Controlling the 'interest' list with a keyboard shortcut for distance cut

This isn't so much as a feature suggestion, but a possible solution to a technical lag problem.

Quite often when I teleport it takes forever to rez in or sometimes when I'm flying I'll enter a heavy prim-lag zone.

FOr example, I'll be tp'd into a room where I really don't care what's outside the walls. It would be nice if I could handily tell my client to just get the stuff close to me rather than everything outside in the sim as well.

One technical solution to controlling the interest list is having a keyboard shortcut which you can automatically reduce the interest list by reducing the distance cut with, and therefore cut down on lag at the appropiate times.

Any thoughts from the lindens?

This particular suggestion doesn't have to be the solution .. maybe you could illuminate what the issues are around interest lists and we can all provide some suggestions that might evoke some inspiration in the labs.

With something as obviously as complicated as SL, how can we hope to provide much in the way of actual useful insight from out here?

Well, SL is complicated, but it isn't rocket science.

And there are lot of very bright people in SL that have PHds and Masters in Comp Sci with over 20 years of programming experience.

Somehow I think we'll be able to help.

I dunno. I see a lot of people trying to offer suggestions from the outside, and it's pretty clear to me that LL very likely would have already thought of something like that. It's like saying, "hey, look, I know this is your full time job, folks, but I think we might be able to tell you how to do it better from out here even though we don't even have the design documents".

It'd be pretty cool if they released the algorithm behind the interest lists and let us take a whack at optimizing it. Then again, I kind of wonder if it might take more effort to write up documentation, communicate with a pool of techie users, filter through their suggestions for feasibility and applicability, and present the useful ones to the company than it would to just allocate time to developing the algorithm internally. Unfortunately, it's not just a matter of coming up with a better method of doing it -- there's implementation and testing too. You don't want them to forget testing, do you?

Oh, fresh perspective and cross polinating with talent which is not insular is a powerful force and probably the leading reason open source development works.

Besides, there are examples where this has worked before specifically in SL. For example, I believe it was Chromal who enlightened the LL team regarding different hardware architectures and their potential impact on Sim performance.

Does that mean Chromal is some kind of uber god who knows all? No. It means he happens to be particular good at that one thing (and may be at others, who knows).

LL has a small development team. It is not only improbable it is, frankly, impossible for them to have sufficient talent which is expert at all things.

Some of us are very very good at our particular things. For example, the unicode font problem. I'm sure if LL advertised it some one from the community could offer them a solution to the problem.

Besides, we don't need a lot of documentation. Vague descriptions of the problem set are usually sufficient.

A lot of us have enough experience to easily fill in the details.

And talking about complex problems can be a very productive way to spend your time. Even if you're not even reading the replies. It helps you formulate in your mind a better understanding of what you're facing.

Hmm, good points. I guess in part I'm just reacting to what I see a lot in the forums: people saying "come on, this can't be that difficult". Especially when someone gives a half-thought-out solution that sounds techie enough to make them look good but is so obvious the Lindens had to have thought of it already, it makes me uncomfortable. I guess that's not what you're after here, though. Well, like I said, this all takes people and time, even to cooperate with us... they made some more noise about open source in one of the recent town halls, so maybe we'll see some stuff opened up.

Well, this isn't really a technical solution. It's a user feature to partially solve a technical problem.

Intuitively, not the first place most technical people start.

Besides, I think you'll find most of us are probably in SL more as users than a lot of the engineers at LindenLabs. It gives us a unique perspective they'd be smart to tap into.

I've wished for something similar... a draw distance (or expanded performance) slider in the menu bar, near the clock. That gray space has gotta be filled with something useful! Keyboard shortcuts could make it even more accessible, yes! MACRO POWAH.

I don't suppose the "Drop Draw Distance if FPS<#" option is quite what you meant, is it?

Unfortunately it isn't, Lee. What I'm personally hoping for is a quicker way to drop and raise the draw distance with a mouse click-and-slide or with keyboard shortcuts, as opposed to having to bring up the Preferences menu and dig within each time.

There is also an empty space above the map and inventory buttons which could have a draw distance selector like Torley mentioned.

I too am tired of hopping into preferences to drop draw difference once I know I'm gonna be indoors for a bit.

Couldn't it be very simple ? Like a second drawdistamce setting in preferences - "indoors" setting.

And a button in the main window (like "minimap" labelled "indoors" which toggles between the two settings.

Would that come some way to meeting the need, or have I misunderstood ?

Is that "minimap" button worthy of the real estate it occupies, anyway? The old cross in the corner was ok for me.

Well, that's cool and I should have mentioned it (I actually forgot about it), but the reason I gave up using it was because it would reduce the cut when I didn't want it to.

Also, it doesn't solve the primary problem of taking forever to rez things in when I TP. Or, like Lindar said, I'm indoors and I really have no need for a large cut.

Another idea is that maybe you could slowly push the cut outwards on a TP?

So, do 25 metres until everything rezzes in, and the 50 metres, etc etc.

How expensive is it client/server wise to change the cut preference?

Intimate knowledge of performance issues of a suggested implementation versus our current methods is a bit out of my league... perhaps this is a better thread for the Feature Suggestions forum?

Re the dropping draw distance thing... it doesn't work because my draw distance is always pegged to the ground at 64. I would assume it won't go lower than that... I tried having it higher and turning this option on, but I think my FPs is so inconsistent that it just automatically was dropping the draw distance. I agree, it'd be cool to have an uber-power-user feature to have quick access to the draw distance Most users probably won't need this.