PDA

View Full Version : Square Sensor Zones



GribbleGrunger
10-30-2014, 09:10 AM
I can't understand why we haven't got them. All you'd need is a set up like the camera where you can adjust the height and width of that four sided zone. It would make adding transparency to layers as the player moves forward, a breeze. I know you could do this with something like the holographic material and a touch sensor, but that takes up more thermo. And I also know you can do this with four sensors back to back and set to 45 degrees, but wouldn't it be much simpler just to add what I've suggested?

Devious_Oatmeal
10-30-2014, 10:37 AM
I always hated placing a zone. An editable zone would be great.

GribbleGrunger
10-30-2014, 10:56 AM
I always hated placing a zone. An editable zone would be great.

For all we know it might actually be in the finished game. The beta didn't have ALL the features, but if it doesn't I'll be disappointed. I just think there are certain things that make sense and don't appear to be too difficult to implement.

Ali_Star
10-31-2014, 10:13 AM
I actually placed this suggestion ages ago for LBP2. It'd be a good idea to have different shapes for the sensors.

Lt_Twinklez
10-31-2014, 08:40 PM
This has been something I've been wanting for a long time. It would always take me forever to set up sensors for a square area. Mostly because it takes five separate sensors to cover one square. Very good suggestion!

f3am4
10-31-2014, 09:46 PM
I would love to see this implemented!

SEWO97
11-02-2014, 12:18 AM
This has been something I've been wanting for a long time. It would always take me forever to set up sensors for a square area. Mostly because it takes five separate sensors to cover one square. Very good suggestion!

It takes you 5 sensors? Should it not be 4? Set at 90 degrees, rotated inwards.

Anywho, my input for this thread: I've myself wondered why we haven't gotten square sensor areas. It seems so simple and obvious.

Devious_Oatmeal
11-02-2014, 02:35 AM
It takes you 5 sensors? Should it not be 4? Set at 90 degrees, rotated inwards.

Anywho, my input for this thread: I've myself wondered why we haven't gotten square sensor areas. It seems so simple and obvious.

Screw square sensors! A sensor area that's editable, like a material and the corner editor would be better!

SEWO97
11-02-2014, 04:01 AM
Screw square sensors! A sensor area that's editable, like a material and the corner editor would be better!

Simple, have both.

Devious_Oatmeal
11-02-2014, 03:35 PM
With the editable sensing area, the default shape is the usual thing we have.

CamostarAtIsabella
11-04-2014, 06:14 PM
LBPK had square zones. It would be nice to see it here.

Devious_Oatmeal
11-04-2014, 09:03 PM
LBPK had square zones. It would be nice to see it here.

Upon reading and thinking on this for the whatever-th time, I think the holo with impact is actually fairly good. I very rarely use sensors that aren't infinite. I can size holo to whatever I want. I can choose the layers it's on, including thin layers. And I can turn the sensing off easily.

Plus, with that dynamic thermo, all that holo lying around won't be such a big issue.

koltonaugust
11-05-2014, 02:11 AM
Upon reading and thinking on this for the whatever-th time, I think the holo with impact is actually fairly good. I very rarely use sensors that aren't infinite. I can size holo to whatever I want. I can choose the layers it's on, including thin layers. And I can turn the sensing off easily.

Plus, with that dynamic thermo, all that holo lying around won't be such a big issue.

But it does get annoying to edit anything within the holo if there's logic to do with the visuals as well. There's a lot of stuff to hide and getting rid of one more thing would be great.

Devious_Oatmeal
11-05-2014, 06:57 AM
But it does get annoying to edit anything within the holo if there's logic to do with the visuals as well. There's a lot of stuff to hide and getting rid of one more thing would be great.

I either move the holo out of the way via grab and back a layer, or I just put the holo in perma-transparent. IF ONLY THAT WAS A THING! >:V

But yeah. I just make it transparent for the edits or move the holo back a layer. Unglued holo is one heck of a helper.

Ryuhza
11-05-2014, 07:03 AM
Square/Rectangular sensor would be great indeed. Rather than controlling it in a similar fashion to the camera-- or, in addition to that method, we should also be able to scale it by numbers corresponding to the x & y axises. The camera zone scaling can be a little funky if you don't have front view toggled. And if I'm not mistaken, you have to zoom out manually if you scale it beyond the borders of your screen.

