Page 1 of 1

To mod or not to mod

PostPosted: 25.04.2020, 04:02
by NEb0
Hello. I'm a little upset about the situation around my client modification.
Looks like, people divided into two sides. Somebody thinks that using it is a real cheat.

Firstly, I want to explain what the modification does and what it doesn't.

The features:
  • unlocked most of the client-depended PRO-features (replays, positions, vector vision, theme changes, and some others). I agree that this is a controversial thing, but I'll discuss it later;
  • some performance optimizations: background caching, tower target calculations, added some indexes and so on. Totally, now it's more or less playable on such maps like CASTLEGROUNDS or RUN on bounty hunter mode when creep count reaches 1000 or more;
  • money and income calculations for replay viewing and spectator mode;
  • greatly improved calculations for tower spots (hotkey: T. works in vector vision similar to the original implementation). It's like in the original client (with pure vector vision), but with a more correct formula of calculations;
  • tower disabling function (hotkey: 8). Despite that i found trivial implementation of it in the protocol, unfortunately, at the original version of the client this event doesn't processed for a some reason (maybe causing desyncs). Because of the algorithm of lives counting, it hard to found an abuse to the original client users. So, if anybody will use it versus an original client, he will just spoil his internal state. So, that feature is only useful at 1p surviviour games;
  • charset bugfixes. Because of improper letter-case processing and mixing utf-8/cp1252 charsets, there were no ability to play maps like
      Code:
    ausreißer
    . A more serious problem, related to that bug: some people, for example, with Turkish locale, cannot play any map with
      Code:
    I
    character in the map's name. Both kind of these bugs are fixed;
  • display improvements: tower priorities, more contrast creep themes integration.
  • configurable creep sending delay (maybe, this is the only feature that is a cheat? really, it's pretty noticable in the stats after the game)

Not modified (and NEVER will be modified):
  • any (state-destuctive) in-game calculations, event handling or any other vanilla in-game behaviour. [spoiler] So, there are only modifications, related to optimization. The engine behaviour is exactly the same, and it tested a lot with a lot of replay files (replay file is just an event queue, so the same calculations are made in-game and in the replay mode) [/spolier]

Known bugs/shortcomings:
  • less visual effects, simplified button rendering
  • some context menus do not display the title
  • some bugs with maps displaying in lobby
  • relatively high memory consumption (this could be optimized)

Motivation
  • I couldn't normally play on my old PC because of game lags becuase of high CPU consumption.
  • Not only me encountered this problem.
  • I wanted to study how to play this game better, learn and understand some basic mechanics. And I needed a user-friendly interface.
  • The game development was almost abadoned, and a lot of people were unhappy with it.
  • I really hate when the browser window forcefully opening after each game (sometimes i get out of memory because of it).
  • I want to make the game more interesting, fun and just more comfortable to play.

More about funs and not funs.
IMHO, it's fun:
  • to know, which creeps are on the board and have a clear and well-distinguishable sprites even for the people who has not the perfect vision, on all maps
  • to have an ability to play the game on pretty old PCs, on modern PCs, on laptops without GPU or on linux, BSD and so on
  • to have somebody, who can handle at least some of the games' bugs
And it's not fun:
  • when the users (in particular, of my modification) called as cheaters by the people who never used it
  • you have no possibility to change the colors/themes, becuase you have to buy unbuyable PRO version
  • you just cannot play 1/3 of all the maps because your desktop's locale is Turkish

I began to share the modification, when a lot of the people were said that "no body cares", "there are a lot of bugs", or just had ideas to improve something.
Now I hear the point "it's impossible to play because of the modified client".

About the enabling features from PRO mode.
The my main argument here is: the PRO is unavailable to buy for a couple of years. But i agree, it's a controversial thing.
In any case, all of them could be blocked (or even totally cutted-off), keeping most of the advantages of the modification.
Also, if there will be some official update, my version won't work in any case. So it's very simple to forbid it.
I didn't remove the logic of the PRO-checking, only the exact features were unlocked (the check
  Code:
CommonClient.isPro()
is still returning false)
So, this point is discussable.
Also, probably, the users of PRO mode cannot use my modification because i had no ability to test it well (I remember that there were some bugs, but I couldn't collect enough feedback to fix it).
If most of the opponents of my opinion are dissatisfied with this, with your help, i'll try to fix the bugs.

