Sheep Tag Community
General Category => Community Discussion => Topic started by: glow on June 09, 2014, 10:26:51 pm
-
IDK how it works, but this is a message to Chakra, Gosusheep, XXXandBEER, WRDA, and whoever edits and makes custom games.
https://www.youtube.com/watch?v=t5g6lpEEuII
Maybe you would like to go make sheep tag there NOW to get good exposure =)
-
IF you could make sheep tag on this: it would be amazing to say the least~!
But Idk if making it would be possible, and if jumps and how building would work (Would BBing work / how would it work, etc).
Also free picture of Katama enjoy!:
(http://i.imgur.com/8Kd1brl.png)
-
Can the MONSTERgeeks of the community fix an st on dota 2 or? Or is there a modding team that create maps only?
-
Anyone is free to mod the game, I believe.
-
http://www.reddit.com/r/DotA2/comments/26gmvt/announcing_d2moddin_a_custom_game_platform_play/
GET THEM TO MAKE SHEEP TAG YEEEEEEEE COMMENT LIKE HELL ABOUT SHEEP TAG!!!
-
http://www.reddit.com/r/DotA2/comments/26gmvt/announcing_d2moddin_a_custom_game_platform_play/
GET THEM TO MAKE SHEEP TAG YEEEEEEEE COMMENT LIKE HELL ABOUT SHEEP TAG!!!
I did mr "PM_ME_YOUR_COCK" :D
-
Don't tell them to make ST lol....
Their idea of sheep tag is some base-version of ROTS.
We need a SUPERgeek from the community to make it, otherwise it's gonna be a big flushing TOILET.
So we need them to hire a supergeek to make it with their platform. This doesn't sound realistic :(
-
Link them a version of modern day revo like this: http://www.youtube.com/watch?v=NcxOep67bHc
and maybe tell them to re make revolution etc.
But yeah I think having a monster geek from our community would also be nice.
I know glitch nova still has some hope even though its going to take a long time, also I haven't spoke to chakra about it in a while so.
-
Aname just dropped out of Glitch Nova. Said maybe 5 years. That being said, I'm still working on my pathfinding algorithms, so...
-
Hi ST community, I actually am interested in making sheep tag. I currently have one mod on D2moddin, Warlocks (http://d2modd.in/mods/warlocks). I believe ST is completely doable. There's already a sheep model, and some building models. I already know how to work with the API pretty well.
Possible obstacles:
- I know nothing about path finding
- I don't the physics of "jumping" off of hills/cliffs in ST (I guess this goes hand-in-hand with path finding.)
- I asked the Element TD Dota 2 dev, and currently you can't get the "building grid preview" before building a building. However, you can show an AoE circle before building it.
So I think what I'll do is put together a basic map, with sheep, shephards, farm creation, and go from there. I'll be posting my progress on my twitch: www.twitch.tv/myll_
-
Hi ST community, I actually am interested in making sheep tag. I currently have one mod on D2moddin, Warlocks (http://d2modd.in/mods/warlocks). I believe ST is completely doable. There's already a sheep model, and some building models. I already know how to work with the API pretty well.
Possible obstacles:
- I know nothing about path finding
- I don't the physics of "jumping" off of hills/cliffs in ST (I guess this goes hand-in-hand with path finding.)
- I asked the Element TD Dota 2 dev, and currently you can't get the "building grid preview" before building a building. However, you can show an AoE circle before building it.
So I think what I'll do is put together a basic map, with sheep, shephards, farm creation, and go from there. I'll be posting my progress on my twitch: www.twitch.tv/myll_
I'd rather move to a standalone game entirely. I just tried to play some custom games HLW, Pudge Wars etc. And that d2mod site has so many bugs and glitches it was almost unplayable. So I'd rather just keep playing on Wc3. Also you have to still connect and play through steam which is annoying as well. But be my guest if you can make it I'll try it, but I can almost guarantee it won't be nearly as good as it is and was on Wc3.
-
The physics of jumping are actually relatively simple.
WC3 uses an entirely grid-based movement and pathfinding system. What this means is that you can subdivide the entire map into a distinct, regular grid. WC3's reduces down to a 32x32 resolution grid system. This means that a normal farm, which is 128x128 units, is made up of 16 unique cells. All buildings are made up of exact, fixed amount of cells, called a pathingmap; the smallest object that uses a pathingmap is an item, which you've probably already seen as taking up 1/16th of a farm or 1/4th of a tiny farm.
Units also are interpreted on this grid system, though they are freely, non-bound agents. Units have a collision size, which simple means radius. What WC3 does is reduce the circle around a unit into the cells that it is occupying and the cells it is not. The cells it occupies are considered by the game for all calculations, not the free circle around the unit.
After you understand that, things actually become much simpler. The movement system no doubt uses some type of grid-based pathfinding algorithm, my guess would be some variant of A*. The actual system you use doesn't really matter, as we, as players, don't really adapt that much to the pathfinder, but rather the jump physics.
Jump physics are separate depending on whether it is mirror imaging or simple unit relocation (i.e., building a farm on yourself). The latter is done by using a regular system that moves your unit to the closest spot you can be placed, and it does this by scanning grid-spaces rather than linear distance. I don't remember the exact number off the top of my head, but if a unit cannot be placed within n spaces, it is simply kept where it is. The important tidbit is that a line of diagonal cells is worth the same as a horizontal cells. This means that if the first open diagonal location is 8 cells while the closest horizontal or vertical distance is 9 cells, it'll place you diagonally. Why does this matter? Because, using a 32x32 cell resolution, going diagonally across 8 cells is ~362, while horizontally across 9 is only 288, so the game prefers to jump you diagonally over horizontally/vertically by a root 2 factor.
The final piece of physics that matters is mirror jumping. What happens you mirror jump is that it takes your current position and facing angle, then picks two points left and right a regular distance from your start position. This distance is freely changeable within WC3, but by default is, I believe, 64 units, meaning a perfect farm space exists between the two if a user images in the open (though one might not be buildable, as buildablility works on a 64x64 resolution grid, so it is likely one of those grids is infested by a single 32x32 cell or two). Once we have this point, the game will then try to find the nearest point that the unit can actually be placed. The cool thing is if this point is on another level (i.e., up or down a cliff or two). This means it must find the closest point on another cliff. This distance it seeks is a bit different than normal jumping, and may be related to the Area of Effect field for the ability, which is set to 1000, but I'm not sure (haven't tested it).
The next thing it does is a near repeating of the above, except now it tries to find a placeable location on the original level (as you can't jump up or down a level, without some active manipulation). This is, of course, a repeat of finding the closest location. The maximal distance this will go is again some unrelated figure to the previous two and, again, may be related to Area of Effect, but I'm not sure. I know, however, that it is not the same as the former calculation, and it seems a bit short to be the first one.
I think that details all the complex physics of ST. The actual figures, once you understand what is happening, is easily testable by playing around with terrain and placing farms (tiny farms would be easiest, though a custom one with item-sized pathing would be more precise, though it should replace an item's pathingmap, as buildings can only be placed on a 64x64 grid).
-
I didn't have a chance to read everything that Beer and Chakra said but I love the fact you're attempting making Sheep Tag on DotA 2, ask for help from chakra and beer as they know about Wc3 physicics. FULL SUPPORT Myll, be back later to post more.
