Jump to content
Search In
  • More options...
Find results that contain...
Find results in...
Janizdreg

Compet-n Resurrected

Recommended Posts

Archy said:

I'm just defending the principles of COMPET-N, and one of those principles has been that Vanilla Doom is used and nothing else.

This is what I meant by assuming the old rules were an infallable dogma. Why did Compet-N demand this in the past? Fairly clearly in order to establish a well-known set of game rules for consistent comparisons. (If someone beats Doom 2 in GZDoom faster than Compet-N's record, it's clearly not a valid comparison since GZDoom modifies the game rules quite a lot.)

PrBoom+ and Chocolate Doom have gone to great lengths to preserve the original rules of the game, to where demo files are compatible between vanilla and them.

Share this post


Link to post
GhostlyDeath said:

Most of them have already been discovered really by people looking at the source release and reverse engineering of the executable itself.

Yes I'm aware of that but since bugs happen rarely, they are rarely tested and often don't produce the same result as they do in Vanilla Doom in each given situation. For a long time PrBoom-Plus didn't properly reproduce one of the bugs, I forgot all of the details but if I come across that post again, I'll link it here.

------------------------Different Topic------------------------

Archy said:

I'm just defending the principles of COMPET-N, and one of those principles has been that Vanilla Doom is used and nothing else.

chungy said:

This is what I meant by assuming the old rules were an infallable dogma. Why did Compet-N demand this in the past? Fairly clearly in order to establish a well-known set of game rules for consistent comparisons. (If someone beats Doom 2 in GZDoom faster than Compet-N's record, it's clearly not a valid comparison since GZDoom modifies the game rules quite a lot.)

PrBoom+ and Chocolate Doom have gone to great lengths to preserve the original rules of the game, to where demo files are compatible between vanilla and them.

If this is the case, when why are DOOM95 demos not acceptable for Ultimate Doom entries? The demo files are compatible between Vanilla Doom (size 715493 bytes) and DOOM95 when it comes to Ultimate Doom.

Share this post


Link to post
DeathevokatioN said:

This is awesome news!

Lol, hearing such a cheerful statement in what is probably the ugliest thread I've ever participated in.

Share this post


Link to post
Archy said:

Yes I'm aware of that but since bugs happen rarely, they are rarely tested and often don't produce the same result as they do in Vanilla Doom in each given situation. For a long time PrBoom-Plus didn't properly reproduce one of the bugs, I forgot all of the details but if I come across that post again, I'll link it here.

Just FYI, some years ago I tested every single compet-n demo for playback issues in prb+, and the few that were found were rigorously investigated. So the "rarely tested" claim doesn't really stand up.

And also FYI, the ugliness in this thread is your own doing. You are posting rather arrogantly and repetitively, while showing a lack of knowledge of many of the issues on which you are trying to make definitive statements. More than anything else, you are managing to sour the relaunch of compet-n, which has no reason to be an especially contentious issue.

I must say that I am a little baffled by the pugnacious feel to this discussion. For about seven years, Compet-n incoming had existed within the Doomworld forums quite happily, and with a reasonable level of activity (growing, I think, but I haven't checked the numbers), either awaiting a possible relaunch, or evolving into a new form of compet-n, where people post here and get their entries archived at DSDA. A new Compet-n site should really be seen as a new stage in this evolution, and if actively maintained, rather a welcome one (and if it flops, then the fallback is to revert to the way it has been for the last seven years). Why it has been pounced upon with the expectation that there should be a huge change in direction at this point, given that the ongoing level of interest hasn't been at all bad, puzzles me.

One question for fx: Did AdamH actually give this venture his blessing? I think that is significant, as it affects how "official" the new site should be seen as.

Share this post


Link to post
Grazza said:

Just FYI, some years ago I tested every single compet-n demo for playback issues in prb+, and the few that were found were rigorously investigated. So the "rarely tested" claim doesn't really stand up.


I didn't think of all the demo playback, just the playing. Sorry for the false statement.

Share this post


Link to post
Archy said:

Note that I'm only talking about bugs that happen internally within Doom, not bugs that can happen from the Operating Environment interacting with the exe.

You can't prove a negative: the idea that there might be a difference hiding in there somewhere is an unfalsifiable concept. But if such a difference does still exist, it's a bug to be fixed.

