Wanted: foolishly easy wesnoth
Moderator: Forum Moderators
Just wanted to throw another voice behind the "we need a REAL 'easy' mode" vote.
While I understand people's viewpoint on keeping the game Hard, the fact remains that this is the reason "Normal" and "Hard" difficulty settings exist. For those of us who like to play on Easy mode, please give us a mode that's actually easy! Whether you're just trying to pick up the game, or whether you're more interested in playing the game casually and can't afford to spend hour upon hour upon hour trying to figure out how to safely reach the Dwarven Gates in the paltry amount of turns allotted to you, or hell even if you're just looking for a romp in the park - please try to cater to all players!
That said, the biggest failure-causing issue I run into is the paucity of turns many scenarios give you before Instantaneous Game Over. While it makes sense to restrict the number of turns allotted to you on the more difficult Difficulty Settings, on Easy mode it would be nice if the vast majority of turn-limits were increased by 50% in most campaigns, if they're not eliminated altogether (with a cap placed onto the amount of money you're allowed to carry forward from scenario to scenario, or something, so as to not introduce that imbalance).
Other than that, some light adjustments to enemy/ally hit-percentages for Easy mode might be good. Not sure if the different difficulty-levels use different AI models, but if not then that's another thing that could help out.
Finally, in regards to the your-units-missing-95%-of-the-time problem, that's one of the issues with using statistically random number-generating algorithms. Given a large enough sample-set, even the most unlikely chain of percentages will occur - basic probability says so. I remember reading about one interesting remedy for this, not too long ago... The idea was that instead of doing a completely-random percentage every time you need to compute a percentage, instead you "draw a number from a hat" (or, alternately, "draw a card from a deck"). Basically, let's say you're calculating percentage-chances-to-hit. Acceptable values are 1-100, with the most common percentages necessary being multiples of 10. What you do, is you take the numbers 10, 20, 30, ..., 100 and put 'em into a "bucket". (Obviously, choose whatever structure you like to hold 'em...) At this point, your bucket looks kinda like the following:
Now, you've got a guy attacking someone else. Percentage to-hit is 60%. Instead of randomly generating a number between 1 and 100, you instead randomly pick one of the numbers in your bucket - for this first calculation, lets say you randomly get 40%. This is lower than the needed 60%, so the attack succeeds. You now REMOVE that 40% from your "bucket", leaving you with this:
Now the next time you need to calculate a percentage, you randomly select a number from the remaining numbers in your bucket - notably missing 40%. Eventually you'll empty the bucket, at which point you need to re-initialize it with the data you started with.
While this algorithm doesn't completely eliminate the chances of a single unit's attacks all missing, it does mean that there is a 0% change of a 90%-success-rate attack missing 10 times in a row. (A straight percentage-chance algorithm, on the other hand, gives us a nontrivial chance that this would happen.) If you want to mix it up a bit, you can add more numbers to the pot - for the example above, either use multiples of 5 instead of 10, or even multiples of 3 or of 1, or else double-up on the values (10, 10, 20, 20, 30, 30, ..., 100, 100).
Anyways, sorry for such a long-winded post. Thanks for reading this far!
--- Mark ---
P.S. Fun game, by the way!! =;)
While I understand people's viewpoint on keeping the game Hard, the fact remains that this is the reason "Normal" and "Hard" difficulty settings exist. For those of us who like to play on Easy mode, please give us a mode that's actually easy! Whether you're just trying to pick up the game, or whether you're more interested in playing the game casually and can't afford to spend hour upon hour upon hour trying to figure out how to safely reach the Dwarven Gates in the paltry amount of turns allotted to you, or hell even if you're just looking for a romp in the park - please try to cater to all players!
