Joints?

Well, what the hell was the point of askin' then?

Perhaps because if there had been more people using them then they would have reconsidered. Of the people that care enough about joints either way to post here we seem to have a pretty small fraction who want to keep joints over getting Havok 3 sooner. If it had been different perhaps they would have wanted to know.

Yyyyyeah, oops. I missed Andrew's post. Bitciness rescinded.

Also, it's looking as if they aren't actually REMOVING joints just yet; just removing the option to make more of them.

The windchimes of the world are safe.

Never know when you might need them, I haven't actually tried it yet. So, I think it should stay.

I can't believe it. I just spent the time to read every single response on this thread hoping to see at least one person pointing out what, to me, seems quite obvious. Maybe it's because I support a software development environment for a living but I know some of you out there are actually programmers.
There certainly was a lot of shit-talking about their need to remove joints and "why can't they just fix them", etc. but no one seemed to care that Havoc is not a Linden product. For all we know, Havoc itself has certain idiosyncracies that make it extremely difficult to implement properly. Back when it came out (and, incidentally, when LL decided to use it for physics) there weren't a lot serious developers making physics engines. It certainly makes sense to me that Havoc 1, the very first iteration of the software, might have a few bugs- a few things that might make the Lindens' implementation difficult.
So, before you go screaming bloody murder about how inept they are and why can't they fix it, I challenge you to come up with a streaming MMORPG that utilizes *HAVOC 1* in a much more smooth and stable manner.
Not to be a chearleader or anything, but maybe this is less the Lindens' fault and more a problem with Havoc itself. Stepping off the soapbox now...

Persig

PS. I am not a Linden alt. This is, admittedly, speculation.

We took them out of that version so folks could see what they think; we're trying to reach as wide an audience as possible on this issue to get feedback on removing the ability to make new joints. Perhaps not everyone who is interested has posted to this forum yet?

The more feedback we get about them, the better we can make a decision. It's that simple. Nothing is set in stone at this time.

Good point Persig. Actually as anyone who's tried making a physics catapult can tell you, too great a force applied against the wall of the bar's rotation couple around the axel will yank the the bar free from the axel altogether and you have a broken physics catapult. Depending on the size of your construct, you need approximately a 0.027m^3 prim's worth of space between the axel and the bar's rotation couple before the bar and its object will swing freely. 0.3m may not seem like much, but when you're looking at it on top, bottom, and all around the axel it's really clear that the physics engine isn't quite refined enough.

I'm told that in Havok 3 with its (comparably) super physics you can build slightly larger than normal hinges for a door with almost no space between the couple and the axel it rotates around. Wouldn't that be worth losing joints for a while?

Um, how does removing joints NOW affect Havok 2+ implementation? If you take them out NOW, provide a way to CREATE/EDIT them with a new system. Don't just mindlessly remove functionality when a suitable replacement is NOT ready! Good god, that's HORRIBLE design!

Better yet, just fix the damn things and make them work correctly like they should've been in the first place...

havok2 wouldn't have any significant improvement on joints (it's just not in there). If you're not moving directly to havok3, don't bother removing features we won't see again until you update physics again (which will probably be never).

Seems my question is answered then - (which was a simple asking the reasoning for taking them out).

If it facilitates the improvment of physics in Second Life (thus enabling me to give you yet ANOTHER heartattack in Burning Life next year) then I'm all for it.

Physics in general in Second Life needs much improving, and if this means we get that AS WELL AS a better joints system due to that - it all sounds good.

But how long is this "transition" going to take place? Theoretically, one build. (Realistically, at least SL 2.) I just don't like the idea of not being able to create/edit joints in the interim--but, according to what Siro writes, I have to wonder if joints would EVER come back if they are removed now. I think Havok can DO joints since Half-Life 2 uses that engine (not sure which version) and I'm sure I saw something jointed in it...but I can't recall what. Just seems odd joints wouldn't be possible in Havok if they're already in SL which uses Havok--unless joints were slapped on top of Havok, in which case I say keep them and get them working right AND implement them in the Havok 2+ implementation.

