Jump to content

source code?


Recommended Posts

well not gonna say never happen but if you did get it who is going to program the new code it would take many to do it and it would need to be done in spare time I just don't see it happening..


 I had many a good time in TAW and would love to see a new version someday but not with TAW..

Link to comment
Share on other sites

Me too...but what are other sims that can grant same possibilities? Falcon bms is fantastic but has a very deep learning curve...falcon Af is more friendly with avionics scalable but is not supported anymore...maybe the perfect solution would be strike fighters 2 with awacs and campaign like total air war...but i dont think it is easy to make...so for this reason i asked...maybe if the source code is at our disposal we can try with our spare time to get a  better game dont you think?

Link to comment
Share on other sites

well not speaking for the others but I am not a codere and could offer no assistance other then testing.. even if we had the source code it would take a dedicated bunch to do their magic.. 1 or 2 guys is not gonna cut it..


but if you think you can get the code who know but we would need the code first before we dream of anything else..other then having the code it is just a waste of time..

Link to comment
Share on other sites

How companies treat source code:

  1. Source code does not make any money. After a game is released, it's worthless for the company. It is treated accordingly.
  2. Source code is always evolving. EF2000 evolved into ADF evolved into TAW. It is likely that it evolved into Typhoon. Talking about “TAW source code”, we mean a snapshot of DID's repository on the day of the release of the last TAW patch. But at some point, they probably switched repositories, databases or development environments and chances are high that this specific snapshot got lost or corrupted.
  3. Few companies keep the source code around for long. Raymond Chen from the Microsoft Compatibility Team has written several articles about maintaining old games, and he repeatedly stated that companies have deleted or lost their sources after spans as short as one year after release (the reason is 1., e.g. here and here).
  4. Even fewer companies keep the source code after being bought by or merged with other companies. The repository (see 2.) typically goes to some hard drive in some storage room, the current version (Typhoon?) is migrated to the new company's system, and as people leave, the knowledge on where to find old snapshots and how to use them is lost.

What we should expect from TAW source code:

  1. It's written in Assembler, C and C++ for an old version (1997?) of the Watcom C++ compiler. C and C++ have evolved a lot since these days, and it very likely does not compile until we fix thousands of bogus constructs. (In my old companies, upgrading to new compiler versions would be deferred until after the next release, just because of the countless subtle problems arising. Imagine skipping 20 versions ahead then.) Assembler code may need to be rewritten from scratch (there's many different syntax flavors).
  2. External dependencies (Glide API, Direct3D 5?) do not exist any more. In general, such dependencies are not included in snapshots (after all, you could get them from Microsofts SDK CDs that every developer had on their desk). In this case, we'd have to do a full port to the latest supported API (cut Glide, use Direct3D 9?) before the source code even compiles!
  3. Having the source code does not mean understanding it. The people that wrote it are long gone. There may not be any comments. One EF2000 developer gave hints into that direction (“pants-down code”, “project from hell”). If we wanted to change things, we'd probably have to rewrite them anyway.

My guess:

  • Even if we got the complete source code in best condition now and we managed to set up a Windows 95 virtual machine with the exact Watcom C++ package they used, I'd have to work some months full-time to migrate it to a state where we can compile and start the game on a current Windows machine. By 2017, we may be able to do the first fixes or changes.
  • If we got the source code in a condition that is typical for most companies, it would be 2017 before the game even compiles and we'd all give up in frustration.

There's a reason nobody created a "better Half-Life 2" even though the source code has leaked. I'd rather continue building our own flight simulator on top of EF2000's and TAW's assets.

Link to comment
Share on other sites

Related: Fabien Sanglard writes articles about old game source codes. I love this site. Seems like he's now working on Oculus Rift support for 1993's Strike Commander.


Most people assume the source codes and gold versions of all finished games were stored in a Vault somewhere at EA. But after getting in touch with people at Wing Commander CIC, it appeared that all the source code was lost when the company closed.


On his first day one developer managed to delete the full 900MB of Strike Commander source tree. The IT team spent 72 hours recovering everything from developers machines. The interview also mention that Wing Commander 1 and 2 code was exchanged via floppy disks: They did not have a network until SC !


Link to comment
Share on other sites

I'm not directly involved in TAW 2.0 development (I'm mostly analyzing data and assisted as a modding tool developer), so I can't comment a lot on this.

There's two sides of TAW: the assets (data: textures, models, HUD layout, …) and the binaries (code: A.I., gameplay, flight model, graphics, …). My personal opinion is:

  • Data modding is close to limit. There may be a handful of features left to realize (DKD's extended worlds, or his more detailed terrain tiles, being good examples of what's still possible) but there's not much waiting for us. Ask the experts on this.
  • Code modding is very difficult (and it would still be difficult with the source code). Reversing the whole thing is no option. Small patches (like the 1024x768 patch, TrackIR support, etc.) require specialists and lots of spare time; big patches (making other planes playable properly) are close to impossible.

My personal plan is:

  1. Go for a rewrite. TFXplorer proved a TAW clone by pure analysis being possible.
  2. Base the clone on TFXplorer. It may take years — but TAW 2.0 took several years, too, and here we are 18 years after release of TAW, so that's not as bad as it sounds.
  3. Keep everything compatible with TAW's data. Once TFXplorer becomes a suitable replacement of TAW's engine, the TAW 2.0 team's improvements will run fine with it and nothing will be lost. (E.g. it's already compatible to DKD's large world experiments, minds some of your user settings, works even with an EF2000 version, …)
  4. Until then, assist the 2.0 team with any problems that might arise.

… and my aim is that we have a playable TAW clone with the most basic gameplay by 2020.

You may not understand all of my euphoria for TFXplorer because the current state is not available for the public. I can promise, though, that we have

  • a stable F-22 simulation like TAW's, plus some improvements
  • basic physics (destroyable buildings, cannons) implemented
  • the ability to play other vehicles besides the F-22, even tanks
  • enhanced graphics (e.g. see this thread)
  • modding abilities only limited by our time.

That's my opinion on my own project. But for what happens to TAW 2.0 until then, you'll have to wait for a reply from the Air Dominance team. I do hope that development is not over.


Link to comment
Share on other sites


  • Create New...