That said, the biggest failure-causing issue I run into is the paucity of turns many scenarios give you before Instantaneous Game Over. While it makes sense to restrict the number of turns allotted to you on the more difficult Difficulty Settings, on Easy mode it would be nice if the vast majority of turn-limits were increased by 50% in most campaigns, if they're not eliminated altogether (with a cap placed onto the amount of money you're allowed to carry forward from scenario to scenario, or something, so as to not introduce that imbalance).
Other than that, some light adjustments to enemy/ally hit-percentages for Easy mode might be good. Not sure if the different difficulty-levels use different AI models, but if not then that's another thing that could help out.
Finally, in regards to the your-units-missing-95%-of-the-time problem, that's one of the issues with using statistically random number-generating algorithms. Given a large enough sample-set, even the most unlikely chain of percentages will occur - basic probability says so. I remember reading about one interesting remedy for this, not too long ago... The idea was that instead of doing a completely-random percentage every time you need to compute a percentage, instead you "draw a number from a hat" (or, alternately, "draw a card from a deck"). Basically, let's say you're calculating percentage-chances-to-hit. Acceptable values are 1-100, with the most common percentages necessary being multiples of 10. What you do, is you take the numbers 10, 20, 30, ..., 100 and put 'em into a "bucket". (Obviously, choose whatever structure you like to hold 'em...) At this point, your bucket looks kinda like the following:
Code: Select all
10 20 30 40 50 60 70 80 90 100Code: Select all
10 20 30 50 60 70 80 90 100While this algorithm doesn't completely eliminate the chances of a single unit's attacks all missing, it does mean that there is a 0% change of a 90%-success-rate attack missing 10 times in a row. (A straight percentage-chance algorithm, on the other hand, gives us a nontrivial chance that this would happen.) If you want to mix it up a bit, you can add more numbers to the pot - for the example above, either use multiples of 5 instead of 10, or even multiples of 3 or of 1, or else double-up on the values (10, 10, 20, 20, 30, 30, ..., 100, 100).
Anyways, sorry for such a long-winded post. Thanks for reading this far!
--- Mark ---
P.S. Fun game, by the way!! =;)
It would be better to have all of the numbers between 1 and 100 in the bucket, not just the ones divisible by 10. And IMHO, the bucket should be refilled when ~75 of the cards are taken out. But other than that, I think the deck-of-cards/bucket solution might not hurt the game too much...
For I am Turin Turambar - Master of Doom, by doom mastered. On permanent Wesbreak. Will not respond to private messages. Sorry!
And I hate stupid people.
The World of Orbivm
And I hate stupid people.
The World of Orbivm
It all depends on how you want the numbers to work out.
It you have 100 players, each playing 100 games, each of which include 100 battles, you've got 1,000,000 individual battles. What this means, gameplay wise, is that the likelihood is that even the most unlikely scenarios will happen to somebody. For a character with an 90% chance to-hit and 4 attacks per battle, there is a 0.1^4 = 0.0001 = 0.01% chance that he will miss every single attack. Statistically speaking, this will occur once out of every 10000 battles, or so. The problem is, with A MILLION battles worth of a sample set, this will happen 100 times! In fact, with such a large sample-set there's almost a guarantee that Ridiculous Things will happen to somebody.
This is where the beauty of the "numbers in a hat" method kicks in. By having a limited set of numbers that can be drawn, you're guaranteeing that the 1/10000 chance thing will not occur. If your "hat" only contains the set of 10, 20, ..., 100, then the worst-case scenario is that the 90% attack misses twice in a row. (And even then, only if 10 is drawn, then the bucket is refilled, then 10 is drawn immediately again.)
One of the dangers to putting too many numbers into the hat is that you're lessening this guarantee, by pumping-up the worst-case scenario. This applies as well to refilling the bucket after it's 75% empty, or something. May not necessarily be a bad thing, but it's definitely something to be wary of. (Also, the "refilling the bucket" idea plays havoc with the numbers, so that it's difficult to mathematically break down the chances of different things happening - again, not neccessarily a bad thing, just another factor to consider.)
