How intensive is it to upgrade graphics for an existing game?

Cryomatrix

I'm not a programmer at all. However, would it be super difficult for a company to let's say, take Star Wars Galaxy, update the graphics to modern standard and then relaunch if the code is similar?

Seems like a cheap way to make money, but I don't know how intensive is it? Do you have to recreate all art assets with a new program? Could you use sharper textures? 

Basically, I don't know, so looking for people who know what it takes so I can elarn. 

Cryomatrix

btdt

You have to recreate all the art assets... that can be simple or complicated, depending on how everything was done.  Then you have to factor in the limitations of the graphics engine being used.  The graphics engine defined the limits under which you had to work.  So it's never as simple as redoing all the graphics... it affects a lot of other things.

It's basically time consuming.  

k4te

It's a lot of work. They need to recreate all of the art assets and some models will have to be recreated to add more texture  as well. 

Let's just say it's near the amount of time needed to create an entire game, minus all the coding, a majority of the debugging and testing phase and voice acting. In short, still a lot of work.

SirAgravaine

https://en.wikipedia.org/wiki/Game_engine#Rendering_engine

It is code intensive because you are changing out calls and using different libraries/techniques etc. Changing from one standard to another would virtually require the entire game to be recoded. Many companies have done some version of this, releasing "remastered" versions. It takes less time to remaster than it does to make the game from scratch in most cases, because all of the content and assets are available for use. In some cases it make take more resources to remaster, such as the case for StarCraft, where they lost the source code and had to basically re-engineer it.

sunandshadow

In many cases with old games, they don't have the original models or animation libraries.

Quizzical

While there would need to be some programming changes, those changes will typically be only a tiny fraction of the game's code base.  They come with the caveat that graphics programming is rather tricky and most programmers can't do it, or at least not without a long learning curve.  If you're grabbing some off-the-shelf game engine, that will be accessible to far more programmers, but may make it so that you simply can't do some of the things you wanted to do because the engine isn't built to do them.

If the original game was single-threaded, then trying to rework that into a well-threaded game in order to properly exploit modern hardware could take a major overhaul.  When you're going back and looking at games that were developed 15-20 years ago, single-threaded code wasn't obviously foolish then like it is today.  So how much work it would be depends tremendously on the state of the code base.

The other programming issue is that if you want to pick up an old, rusty code base, you don't have anyone sitting around who already understands it like they would with a project that they had just spent the last three years of their lives working on.  There is a saying that adding more programmers to a late project makes it later, as it takes a while for the new people to get up to speed.  Well, if you want to pick up an old code base and start development on it again, it takes a while for everyone to get up to speed.  If it's horrible spaghetti code that is completely undocumented and no one understands, changes that only truly require modifying 1% of the code can sometimes be mostly easily done by throwing out the old code entirely and starting from scratch.

The real expense is in completely redoing most or all of a game's artwork.  That costs a fortune.  You'd end up spending a considerable fraction of the cost that it would take to develop a new game, without the payoff of having a new game to sell at the end of it, as people who previously bought the game probably don't want to pay a new game price again just for updated graphics.

Acornia

To remake a new SWG game you would very much have to start with a blank slate.  SWG was made way back in the day of Cobalt, Fortran, and basic version of C, so about all you could use is the disign documents if you can get your hands on them. 
All the art work and graphics from 32 bit to 64 bit to met new post Nvidi 390 driver where they are droping all 32 bit support.

Melecon

SWG isn't as hard as people here would lead you to think it is. There is an open source project called SWGEMU some of you might have heard about it. This project has taken the back end (Server stuff) and recreated it to work with the current engine (SWG Engine), and it is open source. You can in turn take that and the free tools to pull all the models out of the SWG TRE files and rebuild them. This method wouldn't take a great deal of time to get to this stage, I would probably give you a few weeks to get it done. The long part is taking those assets and making them better and cleaning them up. The most time you will need to spend is on the models themselves as you will need to start from scratch as most modern 3D tools like Maya, Max, and Blender like to play with "Quads" where when you incorporate a 3D model into a game it converts all the faces into "Tries" so cleaning it up is a pain. You can just do a gloss coat on the assets to convert the textures to make them look better but in the end they still look a bit dated. Finally you would need to then take all this stuff and put it into a "Modern Engine" or at least an engine you know or is well documented. Lets pick on UE4 as there are a few underground projects that is trying to accomplish this very task, you would now have to incorporate the server calls back to the server from the new engine this again is very do able and can be done in a few months.

Lastly though and the stumbling block that all the projects are running into is the fact that you can't get the SWG animations out of the TRE files into a usable format for a modern engine to use. I am not saying it is impossible or not doable as there are some people out there that have done it just not sharing the results.

So the TL;DR of it all. It is very doable, it wouldn't take nearly as long as some would think it does, as long as you have all the documentation on the files formats and can easily get them into a state that is readable by another engine or tools to enhance them.

Also most times when a developer teams say that the old code is lost to the neither they are normally telling the truth as no one individual knows the whole code base and the documentation for these things is non-existent or scare (with it being unlegible).

Sorry for the wall of text,

Melecon

Nilden

I know as a hobby programmer I would much rather start from scratch with modern code than work with legacy code, for a plethora of reasons.

Quizzical

Nilden said:

I know as a hobby programmer I would much rather start from scratch with modern code than work with legacy code, for a plethora of reasons.

I think this gets at something important.

For the original poster, I think your real question is along the lines of, "I would like for them to remake SWG with modern graphics.  Why don't they?"  And that's a reasonable question to ask.

But the better question is, why would they?  Even if you had legal access to all of the art assets, source code, audio, marketing, trademarks, and everything else, why would you want to resurrect a long-dead project rather than making a new game, perhaps a sequel to the original?

Whenever you code a large project, you make some mistakes.  You make decisions that the game is going to be structured this way and not that way, and that later limits what you can do.  There are many things that are easy to do if you know that you want to do them before you start, but can be nearly impossible to patch in later.

For example, if you start out thinking that you're going to make a purely PVE game, and then sometime during beta testing, think that it would be nice to have PVP servers where players can attack each other, it's too late.  You'll have to make so many changes in so many places and you don't entirely even know where in order to make PVP work that it's not worth the effort.  If you know from the start that you want to have both PVE and PVP servers, it's not really that much more work to add the PVP options as you go along than if you wanted a purely PVE game.

So why not just add the PVP options as you go along?  Because there are many thousands of other such things that you could do that also wouldn't be that much more work to add them as you go along.  And you simply don't have the time or resources to do all of them.  You don't have the resources to do more than a small fraction of them.  Some of them will be mutually contradictory, anyway.  So you have to set priorities and make decisions.

A while after the game is launched, you'll have some regrets.  You wish that you has chosen to do this instead of that.  This mechanic that seemed so good in your head and even in preliminary testing just doesn't work very well once you let the min/maxers try to break it.  But it's too late to go redo it if it's anything sort of game-breaking, as fixing it now will take massively more work than if you knew ahead of time what changes you'd eventually want to make.

If you pick up some rusty, legacy code from a long-dead game, there are a zillion decisions like that baked in already.  And worse, they're not your decisions.  The original developers chose to make it possible to do the things that they wanted to do, not the things that you want to do.  Why would you want to accept that rather than going to a blank slate where you can build the game that you want to build?

Cryomatrix

great comments here. 

Thanks for teaching me. 

Cryomatrix

AAAMEOW

might be off topic but star wars license is probably quite expensive.  Can't seem to find the licensing cost on google.

Octagon7711

Fans want a recreation of something but developers want to be creative and not simply recreate something that's already been done, they want to be allowed to be creative.