Here are the separate files of my themes (which can be absolutely legally used in ANY vanilla PRO client):
highcontrast.zip
(1.02 MiB) Downloaded 255 times

highcontrast2.zip
(1.02 MiB) Downloaded 256 times

hc3.zip
(9.5 KiB) Downloaded 266 times


In conclusion.
I'd want to dot the i. Is my modification a cheat or not? Does it make the game more or less fun?
If it is a cheat, that what exactly features should be disabled? I want to find a compromise, to keep everybody happy and so that everyone enjoys the game.
Maybe it's difficult to install, maybe it needs a "normal" installer like vanilla client?
Or maybe it's too unfair to see the creeps on the board? Maybe it's more fun to send demeters on MONOBLUE because not everybody can see it? If that is really fun for somebody, maybe it will be interesting to make a special mod with poorly-visible (or totally invisible) creeps?
Unfortunately, I couldn't make any decisions of the "cheatability" before I shared it to some people. Now, there are a number of people who tested it. But some of the people are unhappy with it. So, if there are a lot of people who really stopped playing because of it, I'd want to know. Maybe it would be better for community, to totally disable it?
Or maybe you have another ideas which can improve the client?
I want to prevent any kind of holy wars between "cheaters" and "too conservative people". Please, be constructive and argue your position.

P.S. I've deleted the original post, the files from my server, and temporary (or permanently?) made the source repository private, for the worst case (really, i don't think that there were a lot of people who successfully launched it without any instructions in any case^^), until I can't decide what to do with it. I really don't want to harm anybody's gameplay experience.

Thanks for the attention and I'm sorry in advance for my English (especially, technical, maybe somebody will help with translation to a human-readable variant?).
Have a nice day :)

Re: To mod or not to mod

PostPosted: 25.04.2020, 17:53
by SoulMan
Tested Your mod, its nice and I think it should be adjusted because it helps a bit too much.

Good things that doesnt need any change:
1) Pure vector vision
2) Position choice
3) recording and viewing replays
4) Seeing money and income of every player in spectator/view replay mode
5) Custom creep themes

Good things that need to be adjusted:
6) Tower strategies display - view of opponents tower strategies used shouldn't be possible, it may help too much, and its pretty hard to recognize it without this

Things that should be removed:
7) setting send delay - it should always remain as in original client

Things I would like to see:
8) custom creep theme, where stronger creeps are in front of weaker, while they are close to each other (for example shark can't hide under group of raiders etc.)
9) custom creep theme using allowed in original custom theme feature - different incoming sound for every creep (16 sounds instead of 4)
10) customizable resolution (battle area of 1080x1080 or 900x900 size would be nice to see, so I can use more than 25% of my screen size and see more details).

Regards.

Re: To mod or not to mod

PostPosted: 26.04.2020, 10:49
by NEb0
SoulMan wrote:Good things that doesnt need any change


Really, most of the things you listed are in the PRO version of the client :D
But, as i said, as i improved performance, most of them are changed a lot from original. From the visual side, one of the changes: in original PRO in vector vision there is a background:
vectorvision-bg-old.jpg
vectorvision-bg-old.jpg (26.84 KiB) Viewed 8062 times

which is rotated (like a radar) that looks pretty annoying. I replaced it with a static color background. I think, i can make a setting for this color (to make everyone's eyes happy, everyone can chose the more neutral color for himself:)).
Also, when i changed the code, I found a way, how to render everyone's board with the same rotation (so, for example, it could look like everyone are on the first position). Probably, it's not what we're expected, but maybe more comfortably for somebody. That can be enabled via the setting too.
Also, it's possible to make a thing which allows to switch vector vision and normal view ingame "on-the-fly". Now it's only possible to switch between games. Would it be interesting for somebody?
One more idea for rendering improvement is to make 1-2 pixel border between players (sometimes, I'm confusing my towers with opponent's ones at the late game). Maybe it sounds as an easy thing, but really not so easy to implement it (a lot of things are in global coordinates). Especially, because there are at least 3 modes: for 16x16 maps, 32x16 and 32x32, where the border should be placed in a different way. But i can try to do it...