So yeah, run through the numbers as best you can, figure out what's acceptable to you. I'd also strongly recommend implementing a "trial battle" system, wherein you can quickly-and-easily match up two user-input combatants. This would allow you to verify that the outcomes of battles, by and large, "make sense" before integrating it into your codebase. Especially at this late stage of the game, you don't want to be putting something in only to find that under "common usage" the 90% attack now seems to be failing far too rarely! =;)
--- Mark ---
It you have 100 players, each playing 100 games, each of which include 100 battles, you've got 1,000,000 individual battles. What this means, gameplay wise, is that the likelihood is that even the most unlikely scenarios will happen to somebody. For a character with an 90% chance to-hit and 4 attacks per battle, there is a 0.1^4 = 0.0001 = 0.01% chance that he will miss every single attack. Statistically speaking, this will occur once out of every 10000 battles, or so. The problem is, with A MILLION battles worth of a sample set, this will happen 100 times! In fact, with such a large sample-set there's almost a guarantee that Ridiculous Things will happen to somebody.
This is where the beauty of the "numbers in a hat" method kicks in. By having a limited set of numbers that can be drawn, you're guaranteeing that the 1/10000 chance thing will not occur. If your "hat" only contains the set of 10, 20, ..., 100, then the worst-case scenario is that the 90% attack misses twice in a row. (And even then, only if 10 is drawn, then the bucket is refilled, then 10 is drawn immediately again.)
One of the dangers to putting too many numbers into the hat is that you're lessening this guarantee, by pumping-up the worst-case scenario. This applies as well to refilling the bucket after it's 75% empty, or something. May not necessarily be a bad thing, but it's definitely something to be wary of. (Also, the "refilling the bucket" idea plays havoc with the numbers, so that it's difficult to mathematically break down the chances of different things happening - again, not neccessarily a bad thing, just another factor to consider.)
So yeah, run through the numbers as best you can, figure out what's acceptable to you. I'd also strongly recommend implementing a "trial battle" system, wherein you can quickly-and-easily match up two user-input combatants. This would allow you to verify that the outcomes of battles, by and large, "make sense" before integrating it into your codebase. Especially at this late stage of the game, you don't want to be putting something in only to find that under "common usage" the 90% attack now seems to be failing far too rarely! =;)
--- Mark ---
Here are some thoughts on this topic:
Debug mode is not really the same as easy mode. Once in debug, you have put yourself in the god position, and not the player position. You can change anything, declare arbitrary victory, etc. The perception of overcoming a challenge which makes up the concept of game playing is gone. As a result, debug mode is deeply unsatisfying. This is true at least for me in all games I have tried it.
Regarding the need to achieve mastery to play the game. I think I don't agree. There are many single player game experience in the world where most players never become terribly competent at them, but yet they manage to stumble through and enjoy the game. This is more tricky in a turn based strategy game like wesnoth, but since it has such a strong emphasis on the single player experience, I think it is a worthy goal.
There are a variety of interface chances that could also make the game much easier. A visual display of expected damage probabilities in both directions for contemplated attacks would be very handy. How do you represent a curve? Not sure. Colored ranges? Or just use endpoints and let us assume (probably a bit incorrectly) a bell distribution, more or less. Additionally a percentage change of death of the unit attacking and levelling of the unit attacking would be some nice reminders. Maybe bright red when your unit's chance of dying is 60% or higher etc.
I also find the experience point distribution a bit lacking. Very little for attacking, quite a lot for killing. I think flattening this a tad would make the game easier for nonexperts.
------------------
And I really really really hate the triple miss with 70% chance each time. Statistically this is not surprising. But when your attack percentage is "good" watching the little guy just miss, miss, miss and then be counterattacked is enraging. Especially as the number of attacks on given units is so limited. I've heard the arguments that the high randomness is just "part of the game" but I really do not understand them. What I end up doing is after I miss 6 or 8 times in a row I just reload the game. And once I'm reloading I'll just try a few times until I kill it with no damage. It just "feels unfair" and so the game experience falls apart.
