ZEQ2 Lite

Wednesday, June 09, 2010

Shane wrote :

AnTycrisT wrote :

jaguar wrote : EVIDENCE :

DOWNLOAD IT, PUT IT IN THE DEMO FOLDER, THEN ENTER THE GAME AND WRITE :

/DEMO DEMO0000

what the heck? then this is your evidence? hahahahaha!!!!!!!!!!!

You're a total cheater to so you can't talk.

cheater? why?... show me...

Wednesday, June 09, 2010

I went on a server once, and your name was Jason or something of that, and you shot me once and-death.
Everyone was complaining how you hacked, or maybe I'm mistaken and it was an addon?

Wednesday, June 09, 2010

thaha - well why didn't I hack all aspects so that MY character was better in everything - but not like that...by far....I mean I just increase it all by little

You accused me only for healing - its lacking I would use only that - or now you are going to accuse me that I use everything else too? - and as I've seen you have healing rate (heals automatically)...

Jag with that video you only proved that you know how to hack the game; not that I hacked

I don't know if you noticed but people are starting to laugh at you...you are looking more and more idiotic with those intellectually lacking proofs you are tossing into this thread

Spare us all your stupidity

As I said before - lets fight and truth will be seen

Wednesday, June 09, 2010

Good Luck Buksna.
Don't lose.

Who on earth is editing posts around here?

Wednesday, June 09, 2010

okay I am going to speak up me and dokumas were both watching this match. He was healing weird fast. and me and him were a little confused just making a point. and I no the difference

Wednesday, June 09, 2010

The arguement has been settled. They will battle about it. No futher instigation is needed.

Wednesday, June 09, 2010

Buksna wrote : thaha - well why didn't I hack all aspects so that MY character was better in everything - but not like that...by far....I mean I just increase it all by little

You accused me only for healing - its lacking I would use only that - or now you are going to accuse me that I use everything else too? - and as I've seen you have healing rate (heals automatically)...

Jag with that video you only proved that you know how to hack the game; not that I hacked

I don't know if you noticed but people are starting to laugh at you...you are looking more and more idiotic with those intellectually lacking proofs you are tossing into this thread

Spare us all your stupidity

As I said before - lets fight and truth will be seen

People are laughing because most of them don`t know what`s it all about, they just come and say that someone is better or no.

That was no addon, it was the same Goku as dokumas was using, I edited his skills and you can edit them how you want, I just used random values.

Calling me an leaping lizard won`t change the fact that you are just hiding behind your attacks, yes you did the very same thing to change that values.

If we fight and I win that doesn`t save the thing that you used hacks, if you win the same.

Anyone knows that, stop playing the silly card because that`s silly, told you.

Any rational guy with a bit of intelligence can understand that Buksna was cheating.

Wednesday, June 09, 2010

Shane wrote : I went on a server once, and your name was Jason or something of that, and you shot me once and-death.
Everyone was complaining how you hacked, or maybe I'm mistaken and it was an addon?

what? hahahaha, I want evidence, I have a conscience quiet, so I do not care....
you have evidence?...
or do you base only in words?...

Wednesday, June 09, 2010

Jaguar. Please, please let it slide.
Buksna didn't want to lose so he either cheated or "cheated".
Just fight him. Enough talk. I have faith that you will both stick too the rules.

Record the battle.

@AnTycrisT

I didn't mention any addons. I don't have any evidence. I didn't mean to accuse you, sorry if I came off as if I did. Apologies set aside, I have one question.

Everyone was complaining how you hacked, or maybe I'm mistaken and it was an addon?

Was I mistaken?

Wednesday, June 09, 2010

I don`t care about the battle, this is not about the battle but about hacking.

This is the subject that we talk about.

Wednesday, June 09, 2010

So Buksna hacked, flat out.
So what?
He must've been messing around.

Wednesday, June 09, 2010

Shane wrote : Jaguar. Please, please let it slide.
Buksna didn't want to lose so he either cheated or "cheated".
Just fight him. Enough talk. I have faith that you will both stick too the rules.

Record the battle.

@AnTycrisT

I didn't mention any addons. I don't have any evidence. I didn't mean to accuse you, sorry if I came off as if I did. Apologies set aside, I have one question.

Everyone was complaining how you hacked, or maybe I'm mistaken and it was an addon?

Was I mistaken?

okay, no problem Shane,I did not use any of that

Wednesday, June 09, 2010

antychrist does not cheat its just his character has weird stats or something

Wednesday, June 09, 2010

Dokumas wrote : antychrist does not cheat its just his character has weird stats or something

no, nothing, is like any other ...

Wednesday, June 09, 2010

Still about this? Who honest to lloyd cares?

Wednesday, June 09, 2010

I'm closing this topic. It's getting ridiculous.