SoulMan wrote:view of opponents tower strategies used shouldn't be possible, it may help too much, and its pretty hard to recognize it without this

Yeah, okay. I can hide it in-game, and leave unchanged for spectator and replay view mode.
I've mostly done it for learning the strategies, when i started to play and didn't understand why priorities are changed so often. Now, the most profitable part of this feature is more accurate priority control of YOUR towers, especially at endgame and near-endgame. Probably, I'd agree that it sometimes makes it possible to abuse the opponents' defence. But really, for some years of playing, I can remember only a few games where it gave some real profit:) mostly, it works when the main def near the beginning of the map, or maybe on short maps (which, i think, many people know, I really don't like at all and often lose with or without viewing priorities or even the opponents' income).

SoulMan wrote:setting send delay - it should always remain as in original client

Yeah, i implemented that feature relatively not long ago. Mostly for testing. Agree, that it's a bad idea to give a chance to send more than 10k creeps.
I agree, that the normal limit of about 100-120 actions (or maybe 150) per round should not be avoided. That's why in any case, I tried to convince every "tester" of the mod to not send more than 100 creeps (or even 80. "spend 40-45k for mantas, or 600-700 for toucans at max") per round. If that limit was reached, other users can know about, at least, after the game on the game statistics page. Just see at the "money for creeps sent" and, more notably, "actions performed" rows (if u divide one number to another, you probably get the number near "500"; that means that the number of actions == mantas sent). Note: If you still cannot view the statistics or think, that it's too hard to get these numbers for a round too difficult, just use browser extension for the stats: viewtopic.php?f=3&t=3012
I think, if anybody will use it too often, that will be bad for his reputation. As the most of the "testers" (i believe) are adequate people, they should understand that the statistics lives for a long time. So one can find the player's games list and see, how often he used the abuse. But I agree, in the final version of the mod that feature should be disabled, because...
I've learned a very important thing from this feature! Really, almost every game could be won, with no sending delay. It's easy: keep your internet connection good, get to about 500k of money and send 10k of mercuries. All the players will get lagged out, because of the not very performant code and can't handle some frames with such number of the creeps. The strat even can be optimized: wait when the cannons(J-s) at the start will be sold. Rockets have a bigger delay. At least, the "global desync" is guaranteed.
For now, it's impossible just for two reasons: because you should render these creeps too (and that's why you will be lagged out also), and because of not very performant networking code. But one can improve the networking code, and not render "small" creeps at all on higher rounds...
I don't think, that this "strategy" can cause DoS (denial of service) on the server-side, because in my experience, the server handling is much easier than even the sending on the client side (as i see at status page, the server uses some epoll wrapper -- it's more than enough to handle 100k+ messages from clients in a second). But for other ingame players, that could make a lot of problems.
Yes, as for me, or my mod -- I can force the delay, and it's okay. But here is no guarantee, that anybody else will decompile the client, change the code, build it, and started to abuse this "strategy". So, it's a some kind of security problem, in my opinion. That's why I was thinking how to avoid, at least, rendering problems and got to the idea that the bunches of creeps could be "clusterized" (so, we render only one sprite instead of 1000 in the same pixel). For a some reason, when I got to this idea, I could not come up with an exact algorithm how to do it. But now, I think, I've an idea for implementation, which can even help with the situation of "multiple creeps in the same position" (when you see only one creep, but actually there are 2 or more of them) in the normal game.
Btw, the solution is partly tied to the rendering
SoulMan wrote:where stronger creeps are in front of weaker
So, maybe I'll solve the both problems at once (although, such kind of rendering is much more simple to implement, so I shall solve it firstly).
SoulMan wrote:different incoming sound for every creep

Is surely possible. Really, it depends on exact sound files. I'm just too bad with sound making, but I can easily pack them to a theme file.
Also, I've got an abstract idea to make a simple theme maker for those, who are bad with ini files. Really, you can just unpack the standard theme files(it's just a zip file), look into their format and make a theme by yourself with the help of any text editor + your images/sounds -- it's pretty simple. But if it is too difficult, maybe it would be useful to make a gui program (or maybe a plugin for inkscape? gimp? i don't know) to make a theme from a number of media files.
For now, I just have no exact idea what exactly to do to help with theme editing.