Debug mode is not really the same as easy mode. Once in debug, you have put yourself in the god position, and not the player position. You can change anything, declare arbitrary victory, etc. The perception of overcoming a challenge which makes up the concept of game playing is gone. As a result, debug mode is deeply unsatisfying. This is true at least for me in all games I have tried it.
Regarding the need to achieve mastery to play the game. I think I don't agree. There are many single player game experience in the world where most players never become terribly competent at them, but yet they manage to stumble through and enjoy the game. This is more tricky in a turn based strategy game like wesnoth, but since it has such a strong emphasis on the single player experience, I think it is a worthy goal.
There are a variety of interface chances that could also make the game much easier. A visual display of expected damage probabilities in both directions for contemplated attacks would be very handy. How do you represent a curve? Not sure. Colored ranges? Or just use endpoints and let us assume (probably a bit incorrectly) a bell distribution, more or less. Additionally a percentage change of death of the unit attacking and levelling of the unit attacking would be some nice reminders. Maybe bright red when your unit's chance of dying is 60% or higher etc.
I also find the experience point distribution a bit lacking. Very little for attacking, quite a lot for killing. I think flattening this a tad would make the game easier for nonexperts.
------------------
And I really really really hate the triple miss with 70% chance each time. Statistically this is not surprising. But when your attack percentage is "good" watching the little guy just miss, miss, miss and then be counterattacked is enraging. Especially as the number of attacks on given units is so limited. I've heard the arguments that the high randomness is just "part of the game" but I really do not understand them. What I end up doing is after I miss 6 or 8 times in a row I just reload the game. And once I'm reloading I'll just try a few times until I kill it with no damage. It just "feels unfair" and so the game experience falls apart.
-
walker
There is one other condition to consider. In "real life" combat experience means a better chance of survival and damage/death to the enemy. That's why brand new young recruits have such a high mortality rate. I was sort of surprised that gaining experience/leveling up doesn't seem to "harden" veterans. This may seem to be accomplished by increased HP and number & strength of attacks. But this doesn't necessarily provide a better skill at attack/defence. Probability is nice for programmers, but in combat, experience is golden. With all due respect to probability, I still maintain that, all conditions equal, three Knights should be able to kill one Horseman. Even if the Horseman kills one of the Knights, he shouldn't be able to kill all three. At least not frequently. It can happen, but not often. And I wouldn't want to count on it.
walker
walker
Wanted: foolishly easy wesnoth
I also wanted an easier mode (sometimes it's fun just to kick some orc butt 
).
As a "do-it-yourself"er, I've learned several ways to improve my chances to win without changing C+/source code (some more invasive than others...try one thing at a time not all or you won't know what doesn't work).
1. Its easy to edit saved games. Save the upcoming scenario and quit before starting it. Open the saved game in any text editor and edit.
* Give yourself more gold. With extra gold, buy a bunch of cheap units and use them to protect more experienced units.
* Edit your units' traits...make them quick/strong, strong/resilient, etc.
* Give your units more experience so that the first hit landed in next game bumps them up a level.
* Insert "get_gryphons=1" or "have_theives=yes" under [variables]. Gryphon Riders can make that fog/shroud lift faster.
* Add units types under "can_recruit=". Mermen with storm tridents can ease the crossing in Ford of Abez.
* If a favorite unit got killed, copy it from a previous scenario and paste it into the next one. ("Raise the dead" like the bad guys do.)
2. Create your own maps. Add more villages for income. Enlarge your castles/camps to recruit/recall more units. Decrease obstacles for the good guys, increase obstacles for evil guys. Put orc castles/camps farther away so you have longer to recall/recruit a stronger battle group.