And if it means anything, I've already regression tested every single demo in the Compet-N archive (all 9,526 of them, over 27 days of gameplay in total) against Vanilla Doom. Granted the output from statdump.exe is a pretty basic measure but there is no detected discrepancy between the Vanilla and Chocolate behavior.

Share this post


Link to post
fraggle said:

You can't prove a negative: the idea that there might be a difference hiding in there somewhere is an unfalsifiable concept. But if such a difference does still exist, it's a bug to be fixed.

And if it means anything, I've already regression tested every single demo in the Compet-N archive (all 9,526 of them, over 27 days of gameplay in total) against Vanilla Doom. Granted the output from statdump.exe is a pretty basic measure but there is no detected discrepancy between the Vanilla and Chocolate behavior.


Of course the LMPs (demos) would play back correctly, LMPs just assign movement and action for each tic. While an LMP working in Vanilla Doom and then desynching in Chocolate-Doom would indicate a bug, or lack of a bug in some cases (though not all cases), that does not mean Chocolate-Doom perfectly imitates all of Doom's bugs.

Example: In Vanilla Doom without -turbo, if SR or SL are >40, then TR and TL cannot be present.
In PRBoom-Plus without -turbo, if SR and SL are >40, then TR and TL can be present.
A PRBoom-Plus demo with SR50 & TR21 occurring on the same tic could still be successfully played back in Vanilla Doom despite being impossible in Vanilla Doom.
So, what if there's a bug - that causes certain tic combinations during certain times or events - in Vanilla Doom that isn't in Chocolate Doom. The demo would still successfully play back on Chocolate Doom.

I'm 99% sure that there's more impossibilities than just this:

The player cannot have GF>50 and GB>50 tic.
The player cannot have SL>50 or SR>50 tic.
The player cannot have SL>40 or SR>40 tic if turning (TL>0 or TR>0).
, they're just more complicated then walking and turning speeds.

You are correct though, these are all unfalsifiable concepts.

Share this post


Link to post
Archy said:

In PRBoom-Plus without -turbo, if SR and SL are >40, then TR and TL can be present.

Without always strafe-50 in the same tic? Cool.

Share this post


Link to post
TimeOfDeath said:

Archy - how do players play vanilla in Win95/98? Isn't the emulated DOS environment actually part of the operating system?

IIRC - the DOS environment in Win95/98 isn't emulated, what you have is a graphical shell running on top of a multitasking version of MS-DOS.

Share this post


Link to post
Looper said:

Without always strafe-50 in the same tic? Cool.

I'm not quite sure what you're asking but in order for TR and TL can be present (in PRBoom-Plus) while SR and SL are >40, movement_strafe50 must be turned on.

GreyGhost said:

IIRC - the DOS environment in Win95/98 isn't emulated, what you have is a graphical shell running on top of a multitasking version of MS-DOS.

Windows 95/98/ME launched from real DOS (though some people would argue that MS-DOS 7.x [the DOS version used with Windows 9x 4.x] isn't "real DOS" because it does not function the same way MS-DOS ≤6.22 functions - but that's a whole different discussion) and then shut off MS-DOS. The command.com used in the Windows 9x 4.x environment is not MS-DOS, (as the system is no longer running on DOS - but a 32-bit file system manager) but an emulated DOS environment. This is why many DOS programs require you to restart the computer in MS-DOS mode ("real" DOS).

El Heggunte said:
Windows 1.0 thru Windows 3.11 ran on top of MS-DOS.
Starting with version 4.0, Windows used MS-DOS as a sort of boot loader.
command.com was able to run in both the 16-bit mode and the 32-bit mode.
Up to Windows 3.11, the so-called "DOS prompt" really was a DOS prompt.
In Windows 95, however, the CLI already was a 32-bit environment,
which "contained" a 16-bit layer where (most) real-DOS applications
could be executed without a problem.

The Old New Thing said:
This [the 32-bit file system manager] was paradise. The 32-bit file system manager was able to do all the work without having to deal with pesky drivers that did bizarro things. Note the extra step of updating the state variables inside MS-DOS. Even though we extracted the state variables from MS-DOS during the boot process, we keep those state variables in sync because drivers and programs frequently "knew" how those state variables worked and bypassed the operating system and accessed them directly. Therefore, the file system manager had to maintain the charade that MS-DOS was running the show (even though it wasn't) so that those drivers and programs saw what they wanted.