-
@Chakra Thanks for the post. Dota 2 also uses a 32x32 grid resolution. One caveat is buildings will be units with circle collision. However, I can completely eliminate all default collision in Dota 2 by giving every unit the Phased attribute (Phase boots in Dota 2). So, I'm thinking I can manually create and continuously update the pathing map for every unit throughout the game. This would be a very powerful solution IMO.
-
Even if they have circle collision, it's entirely possible their pathing system still reduces them to grids. I.e., you might not have to worry about actual units, and since buildings don't move, you only need to update on creation/removal.
You can test if it reduces them to grids quite easily. As you said the system uses a 32x32 grid, just create a narrow area and place dummy 1-collision units in the way. If it requires ~32 of them in a line to block it, then it uses a vector-based pathing (or a very small movement grid). If it's like WC3, then just one or two units will work.
I.e., in WC3 you can only fit 16 1-collision units in the space of a farm. That's because the unit can either take up zero cells or one cell, and a farm is made up of 16 cells.
-
Say the farm in Dota 2 does reduce to a grid. Now how do you deal with the missing cells? In WC3, the cells of the farm would occupy a square right? Well the cells of the farm in Dota 2 will occupy a circle shape. Wouldn't it be important to add more cells to make it a square shape, to mimic WC3?
Ex. I think a farm in Dota 2 would look like this: http://www2.stetson.edu/~efriedma/squincir/12.gif
-
I think the simplest solution would be to simply forsake the special case of items and define a normal farm as using 4 cells rather than 16.
What WC3 does is this: http://i.imgur.com/Tx84yul.png
The green represents object: a sheep, a shepherd, three farms, a tiny farm, and an item. Pink means that cell is not pathable (i.e., something is there), blue indicates it is not buildable, either directly or effectively (blue around units are direct, blue around items are effective). Direct means it shows up as red when trying to build there, effective means it's green.
Note that the tiny is made up of four, but you can only place it where it is or two cells over. It cannot be placed midway.
-
Myll how is the progress? I would personally love to hear from you.
Did you figure out the farm details?
-
Been working on the custom grid system mainly.
(http://i.imgur.com/zhgjblK.jpg)
-
Dang! Actually doesn't look to bad. Keep up the good work. How are the physics on it?
-
Didn't add the physics yet. But, once I'm done with this grid system, I'll have control of every single cell in the game. Meaning I can tell if a cell is controlled by a sheep, shepherd, is open, closed, is occupied by a farm, etc. So once I have this system in place the physics will come after.
-
Very impressive work myll, its complicated work so I can't help you much myself, but you seem to doing excellent on you're own making this.
You still have my full support in this project!
-
Here's a vid of what's done so far: https://www.youtube.com/watch?v=fq5wxJSVuEI
The current map is just a placeholder. Let me know if you have any questions :)
-
**** Sick Myll! :D
-
Wow thats VERY impressive, nice job! I noticed when you built farms it moved axe or who ever it was in a different way than WC3, very interesting how and what you could do with this.
If you need help with creating terrain -> look at old terrain versions (Revo 6.7.5) etc.
Questions:
What will you do for wolves / sheep?
I remember you mentioning something about a sheep model so sheep is good but what about wolf?
-
Nice stuff. All farms together are more than 10, that means also more than 10 hotkeys. Is there any chance to do the same idea like at warcraft III sheep tag? That you use " b " for build and then it opens a new menu where every farm is mentioned? I doubt it but I got no idea.
-
Wow thats VERY impressive, nice job! I noticed when you built farms it moved axe or who ever it was in a different way than WC3, very interesting how and what you could do with this.
If you need help with creating terrain -> look at old terrain versions (Revo 6.7.5) etc.
Questions:
What will you do for wolves / sheep?
I remember you mentioning something about a sheep model so sheep is good but what about wolf?
Well I hope the jumping is similar to WC3, that's the goal. There are plenty of wolf models/horse riders in Dota so that won't be a problem. The sheep model atm is kind of screwed up, since Dota is trying to combine Axe's models with the sheep's. Hopefully I can fix this. The terrain will be very boring at first, since I'm not skilled with the map editor. However I'll ask a couple mappers in the community if they're willing to remake ReVo.
Nice stuff. All farms together are more than 10, that means also more than 10 hotkeys. Is there any chance to do the same idea like at warcraft III sheep tag? That you use " b " for build and then it opens a new menu where every farm is mentioned? I doubt it but I got no idea.
I can't bring up a whole menu like that, but what I can do is have the last skill be a gateway to 5 other abilities. Also, if sheep tag doesn't have too many items I can use some item slots as abilities.
Thanks all for the feedback!
-
Can you not do a keyboard/mouse hook? A mouse hook would possibly allow the grid for building and a keyboard hook, obviously, would allow advanced menu type deals.
-
@Chakra Nope, can't do that unfortunately.
So what I'm doing for the jumping is searching a rectangular area around the sheep for open squares. It then adds the open squares to a table, and calculates the distance from each open square's center to the sheep's center. The open square with the lowest distance is the one the sheep jumps to, and it jumps to the center of that open square.
Edit: Found a big part of the problem. Like Chakra said earlier, the game treats diagonal movement = to horizontal/vertical movement.
Edit2: So the main problem is there are multiple equi-distant diagonal paths where the sheep can jump, as illustrated in these images:
(http://i.imgur.com/9DTOtyp.jpg?1)(http://i.imgur.com/7366ChW.jpg?1)(http://i.imgur.com/zeUWNIm.jpg?1)(http://i.imgur.com/ZKNdDrG.jpg?1)
Bigger pics: http://imgur.com/a/nKPoQ#0
The question is, how do you know which path the sheep will take?
-
Impressive work Myll, great job ;D
I can't really add much to the technical discussion, but here's a question and I hope we're not getting ahead of ourselves:
Should we call it Sheep Tag?
Some negatives to keeping it sheep/wolf/ST.
- People already know ST, and dota players are probably familiar with it, have a negative opinion.
- Sheep tag sounds childish, makes it less likely to be taken seriously.
Only positive is tradition basically. OFC changing the units from sheep / wolves to something else will change the feel of the game, but the feel would already obviously be dramatically different as it's on a different platform.
I don't think we should call it sheep tag, or anything with the "Tag" label. We should rebrand it and give it a fresh start, although I lack the creativity to come up with a name that is both descriptive and cool sounding.
-
We've had a few projects like this in the past though they've been run by mentally unstable crackheads. This seems promising.
Myll - out of curiosity - what is your background and have you played sheep tag previously? You appear to have knowledge that an outsider typically wouldn't (i.e., jump mechanics).
-
Myll - out of curiosity - what is your background and have you played sheep tag previously? You appear to have knowledge that an outsider typically wouldn't (i.e., jump mechanics).
I've played sheep tag previously ofc. I've participated in a handful of inhouses, but I'm not good at all :P I do respect that this game is competitive though.
@Drew I do agree that sheep tag is a childish name, and doesn't resemble the competitiveness of this game well. Maybe Sheep vs Shepherds? I don't think Dota players would see this game negatively at all. It's a wc3 classic.
Regarding the timeframe for this, I'll get as much as I can done before I go back to college (Aug 18). I've no idea when I'll have something playable though. Could be before then, could be in 2 months.