GribbleGrunger
11-05-2014, 08:26 PM
I don't think the hologram solution is deep enough to compete with what a square sensor could do. Sensors can detect signal strength for instance. If I wanted to create a parallax scrolling scene, I could have a flat edged sensor (Square) at either end of my parallax scrolling scene and then it would just be a simple matter of taking a closeness value and transferring it to pistons moving the scenery in the background. You couldn't do with with a round sensor area detect because the value would be inconsistent when sackboy jumps, were as a square sensor would always take the value of sackboys movement left and right perfectly. With a circular zone you could jump up and forward a bit and NOT change the value until you start to fall (because takes the closeness value from the very centre of the zone), meaning every time you jump up and forward, the pistons moving your parallax scrolling would fall behind a fraction each time. With a square, it wouldn't matter what angle sackboy jumps (because the value would be take from a 'line' going between the top and bottom of the sensors limits), he would ALWAYS produce a value at exactly the same point he would if he was just running or walking in a straight line.

Thinking about that, I wonder if a better solution would be to enable a circular sensor to take a 100% value ONLY when you reach a point at which a line intersects the centre, and at which angle that line is is up to you?

Devious_Oatmeal
11-06-2014, 12:00 AM
I don't think the hologram solution is deep enough to compete with what a square sensor could do. Sensors can detect signal strength for instance. If I wanted to create a parallax scrolling scene, I could have a flat edged sensor (Square) at either end of my parallax scrolling scene and then it would just be a simple matter of taking a closeness value and transferring it to pistons moving the scenery in the background. You couldn't do with with a round sensor area detect because the value would be inconsistent when sackboy jumps, were as a square sensor would always take the value of sackboys movement left and right perfectly. With a circular zone you could jump up and forward a bit and NOT change the value until you start to fall (because takes the closeness value from the very centre of the zone), meaning every time you jump up and forward, the pistons moving your parallax scrolling would fall behind a fraction each time. With a square, it wouldn't matter what angle sackboy jumps (because the value would be take from a 'line' going between the top and bottom of the sensors limits), he would ALWAYS produce a value at exactly the same point he would if he was just running or walking in a straight line.

Thinking about that, I wonder if a better solution would be to enable a circular sensor to take a 100% value ONLY when you reach a point at which a line intersects the centre, and at which angle that line is is up to you?

Hmm... Well a follower and tag sensor always worked for me to get that left and right detection without Sackboy's angle skewing the results.

And with pistons now having outputs, or so I have heard, I can think of solutions using pistons and a follower.

GribbleGrunger
11-06-2014, 02:10 PM
Hmm... Well a follower and tag sensor always worked for me to get that left and right detection without Sackboy's angle skewing the results.

And with pistons now having outputs, or so I have heard, I can think of solutions using pistons and a follower.

A tag sensor has a definite 'centre' though and obviously the results are going to stem from that point. Let's say you have a level that you jump up platforms but you can still see the parallax background. Regardless of how big you make the sensors range, the higher you climb the further away or towards the sensor centre you would become. It wouldn't make MUCH difference if you set it to maximum range but it (perhaps) would make it enough to eventually offset the pistons slightly and cause them to start working out of sync with what you intended. We would need a reading from a vertical or horizontal line for the signal strength to be 100% consistent.

Try it for yourself by exaggerating the point I'm making. Take a sensor and make the range pretty short. Then stand just on the inside of the ring and jump up. You will see that sackboy actually moves further away from the sensors centre as he jumps and then gets closer as he falls. If you tried to tie this into a piston moving scenery in the background, the pistons would move even though you'd only jumped up and down and hadn't moved forward (which is really the only time you want the parallax to work).

This is an exaggeration but if you are trying to make something 100% consistent, you don't want this process to take place even if it is only fractions at a time.

Bonnell7
11-06-2014, 04:34 PM
Square zones seem great, but not in the manner in which you guys are referring. You guys apparently just want the digital nature of a sensor. But I, and I'm sure others, would want an analogue possibility. Especially one that would otherwise be a pain to create in LBP2.

I want to see an axis sensing range. Where the edge of the rectangle running through the sensor acts as the sensor itself.
So instead of how close the tag is to the sensor, it'd be how close the tag is to this edge.
Such a sensor would make xy coordinates a breeze. It'd also make 3d room rotation more efficient.

GribbleGrunger
11-06-2014, 04:49 PM
Square zones seem great, but not in the manner in which you guys are referring. You guys apparently just want the digital nature of a sensor. But I, and I'm sure others, would want an analogue possibility. Especially one that would otherwise be a pain to create in LBP2.

I want to see an axis sensing range. Where the edge of the rectangle running through the sensor acts as the sensor itself.
So instead of how close the tag is to the sensor, it'd be how close the tag is to this edge.
Such a sensor would make xy coordinates a breeze. It'd also make 3d room rotation more efficient.