I won't delete it though, cause it has valuable information about the game which is explained by Void . But I won't allow a flamewar going on in here anymore. Unless Brad says to me that I have to re open it (cause Brad interprets some discussions on a other way then me and might find it valuable to keep it open for a reason), this is closed.

Thursday, June 10, 2010

I'm going to allow this be re-opened under one condition :
No one is allowed to discuss whether player A cheated against player B whatsoever. General statements/comments only!

By opening this again, I would like to clarify several points and allow more relative development/handling discussion to be brought up as long as it can be done so without instance-oriented specifics being brought into play.

Well technically it is cheating. It's similar to using bright skins, wall hacking or aim bots.

Consider for example replacing all model files with a shiny glowing red box exactly measuring up to the bounding box of quake 3 with a big green arrow above the top. That'd make aiming a thousand times easier for yourself while other people would have a lot of trouble firing at krillin for example who's a smaller target than his own bounding box.

Bright skins or modified mesh files that are scaled are used to locate distant players and thus really only relative to the "discovery" process. Since ZEQ2-lite characters are already maintaining outlines at distances and have fairly contrasting textures, a modification would hardly assist this. Additionally, an "aim bot" has zero impact in a game where hard lockons are part of the core mechanics.

Aiming has very little to do with gameflow at the moment whereas it does in a first person shooter.

It's easy to prevent though, unfortunately for some reason ZEQ2 Lite isn't using pk3 files and pure servers which were specifically introduced to Quake 3 because of this (plus the added bonus of making addon characters easier to distribute by just copying / deleting pk3 files from a directory).

We chose not to go with pk3 files for several reasons :

• It's more difficult to manage through subversion. Being as they are packaged up, managing multiple revisions becomes cumbersome.
• It's inefficient to edit when archived as compared to just raw files. Just as there is an issue with Subversion, editing mesh/art/script assets that are packaged up are just impractical in an active development environment.
• At some point in time, we had several issues with getting pure mode to run server end with our varying setups. This likely had to with all of us (as developers) having different assets/configs/characters that we were simultaneously working on but had not yet committed due to incompleteness.
• We're only at the first public beta release. This was intended to be more of a sandbox "get your feet wet" kind of environment and not an actively used, actively modified setup. Again planning and release conflicts caused part of the mess, but honestly we didn't expect a swarm of players in such a mediocre and lacking release.

As it stands right now every programmer with a tiny bit of knowledge can modify the cgame.qvm to code a bot that automatically dodges and fights too.

Sure anyone can modify the cgame.qvm to try and accomplish exploits, but there's honestly not that much that can be modified to give any kind of real advantage. I explained already that any visual enhancements to help with location/aiming are fruitless with the mechanics that are in place. Since the actual playerState data is ONLY passed to cgame for the active player, there's not really any kind of statistical pool you can reach from. There are a few centity sync-ups to make sure the right client-side effects are displaying properly, but very little to no data that could be used maliciously.

On an unpure server (which is every ZEQ2 Lite server unfortunately) it is possible to create a "child" character "gokuu" inheriting all of "Goku"'s data unless overwritten by new data in the "gokuu" folder. This works by Quake 3 automatically searching for a similar name when "gokuu" isn't found, since "Goku" has 4 of 5 letters in it it's a hit.

I'm not familiar with the like-string search process you described being whatsoever implemented into ZEQ2-lite (at least). I'm not saying it doesn't exist, but I know it not to be actively true for a lot of the config-based properties as this is not how I built them to behave.

The tierDefault.cfg file is first loaded up to set basic statistics regardless if an overriding tier.cfg files exist per-tier. There are no default phys and grfx used per-character thus if they do not exist, the character will not have any skills present.

I think Jaguar was actually just using "gokuu" as random arbitrary example rather than saying it was specifically that name, thus it could have applied to any variation whatsoever -- meaning it would be treated as an addon as far as the server as concerned. Yes, this is exploitable, but ONLY if the server has modified files. You should always trust the server you connect to before playing anyhow. I'm fairly certain that even if you do not have a model and they do, you CAN still change to that model (as long as you know its name). The client-side loading of effects will fail, but server-side your attacks, statistics, and their functionality will still remain.

I'm not fully certain, but there SHOULD be a console command to find out information on a certain player or the models available on the server. I know the server can use dumpuser <user> for this, but I recall their being a client-side equivalent as well.

The best solution would be to distribute official releases with pk3s while keeping individual files for SVN/developers/contributors.

This may be an ample future solution for releases, but overall the goal (as mentioned) for this initial release was ease of modification/management over security of mechanics.

Additionally it doesn't seem like ZEQ2 Lite is intended for really balanced competetive gameplay AS OF RIGHT NOW anyway.