-
Good question by Sidey. You should stop by and play with us on Europe afternoon EST around 2 PM - 6 PM. Might give you a better grasp of the game to help with map. Honestly no rush to release anything. We've been teased about the potential of something this for years so its worth waiting for a quality release that will hopefully be exposed to a larger community and bring back old players.
-
Please make dota 2 version based on hunger games... instead of wolves are the deadly alliance 'wolf pack' of the inner districts. Sheep are the weak, pathetic survivalist people 'sheep for the slaughter' of the outer districts. Maybe instead of wolves randomly spawning 15 sec later or w/e later they get to choose a starting weapon in the middle... while the weak **** go build their camps ... granted this sounds like some rots **** but it's just a metaphor... you can make it all same mechanics otherwise. just call it the hungry games or some crap lol. then we get MASSSS players np
-
u better add stack farm
-
New update:
https://www.youtube.com/watch?v=RLyi3ze7hqc
@BeeH It will probably be added eventually..
Pretty soon I'll let you guys actually try it out, and get a feel of how building stuff works.
Or, I can use these for farms. Another thing I can do is use multiple, similar looking models for normal farms, to add some variety.
(http://i.imgur.com/zNLgmdv.jpg)
The big ones are the hard farms. Just added them. What do you think?
-
yeee Hardfarm + normal farms looks better than units imo.
Looking pretty good so far, maybe also create a building process. Just simple increase HP over time?
-
Looks awesome. I am always up for a try-out, tell me your steam name and I'll add you.
-
This looks great, good job!
-
Just felt like I had to reply to show my support. I'm honestly amazed to see how much effort you put into this. Much respect, man! ;)
Personally I think that those buildings in your last screenshot would do perfectly as farms. Hard farms don't look bad either. Great work!
If you need playtesters at any point, I'm definitely down. My Steam name is VsKatshuma.
I'm also studying Dota 2 modding myself but I'm not anywhere close to this level ^^
-
Thanks guys. this community's enthusiasm is really fantastic. Never felt this welcomed into a community before ;D. Join this steam group: http://steamcommunity.com/groups/sheeptag
@VsKatshuma prior to 1.5 months ago i had 0 experience with modding all together. And, I'm no expert programmer either :P. Basically spent my whole summer learning this stuff. I'm happy to have found passion in this.
I wish I would have started with sheep tag from the start though. Could have had a full version out by now. I started with remaking wc3 warlocks (it's rubick brawl now on d2moddin), but I didn't expect the original devs of warlocks to jump on dota 2. I was in their dev group for their dota 2 version at the start, but I didn't feel comfortable with them. I was basically at the bottom of the totem pole in that group. Not a position anybody wants to be in. Not to mention, I wasn't keen on throwing my hard work under the table to work on their version.
-
I registered here just to say this:
Awesome!
/LL
Btw why would you need to do path finding? Dota 2 already has this built in.
-
only me who thinks a toiletname like "sheep tag" is way better than creating some new "cool name that makes the game look serious"?
anyway, keep up da guud work!
Edit: Thought i'd elaborate to save some time. shoot me if my reasoning is bad. THEORY:
the best way to get new people to play ST = keep a retarded name like "sheep tag".
noone cares about the name after they've tried 1 game. they're left with a perseption of gameplay at that point.
the function of the name is to make people play their first game. only important question = what type of name will make people most likely to give the game a trial.
i think a name that gives off a "this is a fun, clowny minigame xD" impression is most likely to make people try the game. on the contrary, i think a name that gives off a "this is a complex game that needs tons of practice and mastery" is least likely to attract people, due to the context (it's a custom game on dota2 -> people aren't looking for games to invest time into, they wanna have some fun and move on).
in short: "clowny" gamename = best bait. gameplay will do the attracting after people decide to give the game a chance. changing the name of the game might be good, but not in the direction you guys are thinking.
bad or good theory?
-
It may attract more people, but what type of crowd are you trying to attract. As you said amir it makes the game seem clowny, but its a highly competitive game. Wouldn't we want to attract competitive gamers instead of casual gamers.
Edit:
ALso what name would you even change it to. In order to sound less clowny?
I was thinking Defense of the Farms DOTF like Dota may attract people idk...
-
That sounds good, Beer! I'm actually kinda liking that name, it makes sense for the content and it will defiantly spark some interest into the game.
-You could mention it's based on Sheep Tag from WC3. Also when the game is released lots of us could join it and start playing like we would on wc3, there we will start to adjust.
Pubs will start to see us playing WHILE developing our game play to the new platform, and boom it will almost be like a race to get better with pubs following behind (in skill).
The execution would have to be perfect, as we would need to get mass development (game play / skill) and games hosted of st.
-
Alright this is almost ready for some playtests. If you want to play, follow these directions: https://developer.valvesoftware.com/wiki/Dota_2_Workshop_Tools/Installing_and_Launching_Tools
I just need to test out how players connect with this new system, it shouldn't take long. :)
-
I have it downloaded. ;) Let's get this party started.
-
Damn I installed it, but when I run it I get this error
https://support.steampowered.com/kb_article.php?ref=8676-EFLX-7453&l=english
I tried doing what they said to do, but none of it worked. :/ oh well
-
Ok there will be playtests tomorrow at noon EST. Join the sheep tag steam group chat at that time, and subscribe to the game here: http://steamcommunity.com/sharedfiles/filedetails/?id=297888525. You can actually try it out by yourself/with others right now if u want.
I'm not sure if it automatically updated to my local files though. Nvm, just updated it.
"How to host
First subscribe to a mod (like this one) by pressing the subscribe button on their workshop page. This will automatically download it and keep it up to date. Now start the normal (not the new 64bit) Dota2 client, press Play -> Custom Game and select Subscribed Game Modes in the dropdown menu. Now rightclick on a mod and press Create Lobby. Set a password on your lobby so people can find when they search for it. You can also invite your friends to your party and then create the lobby which will automatically place them in the lobby. Your friends will also be able to join you through the custom games tab. "
Only my steam friends can subscribe to the game, so add me on steam: http://steamcommunity.com/id/Myll/
@Beer maybe you can find a solution here: http://steamcommunity.com/workshop/discussions/?appid=570. That problem is common.
-
I am extremely unfamiliar with steam as I never use it. How can I get Sheep Tag into my library so I can play it? I added you as friend and subscribed, but I cant find where to play this.
-
Okay, so I just playtested the map Myll is working on for the first time. I played 1 game of 1v1 with Polar. We fooled around a bit and tested how the map feels like. I will now write a post about everything that came to my mind while playing.
Myll, I don't want this post to sound like a list of orders about what needs to be fixed and what sucked, but I want to list every little detail for polishing purposes and also historical documentation. :D I really think you are doing amazing work. ;) Don't forget that.
Major things:
First of all, the very first round was totally bugged. Every round that came after that worked just fine. But during the first round only, both the Sheep and the Wolf stayed stunned for the whole 60 seconds duration. So we had to wait out the 1st round before we got to play at all. Also the Sheep seemed to not have the abilities learned by default during that 1st round, but they all were available from the start during later rounds (like they should be).