Yeah, that's what I was struggling to say here:


Thinking about that, I wonder if a better solution would be to enable a circular sensor to take a 100% value ONLY when you reach a point at which a line intersects the centre, and at which angle that line is is up to you

Devious_Oatmeal
11-06-2014, 11:24 PM
A tag sensor has a definite 'centre' though and obviously the results are going to stem from that point. Let's say you have a level that you jump up platforms but you can still see the parallax background. Regardless of how big you make the sensors range, the higher you climb the further away or towards the sensor centre you would become. It wouldn't make MUCH difference if you set it to maximum range but it (perhaps) would make it enough to eventually offset the pistons slightly and cause them to start working out of sync with what you intended. We would need a reading from a vertical or horizontal line for the signal strength to be 100% consistent.

Try it for yourself by exaggerating the point I'm making. Take a sensor and make the range pretty short. Then stand just on the inside of the ring and jump up. You will see that sackboy actually moves further away from the sensors centre as he jumps and then gets closer as he falls. If you tried to tie this into a piston moving scenery in the background, the pistons would move even though you'd only jumped up and down and hadn't moved forward (which is really the only time you want the parallax to work).

This is an exaggeration but if you are trying to make something 100% consistent, you don't want this process to take place even if it is only fractions at a time.

I understand what you meant. I had the same issue when messing with parallax as well, but actually. I meant using the sensor on the follower attached by the piston, not the player. The follower would follow the player, but be locked to just be able to go left and right. That eliminates the jumping issue you are describing.

Of course, this is useless to say if the other options are available in LBP3 and we get the good tools! But for a 'just in case' sake, i like to say it's possible.

GribbleGrunger
11-07-2014, 07:23 AM
I understand what you meant. I had the same issue when messing with parallax as well, but actually. I meant using the sensor on the follower attached by the piston, not the player. The follower would follow the player, but be locked to just be able to go left and right. That eliminates the jumping issue you are describing.

Of course, this is useless to say if the other options are available in LBP3 and we get the good tools! But for a 'just in case' sake, i like to say it's possible.

Well blow me down! I never thought of that. This just gave me a decent solution for a HUD:

Limit the height of the level. Place two strips of material vertically from top and bottom. fit these to rails at top and at bottom. Place a square of the same material within the two strips and place a follower on it. Make it nice and slippy and then turn everything invisible. Attach your HUD frame to the small square. That should give you quite a 'stable' HUD.

edit: Oh wait, isn't there instances when the camera doesn't follow sackboy exactly as he jumps and moves through the level or is there a way of making the camera move with him in the centre at all times. I can't remember. I hope there is! Doesn't the local setting do that with cameras? Couldn't you then just attach a game camera to the little square piece of material too?

Devious_Oatmeal
11-08-2014, 02:55 AM
If you're talking about a sidescroller where it followers the player's left/right movement and not their jump movement, then that's very easy to do with the locked follower.

If you just want a HUD on the character and using a holo and follower, it's just like making the controllinator follow a sackbot, and then either a game camera, or a movie camera. I prefer a movie camera myself, since it locks in place. But the movie or game camera never want to cooperate anyways, so whatever! >:V

Anyways, then you stretch that piece of holo and put stickers or decorations all over it. Even logic if you want it to move and stick.

The issue though! Is angles. If you have a set angle, it's sort of fine because you can stretch the holo and move the stickers to work with the angle and make it look as if they are actually flat towards the screen. But this is very nearly impossible to do with angle changes throughout a level. UNLESS, you like being tortured and work the HUD to turn and twist to the right direction to match the angle of the camera. And that is no fun at all.

Mr_Fusion
11-15-2014, 08:44 PM
And with pistons now having outputs, or so I have heard, I can think of solutions using pistons and a follower.

Pistons, bolts, bounce pads, and... I think pretty much everything has an output now. :)

amiel445566
11-15-2014, 10:13 PM
Pistons, bolts, bounce pads, and... I think pretty much everything has an output now. :)

everything has an input too, like every single material, its ridiculous, and amazing

koltonaugust
11-15-2014, 11:12 PM
everything has an input too, like every single material, its ridiculous, and amazing

It does make things easier (although sometimes the camera whacks out and it becomes hard to select a specific input but I still say I love the change)

Mr_Fusion
11-15-2014, 11:34 PM
everything has an input too, like every single material, its ridiculous, and amazing

I din't think to check but the camera and infinite scroll sensors are both square, so maybe there is potential in them to make a square zone...