This is completely true. This release of ZEQ2-lite was FULLY intended to be tinkered with and mucked about for a small amount of time, NOT played competitively. We hadn't even BEGAN to go through the process of balancing the mechanics in any way whatsoever. The idea was merely to provide a rough framework and technical demo of where we stood so others could get their feet wet and take a stab at development too.

That'd pretty much require a full time game designer making up tons of formulas for each character, attribute and attack and requires the weapon physics system to impose hard limits on attacks. (Example: The first attack can have 10 attribute points, so a total damage of 8 for example and a firing rate of 2, while the ultimate attack can have 50 points and thus a lot higher damage etc)

I had considered point-based allotment to core skill statistics (which is one of the best ways to ensure pure balance in any environment), but again this has not yet been explored in the early early form of the build.

To put it into an analogy, we gave everyone a broken hover-car along with its blueprints so they could improve transportation in general. What's the first thing they do? Pop on some wheels and rev it up for street racing. Come on guys!

Thursday, June 10, 2010

I agree with almost everything you said. However you're underestimating the power of client-side modifications. Granted an aim bot wasn't the best example, but I felt it'd be the only one which accurately describes the level of modification that is possible without getting to technical.

I can give you another two examples though:
- Wall hack, I won't go into details how to do it for obvious reasons but since it's so slam easy here's a proof of concept (pretty sure you and Dave know how I did it anyway)
- Automatic dodging. You're underestimating the amount of data available to the client, everything that needs to be drawn needs to be known and is therefore available including beams and players, normally this is limited to the current PVS, but in ZEQ2 Lite there's always only one PVS to begin with (that's also why wall hacking works so well here while Quake 3 itself is largely unaffected) so the whole map can be detected :p

Thursday, June 10, 2010

- Wall hack, I won't go into details how to do it for obvious reasons but since it's so slam easy here's a proof of concept (pretty sure you and Dave know how I did it anyway)

Right, there are plenty of ways to easily modify it so you can see through objects/walls, but my point was that this offers NO real benefit in a game where you have hard-locks and more relative to your example, a sensing area to see where players are.

Being able to easily find and look at players isn't a significant gain in the case of our present mechanics (even if you consider the comical hide and seek antics).

Automatic dodging. You're underestimating the amount of data available to the client, everything that needs to be drawn needs to be known and is therefore available including beams and players, normally this is limited to the current PVS, but in ZEQ2 Lite there's always only one PVS to begin with (that's also why wall hacking works so well here while Quake 3 itself is largely unaffected) so the whole map can be detected :p

I'm going to assume that PVS is ... player visual state (gosh, I dislike acronyms). Sure, you could build some simple code that analyzes the player centity states and attempts to make movement efforts based on that data, but there are several reasons why this also isn't a significant hack and simply would not function with the current gameplay.

Since attack information is not available (only the visual state), the code won't know how to properly respond. Some attacks are better to be outright zanzoken'd, some blocked, others fired back at, and yet others ignored entirely (using the explosion shell/smoke as coverage). Simply dodging an attack requires knowledge about your opponent in the form of variables that are NOT exposed client-side and moreover LEARNED by a skilled player.

Trying to just zanzoken every attack you can will drain your energy too quickly. Trying to just sidestep will cause you to not move quickly enough a lot of the time. Firing back all the time could leave you out of energy or underpowered. Taking the attack all the time is a good way to leave you dead.

Whether it's optimal to fight, flee, or use an advanced tactic depends on properties of the attack that the cgame simply does not have access to.

Thursday, June 10, 2010

Potentially Visible Set

Small benefits do help more than you think. It's much more accurate and way easier and faster to visualize than a sensing area. Not to mention that it'll reveal any attack being charged or power up.

No matter how you put it, it's an unfair advantage that's limited to individual people and as such cheating. Even if you don't consider it a big deal the game is no longer fair at the point someone is using that.



Also how is attack information not available?

You know:
- Your opponent's position
- Your opponent's approximate power level
- If your opponent's boosting
- Your opponent's animation
- The type (Kamehameha, Spirit Bomb)
- The time it started charging
- The time it'll be ready
- The time it'll travel

You could even track your opponent's fatigue if you wanted. Sure it's not 100% accurate but it's still better than what a player himself would estimate when he has to make the very same decisions the bot has to make (it's AI after all).

Domi, jaguar and Buksna are pretty much "bots" in this regard too. In most cases their strategies will follow patterns which can be translated to code. Even if everything you said was right (it isn't ) the bot would have an advantage over them since he can see and hear everything on the map regardless of distance or angle and has an instant reaction whereas a human has a delay of x seconds for video and why seconds for audio. Aren't the Quake 3 bots on the highest difficulty setting enough proof? They don't work with more than cgame'esque information either.


I'll take one of your arguments to proof that it's even more powerful though: Side-stepping.

Your point was that it's often not possible to dodge simply by that.