Sheep and Wolves are both overpowered in their own way. I get that the building-mechanic will be completely different from WC3 with the missing buildgrid and the new crazy jump mechanics but the Sheep can build new farms so easily and so fast that its almost ridiculous. In WC3 we have these "massing strategies" that have been in development for years and you have to position yourself perfectly for maximum farm massing speed. Here I just spammed Q and the screen got filled up with farms in a matter of seconds. This is not necessarily a bad thing, it makes the Sheep exciting to play. But if the games are like this I don't see the Wolves ever winning in 5v5 games. Obviously this will change when Wolves get their abilities and items, which are not in the game yet. So I guess it's best to wait and see before we do anything about this.
If anything, the new massing mechanic seemed super FUN! :D It was REALLY hard to figure out where the farm will be placed without the grid but the pathfinding and collision detection feel INCREDIBLE for a before-alpha release. Unbelievable work, Myll. Just want to get that out there. I want to see this map become a reality! :D Also the pure power and speed the Sheep have in this game felt super. ;)
The Wolf on the other hand, was way too fast. In the current version the Sheep had 410 MS and the Wolf had 470. The Wolf is supposed to be faster than the Sheep, yes, but only a minimal amount. With the speed difference this high (60 MS) there won't be any runtagging. I would suggest slowing down the Wolf to at least 440 or even 430. This would enable the Sheep to survive a bit longer when running outside of farms.
The way you shift through "building menus" by pressing R felt super weird at first but I guess it really is the best alternative we have compared to the "real" buildmenus in WC3. A nice innovation. I guess it would only take a bit of practice to get used to. And it's really nice to have more than 6 different kind of farms. =)
The games last only 60 seconds at a time? That's WAY TOO SHORT. A normal game of Sheep Tag lasts 10 minutes in WC3, and a tournament one 20 minutes. I understand that we're not making a carbon copy of ST here but a map "inspired" by it but 1 minute is just insulting. Longer games pleaseeee :D
The last big thing that I want to talk about is the farm size. The normal farm model is a bit too big in my opinion. They take up so much space that a 2x2 space in between 2 farms really looks like an 1x1 gap to my eyes. I thought that I was safe but the Wolf ran right through that space. So make them a bit smaller, maybe. =P Also, the tiny farms could be a bit larger, just a tiny bit. They looks so small right now. I guess everyone just has to get used to playing without the WC3 grid. =)
Minor things:
1. Make 5 separate Sheep heroes and 5 separate Wolf heroes? The picking phase seems to be a bit bugged when there's more than 1 player per team.
2. In the loading screen it said SHEEPTAG. Can you separate the words and remove the ALL CAPS?
3. Tiny farms were 2 hits. They're supposed to be 1 hit by the Wolves.
4. Hard farms were 2 hits too. They're supposed to be stronger than that. Can't remember how many hits exactly, though.
5. If you tried to build something when you didn't have enough gold the Sheep just stopped. Can you make the game give the player a message about not having enough gold?
6. Why are you using a freaking Pudge model for tiny farm? xD
7. Every farm was named npc_dota_farm_name. I know that this is the default placeholder name. It goes on the fixlist.
8. Strong farm had the same name as an upgraded farm. I guess you copy&pasted and forgot to change it. ;)
And to close up this post, there's the thing about swapping teams that we discussed about. I'm just going to mention this here so others know that I said something.
At the moment, the game is played by 2 teams, that get swapped after every round. This means, that a 5v5 is the only fair way thing map could be played. In Sheep Tag, a 2v2 is not balanced. Sheep win that kind of game every time. But if you made a 2v3 or a 2v4 in this map, it would become 3v2 / 4v2 every other round. I'm not sure how we could go around this problem in the Dota 2 engine, though.
What I do know is that it is the SINGLE MOST distinct feature of a good map in Warcraft III Battle.net that the game can be "rehosted" in-game without actually having to end the game and go through the lobby again. (Starting up a Dota 2 game takes forever anyway.) Is it possible to make the game work like Revo in a way that it wouldn't matter which slots you join in the Dota 2 lobby? If we could form teams in-game before starting the actual gameplay and possibly shuffle through the players that get to play as Sheep in a 2v4 game so that the amount of Sheep stays constant through every round it would make for a much better map. I do get that this is hard to implement in the Dota 2 engine, though. And I do understand that even creating the system as it is now has to have taken some amount of work. I just want to put this out there, because in my opinion it is the single most important thing. We can't have Sheep Tag in Dota 2 be a game that constantly needs to be rehosted to make new teams. People would get bored really fast that way.
But still, awesome game, Myll ;) Happy to help you!
EDIT: I forgot to mention the killing range of Wolves. It felt completely different (duh, lol). The attack range of Wolves has been a matter of discussion for... ever I guess. I don't know what the current ATK range is in WC3 but I know that it has been experienced with in the past, and there will ALWAYS be that one guy who gets killed and shouts "RANGE WTF!"
In your game, the attack range seemed just okay while the Sheep was isolated (maybe a bit too long even). But while the Sheep was inside farms, it was a different story. As long as the Sheep was staying inside the 1x1 gap of the farms, it couldn't be attacked. No matter how close. The Wolf and the Sheep were standing right next to each other, their models literally touching, but the Wolf could not target the Sheep for an attack. I guess it is the pathfinding mechanic somehow interfering with the attack order. That's my best guess. Or then the angle was just really weird. Anyway, we need to look further into this in the future.
-
Sorry for not being part of the discussion or being around and testing as of late, for I am moving into a new house that does not have internet yet. But now back onto the subject at hand.
First of all, I still think everything you've done so far is amazing and I thank you for it.
I definitely thinking finding a strong median in between warcraft 3 and dota 2's sheep tag gameplay and design will work very well, warcraft's genius brought in essence to a more modern system would be the best way I can describe it.
I would say katshuma is definitely on track with just the right kind of advice. Him having play(ing) it gives him as others to fine tune the game very fast.
By the time I return maybe I can play some fattie dota 2 sheep tag ^^
As always good luck and keep things updated.
-
Hey Katsh, sorry I haven't been able to make a big reply. I just don't have the time. Thanks for the amazing writeup, all your feedback has been noted, and changes will be made accordingly.
Regarding the sheep/wolf player shuffling, I don't think it can be done atm in Dota 2. The main problem would be vision. The players from the start are locked into either radiant or dire. So if you're on the Radiant from the start, you'll always be able to see every other Radiant player on the map. If you have a sheep and a wolf on the same team, they will be able to see each other.
Regarding the time limit, I think 10 minutes is far too long for a single round. I don't want games to last more than 15-20 minutes, and I want people to alternate between wolf and sheep several times throughout the game. That being said, the farm bounties will be higher and gold will be given at a higher pace so people have access to the high-end stuff much earlier.
Atm I'm optimizing the grid system to use much less memory and to increase performance. Also I'm working on a library revolving around this system for Dota 2 modders who want to make RTS style maps and tower defense games.
Also, if somebody wants to remake the ReVo map using the tile editor in the Dota 2 Tools, feel free.
-
Regarding the sheep/wolf player shuffling, I don't think it can be done atm in Dota 2. The main problem would be vision. The players from the start are locked into either radiant or dire. So if you're on the Radiant from the start, you'll always be able to see every other Radiant player on the map. If you have a sheep and a wolf on the same team, they will be able to see each other.
That's not really a problem. Vision has always been a relatively minor thing in ST. The bigger issue would be aggression towards enemies, i.e., auto-attack, blah blah.
-
yep that would a major problem.
