Camera to LCD Screen

Imagine a security room that watches a bunch of turreted cameras.

I don't think adding a "camera view of the screen" will be easy in a game environment. For this to work properly, two (or more) scenes need to be rendered at the same time - both the scene of the room or area with monitors, and the scene of the view from each camera. And a transformation of the camera view to the monitor view is done.

See this is something that would also be amazing in se-1 but the modding API doesn't have support for it, so would love to see this added, and in reference to the rendering of two scenes you could significantly lower resolution based on distance or even disable the feature based off of distance, and add client settings to change the minimum distance required to view.

I would be happy if the camera could take a storable still shot.

I would have remote surveillance drones spy on enemy bases and take photos.

Photos could easily be displayed on an LCD.

You could find a mysterious photo on a Datapad in an abandoned wreck.

It would have metadata on the photo to locate its origin.

In response to Jamie Bristow the other thing is you could improve performance even more because the screen doesnt need to have 60fps or even anywhere near it. You could get away with 10fps or even lower, because if you need actual real-time feedback you would be looking out of the camera directly.

So you can do this in SE 1, but not with the mod. its a plugin and requires the plugin launcher. Its useable, actually made a ship that uses a large 5x3 lcd and a camera on the front to drive from. It works well enough until it becomes night time at which point it can't detect the lighting for some reason which i believe is a known issue and the originally creator isn't updating the plugin as far as I'm aware. in terms of rendering i didn't notice much of a difference, but I have a good system so other peoples experiences my vary.

@BestJamie

I think there should be an option to enable priority to a screen. Perhaps it would be linked to the cockpit/control seat, so that it only activates priority when you sit in the seat. This way, you could control a drone with a higher frame-rate and still see all of your auxiliary screens and indicators without going fullscreen into the camera view.

You might still need to have a bind to go into a fullscreen view of the camera, but having a priority mode for the screen linked to your control seat would reduce the need to do so.

Yep, already possible in SE1 with plugin:

https://github.com/austinvaness/CameraLCD

I use it with multiple screen pilot seat.

Clearly a must have as vanilla in SE2.

This was done in the half-life one engine almost 30 years ago. It's technically viable it just needs to be coded in at the appropriate level for optimization.

I've been working on this massive rover project for some time, and the idea of it in my mind is that it's intended as an all environment long mission rover. Because of the possibility for hostile environments I don't like having too many windows on it. But because of the limits in SE1 I can either have a single Camara with limited view, or i can have windows for the pilots to see out of. Ideally what I want is for there to be a series of cameras connected to a wall of lcds on the inside, providing a simulated cockpit view. So I would also like to see this feature implemented as well.

This could potentially allow you to discover issues with various automated systems without having to travel all the way out to observe them.

That would sorta defeat the purpose of doing it in the first place since it wouldn't be a live feed. as for consuming resources if you turn the monitor off it won't consume anything. so an option for that would be that it only provides a feed if you are sitting in the control seat the screens are connected to. if its instead sent to LCDs just use the setup action so that when that specific seat is occupied it turns the displays on and when its empty they turn off. very simple solution for conserving processing resources. spec wise this should be a feasible solution, but I can't test it as my system far exceeds the recommended specs.

It wouldn't work as a client side only feature not when using it on a server. That's specifically because the server is doing the calculations and the GPU on the user end is providing the rendering. (Probably not the best explanation)

> Do you really want to sit in a seat and watch screens all of the time, when you are sitting in a seat watching a screen?

I like to fly around a bit and do things.

Personally I would like the screens just to have different view angles from my ship so I can, for example, see where I'm flying at the same time I'm visually scanning terrain below, which can be done using third person but I would find it neat to be able to do so without it.

You are correct, no one is talking about recording camera feeds, but they will, and it is ridiculous. More..is a current theme on SE2. I do like the the idea of cameras on LCDs, I can see the need for a surveillance room with scripted role play as a setting. I do not see much loyalty for standard use when nothing else is happening, see the famous trope of the bored security screen watcher.

When the action is happening and all guns are blazing, watching the screens would have great value.

What level of screen quality would you see as acceptable?

I am not against camera LCDs, but have doubts on expectations, use cases, bloat and whether some simpler tech would be more appropriate for some needs.

I am inspired by the comments on these posts and my thinking is expanded. A little negativity can provoke the debate, too much does the opposite, a balanced view is my goal.

It doesn't matter. The end user renders it, not the server. It's slightly more network overhead at most. Let the user decide how good they want the cameras to be with their own client settings.