To make a square zone you actually only need two sensors on opposite corners to do it (one in lower left, one in upper right). So it's not that big of a deal. It would be faster selecting circle or square from a tweak though, and also a lot less little bits of logic lying about. Squares of Invisible Sticker Panel/Holo material pretty much has no impact on thermo, so there is no real concern there.

You can actually make a very fast to deploy square zone set up with two bits of sticker panel, each with radii set to 45˚ and another bit of sticker panel with a microchip containing an AND gate. You just need to position the two sensor blocks and set the ranges to overlap the opposite block, you're good to go!

On the subject of sensors, inputs and outputs, I wanted to mention a beta test I did. It was copyable and GribbleGrunger played it so maybe he has something to add... Anyways; I had the seasons change as you played a level.

The sensors do need to be able to accept signal strength (one of the limitations of using invisible material set to impact for zones). But with clever use of positional layers, the new material inputs, and outputs, I had it so that grass slowly browned off, then disappeared, and water fall goes away with the wet rock behind it slowly drying up. Walking in snow made you walk slower.

I plan to refine this in release version to have the ground get muddy if it rains, and it'd get slippery, or sticky or even make you slowly sink.

So yeah... not too worried about not having a square sensor, LBP2 was pretty impressive when it came out, but the new options to have nearly everything react and all the options for sensors to give feedback on everything (like even telling the difference if a player is jumping of falling), combined with the Broadcast Microchip means that there isn't really a creative limit to sensing what is happening in you level, and having logic react accordingly.

SEWO97
11-16-2014, 06:20 AM
To make a square zone you actually only need two sensors on opposite corners to do it (one in lower left, one in upper right). So it's not that big of a deal. It would be faster selecting circle or square from a tweak though, and also a lot less little bits of logic lying about. Squares of Invisible Sticker Panel/Holo material pretty much has no impact on thermo, so there is no real concern there.

You can actually make a very fast to deploy square zone set up with two bits of sticker panel, each with radii set to 45˚ and another bit of sticker panel with a microchip containing an AND gate. You just need to position the two sensor blocks and set the ranges to overlap the opposite block, you're good to go!

On the subject of sensors, inputs and outputs, I wanted to mention a beta test I did. It was copyable and GribbleGrunger played it so maybe he has something to add... Anyways; I had the seasons change as you played a level.

The sensors do need to be able to accept signal strength (one of the limitations of using invisible material set to impact for zones). But with clever use of positional layers, the new material inputs, and outputs, I had it so that grass slowly browned off, then disappeared, and water fall goes away with the wet rock behind it slowly drying up. Walking in snow made you walk slower.

I plan to refine this in release version to have the ground get muddy if it rains, and it'd get slippery, or sticky or even make you slowly sink.

So yeah... not too worried about not having a square sensor, LBP2 was pretty impressive when it came out, but the new options to have nearly everything react and all the options for sensors to give feedback on everything (like even telling the difference if a player is jumping of falling), combined with the Broadcast Microchip means that there isn't really a creative limit to sensing what is happening in you level, and having logic react accordingly.

You only need 2 if you're making an actual square. I think more people would be making rectangles.

- - - - - - - - - -


To make a square zone you actually only need two sensors on opposite corners to do it (one in lower left, one in upper right). So it's not that big of a deal. It would be faster selecting circle or square from a tweak though, and also a lot less little bits of logic lying about. Squares of Invisible Sticker Panel/Holo material pretty much has no impact on thermo, so there is no real concern there.

You can actually make a very fast to deploy square zone set up with two bits of sticker panel, each with radii set to 45˚ and another bit of sticker panel with a microchip containing an AND gate. You just need to position the two sensor blocks and set the ranges to overlap the opposite block, you're good to go!

On the subject of sensors, inputs and outputs, I wanted to mention a beta test I did. It was copyable and GribbleGrunger played it so maybe he has something to add... Anyways; I had the seasons change as you played a level.

The sensors do need to be able to accept signal strength (one of the limitations of using invisible material set to impact for zones). But with clever use of positional layers, the new material inputs, and outputs, I had it so that grass slowly browned off, then disappeared, and water fall goes away with the wet rock behind it slowly drying up. Walking in snow made you walk slower.

I plan to refine this in release version to have the ground get muddy if it rains, and it'd get slippery, or sticky or even make you slowly sink.

So yeah... not too worried about not having a square sensor, LBP2 was pretty impressive when it came out, but the new options to have nearly everything react and all the options for sensors to give feedback on everything (like even telling the difference if a player is jumping of falling), combined with the Broadcast Microchip means that there isn't really a creative limit to sensing what is happening in you level, and having logic react accordingly.

