Archive: 23 posts

---

Using the same principals to create an incremental bolt I managed to create an angular tracking device without the use of a 3 way switch. http://farm3.static.flickr.com/2779/4396914483_187e0c0ca1_o.jpg The device is composed of two layers, one thick and two thin. The first layer is composed of dark matter to hold the object in place. The second layer is a thick circular piece of cardboard with a motor bolt in the center. The last is a thin layer of circular cardboard with mag switches/proximity switches and another motor bolt. http://farm5.static.flickr.com/4022/4397680626_c998a6147c_o.jpg Here are the details: 1. Set the back motor bolt to turn clockwise and the front motor bollt to counter clockwise. 2. Depending upon the application, the speed setting may be fast or slow. I used 7 for smoother tracking. 3. On the front layer, place two proximity (or mag) switches 180 deg. apart and set their radii to 180 deg. as well. 4. Next set the switches to speed and invert. 5. Turn the wheel so that the two switches are horizontal. Connect the left switch to the front layer and the right to the back. 6. Tweek to your specs. http://farm3.static.flickr.com/2677/4397680668_cce54e052b_o.jpg Here's what the finished product should look like. I added a light to show the tracking. http://farm5.static.flickr.com/4015/4396914555_488ccd067e_o.jpg The Pro's Decreased latency Fewer components Can output SPEED, DIRECTION, or ON/OFF Pin point Accuracy Con's It takes up a bit of space so this may or may not work some applications. Tracking moves in increments, therefore it's not suitable for lights.

2010-03-01 05:31:00 Author: M_R_Enigma Posts: 161

---

You forgot to mention that latency is reduced over using a 3-way switch, so it's win-win all the way One thing that might interest you as a little experiment when the PS3s wake up again is to spread the proxs so they are much wider apart, set them to 360 degrees and speed. It's kinda impractical for many applications due to the fact that the sensors have to be wide, but the behaviour is rather nice and they hadn't invented angle sensors back then....

2010-03-01 18:58:00 Author: rtm223 Posts: 6497

---

Well the wide spacing of the proxes allows for the speed settings to be effective. Or at least that's what I found. But then I was using 360degrees with overlapping areas to get the speed up to (sort of) occur based upon the angle away from center, rather than distance from the device (unless I missed something).

2010-03-01 19:33:00 Author: rtm223 Posts: 6497

---

Using the same principals to create an incremental bolt I managed to create an angular tracking device without the use of a 3 way switch. Interesting. And if I wasn't too scared to turn on my PS3 right now, I'd have a play with this. I've certainly used the double-bolting technique to create a multi-speed linear tracking device that doesn't need a three-way*, then used a zero-strength piston to convert it to an angular tracker, but I've never tried an angular tracker with this technique. The problems I've found in the past with using a tracker to create a follow-spot, is that they either can't keep up with sackboy if he goes from standing still to running at full speed in the blink of an eye, or they have a noticable 'bouncing' effect when he stops suddenly. I've certain never come across anything as smooth as the one in the Introduction level, but after discussing this on LBW, we suspect that's a cheat by MM, since it's capable of only tracking a single sack when there are multiple sacks in the level. * I'm actually not too fond of the remote three-way technique - always seemed a bit too hacky for my liking, and definately no use in a speed switch configuration - you can never get the thing to stop outputting a value once it's started.

2010-03-01 19:53:00 Author: Aya042 Posts: 2870

---

If it you need to you can reduce it to two planes by using the checkpoint and golfball layers for the card components and the thin layer behind for dark matter. Looks really nice though, I look forward to giving it a try myself. I like the compact nature of it in that it's very much a self contained piece of logic. Great job!!

2010-03-01 22:07:00 Author: croissantbuncake Posts: 572

---

Right. I just had a little eureka moment when I realised exactly how this contraption behaves. I was there thinking: but if you are further away it will move slower. Which is of course accurate, but I forgot to factor in that the distance from the tracker will multiply this and effectively make it track at the same speed at any distance. I like. I have a feeling it'll screw up a bit in multiplayer, so it might be simpler to use directional sensors, although... what does "require all" do with speed? When I can get back on I'll post the angular tracker I made back when angle sensors weren't invented (the one that I didn't explain well above). It essentially subtracts the speeds from two sensors in such a way that they seek an equilibrium point that creates an isosceles triangle with the player (or, in multiplayer, a trapezium, which ruins everything lol). It does a similar thing to yours in that it will move faster if you are closer, but it will also move faster the further off target it is. So if you are at 90 degrees it will whip round waaay faster than normal. But it needs the prox's spread out, making it pretty much useless for a lot of things