SoulMan wrote:customizable resolution

Oh, I think, some kind of that idea I already saw on the forum (viewtopic.php?f=3&t=2167). dcode is right, all the images are in raster formats, much of the logic of rendering is tied to pixels.
The only thing I see, I could implement without re-making the all things, is to add global scale to all the window. That should be not so easy, but is possible... Maybe I even can render some of the widgets (sidebar, lobby and so) for higher resolution, not using background images (or re-scale them)... It's pretty hard, but it's possible. But the main widget ("game view") is very hard to scale; the maps' backgrounds, creeps, towers have the exact resolution.
If you don't understand the difference between raster and vector graphics, just imagine a 1x1 pixel image. I can copy that pixel 4 times, get the 2x2 square and so, it will be "2x2 image". You can even make an "NxN" image with that pixel, but you don't get any new details. Probably, with good algorithms of re-scaling, 800x600 image could be rescaled into 1024x768 or even more. But it's the same: you won't get any new "details", it's just an approximation.
So, the one solution is to rescale everything: probably, that would be not so difficult to implement. The another solution: rescale creeps, backgounds, and other raster images, but re-draw sidebars, lobby, player lists and so on. For some reason, they are tied to absolute coordinates too (really, now most of the graphics frameworks contains layouts for making the GUI scalable, but they didn't used). So, the task is to rewrite almost all the gui code. I've already tried to do that, when i tried to reduce cpu consumption at least, in the lobby screen. But I got some bugs in very unexcpected places and it was too hard to debug all of them.
Even harder is to deal with the game view: again, it could be rendered as now, but with a some scale coefficient (transforms allow do it), but I think it will be looked not very well.
Probably, sometimes I'll get some "inspiration" and try to make it again, but i'm not sure. It's just a lot of work which really isn't visible (at least, for me. I like the windowed mode, and the resolution is okay for me:D).
As a case, you can try the thing: record the video of any game with good quality (for example, with obs). And then try to watch it in full-screen. Are the details okay for you? Or simpler: just find some creeptd videos on youtube having 720 or 1080 resolution in source. If that quality is pretty good, I'll probably try to rescale the interface (in the similar way as youtube does).

Re: To mod or not to mod

PostPosted: 26.04.2020, 11:40
by NEb0
One more note about creep sending. As i said, i agree that creep flooding is a bad idea at all. Really, i don't understand why the investment feature is disabled at AvA. That would solve a lot of technical problems with rendering, time wasting with creep sending and so on. It even would solve the problem that the client can count the creeps sent, and know everyone's income. Investment request are invisible to other players. That's why nobody could exactly be sure, does the player saves money, or he invested in income.
And yes, I tried to send investment request to the server at AvA -- it's not allowed, and not working.

So, the only thing we can do is to send. Original client provides an ability to send waves. So, i understand it as simplification for a player: don't distract with sending, just press shift+4 and you have some time to upgrade towers. In the original code, there are actually two different delays for sending. One for "usual" sending and another is for "waves" sending -- there are two unrelated parts of the code. Also, original code creates a thread for every wave of sending. This, at least, is wasting resources, not very effective, and makes some randomness in sending.
I'm thinking about the similar implementation, but "more fair", not so tied to random factor. And i want to avoid two problems:
1. standard "waves" are sending too fast. When i'm pressing shift+4, and try to upgrade or change priority of something, i just often can't be in time to do it, because i need to put my fingers again at shift and 4.
2. standard "waves" are sending too slow. When i have about 2-3k of money, and 1-2 seconds until the end of the round, I press shift+4. Sometimes, round is ending before the whole wave were sent, and some money left in the end of the round, the wave continues sending in the next round. A lot of times that isn't what i wanted.

