:30:00

Author:
Bridget

Posts: 334

One I'd like to add that goes along with the talk of the thermometer up above:

Dont use too many different materials all in the same level!

Im not sure that people realize this one right away; I know I sure didnt, originally.

Every new material you add to the stage will add a bit to the thermometer. For example, if you hadnt had the Aztec Jade stone material in your level already, and you go and stick even a tiny block of it in, the thermometer will jump up a bit, likely to keep space to process the textures and whatever that goes along with the material.

It's important to keep a theme going with those (and looks nicer too). And if you really must change up appearances but dont want to waste thermo space, you can use large stickers to alter the way they look (mostly the "color" stickers).

While this is good advice and truly should be heeded, it is not completely true.

The thermometer really is a counter of many things, amount of materials/stickers used, total object complexity and other things. These different things do NOT count up. The thermometer always shows the most critical part of it. An example to clarify this.

Say you make a level with just 2 materials. The first time you put down those materials, it fills up the thermo a bit. If you make the level truly complex with a huge amount of jagged edges, custom brickwork etc etc. Basically use a lot of polygons. This will cause the thermometer to fill up.

If you then start to add new materials, it will not increase the thermo. This is because the thermo is based on the complicated shapes you have, not the memory used for the materials. So, depending on the level you have at the moment, different things may cause different effects on the thermo.

2009-02-24 08:49:00

Author:
Wyth

Posts: 263

While this is good advice and truly should be heeded, it is not completely true.

The thermometer really is a counter of many things, amount of materials/stickers used, total object complexity and other things. These different things do NOT count up. The thermometer always shows the most critical part of it. An example to clarify this.

Say you make a level with just 2 materials. The first time you put down those materials, it fills up the thermo a bit. If you make the level truly complex with a huge amount of jagged edges, custom brickwork etc etc. Basically use a lot of polygons. This will cause the thermometer to fill up.

If you then start to add new materials, it will not increase the thermo. This is because the thermo is based on the complicated shapes you have, not the memory used for the materials. So, depending on the level you have at the moment, different things may cause different effects on the thermo.
Actually, you're BOTH right.

Thermo is based on the following things:

Number of different resources (different materials, music, objects) - each individual separate entity you place in your level will cause your thermo to go up. They must be stored in your level when playing.

Number of points on your objects - the more complex the object, the more thermo.

Objects stored within emitters

Number of separate objects that exist in your level

ALL of these things separately take up thermo. The thermo is a little tricky, however, because it doesn't seem to immediately move up or down EXACTLY when you use something all the time.

My suggestion is that you use limited materials and limited complex objects until you know you are going to have enough thermo to finish what you are doing. Then, make it more complex afterwards.

By the way, just to illustrate something I ran into - Early on in the development of Splat II I had a skateboard as a joke in the level. When I placed the skateboard my thermo did not appear to use ANY resources. However, later when I decided it was a bad idea I highlighted the skateboard and deleted it - and my thermo suddenly dropped by 30%!!!

So, it's a little hard to experiment with what affect multiple materials have on thermo. But they do use it regardless.

2009-02-24 15:06:00

Author:
CCubbage

Posts: 4430

Ahh well I think I know how that happened and it is consisten with my theorie. What happened when you added the skateboard is dat the complexity section of the thermo at that moment was higher than the memory section. This way the increase in memory (used by the skateboard) didn't show up. Aftwerwards you likely increased the memory used in your level, rising that section of the thermometer above the complexity section. Then when you removed your skateboard, it lowered the memory section, visibly because that was the critical component at that moment.

I think it works this way, we should experiment a bit with this to confirm, could be interesting.

Would be nice I think if MM made the thermo 2-part so we can see what happens (if this theory proves correct).

Might be that the skateboard counts on the complexity AND the memory part though...

2009-02-24 15:16:00

Author:
Wyth

Posts: 263

I get the feeling the thermo logic is more complex that we can visually see. It's not always "cause-effect" where you can immediately see the exact result of what you've done. But in basic resource terms, every little separate texture, shape, complex base object, and music resource you use MUST be stored in your level during play. So, a basic rule is to not over use any one of these beyond what you need.

edit: But that's not to say you're wrong, Wyth... you may be correct in the way the thermo is displayed, but keeping things simple until all the mechanics are in place may still be a great solution.