Note also that those state variables were per-VM. (I.e., each MS-DOS "box" you opened got its own copy of those state variables.) After all, each MS-DOS box had its idea of what the current directory was, what was in the file tables, that sort of thing. This was all an act, however, because the real list of open files was kept in by the 32-bit file system manager. It had to be, because disk caches had to be kept coherent, and file sharing need to be enforced globally. If one MS-DOS box opened a file for exclusive access, then an attempt by a program running in another MS-DOS box to open the file should fail with a sharing violation.

Windows developer Raymond Chen said:

MS-DOS was just an extremely elaborate decoy. Any 16-bit drivers and programs would patch or hook what they thought was the real MS-DOS, but which was in reality just a decoy. If the 32-bit file system manager detected that somebody bought the decoy, it told the decoy to quack.


If you want better explanations, talk to myk.

Share this post


Link to post

I must admit, I never quite understood the veneration of doom2.exe, even back in the old days when I debated it with Heygi and others.

I don't think it's really a rational thing. It comes from the strong 'old school' bias of some (most?) players in the demo scene. Their feeling that the original executables and their gameplay possess some ineffable purity that source ports cannot, by their very nature, reproduce. It's like people who argue about music on CD vs vinyl. It doesn't matter how much you debate the facts, because the objection derives from emotion, not reason.

Cheating is an irrelevance regarding the rules in this matter. Girlich and Winterfeldt were able to cheat, using speed hacks and probably demo stitching, long before the source was ever available. And any rule regarding use of source ports will have no effect on cheaters, as they're willing to break the rules by definition. Internet-based verification might help, but short of running the entire game on the server and streaming the framebuffer to a dumb client, I don't see you'll ever make cheating impossible.

Frankly though, I don't think trying to resurrect Compet-n is particularly good idea at all. Simon Widlake is dead, most of the old players are long gone, and the competition model it represented is a relic of the past. It should be cherished as a part of the community's history, but I don't think trying to bring it back is the right way forward.

I'd much rather see a new competition. One that accepted uploads of any demo, which users would tag to indicate what source port they recorded it with and whether they used tool assistance (yes, people could lie, but they could, and did, do that with Compet-n as well). All this would feed into a single central database, which could provide a customised views such that, if you wanted to, you could disregard all non-vanilla demos, and see what the records tables just for doom2.exe.

You would have the best of both worlds. An open competition, but with the ability for those concerned with port purity to have their own internal competition.

Share this post


Link to post
Jonathan said:

I'd much rather see a new competition. One that accepted uploads of any demo, which users would tag to indicate what source port they recorded it with and whether they used tool assistance (yes, people could lie, but they could, and did, do that with Compet-n as well). All this would feed into a single central database, which could provide a customised views such that, if you wanted to, you could disregard all non-vanilla demos, and see what the records tables just for doom2.exe.

You would have the best of both worlds. An open competition, but with the ability for those concerned with port purity to have their own internal competition.

Might I say DSDA?

Share this post


Link to post
Archy said:

Might I say DSDA?

DSDA is along the right lines, but its design, and the ability to search and filter leave a lot to be desired, in my opinion. It also seems to be missing a lot of the older Compet-n demos.

Share this post


Link to post

DSDA is good place for C-N demos, and some of the demos are missing because Andy is busy uploading them as far as I know.

I don't think Compet-n can be as glorious as it was 10 years ago. That's why more modern competition would be cool with vanilla compatibility and source ports allowed to attract new players. Limiting the amount of wads would make the competition more intense. Also, no more headache with dosbox or/and 320*240 resolution, unless you want to.

If you ask me, dosbox isn't good enough for emulating dos because the quick start is f*ed up. You lose too many tics which is a catastrophe in UV-speed runs, and I'm not going to buy some shitty old computer just for doom, I mean just for C-N tag @ DSDA.

When I saw the topic "Compet-n Resurrected", I thought someone has finally figured the issue with cheating, and maybe modernized the competition, but no, it was just a wall of words, nothing else. I hope you guys don't put too much time on doom.com.hr, unless you change it somehow. Right now it has pretty much no purpose.

fx02 said:

With PrBoom+ I can beat in a day for example every Sedlo's record. That's ridiculous.

