Results 1 to 24 of 24

Thread: Solving the Loadgame AI Bug - A suggested approach

Hybrid View

Previous Post Previous Post   Next Post Next Post
  1. #1
    EBII Council Senior Member Kull's Avatar
    Join Date
    Jan 2003
    Location
    El Paso, TX
    Posts
    13,502

    Default Re: Solving the Loadgame AI Bug - A suggested approach

    Quote Originally Posted by Sundjata Keita
    Am I missing something here?
    No, it looks like I'm missing something here! Before announcing a "workable solution", you'd think I'd actually test it out first, right? But nooooooo. I mean what could possibly go wrong?

    Well, I just finished setting up dual boot on my system, and there's a problem. In a normal dual-boot installation, you eventually get a menu that allows for the selection of the desired O/S. And this works fine during a normal start-up. However, after hibernating and restarting, the boot sequence DOES NOT GIVE YOU THIS OPTION! You can force things using the F8 key, but that simply generates a screen with two choices - continue with restart from hibernation, or delete the hibernation data and go to the O/S selection menu!! Damn!! That is soooo frustrating!!

    Hopefully there's some way to work around this, but for now I'm back to square one. GRRRRR!!!! (Note: I'll need to make some edits to the previous "good news" post - no sense getting people's hopes up)

    And many thanks to Sundjata Keita for calling me out on this!
    Last edited by Kull; 04-03-2005 at 18:06.
    "Numidia Delenda Est!"

  2. #2
    EBII Council Senior Member Kull's Avatar
    Join Date
    Jan 2003
    Location
    El Paso, TX
    Posts
    13,502

    Default Re: Solving the Loadgame AI Bug - A suggested approach

    While burrowing through all the commentary in the Application Hibernation thread (the link was posted earlier), I came upon a few gems that you might enjoy:

    1) "I guess (being in games) I always hated games that saved games by writing out all of the allocated memory, and restored them by reading it and fixing up the pointers (often missing a few of the pointers in the process). Serialize everything, and you've solved both saving and reproducing any kind of state you need for your application."

    Hmmm, I wonder if this is how we get the "Loadgame AI" bug?

    2) "I want this feature. Always have. [snip] I also have Rise Of Nations open and minimised and it's been minimised for 5 days. [snip] I want RON and the IE windows to go away from my taskbar for a while. And to survive a reboot (as 5 days ago I rebooted with another RON minimised)."

    And here's the absolutely classic reply:

    "I thought most games had a "save" feature anyway?"

    I will now die laughing!!!
    "Numidia Delenda Est!"

  3. #3

    Default Re: Solving the Loadgame AI Bug - A suggested approach

    I don't think the dual boot option is a very good idea either, since it means you have to stop everything else you're doing in order to play.

    VMWare might be the best option, if it has improved any since I last used it a long time ago. Unfortunately back then RAM certainly wasn't the only limitation. It rarely worked for 3d apps, and when it worked it was awfully slow.

    I will give it a shot though.

  4. #4
    EBII Council Senior Member Kull's Avatar
    Join Date
    Jan 2003
    Location
    El Paso, TX
    Posts
    13,502

    Default Re: Solving the Loadgame AI Bug - A suggested approach

    Quote Originally Posted by bouis
    I don't think the dual boot option is a very good idea either, since it means you have to stop everything else you're doing in order to play.
    I'm not particulary thrilled with it either, but the goal is to develop workable solutions to the bug. Since we can't get into the source code and CA is unwilling to address it, we're left with these sledgehammer approaches. But any choice is better than no choice. Speaking of which.....

    VMWare might be the best option, if it has improved any since I last used it a long time ago. Unfortunately back then RAM certainly wasn't the only limitation. It rarely worked for 3d apps, and when it worked it was awfully slow.

    I will give it a shot though.
    Many thanks for taking this on, and I look forward to hearing the results of your tests! Good luck!
    "Numidia Delenda Est!"

  5. #5
    EBII Council Senior Member Kull's Avatar
    Join Date
    Jan 2003
    Location
    El Paso, TX
    Posts
    13,502

    Default Re: Solving the Loadgame AI Bug - A suggested approach

    I've been experimenting with Boot Loader programs, and they do allow for a successful dual-boot hibernate strategy. I'm always leery of recommending anything like this, since these programs usually install to the Master Boot Record, and since every system is different, what works for one person could totally hose someone else.

    However. There is a freeware program called Gag45 that can be installed ONLY a floppy, and which successfully enables the Dual-boot Hibernate strategy. It can also be installed on your hard-disk, but that is strictly optional (and as noted above, "user beware"). If anyone is interested, I can provide more details on a safe dual boot-hibernate strategy using Gag45.

    http://gag.sourceforge.net/download.html
    "Numidia Delenda Est!"