2010-03-01 22:53:00 Author: rtm223 Posts: 6497

---

To clarify: Before Cornish Yarg and Creator Pack 1, switch angles couldn't be tweaked so PTS's were hard or impossible to make. I made an advanced angular tracker too, but unlike most, it used stacked motor bolts. There were three wheels with stacked motor bolts on top of each other, and there was a line of seven proximity switches angled out differently. It also had a dark matter emmiter, which, although simple, worked phenomenally. If the device was pointing at you, only the middle proximity switches that activated the lever of the slowest wheel would make the tracker move. If you went farther away from it pointing at you, the more far-out proximity switches would sense you and activate a faster wheel. I haven't had time to perfect the design but with good tweaking I think it could be flawless. You could make one if you want. Brilliant tracking idea too.

2010-03-01 23:07:00 Author: Incinerator22 Posts: 3251

---

Wow... I would love to remove the old-school tracking devices from some of my levels and throw in a beaut like this... especially for False Idols III's sun ray. God, I want to rebuild everything so much when I get back... you guys are moving at light speed and this stuff is so awesome.

2010-03-02 05:58:00 Author: Unknown User

---

...if I wasn't too scared to turn on my PS3 right now, I'd have a play with this. Now the apocalypse is over, I'm finding this to be quite jerky, so probably not much use for a follow-spot, but would be fine for some sort of auto-targetting system. Frankly, I'm not sure what the point of setting the sensor switches to 'speed' actually accomplishes - at the sorts of ranges they'll likely be used at, it just means you have to consider two variables (motor bolt speed and sensor radius) to determine its sensitivity instead of one, were you to set the sensor switches to 'on/off' instead. The example in my copyable level "Follow-spot Example for Sehven" will actually track at a variable rate, but it doesn't react well to quick changes in sackboy's velocity. I don't think it's possible to achieve a smooth variable-rate tracker (linear or angular) which can actually keep up with sudden changes in sackboy's velocity, but I'd very much like to be proven wrong. However, you definately get full marks for thermo-efficiency.

2010-03-02 12:22:00 Author: Aya042 Posts: 2870

---

It's threads like this that really make me wanna get back to lbp but I musnt... As for "turret tracking" couldn't that technically be achieved with 1 thin 1 thick a motor bolt and a single prox set to directional?

2010-03-02 22:18:00 Author: Kern Posts: 5078

---

If you want it really, really jittery Although you could lock it in place using emitted dark matter when it finds its target...

2010-03-02 23:47:00 Author: rtm223 Posts: 6497

---

But surely many guns are jittery so it would add to the effect...

2010-03-02 23:54:00 Author: Kern Posts: 5078

---

I found that the distance at which i used to make my calculations was pretty spot on for smoothness, but like others have said it is very jittery at other distances. Perhaps I can find a formula for distance/speed for optimum tracking smoothness. The reason that I used speed over directional is that I found it to be a lot smoother. When you trigger a proximity set to Directional, the motor bolt will instantly jump to the maximum speed you set it at. Ultimately, it will center it self on the player but the tracking is jerky. Conversely, when the switch is set to speed the result is somewhat like a a variable output transmission. When no players are detected, the speed of the wheels will output at say 10 in opposite directions. The result is that the sensor doesn't move because they are of two equal speeds. Now say that a player in range makes one wheel turn at a speed of 8. This in turn, result in the wheel turning at a speed of 10 to overpower the wheel at 8 making the device move at a speed of 2. It would be best if i could show you through video but I can't. Yes this device requires a bit of tweeking to make it spot on depending on its distance from the player. My advice would be making the proximity radius larger than the vertical distance so that it moves smoother.

2010-03-03 00:22:00 Author: M_R_Enigma Posts: 161

---

It would be best if i could show you through video but I can't. Could you publish a tech demo level in which you've tweaked all the settings as best you can?

2010-03-03 01:10:00 Author: Aya042 Posts: 2870

---

Sure, no problem. To be honest, the reason i built this object was because I needed a fast tracking device. When you guys see the tech demo, you'll know what I mean. EDIT: Alright I've done a bit more testing and I came out with two completely different behaviors. On one hand, I was able to create a tracker that was extremely smooth, even at a speed 60, but the precision was horrible. The second device was the total opposite. It moved in minute increments so the tracking wasn't as smooth as I hoped it would be. On the other hand, the device moves with pin point precision. Overall, I thought this device was excellent for a turrent style weapon, but not as a spotlight. Like RTM stated earlier, setting the proximity switches so that the perimeters create a triangle yields the smoothest result.

2010-03-03 01:44:00 Author: M_R_Enigma Posts: 161

---

I've know how to do this and I have to say one thin layer and one thick layer dosn't work because the thick layer has more power over the thin layer... don't know why.