Go run PL11-048 pacifist then :p (without auto strafe50, of course)

Share this post


Link to post

TimeOfDeath said:
Isn't DOSbox an emulator - a replica of DOS? Isn't "DOSbox to DOS" the same as "ChocolateDoom to doom2.exe"?

You use Chocolate because it lets you set up key bindings easily that are normally not available in vanilla and let you place crucial weapons close to your movement keys, that you got used to using advanced ports. If they were the same, you wouldn't even be complaining because it wouldn't matter to you.

Jonathan said:
I'd much rather see a new competition. One that accepted uploads of any demo, which users would tag to indicate what source port they recorded it with and whether they used tool assistance (yes, people could lie, but they could, and did, do that with Compet-n as well). All this would feed into a single central database, which could provide a customised views such that, if you wanted to, you could disregard all non-vanilla demos, and see what the records tables just for doom2.exe.

It already exists "unofficially", as people have created threads on IWADs and PWADs already in Compet~n but to compete with any (vanilla compatible) engine.

Saying "lets change Compet~n to allow other vanilla compatible engines" only means that there would be no more vanilla-only competition, or some would just ignore the initiative and there would be two Compet~n competitions at the same time, with the same name. For that reason, it's better to give the new competition, which would include old and arguably also current "real Compet~n" demos, its own name. Maybe Compet~n+?

fraggle said:
Sorry if I sound slightly pissed off in the comments I'm making here but the fact is that I've spent several years now crafting what, if you love Vanilla Doom so much, ought to be your favorite source port.

I love vanilla but my favorite source ports are the online engines (ZDaemon by historical usage, but Odamex is replacing it.) DOSBox and Chocolate are two alternatives in modern systems to vanilla and DOS, but you can't just assume that Chocolate is a better option than DOSBox for everybody. Without considering the Compet~n debate, not everyone needs the benefits Chocolate is adding, some want to use their DOS knowledge or habits, and DOSBox has DOS batch file support and allows hacks like Doom+ or others, and supports old (DOS) versions of the game.

Share this post


Link to post

Apparently DOSbox lets you use custom keys for weapons too. I haven't tried DOSbox, but for me choco was easy to set up (except I don't understand the mouse sensitivity options).

Share this post


Link to post

Yeah, but that's why I said "easily" and "not normally". One needs to alter the underlying keyboard mapping (which is an option for any port or program) to allow it, which is less practical and may interfere with things, such as typing commands in the command line. I don't recall Compet~n runners using "port-like" key settings yet I've noticed more than one person now that takes advantage of the Chocolate feature.

Share this post


Link to post

Compet-N players type in the command line for each of their thousands of attempts? ;)

I don't know everybody's key configs, but anyone who uses DOSbox for CN has the opportunity to use a custom config for weapons.

Share this post


Link to post
TimeOfDeath said:

Compet-N players type in the command line for each of their thousands of attempts? ;)


With DOSKey, all they have to press is the "Up arrow key" and the "enter key."

See this thread to get quick shortcuts for demos (example: to record a UV -RESPAWN demo of Requiem map 21, with the shortcut, type in "doom2 <s @rr 21" - a command much much easier than "doom2 -file REQUIEM.wad REQMUS.wad req21fix.wad -warp 21 -skill 4 -respawn -maxdemo 999 -record RR" which you'd normally have to type in without the shortcut.

TimeOfDeath said:

I don't know everybody's key configs, but anyone who uses DOSbox for CN has the opportunity to use a custom config for weapons.

So do those who use Windows 95...

Share this post


Link to post

Try it. I swapped only the Q with the "console key" for some time to avoid quitting by accident and it was already a hassle, and I was lucky enough to have a keyboard with removable keys. Every time I picked up another computer and I'd try to write a word with Q or try to open a game console I'd type the wrong character out of habit. It's annoying to stick with or enable every time you play, and an obscure "feature" based on the operating system (and DOSBox acts as one) that may require programs or hacks. In Chocolate Doom, on the other hand, it's integrated into the engine, accessible though the setup program, has no side effects on other stuff, and to use it on any system you need to do little more than import your CFG files.

In vanilla it's awkward and it's unclear if it's a cheat or "tools assistance", in Chocolate it's readily accessible and part of the game settings.