My point is the bot will know when this is the case.

cgame knows exactly when the attack started charging, whom it's aiming at, when it will be ready, when it is being fired and how fast it is travelling (also if it's being boosted). It can make a very good guess about the power level of that person too.

It also knows it's own power level, fatigue and speed.

As such it's an easy mathematical function to calculate when the attack will impact and if that's enough time to escape the radius by just outrunning it or if a zanzoken would do the job. It can also do a pretty good estimate (much more accurate than a human anyway) on how effective blocking or struggling would be.


Oh and if that's not enough for you let's get back to the idea of an aim bot:

The second a player respawns it fires an attack at the location where EV_SPAWN occured. Since you yourself think side-stepping isn't a viable strategy, how many people exactly do a zanzoken right the second cgame starts responding to user commands? Cause that'd pretty much be the result of what you said yourself.


The limit of the ability of a bot is what is available via cgame, that's right. But that's much more than what is available to a human. No player will remember every attack used, every second someone powered up or rested during a game. A bot will and takes a single frame to make an approximation of the current situation and reacts at the same time.

Thursday, June 10, 2010

I was on the server yesterday , (NONADON) EVERY ONE HAD 32K HIGHEST and I had 2.6 millionk with normal Goku. and I had 32k with other models , what the heck was that

*stunned*

Thursday, June 10, 2010

void wrote : Potentially Visible Set

Small benefits do help more than you think. It's much more accurate and way easier and faster to visualize than a sensing area. Not to mention that it'll reveal any attack being charged or power up.

No matter how you put it, it's an unfair advantage that's limited to individual people and as such cheating. Even if you don't consider it a big deal the game is no longer fair at the point someone is using that.



Also how is attack information not available?

You know:
- Your opponent's position
- Your opponent's approximate power level
- If your opponent's boosting
- Your opponent's animation
- The type (Kamehameha, Spirit Bomb)
- The time it started charging
- The time it'll be ready
- The time it'll travel

You could even track your opponent's fatigue if you wanted. Sure it's not 100% accurate but it's still better than what a player himself would estimate when he has to make the very same decisions the bot has to make (it's AI after all).

Domi, jaguar and Buksna are pretty much "bots" in this regard too. In most cases their strategies will follow patterns which can be translated to code. Even if everything you said was right (it isn't ) the bot would have an advantage over them since he can see and hear everything on the map regardless of distance or angle and has an instant reaction whereas a human has a delay of x seconds for video and why seconds for audio. Aren't the Quake 3 bots on the highest difficulty setting enough proof? They don't work with more than cgame'esque information either.


I'll take one of your arguments to proof that it's even more powerful though: Side-stepping.

Your point was that it's often not possible to dodge simply by that.

My point is the bot will know when this is the case.

cgame knows exactly when the attack started charging, whom it's aiming at, when it will be ready, when it is being fired and how fast it is travelling (also if it's being boosted). It can make a very good guess about the power level of that person too.

It also knows it's own power level, fatigue and speed.

As such it's an easy mathematical function to calculate when the attack will impact and if that's enough time to escape the radius by just outrunning it or if a zanzoken would do the job. It can also do a pretty good estimate (much more accurate than a human anyway) on how effective blocking or struggling would be.


Oh and if that's not enough for you let's get back to the idea of an aim bot:

The second a player respawns it fires an attack at the location where EV_SPAWN occured. Since you yourself think side-stepping isn't a viable strategy, how many people exactly do a zanzoken right the second cgame starts responding to user commands? Cause that'd pretty much be the result of what you said yourself.


The limit of the ability of a bot is what is available via cgame, that's right. But that's much more than what is available to a human. No player will remember every attack used, every second someone powered up or rested during a game. A bot will and takes a single frame to make an approximation of the current situation and reacts at the same time.

So are you implying that You could code my pattern of playing? I don't always use the same strategies..? Sorry I don't follow you entirely :p

Thursday, June 10, 2010

He doesn't mean you use the same strategies, just that they are in a pattern.
( Kiblast-Kiblast-Jump, Kiblast-Kiblast-Jump)

Thursday, June 10, 2010

Your strategy doesn't always need to be the same.
Think of it as playing chess. Not every game is the same but there's always good moves and bad moves and a bot can be taught to (even randomly) only choose from the good ones.

That's also how the difficulty settings in games work. On the hardest setting the decision making will always choose one of the most effective strategies while on easy the reaction might be completely random.

Your strategy on what to do when an opponent uses attack A will likely involve a few options from which you choose and you'll know another few which are just plain lacking and a bot does exactly the same.

This doesn't mean the bot would fight completely like you do but in one specific situation it'd do the same.



EDIT: Shane is somewhat right. Something like that might be a good tactic in some or all situations. A bot can identify such an scenario and then switch to using that.

Thursday, June 10, 2010

oh my freaking lloyd

I'm out of here
too much text

jaguar handle yourself *laughing out loud*