I use the plugin now in SE 1 and while I do have a very good computer. I don't see any performance issues. Plus if it's implemented like the plugin you can configure the settings to reduce load.

At the end of the day it would be the end user rendering it so it would be up to them on if they even use the function or not. It's also easy enough to have it set up so those LCDs only displays the camera feed at certain moments. Like when you sit at a certain control seat or enter a specific room

Would performance be an issue if trying to patch another players camera feed onto your LCD or would you be simulating a copy of their view?

Does this depend on the connected server or local host?

Currently it only works by having the plugin in SE 1 and using the plugin launcher. As for how you tie it in. You just have to call the camera in the custom data if I remember correctly (haven't been on SE 1 in a few months). As for patching into someone else camera it doesn't matter who's camera it is as long as you have access to it such as being part of the same faction or it not being owned by anyone.

For servers however, they also need the plugin, but unlike a mod you don't have to install the plugin to play.

For SE 2 it would be different because at present the hope is it's a built in feature. Because of that I would assume in the LCD drop down where you can select what to display. You could choose camera and then below that it would allow you to add the camera you want to feed to the monitor. Not sure how in depth they would take it, but in theory they could have it so you can select several cameras and have it cycle through them on a timer.

Where people might have issues is if one person on a server has a really good computer so they can handle running 20-30 of these camera to LCD feeds, but if a faction member that doesn't have a good PC gets to close they may see major frame drops as the GPU tries to render 20-30 of these systems plus whatever it is they are actually doing. Especially with SE 2 having Ray tracing, but even now their game seems fairly optimized and this is a very bare bones alpha.

A fix for that though which I've stayed a few times. Is to set it up so the LCDs are only on when certain conditions are met. Better still we might get a combination camera + sensor that only turns on the camera feed when it detects a player, grid, or debris. I could also see a sever side limiter option that controls the number of these feeds each faction can have. Though that would be a little weird and would likely work better as a per grid limiter. I'm just not sure how that would be coded

This leaves me with a question relating to game optimisation.

If a tree falls in the forest does it make a sound if it is not observed?

If a remote grid is otherwise in a steady state, and it is at a distance from interference, do the Havok physics process?

Do they need to?

It's calculated, but not rendered. Servers don't have to render something to understand what's happening. It's just using positional data along side velocity to determine interactions. This is different from say a client side save where your PC/console is rendering and doing calculations. In this instance, typically the PC records spawn data, but it doesn't actually place anything till the play gets close enough for that thing to come into render.

An example of this is trade outposts. If you land on a plan and pull the data pad it gives the coordinates of a trader. If you enter spectator and teleport to the location in spectator. The trader won't be there because the player model isn't in render distance. But the moment you teleport the player to that position the trade lost will appear. In some cases this is referred to as a hibernation state in which areas that aren't being rendered don't change. That cuts down on performance requirements.

In the case of SE 1 I believe the do utilize a hibernation state in solo player, but the things set to hibernate are npc grids. Player grids as I recall continue to update. That's why a pam miner can make a trip to the moon then return to earth. Its also why the random encounters spawn within a certain distance of players. That way it conserves processing power otherwise the save would be overrun by npc spawns

Some kind of stand-in for docking would be a brilliant middleground. Could make it akin to those old fighter jet radars, where it sweeps back and forth and gives a monotone gradient of "there's a... something... kind of, over that way." It gives you just enough information to go off of without being super resource intensive, and if you need additional information, you can always have an *actual* camera there to check in with at your security office or intelligence room or what-have-you a player might want on their ship.

Short range, you could build a + at a docking bay and when the cross is in the middle of the LCD screen it means you're lined up. Long range, you know there's a distant planet or nearby asteroid or, if it's moving and you aren't, potentially a ship.

I'd love to have a live feed from cameras displayed on LCDs. Even if it was just an image taken every few seconds. Though of course some situations like docking or spotting cameras (perhaps cameras on the left and right of a large rover to avoid hitting objects) would need probably 2 FPS or more to be serviceable.

In one of my creative games, I made a wall of LCDs with static images taken from various security cameras. It would've been a lot cooler if those images were updated by the game with real images taken from the cameras, whether it be a 30 FPS live feed, or 0.5 FPS like old-style VCR surveillance recordings...

I strongly encourage you to look into the camera to LCD plugin available for SE1. Specifically you'll need the plugin launcher and the. You can get the plugin through the plugin client. You can then look at all the settings options it has to offer in game. I don't remember off the top of my head what all is offered, but I believe you can set a max fps and a few other options.