Again, as I said, the game provides a way to "comfortable" sending a lot of creeps, it just implemented badly. So, what about to improve it?
I'll think about the ideas by myself, but at least, i have an idea of "sending switch": for example, you press Alt+4 -- and mantas are starting to send in the background, press Alt+4 again -- and they stop. Also, the wave is stopping if any other sending is performed.
Another possible variant is to increase the wave creep count, but stop sending the wave at the end of the round.
What do you think?

Re: To mod or not to mod

PostPosted: 26.04.2020, 23:39
by Seraph07
Giving players more visual and audio ques what creeps are on the field is some unfair advantage. I understand that you want to see the creeps, thats why I changed some maps and I really like giving the PRO features to everyone in case of visuals at least.

Re: To mod or not to mod

PostPosted: 28.04.2020, 14:49
by NEb0
Seraph07 wrote:Giving players more visual and audio ques what creeps are on the field is some unfair advantage. I understand that you want to see the creeps, thats why I changed some maps and I really like giving the PRO features to everyone in case of visuals at least.

Sounds are related to the themes only. Just in the classic/default variant there are 4 sounds only, you can add more sounds if you wish, to the standard PRO client.
As i were recieved the sound files (made by Dieego98), i made a sound theme:
piano.zip
(1.37 MiB) Downloaded 268 times
. You can also test it in your standard client (just place the file into "themes/" folder), it should works properly.
Not sure, that these piano sounds are exactly what SoulMan wanted, but you can send me another sounds, i can pack them into a theme.
Maybe, you just didn't tested the mod: most of things are related to performance optimizations, tower position calculations, things to improve analyzing replays, and so on + creep themes (usual themes, like in any PRO client).
For about tower position calculations: nobody cannot forbid to print all the maps' positions for any map, make a list of them and it will be MUCH more comfortably to use, than the way it implemented for now :) (i think, i can create that list in a couple of days, if anybody interested) It's also exist in PRO version, but it doesn't count correctly. I think, it's a bug, not a feature :)

Re: To mod or not to mod

PostPosted: 28.04.2020, 15:09
by NEb0
Today, I've implemented a basic clusterizing implementation. This includes an implementation of rendering creeps from weaker to stronger ones (so, sharks are always on top of vultures, raptors are on top of everything except mothers and so on). screenshots:
676.jpg
676.jpg (117.65 KiB) Viewed 8032 times
677.jpg
677.jpg (126.24 KiB) Viewed 8032 times

Stress testing:
53k creeps:
678.jpg
678.jpg (115.95 KiB) Viewed 8032 times

123k creeps:
680.jpg
680.jpg (119.1 KiB) Viewed 8032 times

Now, at least, the creep rendering is not the main hotspot :)

Clusters with less that 5 creeps, have the merged healthbar (1px height for one creep). Still not tested ingame, not sure that exactly that implementation is very good, but i hope, you understand the idea.

Re: To mod or not to mod

PostPosted: 14.05.2020, 21:05
by PlixFiz
dont invite me to any game or play in any game with me while using this hack. thanks

Re: To mod or not to mod

PostPosted: 14.05.2020, 21:11
by PlixFiz
theres already been instances where players will literally make a fake account. act as a whole different person, and spectate just to see everyones income. @Dieego98. Also one night we all were playing i was saving for vulture to send next round, dieego was drunk, paused the game and was like "vulture!?!" everyone was like uh what? and yea i had 5k inc 10k saved he knew exactly what i was doing how much income i had just from having a fake account to spectate with. It's a shame that players really get that offended when they lose because they're garbage at the game that they have to resort to cheating. on fucking CreepTD lmao. pathetic. There have been other instances too as well shit just annoying at this point.

and what is this "Improved tower calculations" and all this other stuff added like wtf lol shit literally sounds like a hack and its turning out to be.

Re: To mod or not to mod

PostPosted: 14.05.2020, 21:22
by PlixFiz
very unfair how you can just see strongest creep above others too. NO. That shits part of the game, if theres some shit like a shark hidden under raiders, well, sucks for you for being garbage and not paying attention to your field. You're trying to make this game easier because you yourself could never master the game like other pros did. like wtf is this lol putting the game on easy mode. its supposed to be a hard game don't play it if you can't keep up with the REAL pro's ;) :lol:

Re: To mod or not to mod

PostPosted: 14.05.2020, 23:44
by Dieego98
Just answering to the personal attack: I have never seen anyone's income while playing, except 1vs1 vs NEb0 for testing purposes.