3. Rewrite "scenario.cfg" files.
* Use "items.cfg" more. Place more storm tridents or staffs of speed in easy to find places. Liberally "sprinkle" holy water around, especially before undead scenarios.
*Have helpers pop up in turn 3 or 8 (like the theives pop up in Elensefar). Have more "Moremirmu and mage friends" (Valley of Death) come to help out in other scenarios. Create a new Dragon or Yeti (Fallen Lich Point) to slow down the other side. Make the cuttlefish/sea serpents '[AI][target]' attack both sides equally. Have long-lost elvish cousins pop up to see what the fighting was about. Write in a fairy godmother (Elvish Sylph?) coming to the rescue.
4. Edit configuration files. If you change some *.cfg and the executable won't load, switch files back to saved copies of original "*.cfg" files. (or if all else fails, reinstall.)
* Edit your unit.cfg files. Give Konrad or whoever extra hitpoints, more hits/damage or extra movement, etc.
* Edit the "game.cfg" file. Ease movement restrictions (speed up good guys in caves) or add restrictions (make it harder for orcs to cross mountains). Lower recall costs, heal faster, change resistance/defense, etc.
*Edit the "traits.cfg"file (or [trait]'s in `game.cfg' in older versions) Make '[trait]id=strong' stronger ("apply_to=hitpoints increase_total=4"). Eliminate the '[trait]id=quick' penalty (change "-10%" to "+10%").
5. Be creative. Now that you've become more experienced with configuration files, write your own scenarios or a whole campaign. Assemble a HUGE army and eliminate evil.
6. Have Fun!!! This IS a game.
Remember this is Open Source software and creativity is encouraged. Wesnoth developers won't send the "Mycrosopht Police" to your door. Dave or Issac or Turin, etal may even commend your creativity.
Thom
-------------------------------------
to paraphrase a US President:
You can 'please' some of the people all of the time,
and all of the people some of the time,
but you can not 'please' all of the people all of the time.
Abraham Lincoln, (attributed)
). As a "do-it-yourself"er, I've learned several ways to improve my chances to win without changing C+/source code (some more invasive than others...try one thing at a time not all or you won't know what doesn't work).
1. Its easy to edit saved games. Save the upcoming scenario and quit before starting it. Open the saved game in any text editor and edit.
* Give yourself more gold. With extra gold, buy a bunch of cheap units and use them to protect more experienced units.
* Edit your units' traits...make them quick/strong, strong/resilient, etc.
* Give your units more experience so that the first hit landed in next game bumps them up a level.
* Insert "get_gryphons=1" or "have_theives=yes" under [variables]. Gryphon Riders can make that fog/shroud lift faster.
* Add units types under "can_recruit=". Mermen with storm tridents can ease the crossing in Ford of Abez.
* If a favorite unit got killed, copy it from a previous scenario and paste it into the next one. ("Raise the dead" like the bad guys do.)
2. Create your own maps. Add more villages for income. Enlarge your castles/camps to recruit/recall more units. Decrease obstacles for the good guys, increase obstacles for evil guys. Put orc castles/camps farther away so you have longer to recall/recruit a stronger battle group.
3. Rewrite "scenario.cfg" files.
* Use "items.cfg" more. Place more storm tridents or staffs of speed in easy to find places. Liberally "sprinkle" holy water around, especially before undead scenarios.
*Have helpers pop up in turn 3 or 8 (like the theives pop up in Elensefar). Have more "Moremirmu and mage friends" (Valley of Death) come to help out in other scenarios. Create a new Dragon or Yeti (Fallen Lich Point) to slow down the other side. Make the cuttlefish/sea serpents '[AI][target]' attack both sides equally. Have long-lost elvish cousins pop up to see what the fighting was about. Write in a fairy godmother (Elvish Sylph?) coming to the rescue.
4. Edit configuration files. If you change some *.cfg and the executable won't load, switch files back to saved copies of original "*.cfg" files. (or if all else fails, reinstall.)