I also don't like a "discussion" about a topic that lasts a mere 3 days before a decision is made. What kind of public relations is that? If you want the community's feedback, GIVE IT SOME TIME (at LEAST a week--sheesh!) before implementing something--not like I see there was much thought of the community in this decision ANYWAY (a familiar tune from LL from what I've read here in the forums...).

Did any Lindens reply back in this thread on the subject?
I don't have time to read all this.
They need to make their answers stand somehow. Without having to go hunt down the topin in another forum list.

Yes they did.

It's very simple, Eep. As I stated at the top of this page, development of Havoc is entirely outside the Linden's control. All they can do is implement Havoc as it's released from its developer. The longer they leave the current joint creation facilities in, the more items that will be created with it. All items created with the current joint creation system will likely be broken when the new physics engine is implemented. They are trying to decrease the amount of breakage when they introduce Havoc 2 (or 3). The reason joints from Havoc 1 might not work in Havoc 3 is not something the Lindens can necessarily control. It's not because they aren't working hard enough or haven't done their homework. It's just not their product. There is, apparently, a great deal of difference between Havoc 1 and the subsequent releases.
You might even say it's MORE "mindless" to rant about functionality removal when you haven't considered that there could be good reasons for doing so...

*And in case I haven't been entirely clear, I fully support joint removal until it can be reimplemented with a more current physics engine*

Persig

I won't miss them.

I will.

I guess there goes all hope of my ferris wheel ever working properly without it costing me thousands of L$ to get someone to script it for me.

Way to go LL. Ask for our input, then make a decision before there's been much chance for feedback from the 1% of the active playerbase who bother to read the forums.



Lewis

Yeah, I'm going through withdrawls too from joints being removed. But on the plus side of things, I no longer have the munchies.

Ridiculous. Like other engines, they can be modified. HavoK (no "c" is most likely modifiable. Why remove joints NOW when, supposedly, SL's Havok 2+ implementation has been going on for YEARS? That just doesn't compute.

...i'll miss beatfox's windchimes, surely, and to lose the swings on my parcel will make me sad...if it means real progress toward an improved physics engine, then do what you need to do, but please don't take the action unless you intend to follow through within a reasonable timeframe...

Except it wouldn't have ever worked properly WITH joints, so.... man what?

Initially I thought, "Breaking existing functionality is bad." But I don't think joints are critical for most people, and on a whole, we will get a lot more out of your time if you take support for joints out while you roll out incremental fixes to get Havok 3 ready for prime time in SL. It is for the greater good. It sucks for people who have windchimes, but think of all the simulator crashes (and hence, lost work, in many cases), sims running at .1 time dilation, etc. that Havok 3 can solve.

Windchimes and swingsets are nice, but not at the cost of crashy/time-dilated sims. If it was something like, "Ability to attach objects," or "Ability of scripts to send and receive link messages," I'd have a real problem with it... but joints? Let's suck it up now and enjoy a more stable environment quicker.

While I have exactly 2 items in my inventory that will be broken by this change, and I sympathize with those creators...

LOSE 'EM!

Bring 'em back some day when the implementation actually *works*.

Let's face it, we've got one talented, wonderful, creator in this thread using joints - the windchimes - that work. While I think we should all tip our caps to the guy who actually made 'em work, and bow before his genius, the fact remain that 99.999999% of content created within SL don't use joints.

If it will improve the physics engine - and we can eventually re-introduce joints - then do so. I don't understand why this is even an issue, to be honest. Every point release in SL has broken more content than this will, accidentally. Its always a shame, but hey, upgrades are always a pain. Trust me, I've had to re-code a LOT of stuff on a regular basis...

I feel for the brilliant geniuses who, at a one in a million odds, got joints to work. For the rest of the SL universe, I hope you understand why an improved physics engine is more important.

Regards,

-Flip

I just want to know, FROM A LINDEN, how removing joints NOW, this late in Havok 2+'s implementation (supposedly) will help that progress (assuming it's FOR that anyway).

I use a joint on my smurf house's hanging pulley rope and am considering using a joint for more hanging objects (well bucket, other pulley ropes, etc). If I can still edit an existing joint that's not too big a deal, but if I can't copy it (is that technically creating a new one?) then that will suck. Do I have to create a bajillion joints before SL 1.9 is released so I can edit them thereafter? :/