2010-03-05 00:24:00 Author: singularik Posts: 130

---

You have issues with the weights of objects attached to connectors. You need dark matter -> heavy -> light objects if you want to stack motor bolts. Of course, thick layers are heavier than thin layers

2010-03-05 00:42:00 Author: rtm223 Posts: 6497

---

You have issues with the weights of objects attached to connectors. You need dark matter -> heavy -> light objects if you want to stack motor bolts. Of course, thick layers are heavier than thin layers I had no idea the system would take into account the weight on an object attached with a motor bolt. In any case, the order of the components are exactly as you stated. On a side note, Is there a thread that contains all the rules of the programming?

2010-03-05 05:56:00 Author: M_R_Enigma Posts: 161

---

I had no idea the system would take into account the weight on an object attached with a motor bolt. In any case, the order of the components are exactly as you stated. Indeed, so you should be okay. Alternatively, if you do have the layers inverted (i.e. thick dark matter -> thin non-dark matter -> thick non-dark matter), you can just compensate by using different materials, e.g. make the thin layer out of, say, metal, and the thick layer out of, say, polystyrene. On a side note, Is there a thread that contains all the rules of the programming? Not sure - I've wanted to have a go at one for a while, but here's a couple of quotes from LBW on the subject. From a post (http://forums.littlebigworkshop.com/lbp/board/message?board.id=creationgen&message.id=41287#M41287) about problems rotating a heavy wheel:- An important difference between LittleBigPhysics and real physics is that, with LittleBigPhysics, the force applied by moving connectors, such as pistons and motor bolts, is proportional to the masses of the two objects they connect. Dark matter (and anything glued to dark matter or to the floor) has infinite* mass, so if you attach your 'heavy wheel' directly to a block of dark matter with a motor bolt, it will work just fine. If you're unable to use dark matter, you might be able to get away with increasing the mass of the object to which you're connecting it. If that's not possible either, then you may have to go with the other suggestions of reducing the mass of the wheel. * After checking, it's either not actually infinite, just very large, or it's only infinite for some calculations, most notably F=ma (i.e. no amount of force can cause it to accelerate). The 'check' was a setup like this... A---B C ...where 'A' is a block of dark matter, connected with a stiff piston to block 'B' of an arbitrary material, and seeing if it was able to push a very heavy block 'C' across the ground. i.e. if the mass of dark matter is infinite, then does the piston have an irresistable force? The answer was 'no', but as you increase the mass of block 'B', it was able to shift larger masses for block 'C'. From a post (http://forums.littlebigworkshop.com/lbp/board/message?board.id=creationgen&message.id=34719#M34719) about pistons bending (same issue, different connector):- After investigating some more, it looks as if it's not the force of gravity causing the horizontal piston to bend, but the relative acceleration being applied by the force of the vertical piston, although this is still a violation of the laws of physics. And since the relative acceleration is inversely proportional to the mass of the connecting objects, then simply using a larger mass at the junction of the two pistons solves the problem. It doesn't even need to rest on anything for stability. C | A---B In the above diagram, assuming 'A' is a block of dark matter, and the lines represent piston connections, then as long as the mass of block 'B' is not significantly smaller than that of block 'C', it all works fine. That thread then goes on where I have an interesting discussion with Sehven about the fact that speed of pistons and motor bolts, and the lengths of pistons also makes a difference to the forces that they apply. From what I can gather, for the most part, the rules are detemined by basic Newtonian mechanics, but there are a few cheats that Media Molecule have used to make the calculations simpler. For example, atmospheric resistance in the real world is influenced by the shape of an object, whereas in LBP, it's based solely on the mass.

2010-03-05 14:05:00 Author: Aya042 Posts: 2870

---

Can't you guys just use floaty material?

2010-03-07 14:05:00 Author: MMLgamer Posts: 183

---

Can't you guys just use floaty material? Do you mean for the circles? its a matter of Thermo Saving, Seen as Pink and Peach floaty isn't used as often it would stand to more sense to build it out of more common materials.

2010-03-07 16:04:00 Author: Kern Posts: 5078

---

If it is for a weapon system couldn't this be achieved with 1 sensor switch a delay and an Emitter?

2010-03-08 17:34:00 Author: Kern Posts: 5078

---

Ugh, it pains me when you guys say that you can't use floatie because it will take up more thermo. No. Materials is a different thermo than all of the others, and they are independent. Try it out. Go into a level where you can't place anything else. Start changing some objects to floatie. Nothing happens.

2010-03-08 17:53:00 Author: comphermc Posts: 5338

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

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.