After seven years playing, you know that if someone doesn't send for 3 rounds, they are saving for something strong. You can do it yourself, you know when you are getting a shark, and you don't need to see anyone's inc for that.

As you yourself said, I was playing drunk... I wouldn't be playing drunk if I cared that much about winning. That day, I lost the three ava we played, as expected :)

Sometimes you win and sometimes you lose, it's part of the game, and you should not be making accusations like that if you don't have proof.

Re: To mod or not to mod

PostPosted: 15.05.2020, 00:32
by PlixFiz
dieego you sound slow stop trying to play it off its ugly. Your acc that you tried to play off was "funkadelic" you even went as far as to play a game with me and Rex while we "taught you how to play" because i thought you were a new player, then you just used the account to spectate games to cheat and see income. You even said yourself, which truly fucks you and puts the cherry on top "funkadelic is Rex's twink"
http://www.creeptd.com/en/games/4y9ugmw ... 0dj0y467hz
but this is the game where you both played and me and Rex taught you "how to play" i was spectating ^ game.
pathetic dieego.

Re: To mod or not to mod

PostPosted: 15.05.2020, 01:38
by NEb0
PlixFiz wrote:@Dieego98. Also one night we all were playing i was saving for vulture to send next round, dieego was drunk, paused the game and was like "vulture!?!"

idk about the exact situation, but there is a lot of situations when you can count when anybody didn't send anything, and so is saving... or just writing random things in the chat and they happens.
Ingame, people don't see income when they playing. also, i'm thinking about to remove that feature for spectating mode, if that thing is considered as a cheat.
PlixFiz wrote:theres already been instances where players will literally make a fake account. act as a whole different person, and spectate just to see everyones income.

Also, they could be "spies" who hint to somebody. At least, i have these arguments against enabling it for specs...
But really, even viewing income ingame doesn't help a lot (you can try it by yourself first). It's useful a lot for analyzing the games: when you see the stats after endgame, you can find the point where you got worse income/accumulative money and look into that point.
Also, I'd like to enable income viewing for everybody (in any case, if i made that, then the other ppl can do the same thing in any case), but looks like it's a too cardinal change (at least, for people with <= 1600 skill) and there aren't too much people who likes to test things. So, i have no idea how to test it to not hurt the community more...
PlixFiz wrote:and what is this "Improved tower calculations" and all this other stuff added like wtf lol shit literally sounds like a hack and its turning out to be.

it's an improvement of standard PRO feature, which you can see on the screenshots like viewtopic.php?f=3&t=775 . (green/yellow squares) now it has 2 modes:
707.jpg
707.jpg (170.73 KiB) Viewed 7849 times
708.jpg
708.jpg (129.14 KiB) Viewed 7849 times

Really, the numbers could optimize your build just by a bit, after you play a map 1-2 times, you should know where to place def (if i still don't know, i just watch the good games on that map, and it's more helpful usually, because the numbers are static, they don't show any towers' cooperation). Also, as i said, i can calculate these numbers for all the maps and make a screenshot for every map. It should cover topics like viewtopic.php?f=15&t=2386 and similar which exists on German forum (i don't remember the exact links).
PlixFiz wrote:dont invite me to any game or play in any game with me while using this hack. thanks

I very rarely invite ppl to the games at all. Never invite ppl who don't use the mod. And usually agree to play from original client, at least at 1vs1, if people are cares about that. The main advantage f the mod is themes, and vector vision enabled, both of these are achivable by using the normal PRO, which is unavailable now. Others are just a minor performance/visual optimizations.
Afaik, a lot of ppl without the PRO had a lot of chances vs PRO users... Who likes standard themes, they could use them, and there will no any advantage (or probably, you will get a little cpu load and some tools for better analyzing the games).
PlixFiz wrote:very unfair how you can just see strongest creep above others too

It's just a bit annoying. Really, here no problem at all most of the time. It is a just a minor improvement for rendering.