Share this post


Link to post
myk said:

Try it. I swapped only the Q with the "console key" for some time to avoid quitting by accident and it was already a hassle, and I was lucky enough to have a keyboard with removable keys. Every time I picked up another computer and I'd try to write a word with Q or try to open a game console I'd type the wrong character out of habit. It's annoying to stick with or enable every time you play, and an obscure "feature" based on the operating system (and DOSBox acts as one) that may require programs or hacks. In Chocolate Doom, on the other hand, it's integrated into the engine, accessible though the setup program, has no side effects on other stuff, and to use it on any system you need to do little more than import your CFG files.

In vanilla it's awkward and it's unclear if it's a cheat or "tools assistance", in Chocolate it's readily accessible and part of the game settings.


so what?

Will that make a person that wants to be accepted as a compet-n player cheat? I don't get it.

Half of this is trying to "prevent" a problem that, in all honesty, never existed in the first place.

This is all smoke and mirrors. I dont even know why this got to 5 pages.

Chocolate Doom is as vanilla as it gets. in fact, it passed every compet-n demo. that means, every shot, frag, imp fireball, passed with flying colors.

I don't understand it. It sounds like excuses to me. All of it.

Share this post


Link to post
myk said:

In vanilla it's awkward and it's unclear if it's a cheat or "tools assistance", in Chocolate it's readily accessible and part of the game settings.


This is getting off topic but I personally don't think using external programs to rebind to keys is cheating (as long as not one single binary digit of Vanilla Doom is changed) and I think we should encourage it so players that simply can't/refuse to adapt like TimeOfDeath could have a shot at C-N. I use DOSBox's key-mapper feature and zdkeymap on my Windows 98 machine to swap the Q key with a useless key.

And TimeOfDeath, there's a lot more reasons I (and I'm sure myk) feel Source Ports should not be allowed here at COMPET-N than just key binding differences.

I'm going to try avoid repeating anything me or myk has said but there really is no substitute for Vanilla Doom. I was playing Chocolate Doom just earlier and could notice an instant difference in the sound. I was playing on Windows 98 so the difference wasn't from the way DOSBox emulates sound cards.

And really (and yes I'm reaping my self now), why are we trying to change COMPET-N into something that already exists? Non Vanilla Doomers still compete with Vanilla Doomers on DSDA and vice versa.

Allowing Source Ports at COMPET-N would just create two DSDAs, except one's called COMPET-N and only allows entries on 10 different WADs - and the other's called DSDA and excepts entries on all WADs including the 10 COMPET-N WADs. The COMPET-N you (and apparently many others) dream of already exists, just under a different name.

I think we should use my Yellow Record/Purple Record system. It shows what the "real" COMPET-N record is but is also shows the faster entries that are not COMPET-N valid. Trust me, many (virtually all) of the C-N players would treat those Purple Records just as seriously as C-N records, I know I do. I didn't submit my E3M3-217 which beats Radek Pecka's E3M3-218 because Kyle McAwesome had a PrBoom+ E3M3-216. I didn't submit any entry till Kyle was beaten, which he was when my E3M3-212 came along.

PS. I do think using an external program to bind multiply keys to one key is cheating, because if done properly, a single key stroke could perform all the functions needed to complete the level in a record time. A less extreme example is binding the Forward Key and the Strafe Right key together, thus creating a Strafe-Run 40 key, which is faster than having to take the time out to press both the Forward key and the Strafe Right key simultaneously.

Share this post


Link to post

Dosbox does have direct impact on how doom plays for me - at starts majority of attempts I usually end up getting twisted in random direction. Kimo already explained why but it's mumbojumbo for me as I'm technically impaired, although I understood that it's a failure on dosbox behalf. That's more than enough of a difference between 'real vanilla' and 'dosbox vanilla'. Sometimes it can result in a whole second or two lost were I to keep all attempts (which is partly why I went for lenghty stuff with dosbox in the first place - shorter demos are entirely impossible for me like this).

I'd most likely vote for choco being treated as 100% valid for c-n if dosbox doom is accepted.

Share this post


Link to post

Archy said:
This is getting off topic but I personally don't think using external programs to rebind to keys is cheating (as long as not one single binary digit of Vanilla Doom is changed)

Whether the executable is edited is not the issue, the effect is. You could edit it in a harmless way and that would not matter, and you can also use many external TSRs or apps that may be considered TAS or cheating.

I think we should encourage it so players that simply can't/refuse to adapt like TimeOfDeath could have a shot at C-N.