-
Regarding the time limit, I think 10 minutes is far too long for a single round. I don't want games to last more than 15-20 minutes, and I want people to alternate between wolf and sheep several times throughout the game. That being said, the farm bounties will be higher and gold will be given at a higher pace so people have access to the high-end stuff much earlier.
you could just replicate the revo map as much as possible, and make adjustment where it's impossible to replicate (like everyone expected you to do). i assume it would just take you a couple of moments to create ''your own version'' after that point, where rounds last 1 minute and income is scaled.
this will basically give you a community (with up to decades of experience) attached to your maps from the get-go. i hope i don't have to go indepth about why this is benefitial for everyone, including you. but basically it will save you an infinite amount of time promoting the game/recruiting players/educating folks etc.
im saying this because i honestly don't think anyone here is interested in the vision you have for ST with minute-long rounds. it changes way too much about gameplay. i assume you play dota, just look at the difference between normal dota and EasyMode (-em). these versions are essentially as identical as you can get, with very slight adjustments to make games a little faster/easier. yet still, they change so many important factors that the majority of dota players don't want to touch the easymode version. risk-taking changes, positioning changes, importance of farm changes, balance changes, and so on. you'll observe a similar (yet far more notable) change in gameplay with the ideas you have for the ST map.
i mean, it seems so obvious that replicating revo as much as possible is the best route for everyone involved, and then trying to improve/change the game after that point, or create alternate versions with massive changes. there will still be lots of room for playing around and pulling different switches - noone thinks revo is even close to being complete. i think you should do this after you replicate revo and move the wc3 community over to dota2.
gl with the rest of ur project, great work so far. can't wait to see how the final version ends up, no matter what you do :D
-
Does any1 want to start a ST team on League of Legends!? - looking for some real muscle here!
-
Okay finally figured out how to play this lol (I wasn't opting in for beta testing on my steam account).
Anyways my thoughts:
First few things not having a placement grid thingy made the game way harder then I thought it would not sure how I'll get used to that. Also the hotkeys I guess we can just fix with Dota 2 custom bindings so that's not too big of a deal. Clearly the map needs some kind of terrain it will be curious to see how the physics interacts with that. Wolves didn't have mirror images if I recall I just saw summon golem. I didn't get to play with someone else so I'm not sure about the speeds of sheep/wolf etc. I agree with Amir and Kats the timer was way to short a standard Sheep Tag 5v5 game is 20 minutes for both teams. 1 or 2 minutes isn't going to cut it lol.
Overall It's extremely hard to judge this game as I know its roughly 20% done as well as custom games themselves are in Alpha. Anyways thanks for making this and keep us posted on the next update!
-
Figured I owed you guys an update. The project is currently on a hiatus for one some reasons.
1) The workshop tools actually ****ed over modders more than it helped. There's no proper lobby system for people to play mods. I myself haven't even had a 10 player mod game since the days of d2moddin. I'd take pre-workshop days over these days any day.
2) There's no building ghosting/building preview. I knew this going in before this project, but it's still a serious problem for this game.
Honestly the biggest reason is #1. It's completely demotivating to continue this project when people can't even play your game.
At the moment in my free time I'll be learning UE4, and probably contributing to this project. (https://forums.unrealengine.com/showthread.php?29159-Community-Project-Real-Time-Strategy-Game-NEW-VIDEO-UPDATE-2)
-
Hi friends,
I am excited to say I will be working from where Myll left off in recreating Sheep Tag within Dota 2. Previously worked on the Slide Ninja Slide Dota 2 port (http://steamcommunity.com/sharedfiles/filedetails/?id=401429935) which is a 1:1 recreation of the WC3 map with some minor changes, so I have no doubt that I will be able to port your map as completely as possible keeping it fun and balanced.
I plan to start off by first recreating the map 1:1 in hammer, granted there are some differences with cliff/river resolutions (This is my progress so far http://snag.gy/YKdsn.jpg ). Unfortunately I was unable to unprotect the most current version of sheep tag by Chakra to view that map in world editor so it is currently being recreated after GosuSheep's 7.1 version instead.
I will try to have a playable test version by the end of the week, so that we may discuss as to where we shall go from there.
-
Beer has the most up-to-date versions. Site has a max attachment size of 512KB, so I've uploaded the files to ChatCraft (http://chatcraft.net/files/Sheep%20Tag%20Ver%20Resp.zip).
-
WHaaaaaaaat?
-
Thanks Chakra! I noticed those maps are a little behind the most recent version (lacking shops and some terrain changes), but its workable in the mean time. The tile editor for hammer is great so it will only take a little bit of time to make the terrain similar.
Edit: Bonus Progress Image of the 'farms'
http://snag.gy/ov3Vt.jpg
-
Looks dope dude! Great work I can't wait to try it out. Add me on steam: Luckysheaven (aka foxy)
-
I uploaded the project to my workshop, but for now it is only visible if you add me as a friend.
http://steamcommunity.com/sharedfiles/filedetails/?id=435131884
http://steamcommunity.com/id/A_Dizzle
Be aware this map is currently super alpha and lots of things don't work yet or break. But feel free to play around and give feedback.
Things I am aware of:
- Hero's have cosmetics. This requires a massive workaround to fix.
- Far Sight reveals ground and not air
- Bomber has no particles
- Stack doesn't work yet
- Beam of Strength has yet to be made
- Farms don't upgrade
- Starting gold and gold rate isn't implemented
- Sheep Tag game mode isn't implemented
- Spawn position isn't right
XXXandBEER, would it be possible to get the current map off you, so am able to update to the 7.7.4 map changes?
-
Yes sir send me an e-mail at jfisher993@gmail.com
-
Minor Update - added missing mechanic https://gfycat.com/SparklingHairyFirecrest
-
Map has been updated so it's is 7.7.4 ready. Tell me what you think. It lacks props but the form is there :).
http://snag.gy/m4zti.jpg
http://snag.gy/aDkR1.jpg
http://snag.gy/AGcnJ.jpg
http://snag.gy/QOHFw.jpg
http://snag.gy/hD5M6.jpg
-
looks legend mate
-
Can it be played? I don't play on Dota 2 much but I'm willing to try and come play it whenever its ready, looks great mate.
-
yo with dota2 reborn out anyone gonna continue or make a new st map?
-
I plan on remaking Sheep Tag in Unity at some point. Just waiting for a few other projects to finish + some more money.
-
Beer you have my word that I would play ST if you made it as much as I could and even try to advertise it. <3
-
Sorry for the lack of communication, I've been sick had exams and been updating Slide Ninja Slide for reborn. Sheep Tag is mostly done except for a few building issues. I don't want to release it till its perfect but you will be able to play it if you add me as a friend on steam.
-
Things that need to be complete:
[Art/Misc]
- Loading Screen / Artwork (High Res 1750 x 1125) Can be made in sections so that the bg and fg elements are all separate png, psd files
- How to play write up - short, simple, include small high quality png images.
- Workshop Image (stylised) - 1024 x 768 png
[S2 programming/mapping/modelling]
- Model and texture all farms (4 types?)- (right now we are using valve building assets)
- Make the building helper functional for our use, right now its missing upgrade features, client side particles, blacklist areas and sometimes fails to place buildings - https://github.com/Myll/Dota-2-Building-Helper
- Fix all remaining items and spells - (80% of these are completed)
- Building units are done besides a few issues. (Upgrading not implemented)
- Map is mostly the same as the wc3 version, slight differences in area sizes due to different tool resolutions - (Props/decorating needs to be done)
-
Hi guys, I'm back again. Valve have added more features allowing me to continue with sheep tags development.