2009-02-24 15:42:00

Author:
CCubbage

Posts: 4430

I would have completely agreed with Bridget until a few days ago, when I too noticed exactly what Wyth is saying.

In my Idol level, I replaced some textures with new unused ones late in the development cycle. Every new texture caused the thermo to jump up.

In my new MGS level, I went to do the same thing (add new textures post-production) and I can, right now, pick any new texture and my thermo won't budge, whether it be through material changing or adding new objects.

Why the difference? I have no idea, but clearly the phenomenon reported by both Bridget and Wyth are true in certain scenarios. One thing I can say for certain is that no objects in my MGS level are really vertex-heavy. Given the mechanical/industrial nature of my environment, most of the superstructure is composed of rectangular objects. My Idol level is the complete opposite... caverns, vertex-heavy shapes.

Perhaps the degree of thermo impact from adding a new material is dependent on how many vertices it has to use in calculating a polygon fill for the new material object. New material on complex shapes would cause the thermo to visibly move, whereas new material on basic shapes would not cause enough of an impact to visibly move the thermo. That's the only way I can unify my observations.

2009-02-24 18:35:00

Author:
Thegide

Posts: 1465

You know... I went home and played with this a little over lunch. I think Wyth is dead on with this one.... it SEEMS like the resources are in separate pools and the pool that uses the highest resource is the one that is displayed - and also matches the specific warning that pops up when your thermo is almost full.

So, if the number of objects in your level is REALLY high and is causing that particular pool to be maxing out the thermo, it will say something like "Simplify or glue your objects - you have too many"

But if you are using tons of different materials, and your material resource pool is too high it will say "Use less materials".

Interesting. This adds a whole new dimension. I'm not sure what it means behavior-wise when creating your level. Even based on this, it's still a good idea to keep all areas simple until you have the base of your entire level completed. That way you know what resources you really have to work with.

2009-02-24 19:09:00

Author:
CCubbage

Posts: 4430

Good work Ccubbage. Now to find out which categories there are and optimizing the thermo will never be the same again! How great would it be if we could see the different thermo btw?

2009-02-25 07:36:00

Author:
Wyth

Posts: 263

Every creation is unique, and complex in different ways. The therm calculates both cause and effect, ie: one object sitting on another. Just putting a boulder on a ledge opens up the kinetic value of the object, and the engine must keep it in memory to allow it to happen as it's meant to and react to all possible actions with the proper reaction. This is why glued always takes less memory than unglued... the unglued state is already open, and then any change in vertices, or surroundings in turn increases the complexity of logic involved in rendering a single object. The greater the complexity of the equation the more substantial each individual variable becomes. There really is no easy answer, as you can put an unglued object on a surface that's physical mass/weight and shape dictates it will never budge, but then create an order of events that increase the the potential for movement and that once "immovable object" suddenly has more value.

Conversely, an oddly shaped object with a high propensity for rolling or interaction with the player (sponge) will inherently retain kinetic energy as a value for the thermometer to take into account. Each material must load in it's behavioral traits... glass is heavier on therm than steel, as the logic involved in dictating that it's slippery and reflective is greater than that of the static traction of woods and metals. The more reflective ceramics and metals will also have greater value than their dull and grainy counterparts.

A level made of all woods can have pretty much every kind of wood in it's equation and barely change the outcome... all the sounds are the same when the player steps on it and walks along, and they are all rendered with the same set of allocated resources... when a new variable is introduced, a new set of parameters, behavior, sounds, and special properties is then allocated a separate set of resources. Another thing to note is that something like sponge and cardboard must retain it's potential for destructible morphing (from explosives, for example).

Managing these individual allocations of memory then requires a bit more processing - it's as if each by itself is an employee. Like minded employees are in the same small business, and thousands of them create a large business but still only require one manager. 4 small businesses, each with their own manager AI, will then require a general manager to distribute the economy evenly and as needed.

Basically, it's what you do with what you use. In "Time is of the Essence" I only have 4 materials... only one of them, glass, is reflective... ever since the inception of the hourglass, it has always told me that "This area of your level is too complex, try using less moving parts or spacing things further apart" when saving - the thermometer is calculating not just the presence of the grains of sand (electrified glass triangles) but their proximity to each other and how they will interact and react to themselves, their surroundings and the introduction of movement via the motor bolt of the hourglass that brings an innumerable amount of probable causes and effects. The individual reflective property of each must be rendered all at the same time, and the management of the unique clinking sounds of all the tiny pieces as well as the hourglass itself must be taken into account - probability is a big equation even with a few simple variables, but is even bigger with more variables and more complex variable.