  6. #6

    Default Re: Solving the Loadgame AI Bug - A suggested approach

    Quote Originally Posted by Kull
    Well, I just finished setting up dual boot on my system, and there's a problem. In a normal dual-boot installation, you eventually get a menu that allows for the selection of the desired O/S. And this works fine during a normal start-up. However, after hibernating and restarting, the boot sequence DOES NOT GIVE YOU THIS OPTION! You can force things using the F8 key, but that simply generates a screen with two choices - continue with restart from hibernation, or delete the hibernation data and go to the O/S selection menu!! Damn!! That is soooo frustrating!!
    Kull, if your "another" system is Linux, than you get an boot-time option to start either
    a) Linux (w/o R:TW obviously)
    b) Windows (w. the option of either de-hibernating your R:TW or deleting the hib. data by means of F8).

    This is tested and working. What I am fighting for is to overwrite the hibernation data with the previously captured one. I realize it is risky as outlined in the post #9 in this thread. But this seems to be the only way to get a kind of the working savegame file, so to say, given the denying approach of the game developers.

    I daresay the Linux bootloader is more forgiving than the M$ one!
    "Only when the human spirit is allowed to invent and create, only when individuals are given a personal stake in deciding economic policies and benefitting from their success -- only then can societies remain economically alive, dynamic, progressive, and free. Trust the people."
    Ronald Reagan

  7. #7

    Default Re: Solving the Loadgame AI Bug - A suggested approach

    sems to me you need to make a emulator,
    And save state,

    Just a option,...
    Programers may be able to look at the way The old snes emulators save state, Then use one of the emulators that linux users use to play windows games,
    And edit them to save state in a similar way,

    Seems phesable to me,
    But dont ask me to try and do it lol:)

    I can let you have a Snes emulator,
    But if you want a game for it Your going to haft to already Own the game b4 i can send you the rom,
    Used to be i was allowed to tell you to delete it after a day or 2 But now you haft to already Own the game.

    Just giving you a nother avenue to explore

  8. #8

    Default Re: Solving the Loadgame AI Bug - A suggested approach

    kull: what's the state of this now? Do you have a working solution with the Dual-boot Hibernate strategy with the Gag45 thing?

  9. #9
    EBII Council Senior Member Kull's Avatar
    Join Date
    Jan 2003
    Location
    El Paso, TX
    Posts
    13,502

    Default Re: Solving the Loadgame AI Bug - A suggested approach

    Quote Originally Posted by harlekin65
    kull: what's the state of this now? Do you have a working solution with the Dual-boot Hibernate strategy with the Gag45 thing?
    In short, "Yes". Let me start with the requirements:

    1) Either a second hard drive or a partition, each with a functioning Operating System. In my case, it's two hard drives, both loaded with Windows XP (SP2).

    2) Download GAG45 (link posted above) and use it to create a bootable floppy. Although you can load GAG45 into the MBR (Master Boot Record) of your hard drive, I don't recommend it. Instead, follow the directions and load it exclusively on the floppy, and make sure you identify each O/S with a different name (so you don't forget which WinXP is the one with RTW running)

    3) Now boot into the secondary O/S using the GAG45 floppy, and load RTW along with any desired patches or mods. To minimize the chance of errors, you probably shouldn't use this O/S for anything other than RTW. Again, you're going to be playing a game without saving, possibly for months, so don't add complexity.

    4) Start up your game, play as long as you like, and then minimize it to the desktop. To go into hibernate, click the "Start" button and select "Turn off computer". The shut down menu will appear - press the "Shift" key and you'll see the yellow "Stand-By" button change to "Hibernate". Click this and your computer will shut down.

    5) To start up your primary O/S, just boot normally. To get into the secondary O/S (with RTW), boot with the GAG45 floppy and then select either O/S, and the system will go through the "restore from hibernate" start-up sequence. When your desktop appears, voila, there's RTW just as you left it!


    Issues/Problems:

    Of course, there HAVE to be problems, right? Probably not a complete list, but here they are:

    1) Sounds disappear. As in, you now play the whole game with no sounds! Maybe the load-AI bug ain't so bad after all? Anyway, others have reported this, and I'm not sure if it's directly related to hibernate or just a consequence of tabbing down to the desktop. If anyone has a fix, please post!

    2) After rebooting into your primary O/S, Windows can tell that something's been going on and insists on running CheckDisk. It won't find a problem, and it can be pretty timeconsuming with big hard drives, so bypass the test immediately. Of course if you aren't paying attention (like me 80% of the time), you won't notice until it's running the test! If anyone knows a safe way to cancel CheckDisk, please post!

    3) It seems odd that the GAG45 floppy ignores your O/S selection and just automatically restores from Hibernate (see item 5 above) - and thus removing the floppy is the only way to boot normally into the primary O/S. So this MAY be a unique feature to my system, which means your results might differ.

    4) Not all systems come with 3.5 Floppy Drives these days, so I'm not sure what you would do if your system is "floppy-less". Ideas anyone?

    5) Be very careful about adding thing to your RTW disk/partition from the primary O/S. When the system reboots from hibernate it expects that things will look a certain way, and something as simple as a different icon on the desktop might cause a problem. I haven't tested this to see what causes crashes and what doesn't, but keep it in mind.

    Hopefully others will test this out on their systems, and perhaps we can fine-tune and improve the process, but for now here is a tested, albeit "less-than-perfect", solution to the LoadGame/AI bug.
    Last edited by Kull; 04-18-2005 at 06:43.
    "Numidia Delenda Est!"

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Single Sign On provided by vBSSO