You can add me on steam to test the map at this link using the Reborn Dota 2 client.
http://steamcommunity.com/sharedfiles/filedetails/?id=435131884
http://steamcommunity.com/id/A_Dizzle
Some issues:
- Frost Farm doesn't have an attack.
- Shepherd pathing is **** (Valve have to fix)
- No scoreboard
- Debug shop is active
- Sheep can buy from the shop.
If you encounter anything else that needs fixing please comment here.
Ie. A lot of variables, for example build time may be off.
-
(http://i.imgur.com/uMVRerZ.png)
-
You have to be my friend on steam to see it.
-
That's a terrible way to get a game going.
-
That's a terrible way to get a game going.
When you are keeping the game private until launch, it is the only way unfortunately.
If it was public it would be bombarded with bad feedback as it is unfinished, unless that is what you want to happen?
To test it by yourself these are the steps you need to follow:
1. Click Create Custom Lobby
2. Add a lobby password
3. Enable Cheats
4. In-game you need to go into the console and type dota_create_unit npc_dota_hero_riki dire within 30 seconds. This will put a bot on the other team you can play with.
-
I have updated the map yet again. This time with Noya's new building helper. The following have been added.
- Queuing (shift)
- Shop restrictions
- Better building placings
- Shop restrictions
-
Accept my invitation. I want to test the map!
-
Added you and Nmcdo. :3
-
The code is available to look at here. https://github.com/ynohtna92/SheepTag
You can find all the ability, unit, and item definitions located https://github.com/ynohtna92/SheepTag/tree/master/game/dota_addons/sheeptag/scripts/npc to see if i have done anything wrong.
Note:
- tiny farms don't work yet as valve do not have 1x1 grid blockers implemented.
- stack farms haven't been implemented in BuildingHelper yet.
-
Please please please please please make normal farms {F], Stack [S}, Tiny farm [T], Hard farm [R], Strong farm [D], Upgraded [G], cash [C], and so on like it was from Wc3.
After years of playing with those keys on wc3 I'm sure not only me but others would have massive trouble getting used to the other keys that are currently set.
Edit: Also from what I remember from playing it, there seemed to be a slight delay whenever you tried to build from after you hit the key. Might just be me.
-
Please please please please please make normal farms {F], Stack [S}, Tiny farm [T], Hard farm [R], Strong farm [D], Upgraded [G], cash [C], and so on like it was from Wc3.
After years of playing with those keys on wc3 I'm sure not only me but others would have massive trouble getting used to the other keys that are currently set.
Edit: Also from what I remember from playing it, there seemed to be a slight delay whenever you tried to build from after you hit the key. Might just be me.
I'll add the wc3 keys in as legacy which will be accessible in the options menu of dota 2. Since majority of the custom games players will have not played it on wc3, it does not make sense to change the default key binds to something people will be unfamiliar with instead of qwerdf.
-
Legacy keys have been added. Please refer to the image to see how they are enabled.
-
I just wanted to hear your feedback on what you want to see in this mod. I want to keep things as they were as much as possible but obviously things will feel different or be changed.
I also have some questions regarding some of the features in the game:
- How important are some of the modes? Practice mode, captains, pick, smart, fair, switch?
- How important are runes to the game, are they used in all the game modes?
-
Runes are a fairly new game mechanic that really are debatable whether should be kept or not. The two most important ones are The rune of speed and Invisibility.
The wolf that usually takes the one of invisibility also takes the one of speed.
Pros:
-Limit's wolves from spamming invisibility
Cons:
Sheep can usually predict when invis is used because it spawns in 1:30 into the game and the sheep will only die to miss-play of risky massing when they know of invis.
Versus Mode (2 Captains pick players) is used for a competitive format so it is pretty important while, for normal play I'd say the most important ones are -random, -smart, -fair, and a runtag mode some people would like. (Runtag mode usually is -view (you can see the whole map) -gold 9999 0 -time 99 -switch x x
I felt like when I was playing there was a strange delay when I hit the key from the time I hit the key to the time the farm appeared. When I tested it.
Edit: Also, you should not be able to build inside the pen.
-
By delay do you mean from when the ability was cast it took a while for the farm to show on the cursor? Or when the sheep went to the site to build it.
As for the blight area, that feature is in progress with the next building helper library by Noya. So as soon as that is done the blight in the pen and stack farms should work as normal.
Runtag is all those commands? (-view, -gold 9999 0, -time 99 (round time?), -switch x x)
-
predicting when invis comes doesn't matter. I snipe kids out like its my job.
-
Those commands make up Runtag, yes. -Time is how long you play. -Switch x x x. The first x is how many seconds of protection the sheep get after switching, the second is always set as zero. The third is how many spirits in the pen that can be saved spawn. View let's you see the whole map, and gold for sheep to build any farm while runtaggimg.
-
Also the delay is to the curser, seemingly.
-
What do you mean by switching? Switching from spirit back to sheep?
I can't notice any noticeable delay to the cursor, it seems almost instant for me. Have you got a video?
-
When you play runtag mode, you switch between Sheep / Wolf whenever the wolf kills the sheep. Don't have a video of the delay. But I checked again today and it's definitely there. Not sure what it is. It mainly happens when you try to make two farms at once, the first one will appear, then you will not see the second until it's already slow massing speed.
-
Where are the strings stored for the Quest log? I see the Jass triggers but have no idea how to access the values it represents. If you can access it can you pastebin all the quest logs?
-
Hi
Just made this account to thank you guys for putting effort into this port, and wish gl with it. ^^
Can't wait to play some sheep tag!
o/
EDIT: Can I test the map?
-
Yes you can add me if you would like to test the map, please tell me who you are before hand since I get a lot of request.
The map is currently on hold again. I'm waiting for valve to fix two issues which currently make the game mode unplayable. There aren't currently any 1x1 gridnav blockers so I can't implement tiny farms correctly. And pathing for units with a custom hull size is broken.
-
Still waiting on volvo for the fix for custom hull radius pathing but in the mean time this map requires assets!!!!
We need artwork for the workshop image which has to be 536 x 402px minimum.
And then we will need a load screen image 1920x1080px.
Also If you want any sound clips to be added for a specific action link it here.
As for any other art work or ui design recommendations please comment below.
-
Finished!
Go play it at http://steamcommunity.com/sharedfiles/filedetails/?id=435131884!
-
Sweet I'll check it out when I get to my computer!
-
So I played a_dizzle's version and I must say that I actually like it (played it for 5 min alone).
Some of my comments:
- Shift-building is way too good. When using shift-building, you shouldn't be able to move to the next building until the current one is done (that's how it works in wc3.
- Even though I know it won't be the same as wc3 revo, please keep the same hotkey;) If possible, it would be nice to have only one build menu.
- When trying to build a farm, if you hover on something (ie a farm), the grid should appear red and not green
Could be nice if some other people would install dota 2 and play with us
-
Shift-building is way too good. When using shift-building, you shouldn't be able to move to the next building until the current one is done (that's how it works in wc3.
Lol bruh... That was a fault in WC3, not a benefit! Good shift building should definitely be a thing.