2009-02-25 16:30:00

Author:
Unknown User

Ouch. My brain hurts.

This is still a game right? (twilight zone theme plays in the background)

2009-02-25 18:04:00

Author:
CCubbage

Posts: 4430

Don't overload a small area with thermo-intensive stuff such as small lights coupled with tons of mag switches and sensors otherwise your mag switches will fail to read a sensor in range and cause unexpected results. An example of this is the beginning to my level, MR STAMP! ii. When you have multiple sackboys grabbing the fire trigger (tap R1) the "win" condition, caused by a mag switch, fails. The mag switch fails to read mag sensors even when they are in range, causing the player to progress even though they shouldn't. Another example of this can be found in my other level, ERADISACK, but it only occurs when you're playing the level with more than 2 players.

Also, I agree that there are probably multiple thermometers and that the thermometer we see is a representation of the highest temperature (which averages out as the thermometer heats up). Checkout this image from E3: http://www.thebbps.com/blog/wp-content/uploads/2009/01/27495_normal.jpg I'm speculating that those bars in the top left box are multiple thermometers; they increase in the E3 trailer when more material is created.

2009-02-25 21:39:00

Author:
snowflakecat

Posts: 102

Great post ninja, may I ask what education you have enjoyed or what kind of work you do?

I agree with the entire thing except one thing: I do not expect that the engine pre-calculates what the materials are gonna do. It'll keep in mind all of the different properties (sound, destructability, memory for the texture etc.) and keep in mind if it's glued, stuck with some kind of bolt or completely loose. I don't think it matters if you put to loose rocks on top of each other or next to eachother. The calculations necessary for that are just to complex I think.

The difference between a triangle shape or a circle is not in the fact that the circle is more likely to roll anywhere and the triangle is not, the diffence is in the number of polygons (or vertexes if you like).

It is however likely that it counts the number and complexity of object in near proximity, because that heavily influences the ingame load on the graphics processor and physics handling.

2009-02-26 09:44:00

Author:
Wyth

Posts: 263

For what it's worth, in Splat Invaders 2 I did some material changes. Since the level was fairly complex with vertices and objects, however only used about 4 types of material - I went in and started prettying it up with additional material types. I started going gang busters in certain areas and added about 6 different material types - some wood, some metal, some stone. My thermo didn't go up at all.

I suspect Ninja's assessment also fits into the equation - another "pool" as you will (for instance, detached objects being in close proximity therefore causing more load on the physics engine), but I'm still thinking the idea of different pools of resources with the highest one being the priority seems to make sense.

Also, I'm not really sure "logic" that would tax the processor is REALLY considered a resource. In general programming terms, processor logic will slow down the software but will not necessarily cause more resource usage. For example - I've seen levels in LBP that tax the processor so much everything is creeping along - but this was allowed during creation of the level.

Normally resources are structures containing data that must exist during execution of the application - for instance - bitmaps, sound, complex polygonal structures. The logic that controls physics must be there ALL the time regardless. Normally the different physics properties would react based on passed parameters to the same physics routines.

But then again.... I didn't program it....

2009-02-26 13:37:00

Author:
CCubbage

Posts: 4430

Can't they just hire a programmer at MM that only gives us nice technical explainations about these kind of issues? Not gonna happen right?

. Anyway the conclusions from the entire thermo descussion are best kept in the back of your mind while programming.

2009-02-26 15:22:00

Author:
Wyth

Posts: 263

LBPCentral Archive Statistics

Posts: 1077139 Threads: 69970 Members: 9661 Archive-Date: 2019-01-19

Datenschutz

Aus dem Archiv wurden alle persönlichen Daten wie Name, Anschrift, Email etc. - aber auch sämtliche Inhalte wie z.B. persönliche Nachrichten - entfernt.
Die Nutzung dieser Webseite erfolgt ohne Speicherung personenbezogener Daten. Es werden keinerlei Cookies, Logs, 3rd-Party-Plugins etc. verwendet.