Why torture TimeOfDeath when he could use Chocolate with less trouble? He's still be recording pretty much the same demos, with maybe a small performance loss due to occasional issues when typing in the command line with a messed up keyboard layout :p

If his (Chocolate or whatever) demos went to the Compet~n+ tables, you could still check them if you want to beat them from Compet~n.

Csonicgo said:
Will that make a person that wants to be accepted as a compet-n player cheat? I don't get it.

They could always cheat. It just shows how technology changes the game and using Chocolate starts to encourage different settings and behavior than using vanilla, changing standards. With Chocolate, you get more people using custom weapon key binds, which overall would improve average record times a little bit, and it would encourage people to rebind their weapon keys for the benefit, making vanilla somewhat obsolete.

Chocolate Doom is as vanilla as it gets. in fact, it passed every compet-n demo. that means, every shot, frag, imp fireball, passed with flying colors.

Another comment (by fraggle) tried this argument and I had forgotten to reply to it, so thanks: So does PrBoom+, but the demo data or format can't be used as a basis. I mean, everything that is TAS also uses it. The written data and interpreters don't reflect input and playing behavior differences. That's not enough.

It sounds like excuses to me.

To me the excuse is to say "lets add X engine to Compet~n"... to destroy vanilla-only competition. That may sound harsh, but why alter the rules of something that has a history and supporters, when you can make a new branch without harming anyone's freedom of choice, and when you can include Compet~n records in Compet~n+ tables? (But not vice versa.) However competitive, this is just a hobby and a game, anyway, so even if we assume that what the Vanilla School argues is irrational or dogmatic, it's still a valid option.

j4rio said:
Dosbox does have direct impact on how doom plays for me - at starts majority of attempts I usually end up getting twisted in random direction. Kimo already explained why but it's mumbojumbo for me as I'm technically impaired, although I understood that it's a failure on dosbox behalf. That's more than enough of a difference between 'real vanilla' and 'dosbox vanilla'. Sometimes it can result in a whole second or two lost were I to keep all attempts (which is partly why I went for lenghty stuff with dosbox in the first place - shorter demos are entirely impossible for me like this).

It is a difference, but so are different mice, keyboards, monitors, sound output devices, whether you use Windows 9x or real DOS and other such things. The Compet~n rules say that only the DOS executables noted may be used, which don't exclude DOSBox which they run on, but do exclude other engines like Chocolate Doom. The rules are arbitrary, defined by historical circumstances, but that is what makes them concrete.

Share this post


Link to post
j4rio said:

Dosbox does have direct impact on how doom plays for me - at starts majority of attempts I usually end up getting twisted in random direction. Kimo already explained why but it's mumbojumbo for me as I'm technically impaired, although I understood that it's a failure on dosbox behalf. That's more than enough of a difference between 'real vanilla' and 'dosbox vanilla'. Sometimes it can result in a whole second or two lost were I to keep all attempts (which is partly why I went for lenghty stuff with dosbox in the first place - shorter demos are entirely impossible for me like this).


I wonder if someone could make a DOS program that forces the screen resolution to become 320x200 regardless of what program in launched after that. The issue you're having is caused by the resolution change from 640x400 to 320x200. I suggested that we just use a 320x200 command line interpreter, but Kimo said that the Doom start up screen would still end up launching in 640x400 - have not tested this or not.

If I remember correctly, this is an error with the SDL libraries, which means once Fraggle fixes this issue with Chocolate-Doom (Chocolate-Doom has the same problem as DOSBox), he might know how to fix it in DOSBox since they both use SDL libraries. Now I'm not tech savvy so I may have just said something completely stupid, and I apologize if that is the case.

PS. If you move and click the mouse at the start up screen, the negative affect you are dealing with will be reduced.

Share this post


Link to post

here's what happens, plays back with pretty much anything.

I've just touched the mouse and the twitch happens. Not always, but usually. It can really disorient and when you consciously realise where to turn, it's usually a second or two gone so speeds or shorter maxes are a definite no-no. It also happens either sooner or later as I press different buttons depending on what and in which order I press them. If I strafe, it usually cancels itself, but the fact is I don't generally need to, so I end up with awkward movement I'm doing myself instead of dosbox doing it for me.

Share this post


Link to post

I remember that happening to me all the time with Win98, never happened in DosBox though.

Share this post


Link to post
×