You only need 2 if you're making an actual square. I think more people would be making rectangles.

Mr_Fusion
11-16-2014, 10:10 PM
Not specifically, you can make a square or rectangle, you just need to position the sensors (the green circles) correctly, so that the 90˚ zones overlap to your liking (I did type 45˚ before though... sorry about that).

http://www.lbpcentral.com/forums/attachment.php?attachmentid=49499&d=1416175968

The AND gate will only trigger if you are standing in the area of overlap (the black rectangle in this example).

SEWO97
11-16-2014, 10:57 PM
Not specifically, you can make a square or rectangle, you just need to position the sensors (the green circles) correctly, so that the 90˚ zones overlap to your liking (I did type 45˚ before though... sorry about that).

http://www.lbpcentral.com/forums/attachment.php?attachmentid=49499&d=1416175968

The AND gate will only trigger if you are standing in the area of overlap (the black rectangle in this example).

Then players are free to exploit the sensor area because not all levels will be created the way you have shown.

49500

amiel445566
11-16-2014, 11:03 PM
http://www.lbpcentral.com/forums/attachment.php?attachmentid=49501&d=1416178776

http://www.lbpcentral.com/forums/attachment.php?attachmentid=49502&d=1416178935

EleoMod
11-17-2014, 01:42 AM
http://www.lbpcentral.com/forums/attachment.php?attachmentid=49501&d=1416178776

http://www.lbpcentral.com/forums/attachment.php?attachmentid=49502&d=1416178935

For the custom area, you can just use the Impact sensor.

amiel445566
11-17-2014, 06:30 AM
For the custom area, you can just use the Impact sensor.

I meant closeness %

Devious_Oatmeal
11-17-2014, 07:56 AM
http://www.lbpcentral.com/forums/attachment.php?attachmentid=49501&d=1416178776

I would have mentioned that as an issue, but then I tried to figure out in what possible situation would I actually have something like that. Like... figure out a situation where you would use it, then apply the square sensor to the same situation. The editable section would be able to match any square zone.

As for actually getting a degree from that, I would think they would make it to where it could never over lap, like editing a material and then it stopping your edit when another object is in the way, or it reads that percentage as if that gap was included in the black space.

tdarb
11-19-2014, 05:50 PM
Not specifically, you can make a square or rectangle, you just need to position the sensors (the green circles) correctly, so that the 90˚ zones overlap to your liking (I did type 45˚ before though... sorry about that).

http://www.lbpcentral.com/forums/attachment.php?attachmentid=49499&d=1416175968

The AND gate will only trigger if you are standing in the area of overlap (the black rectangle in this example).

That works for most purposes, but if you need X and Y position, it's still not a good fit. Unless I am missing something, the closest you can get with that setup is the two points where the distance arcs intersect between the two sensors. Using other methods (pistons, movers, followers, etc), you can get specific X and Y values for the position of an object within the rectangle.

I suppose it just depends on the output you need. A square sensor would be ideal.

P.S. Looking at your sensor setup, this may be ideal for something I am doing in which I need to know where both points are. Your setup hadn't occurred to me. Thanks!

modernknight
12-07-2014, 12:13 AM
A tag sensor has a definite 'centre' though and obviously the results are going to stem from that point. Let's say you have a level that you jump up platforms but you can still see the parallax background. Regardless of how big you make the sensors range, the higher you climb the further away or towards the sensor centre you would become. It wouldn't make MUCH difference if you set it to maximum range but it (perhaps) would make it enough to eventually offset the pistons slightly and cause them to start working out of sync with what you intended. We would need a reading from a vertical or horizontal line for the signal strength to be 100% consistent.

Try it for yourself by exaggerating the point I'm making. Take a sensor and make the range pretty short. Then stand just on the inside of the ring and jump up. You will see that sackboy actually moves further away from the sensors centre as he jumps and then gets closer as he falls. If you tried to tie this into a piston moving scenery in the background, the pistons would move even though you'd only jumped up and down and hadn't moved forward (which is really the only time you want the parallax to work).

This is an exaggeration but if you are trying to make something 100% consistent, you don't want this process to take place even if it is only fractions at a time.
Followers can be set to restrict vertical movement. Rotate it 90 degrees and it restricts horizontal movement. Put your sensor on a hologram that follows sackboy on the y axis by restricting horizontal movement. If you jump the follower ensures that the sensor stays on the same plane as sackboy and your parallax is preserved