I may try it out if it's totally free and stuff. Any special directions or is it just straightforward download Dota 2, run, play Sheep Tag?
-
There are no special things needed to run custom games in Dota 2 any more. Just dl and launch the game and download the custom game from the 'Arcade' to play.
I am aware of the hover issue over other farms and it is a problem with syncing the client to the server that I haven't done yet due to possible performance issues.
Legacy keys should work if you select them in the options but it will fail to work for units. Only heroes get legacy keys given to them. Valve please!
Yeah it would definitely be nice to get some more players! And some videos so we may get more appreciation for the game. It is currently at a 61% positive rating compared to my other custom game Slide Ninja Slide which is at 80%. I asked Baumi to give it a play through and he didn't like it due to 'its just impossible to do anything as sheep, one side chases the other'. I assume he had never played sheep tag in wc3 and doesn't understand how the game should be played.
-
You're basically going to have to do with Sheep Tag what should have always been done: depending on the skill level of the players on either team, change the gameplay.
-
I agree with 008 that shift build shouldn't be faster than actually massing, or even really comparable, regardless of what was intended in wc3. It would make blindly massing even more boring/easy than it already is.
About legacy hotkeys, I'm guessing there is some reason sheep cant just be made into a hero unit?
As for people finding it boring, I'd suggest just adding some fluff (more interesting farms/items/abilities) that can be deactivated if desired.
Downloading dota 2 now to give it a shot.
Edit:
After testing, just as sheep, it feels and looks pretty nice. A few issues I noticed:
-After building a stack farm and deleting it, I was unable to build any other farm in that location ever again (I used C to delete).
-The sheep can build from too far away. I didn't test with a shepherd, but I'm fairly certain that farms can be built outside the mass without ever being able to be attacked and I know farms can be built over two farms that meet at one corner.
-Turn rate is really fast, though this is just preference.
-
The Sheep are heroes, the farms are not though so legacy keys won't work on them. Hopefully valve fixes all the hotkey issues when they eventually release their panorama hud ui.
What do you mean by building over two farms at corners? Units in dota do not care if a building is in their way so they will try to attack through it. I'm not sure how to remedy this since if it is too small they won't be able to attack large buildings, since they attack at the center and not the edge.
I'll check out those issues thanks!
-
Hmm.. Hello old friends. Guess I picked the right time to try and reconnect with the ST world again. Excited to try the map.
Add me on Steam: JoshRoo "Rival"
-
(http://imgur.com/JPlgCU8)
http://imgur.com/JPlgCU8 (http://imgur.com/JPlgCU8) (Image isn't working for me)
In the first part, I built the two farms that touch at their top right/bottom left corners, then was able to build the third over them.
In the second, I was able to build the farm not in grid from the position the sheep is in, where the shepherds shouldn't be able to reach it.
-
I don't understand the second one. What is wrong? You should be able to build it like that.
-
The problem is a sheep can build a farm like that without ever being vulnerable to attack. This would make it extremely difficult for a shepherd to prevent a sheep from escaping isolation.
-
I also cannot fix the stack farm issue until this is fixed. I tried to fix it but placing and removing a stack farm allows you to place another farm in its place even if it is visibly blocked.
http://gfycat.com/SpeedyFickleCat
http://dev.dota2.com/showthread.php?t=188073
-
I agree with madhatters point and I forgot to point it out in my first post. Currently you are able to build a farm from way to far from your sheep. When building a farm you need to touch the point where you are building so I shouldn't be able to build a farm from x distance from my sheep.
-
I'd much prefer being able to build on the place the stack farm was placed even if you shouldn't be able to than not being able to build in unknown places.
That being said, can you force all farms within X range of a killed/removed stack farm to be repathed? or even just replace them with a new farm? Not the best solution, but better than the other two options. Or maybe just remove the stack farm entirely... it isn't a huge deal to not have it and was even removed from revo for a while.
-
http://steamcommunity.com/sharedfiles/filedetails/changelog/435131884
I fixed the map edges so they don't hide units anymore.
Adjusted the values of the cast range of buildings to get them as low as possible. You still can build through if you are right in the corner but it should happen less likely. I could make it smaller but it makes the new sheep position wrong when you build it from a corner.
Stack farms are fixed however you can now build where they originally were even if there is a building in the way. It will be like this until valve fixes the issue. However i could reapply all gridnavs in area to maybe fix it. I will see how important it is to be fixed.
-
I really am a fan of the map and the effort. I just dont think people that have played st so long on wc3 could learn to love the style that the D2Reborn map has to offer. I get that it is new and still not 100% but there are alot of small things that i dont think can be mimicked. Things like the small cliff jumps or front builds to juke wolves and stuff of that sort. Its gotta just be stuff like open run jukes and backbuilds to stuff wolves when breaking free from a mass of farms
-
I took a look at the code you're using to build, and it looks like it'd be pretty easy to fix the range issue with building from corners... you just need to check distance from any part of the farm (rectangle/square) instead of distance from the center of the farm (point).
So the code in buildinghelper.lua should be changed from:
builder.state = "moving_to_build"
local distance = (location - builder:GetAbsOrigin()):Length2D()
if distance > castRange then
return 0.03
else
builder:Stop()
to something like:
builder.state = "moving_to_build
local pos = builder:GetAbsOrigin() //Some sort of point object?
local x = max(location.x - castRange - pos.x, 0, pos.x - (location.x + castRange))
local y = max(location.y - castRange - pos.y, 0, pos.y - (location.y + castRange))
local distance = sqrt(x*x + y*y) //The distance from the builder to any part of the farm
if distance > [Some constant value] then
return 0.03
else
builder:Stop()
The castRange variable would need to be half the length of whatever type of farm being constructed so that the (location.x/y +- castRange) gives the min/max x/y of the farm being built. I got the distance from point to rectangle formula from http://stackoverflow.com/questions/5254838/calculating-distance-between-a-point-and-a-rectangular-box-nearest-point (http://stackoverflow.com/questions/5254838/calculating-distance-between-a-point-and-a-rectangular-box-nearest-point) if you're interested.
Never used this language so I don't know the syntax and all, but it should work (also possible I have no idea what I'm doing and am not even in the right section of code).
-
That won't fix it. It needs to be so that it is the pathing distance to the center. Otherwise it will build through buildings. I'll leave it as is for a while though since it is so minor, and will take some time to fix.
-
I was looking to fix the not having to come out of the mass to build part (when I said building from corners, I was referring to "I could make it smaller but it makes the new sheep position wrong when you build it from a corner"), like you said, building over corners isn't a big deal.
Edit:
Thinking about it now, wc3 actually bounces your sheep if you build from the corners (easy to see on hard farms), so it might be best to just lower the cast range like you were saying/did.
-
Anyway most issues should be fixed as good as they are going to get for now, so if someone could get a full stack together and play and record a game that would be great!!!
-
Okay so we played around a bit. Found some noteworthy issues:
Wolf attack animation speed is way too slow. You can't kill a sheep that is running away from you, even if it's standing still when you start the attack.
Wolf range is way too high. You can swing on sheep well inside of their mass (not that you'll hit them ^^).
A lot of the items don't work (guessing you know this), and necklace of immunity causes wolves to miss 100% of their attacks.
