Future Tutorial Ideas ...

[Urre]

It's not like it's a particularly complicated model to embed. I bet it generates it on demand, even.

[MeTcHsteekle]

hey baker, in the centerprint logging and anti-wallhack tutorials i keep getting errors that say i need to define stuff... where do these definitions go? i tried he top of the files where the errors arise but it doesn't seem to work :S

perhaps a tutorial that teaches you of definitions and stuff like that will help

aright her some real suggestions

1] fixing the alt-tab thing where the game gets pushed to the bottom of the screen and u can only see the vary top, the rest of the screen is all balck

2] fixing the washed out kinda look, like make it so it can be darker or something /hopes you understand what he means w/o images

3] adding .loc file support and the %h %a %w/e stuff..thing..

4] stainmaps

[Baker]

hey baker, in the centerprint logging and anti-wallhack tutorials i keep getting errors that say i need to define stuff... where do these definitions go? i tried he top of the files where the errors arise but it doesn't seem to work :S

perhaps a tutorial that teaches you of definitions and stuff like that will help

aright her some real suggestions

1] fixing the alt-tab thing where the game gets pushed to the bottom of the screen and u can only see the vary top, the rest of the screen is all balck

Here is that one:

This keeps the window from drifting off center, a flaw in the Nvidia OpenGL drivers.

if (fActive) {
if (modestate == MS_FULLDIB) {
IN_ActivateMouse ();
IN_HideMouse ();
if (vid_canalttab && vid_wassuspended) {
vid_wassuspended = false;
ChangeDisplaySettings (&gdevmode, CDS_FULLSCREEN);
ShowWindow(mainwindow, SW_SHOWNORMAL);

// Baker 3.80x - Fix for alt-tab bug in NVidia drivers from JoeQuake
MoveWindow (mainwindow, 0, 0, gdevmode.dmPelsWidth, gdevmode.dmPelsHeight, false);
}
} else if ((modestate == MS_WINDOWED) && _windowed_mouse.value && key_dest == key_game) {
IN_ActivateMouse ();
IN_HideMouse ();
}
}

[/quote]

The centerprint logging tutorial is "bad" (doesn't work for someone new as-is) and I won't have time to revisit it anytime soon.

Loc support can mostly be added by adding location.c and location.h from ProQuake to a project and finding the few instances in the source code where functions get called like on map load and I'm thinking client.c (doesn't sound entirely right to me ... cl_parse?) where the macro gets translated.

I'd like to be around more and at some time I will be again, but for now I won't be.

[Spike]

wonder how hard it would be to implement basic csqc support...

Mneh, someone choose.
stock, FitzQuake, or ProQuake?

Or would some other engine be a more practical base for such a tutorial (aka: wall of copy+paste instructions)?

[MeTcHsteekle]

wonder how hard it would be to implement basic csqc support...

Mneh, someone choose.
stock, FitzQuake, or ProQuake?

Or would some other engine be a more practical base for such a tutorial (aka: wall of copy+paste instructions)?

dymn, that's like asking if you want a rocket, a helicopter, or a fire truck :S

there's only one logical course of action: http://legorobotcomics.com/?id=23


//leaning more towards stock or proquake :S :S, but fitz quake..."damn, we're in a tight spot"



.... but i guess if yo do stock it might be able to be ported to the others ... maybe ....

...stainmaps

[leileilol]

(aka: wall of copy+paste instructions based on others work but with the tutorial authors name plastered over it like a savior)

fixed

[Spirit]

stock, FitzQuake, or ProQuake?

Fitzquake please! It's got so many good things, fixes and useful features in comparison to the other two. Also it is the defacto standard for any new maps released. [insertnongayhugsmiliehere]

[MeTcHsteekle]

stock, FitzQuake, or ProQuake?

Fitzquake please! It's got so many good things, fixes and useful features in comparison to the other two. Also it is the defacto standard for any new maps released. [insertnongayhugsmiliehere]

man thats true too, i cant decide :S im no help

[metlslime]

if it's just a tutorial, why no base it on standard quake? Seems like that would be applicable to the largest number of engine developers, since most people base their engines on stock quake.

[Spike]

if it's just a tutorial, why no base it on standard baking methodologies? Seems like that would be applicable to the largest number of bakers, since most people base their baked goods on regular flour.

The problem is that ClientSide BakeC is such a large adaptation to baking that a genuinely informative tutorial would take so much describing that the time spent fixing the errors introduced by the follower skipping bits greatly outweighs the time spent actually following the tutorial. Thus the follower must have reasonable baking experience.
Basing the tutorial on regular flour is thus less important than it would normally be.

If CSBakeC is to ever be widely usable, it has to be in the flours that are widely used.
Thus by basing such a tutorial on an existing, used, flour, I get two birds with one stone.

I did not list joequake or qrack as I know both these engines have extensive changes. I am unsure how much Fitzquake has really been modified.
For QW, fuhquake is a tempting base, but I doubt anything would come of it that wouldn't come from using proquake as a base, while the qw/mods of it would hinder non-qw adoption.

I just don't want to spend my time working on something that only a handful of people will ever taste.

The 'tutorial' would likely be in the form of modified version of the base flour with extensive precompiler usage inside the modified files. A follower can then read the additional comments in the modified code sections to understand why it is needed.
Then it can be used either as a base for other tutorials, or as a guideline for what needs changing in others.
The advantage of a complete solution is simple: csbakec changes are likely more than most 'small' custom engines require.

If I did one and someone else took that and did the other, we'd have the best of both worlds (the precompiler allowing it to be disabled helps integration into the respective official trees).

[Sajt]

Don't do it Spike! Go outside and smell the respective official trees! There is still a life for you to live!

Or maybe you could make a tutorial about adding Duke3D map support to a Quake engine.

[MauveBib]

Or Pacman map support.

[r00k]

I'm still unsure about where I stand on CSQC. While its a cool feature technically, and can be used to make a nifty HUD, how much of a deterrent is there from say creating a Quad timer, or an aimbot or other hacks? Or is CSQC limited from those possibilities?

How about a tutorial to add client-side movement prediction to netQuake
{Client-side prediction is the use of dead reckoning with application to disguising the lag time inherent to online multiplayer games.}

[Spike]

How about a tutorial to add client-side movement prediction to netQuake ;)

CSQC has that capability.

The reasoning behind cheat protection with csqc is that its no easier to add a cheat to the engine than it is to add a cheat to the csqc.
Sure, if you know a little QC, you can set up a timer to time quads.
But on the other hand, if you know a little C you can set up a timer to time quads...
But if you want to add a cheat in CSQC, first you have to get an engine that'll load any CSQC mod, rather than only ones with the exact same hash as reported by the server. Which requires C.
The original CSQC spec has no provision for finding out the location of entities other than ones the server explicitly tells it about. So no aim bots. No wall hacks either.
A quad timer is easier to implement using a really long wav file.

You could make a wallhack by setting bit 8192 of other player's effects field thanks to a DP extension. But you could do that more easily in C than with CSQC, as you would first have to get the engine willing to load your modded csprogs. And yeah, that takes C.

But yeah, the point I'm trying to make here is that its not really any easier, it just has a smaller code base.

[r00k]

So basically CSQC opens possibilities to the enduser without having to mess with the actual engine port, while still allowing the same level of security that any other client has based on the authenticated info from the server itself. So, any client capable of load a csqc progs.dat could have clientside movement prediction without any change to the server's netcode. Nifty!