Again, if these annoying parts are the part of the games, and the design is good for everybody but me, then probably it will be more fun to have games with invisible(or just bad-seeing) creeps? If this is not a bug, but feature, i'm serious can create a theme for these kind of games. But in my opinion, it should be a feature, not a "bug of design". You can name the game like "invisible creeps" and i'll think about how to control that everybody uses such "bad" themes.

The another thing, which worries me: there's a lot of people who was upset that there's no development. if you don't like the features i made, maybe suggest the things what would you like to see from dcode?

Also, as you already have the PRO, i can send you the mod to test the things by yourself. (as i said at the beginning, i frozen sharing it in total, until that wouldn't be clear, does the changes make good or bad things for players). Also, i think, it will be fair for other pro, especially for whose who thinks there are some magic-cheating features, at least to show that there are no:) Just write me or the current testers (soul, diego) a pm.

I'm in hope to create a some kind of a new version of the client, which could be a standard, at least, while there is no official development. Not the thing that looks like a cheat. Also, i call up to test it by other players, to make a decision.

But looks like there are 2 YES/2 NO main answers for now... Okay, i'll wait for 2 months more. I've created the vote for 3 months, in hope that most of the active players who care, will answer the question. Looks like i need to delay all the final preparations for release (installer/scaling modifier configuration) and better do my works instead.

Have a nice day!

Re: To mod or not to mod

PostPosted: 15.05.2020, 02:00
by NEb0
also, that's not a topic for the war plix vs diego.
I don't want to judge, who is right or wrong in your exact situations. There are another topics i think, to write about people who cheated. IMHO, if a person WANTS to cheat, he could find a lot of things, a WAY better than viewing income, to cheat here. It's not a problem of the mod, it's a problem of the original implementation.
But okay, as i said, that's why i created THIS topic, to decide, which features could be considered like a cheat, and which are not, and get the most people point. In the worst case, i wanted to see the numbers like 2/20 votes against the mod, then i have no questions, just stop any development of it. Now i suggest finding a compromise

Re: To mod or not to mod

PostPosted: 15.05.2020, 04:21
by PlixFiz
wait, so, your mod also shows the best tower postions down to exactly how many times a tower will attack the creep? :roll:

Re: To mod or not to mod

PostPosted: 15.05.2020, 05:00
by NEb0
not exactly. it's just an estimation, like in original implementation you have in the PRO client, in vector vision mode. (like on screenshots i linked above). Just original implementation uses too simple formula.
really these numbers are path coverage for the range of the tower divided by some constant, and they show relative "usefullness" of the tower with an exact range from point to point.
I thought about to calculate the number of times the tower which shoot, including cooldown... but these calculations are much harder, but actually the accuracy won't be a lot higher.
Also, the "count of times to shot" is too difficult to implement because it depends on the creep speed. For sure, you can take an "reference" creep, like shark, or mako, but these numbers will be almost the same as just the "path coverage" (really, they will be a bit different, because cooldown is still "recovering" when the creep is temporary out of the range of the tower)
Btw, for validating my formula i also implemented a simple "tower statistics" box, which shows how much damage the tower did. It shows when you click on the tower instead of "mini-tower" view. But this feature is so minor, that i forget about it^^
Also, in a real game, you need to take into account much more factors for tower building than just the "perfect" spots. I think, it's an obvious thing.

Re: To mod or not to mod

PostPosted: 17.05.2020, 15:00
by Seraph07
I don't bother sharing the PRO-Features with everyone, only things which provide more benefits than that. Please cooperate with Knappe8 or someone and redesign the game as a whole. Adding mods on this shipwreck won't do much in the end. You need a solid ground for your advancements and you will learn more by starting fresh than fixing those bugs here. Also you will get a way more active suggestion community for a new game. I don't wanna have a clientmod with extra features which will seperate basic and mod players. A new game where the host can set stuff like:
startingincome
gamespeed (map or income)
lifes
... and so on will be better for everyone in the end. Creepsmash was lying dead and dcode made CreepTD as a fresh start. It takes some commitment and will take some time to finish it but I hope it can work out.