A few things about mirror image: The ability animation is really slow, so using it to jump farms for a kill would be basically impossible. A couple of times the ability resulted in the shepherd/its illusion being stuck between farms. Images are kinda hard to tell apart from the real unit. Using image at the bottom of the map while facing down caused my image to go on for a really long distance before eventually (like 10 seconds) just teleporting me back to the same spot, though this may have been because i was surround by some farms at the time. The ability can also travel over hills.
Frost farm missiles can't be dodged, making it seem overpowered. I'd also like to see some sort of construction animation if possible.
Sheep speed is pretty low (360 vs I think 380 of revo), very noticeable when runtagging.
When building a farm, the sheep sometimes seems to teleport the last little bit to the farm, might just be a visual thing, but it looks choppy and sometimes the sheep still stays pretty far away. Hard farms and wide farms can be built from very far away.
008 disconnected during the start of one game and then reconnected after I spawned as wolf, but wasn't allowed to play and the round didn't end, so we had to restart the game (a -cancel or -restart within the first minute or so would be nice).
It's really hard, for me at least, to see where you can build/walk near the edge of terrain (especially since the buildings don't always turn red when they can't be built). And you can jump down/up levels of terrain really easily (not necessarily a bad thing).
Sometimes an attack command on a standstill sheep caused the wolf to run back and forth in the same area indefinitely.
-
Wolf attack speed should be the same as it was in 7.7.4, so should the sheep movement speed. I'll double check these though.
I'll try to reduce the attack range of the wolf but it can only go down a small amount since it is already struggling to hit large buildings on corners without running to a side. Also if the sheep isn't within 185 units from the shepherd when the attack lands it will miss.
Which items don't work? They all should work correctly. I'll check the necklace item.
How fast should the illusion move? It currently moves at 500 units per second. I wasn't sure of the mechanics for the illusion so as long as the terrain is blocked it will continue to move forward until it finds a spot or 10 seconds. Should it not go uphill? The illusions should be blue but when I add a nodraw to it for illusion stuff it gets removed :(.
Frost projectiles are tracking are they not?
I'll look at the revo version again to see how far large farms are built away. It is hard to build rts mechanics in a game that isn't one.
I'll thought I fixed all the disconnect issues. I'll add a reset command if i can't figure it out.
Edge of terrain? The buildings will always turn red when it is hovering over in-traversable area.
Attack command thing is a pathing issue valve has to address with larger units.
-
Wolf attack speed should be the same as it was in 7.7.4
Both attack speed and back-swing animation delay matter. I think the back-speed is something like 300ms.
Also if the sheep isn't within 185 units from the shepherd when the attack lands it will miss.
That seems like a random number. It's either 192 or 224 in WC3 (check range motion buffer, which I think is added to the range to determine how far the unit can be when the attack lands).
How fast should the illusion move? It currently moves at 500 units per second. I wasn't sure of the mechanics for the illusion so as long as the terrain is blocked it will continue to move forward until it finds a spot or 10 seconds. Should it not go uphill?
You can check the speed in WC3. It'd be listed under the ability. The mechanics for imaging is relatively complicated. First, it scans the area from origin for a viable area to place the unit, up to a max of like 7.5 farms (scanning in a spiral grid). Then it checks if the area is the same level as the source. If it isn't, it then scans from there up to a max of 5.5 farms for a viable area of the same level. Illusions can only land up hills if a space becomes occupied while the image is trying to be placed. The Bulldog map that makes testing image physics relatively simple (the one with the long hill jump).
I don't exactly remember all of these numbers and don't have WC3 installed anymore. But they should all be verifiable.
-
http://steamcommunity.com/sharedfiles/filedetails/changelog/435131884
I have updated the following:
- Fixed spell immunity causing you to not be able to hit sheep
- Adjusted attack range to hit sheep from 185 -> 224
- Adjusted far sight from 600 radius to 1200 radius
- All Shepherd aura items are global now instead of 600 radius
- Adjusted sheep movespeed from 360 -> 380
- Adjusted movespeed of golem from 420 -> 300
@Chakra do you have a more detailed psudocode or code for the mirror image function? I can't seem to find it.
I didn't change the attack rate since it should be fine at 1.5 secs, animation back-swing points at 0.3 secs do not affect the overall attack rate of 1.5.
I'll try to fix up the range of units so building is closer as well as the distance shepherds can attack from but it may never be the same as it was in wc3.
For future reference all game definitions for items, abilities and heroes/units can be found here https://github.com/ynohtna92/SheepTag/tree/master/game/dota_addons/sheeptag/scripts/npc
-
The attack rate is fine, the animation delay is the problem. From your hero file:
"AttackAnimationPoint" "0.6" // Normalized time in animation cycle to attack.
//"AttackRange" "150" // Range within a target can be acquired.
"AttackAcquisitionRange" "150" // Range within a target can be acquired.
//"AttackRangeBuffer" "74" // Extra range the target can move without canceling the attackI'm not sure why range/rangebuffer are commented out, but it seems like a shepherd can start an attack on a unit within 150 range, and takes .6 seconds from starting an attack to actually dealing damage. During this .6 seconds, the unit being attacked must stay within (150+74 = 224) range of the shepherd. Sheep move 360 * .6 = 216 range in .6 seconds, so if they are moving you are never going to hit them. The values in wc3 are:
"AttackAnimationPoint" "0.3" // Normalized time in animation cycle to attack.
//"AttackRange" "32" // Range within a target can be acquired.
"AttackAcquisitionRange" "650" // Range within a target can be acquired.
//"AttackRangeBuffer" "192" // Extra range the target can move without canceling the attackAcquisitionRange in wc3 meant the range a unit would recognize an enemy and automatically attack them (if they weren't a worker or doing something already), not sure what it means here.
Frost farms didn't track in revo, they had 125 area splash damage on a point target (600 missile speed, 600 range).
Is it possible to have two attacks? One for sheep, one for buildings?
Images travel at 1000 speed, and I think the pseudocode is something like:
p = casting point
level = terrain level at p
p2 = p offset by 128 range 90 degrees from casting unit facing angle
level2 = terrain level at p2
perform some sort of BFS from p2 on cells to find the nearest point at which a shepherd sized unit can be placed,
ignoring all cells that aren't at terrain level <level2>, and breaking after cell distance from p is > some value
(maybe 1280ish).
p3 = the point returned by BFS, or p2 if no valid cell was found
if (level != level2)
perform the same BFS, except from p3 and excluding cells not at terrain level <level>,
the point found is the final destination or there is no valid destination.
p3 = point returned by BFS or null if no valid destination
else if (p3 is not a valid cell) // The first BFS didn't find a valid cell
p3 = null
If (p3 == null)
put shepherd at the original spot (p), I guess. wc3 had some sort of image cannot be created error that might apply in this situation
else
start the missile animation towards p3, constantly checking to make sure p3 is still a valid location, and BFSing from p3 to find a new valid location if it isn't.
repeat for the image (mark p3 as occupied in the image's code)
(Emphasis on pseudo, and should be double checked)
PS. I would argue for making the image always get sent to the nearest of the two final destinations so the real would always be the one traveling over farms/jumps/fences.
-
I'll fix the frost tower and image next but here you go for attack range, attack speed and hull size changes.
http://steamcommunity.com/sharedfiles/filedetails/changelog/435131884
The attack rate is exactly as it was in wc3 but it feels soo much faster. If you ever get swung at you will never be able to run out of range.