* Edit your unit.cfg files. Give Konrad or whoever extra hitpoints, more hits/damage or extra movement, etc.
* Edit the "game.cfg" file. Ease movement restrictions (speed up good guys in caves) or add restrictions (make it harder for orcs to cross mountains). Lower recall costs, heal faster, change resistance/defense, etc.
*Edit the "traits.cfg"file (or [trait]'s in `game.cfg' in older versions) Make '[trait]id=strong' stronger ("apply_to=hitpoints increase_total=4"). Eliminate the '[trait]id=quick' penalty (change "-10%" to "+10%").
5. Be creative. Now that you've become more experienced with configuration files, write your own scenarios or a whole campaign. Assemble a HUGE army and eliminate evil.
6. Have Fun!!! This IS a game.
Remember this is Open Source software and creativity is encouraged. Wesnoth developers won't send the "Mycrosopht Police" to your door. Dave or Issac or Turin, etal may even commend your creativity.
Thom
-------------------------------------
to paraphrase a US President:
You can 'please' some of the people all of the time,
and all of the people some of the time,
but you can not 'please' all of the people all of the time.
Abraham Lincoln, (attributed)
-
Elvish Leader
- Posts: 49
- Joined: April 17th, 2005, 8:18 am
Wesnoth is not hard. Pathways into Darkness is hard. Super Ghosts & Goblins is hard (really hard, actually).walker wrote:I joined this forum on June 8, 2004. I had just installed 0.7.10. I'm still playing it, and I just got to Test of the Clans (I will eventually upgrade). Even on easy, this is probably one of, if not the, hardest games I've ever played. This is not how I want to LEARN a game. Little things like one of the AI's level one units killing 2 or 3 of my level one units can be very frustrating. I mean similar units like Horsemen on grassland during the day. For that matter, the AI's level one Horsemen are a match for my level two Knights! That's a little like saying a WW II tank is a match for any current Main Battle Tank.
My point is that I agree that a reasonable way to learn the game with a lower frustration factor is a good idea.
walker
What is true is that wesnoth is aggravatingly frustrating, in that people have a natural inclination to become attached to special units they hire, and almost treat them like rpg characters. And then they die, as is so easy to have happen in wesnoth, and people flip out.
Wesnoth is also frustrating because of two other things - it's often tough to put a unit into a condition such that they are perfectly safe. If wesnoth had this, it would eliminate many of the problems people have with experienced units dying - the worst problem people have is that there is nothing they can do to save the unit. I speak of this with regards to a unit which has only a few hitpoints left.
Even dragging an elf into forest only gives you a 60-70% chance of not dying. That's nuts - there's no way I'd bet my life on that. And what hurts even deeper is that the one means of truly protecting such a unit - locking out all enemy access using ZOC, is often impossible to perform, due to how few troops are on the board, and due to dastardly things like the "skirmisher" ability.
Something has always been rotten in wesnothville - a mistake that other games like warcraft fail to make. Wesnoth is a lot of fun in spite of this fundamental design problem, but that still doesn't make it any less of a problem.
Let me define "persistence value" to be the "worth" accumulated in a unit, based on a mixture of both how much time has been spent cultivating the unit, and how much more powerful it has become.
I'm not sure exactly what the problem is, but I'm getting quite closer to being sure. If you compare wesnoth to Warcraft III, a few things are different. First, it is quite feasible to command your army in such a way that you almost never lose a unit. If you do lose a unit, hey, all of your basic units have no persistence value - you can hire a new unit that is basically identical to the old one. Finally, even if you do lose a unit with persistence value, a hero - you lose none of it, because you can resurrect the unit.
Wesnoth would be a lot less frustrating if its persistence value was not lost so often.
