First Person perspective, without Mouse Move

I've done a few quick searches and I haven't seen this anywhere. However, I have seen a lot of complaints about the relative size of objects in SL.

If we could get into first person perspective mode while still having the normal interface, that would help this situation a ton. Right now, you can't do anything but touch something when in first person mode because it's mouse view mode.

A first person mode without mouse view would mean more people would be in this mode and thus things could be made smaller. Most people consider a 20mx20m plot to be small, but in reality it's over 4300 sq ft! That should be room enough for dozens of people. In RL, we've hosted 20-30 people in hour 1300 sq ft townhouse.

The effect of this would be an increase in land value, which is not a bad thing at all.

(And if this does exist somehow, call me blind -- I haven't seen it.)

-Zax

Yes indeed, Zax.

Basically, the camera and interaction model is totally borked by design at the moment. There should not be an "either-interaction OR 1st-person" paradigm in the UI at all --- the issues of interaction and camera view are entirely separate issues. I know of no other world with this restriction, it's pretty nuts.

Whether we want the menus and chat windows etc of the UI to be visible or not should have absolutely nothing to do with whether we are in first person or not. It should be a separate control.

And whether we are in one camera mode or another should have absolutely nothing to do with whether we have chosen to hide the mouse cursor or not, and whether we want to be able to Edit objects or not. Again, that should be switchable separately. Some of the larger MMOGs have up to 5 camera modes selectable by the player, depending on the situation. It's not rocket science.

The best way that I've come up with so far for trying to head in the direction of UI sanity is to propose that they XMLify the whole UI (a small part of the UI already uses client-side XML files in this way). This would allow us to decide for ourselves when to switch various modes on or off, so perhaps there would be some light at the end of the tunnel somewhere.

It's pretty awful at the moment though, alas. It's not just the problem that you highlight. Many other difficulties stem from this same issue.

You make good points, thanks for adding to this.

I think one thing we get caught in is the difference between what "we" see, as a sort of creator being, and what our avatars see. I would not want to remove the ability of moving the camera independent of my avatar -- build would become terribly difficult.

What you elude to is essentially switching between a "meta-world" mode where you and your avatar aren't quite the same and an interactive mode where only non-edit items are available and you _are_ your avatar.

Part of SLs goal was to make this difference as fluid as possible, but the end result is that almost no one ends up "being" their avatar -- rather they watch their avatar.

-Zax

Darn right. It's hard to not step in dog s*** if your not using the third person camera.

The 16x16m parcels seem a little underscaled, I may add. They seem about 1/3 - 1/2 that size.

No, seriously, although the camera can get wonky and I agree having a first person with the regular interface would be nice, it'd also help if most people didn't make 7' tall AVs. If LL scaled all AVs down even 15%, this would be a huge world.

Also, realize you can hit -tab- and you'll go to close-over-the-shoulder cam on your AV at the person who spoke last.

Granted, it'd be nice if it would zoom out as soon as you start moving again.

It's just two camera modes, "the camera" meaning the thing that's focussing the view of the landscape that you see on your screen. These two camera modes would be fine (if a bit limited compared to other world clients) if only they weren't so poorly implemented, especially in the immediate loss of camera position as soon as the av moves or when you switch between modes.

In 1st-person camera mode ie. mouselook, the camera is permanently affixed to the avatar's eyeballs. That's simple and not too bad by itself, but it's let down by not actually being able to do anything much when in this mode because they've swiped the normal UI controls away. In other popular virtual worlds, you always retain all your abilities when in 1st person, and you can travel in the direction you're looking at by using just the mouse alone. Our 1P is a shadow of what it could be.

In 3rd-person camera mode, the SL camera starts off by default a little behind and above the avatar and looking in its general direction, but it can also be moved away from the avatar within some limits of distance. The movement is very flexible but highly idiosyncratic, for example you can only move the camera if you have a physical pivot for the movement. It's not too bad on its own, but the tie-in with avatar movement is pretty dire, usually causing total loss of 3P camera position when the av moves. The notion of camera independence got lost somewhere when they implemented it.

There's also another issue that is relevant here, camera continuity, because we have the potentially very nice concept of smooth 1P-3P transitions in SL: your mouse scrollwheel can take you from 1st person to 3rd person camera modes fluidly like a zoom. That would be very nice indeed and a salient SL feature, but unfortunately it's badly borked --- I've written a whole thread about it, so I won't repeat the issues here. See the camera link in my sig if you're interested.

I'd summarize by saying that it's very early days for the SL camera. On a scale of 1-10, it's at 3. Camera modes have progressed enormously since the first 3D virtual worlds came out, and we have a lot of catching up to do. I just hope that LL listens.

I forgot to mention the average AV is about 6ft tall, instead of 5ft 6 makes a big difference.

Somewhere else someone posted the average as being closer to 7ft. Real averages depend on gender and are closer to the 5-9" range, and on the rise.

This makes a big difference in perception. Take an average sized car, such as a Honda Civic. If you drive around in a Honda Civic yourself, it'll look pretty normal. However, if you drive around in a large SUV that has a seat high off the ground, that same Honda Civic will look quite small. Alternately, if you drive around in a Mazda Miata that same Honda Civic will look quite big. Typically, when looking up at something it's bigger than when looking down on something.

When we're in 3rd person mode we're usually another meter or two above our avatars head's. This makes things look even smaller both because of the perception of looking down on things and the fact that we're also a few meters farther away than our avatar is.

It doesn't help that our screens can't emulate the field of view we normally get unless we zoom out, as well. This is all very similar to the problem that racing games have where showing the effect of speed (so people can see better when to break, like in real life) is very difficult.

A goggle interface would be great for SL. But I'd also want gloves to access floating UIs. *sigh* Everything has to have a starting point, though...