Wesnoth would also be a lot less frustrating if it is was more reliable - in a game like warcraft, certain actions are infallible. Attacks are set to almost always hit (the ability to make your foes miss is actually a special ability possessed only by a few heroes), and the randomness in damage is only about 20-30% of an attack's damage rating. So, if I send a (warcraft) knight to fight a ghoul, one on one, I know, beyond any doubt, that the ghoul will die and my knight will win. Strategy comes of larger troop movements - for example, in a skirmish, I could send a few knights around the central melee to destroy the enemy's artillery. In wesnoth, sending an experienced troop to "deal with" a low-level enemy is not always a winning proposition.
In warcraft, when I have a unit near death, I can do something to bring it back instantly, before it dies. Strategy comes of making that happen as rarely as possible. It's sort of a deus ex machina mechanism, in which you can enact divine intervention on the fate of your units. You don't have to use it, and if you run out of it (a thing which you have the same amount of as your opponent), your units start dying and you start losing. It keeps you from losing any persistence value, though.
Wesnoth's strategy is not inherently better, it's just much more reliant on luck.
-----
We have, however, decided on our game model, and it stays the way it is. I know I'm not going to repeat the mistakes made here, but it's been fun learning them.
Jetryl, don't you think it's more than a bit arrogant to describe the design decisions in Wesnoth as "mistakes" and "problems" and "rotten", rather than just things you personally don't like? I'm very sure that I like Wesnoth more than I would like your alternate design in which risks can always be avoided.
Yeah, I think it comes down to preferences for different types of games, at least to a certain degree.
Some players like games with lots of chance, and having to juggle probabilities, and some don't. Part of the aim of Wesnoth was to appeal to players who feel that 'mainstream' games don't fulfill their needs.
If we had made Wesnoth just like Warcraft, we'd have ended up with a game that would only be played by people who were too cheap to buy Warcraft, and too honest to acquire it illegally.
Also, it's a little hard to compare Wesnoth to Warcraft, since Warcraft is real time. Even a completely deterministic real time game plays out in a way that doesn't seem at all deterministic, due to human error in such situations. A turn-based deterministic game plays out completely differently.
Wesnoth tries to be different, and for that reason some people love it, and some people really, really hate it. I'm much happier that we went down this route though, rather than trying to please everyone -- and in doing so making our game mechanics a boring copy of most other modern games.
Oh, and FWIW I really don't like Warcraft much at all. I think it's really rather boring. (Not that I've played Warcraft III much at all).
David
Some players like games with lots of chance, and having to juggle probabilities, and some don't. Part of the aim of Wesnoth was to appeal to players who feel that 'mainstream' games don't fulfill their needs.
If we had made Wesnoth just like Warcraft, we'd have ended up with a game that would only be played by people who were too cheap to buy Warcraft, and too honest to acquire it illegally.
Also, it's a little hard to compare Wesnoth to Warcraft, since Warcraft is real time. Even a completely deterministic real time game plays out in a way that doesn't seem at all deterministic, due to human error in such situations. A turn-based deterministic game plays out completely differently.
Wesnoth tries to be different, and for that reason some people love it, and some people really, really hate it. I'm much happier that we went down this route though, rather than trying to please everyone -- and in doing so making our game mechanics a boring copy of most other modern games.
Oh, and FWIW I really don't like Warcraft much at all. I think it's really rather boring. (Not that I've played Warcraft III much at all).
David
“At Gambling, the deadly sin is to mistake bad play for bad luck.” -- Ian Fleming
I can confirm that for some players it can make the game much more fun if able to resurrect important units (by playing with a version of Wesnoth patched accordinglyFinally, even if you do lose a unit with persistence value, a hero - you lose none of it, because you can resurrect the unit.
). I'm just in the process of play-testing it a bit (i reduced the cost from 100+unit-cost, to 40+unit-cost, to now 20+33%-of-unit-cost in the last 2 weeks or so). So I can now resurrect a unit which costs 100 gold (e.g. a L3 Arch Mage) for 53 gold. That's very cheap compared to the initial costs I had assigned, so I probably made the game not just more fun for me now, but also more easy Not sure though, it still is hard, but at least I can now sometimes afford resurrecting a unit, without then having no more gold for recalling..I agree about the probabilistic fighting, it's something I also don't really like. But then, I have no idea what consequences it would have on gameplay if it would be changed - maybe that should be my next try at hacking the code? But then, it would be much more difficult to change that, so an exercise better left to the developers, after 1.0 is outWesnoth's strategy is not inherently better, it's just much more reliant on luck.
Believe me, that would be a huge mistake, too - I was merely pointing out some things that they got right, and we got wrong.Dave wrote:If we had made Wesnoth just like Warcraft
You may not saveload, but most players of wesnoth do, and that they would be inclined to do so is a mistake on our part.
-----
One of the really bad things in wesnoth's game mechanics is the lack of things like "hit and run". It's simply not possible in wesnoth, and is part of why the game is so unforgiving. If the attack backfires, you cannot pull out your unit - it is done moving. One might think that having your unit grievously wounded would be enough of a punishment for a foolhardy attack, but no - the unit is pinned there and is at the mercy of the enemy next turn.
Essentially, by attacking, they commit themselves to - not one, but TWO or more combat cycles.
Here's an idea: How about, after a unit launches an attack, instead of losing all movement points, they have 50% of the movement points they had before launcing an attack? For most units, this would be only 1 or two moves - for cavalry, this might be four or five. I think this would add a new dimension to the game. First, cavalry units could hit and run. Second, most foot units could take a step back and let a healthy friend take the brunt of the next round's punishment. That last line, right there, is key, and I think would make the game much more forgiving without harming anything.
This would make certain things like flanking, more relevant - it would also make more traditional battle moves, such as using cavalry to hunt down enemies retreating from a battle, meaningful. (heck, it would make "retreating" relevant in wesnoth, an idea which currently is quite anemic in our game).
Certainly not an idea for pre-1.0, but one that's been brewing in my mind for some time now.
Warcraft I and II are very different from Warcraft III - to be honest, I really did not like Warcraft II. Warcraft II was far too quick-paced, and had an absolutely brutal AI. Warcraft I was a fun game, in its time, but had quite a number of pacing and AI issues.Dave wrote:Oh, and FWIW I really don't like Warcraft much at all. I think it's really rather boring. (Not that I've played Warcraft III much at all).
Warcraft III is as fun as it was, to many people, because its skill envelope was so well-rounded. A novice could still compete, and quite well, but a master could do much better at the same job. The game had a lot of features designed to alleviate the usual "clicking skill" problem with RTS games, which worked very well for novices, but they were not implemented in such a way as to remove the benefits of being good at the game.
One great example was "auto-cast spells". Units such as healers only had up to three spells, and you could set them to do their thing automatically - if a unit was wounded, they would heal them with no input on your part. Units that cast protective spells would do so automatically - also, they added in the idea of auras, effects that radiated based on nothing other than a unit's mere presence. I wish they had gone further down that road, but they went quite far enough to eliminate most of the problems with warcraft II.
Last edited by Jetrel on June 19th, 2005, 10:35 pm, edited 1 time in total.
Oh yeah - one other thing that Warcraft does a hell of a lot better than we do: The average, normally-scaled game with 4-6 players takes less than an hour (typically about 30minutes to an hour). A 4-6 player game of wesnoth takes about ... three hours, sometimes. It can take all night.Playing one on one is fine, but isn't especially interesting.
-
Dragonking
- Inactive Developer
- Posts: 591
- Joined: November 6th, 2004, 10:45 am
- Location: Poland
I think it isn't problem Wesnoth <-> Warcraft, but real time <-> turn based strategy games.Jetryl wrote:
Playing one on one is fine, but isn't especially interesting.
This is a block of text that can be added to posts you make. There is a 255 character limit