Messages

16

Suggestions / Re: Base defence - lights out

« on: September 30, 2013, 08:45:35 pm »

well... riddick didn't need big eyes to see in the dark

17

Programming / Re: A bug in battlescape and a fix for it

« on: September 30, 2013, 08:14:45 pm »

well I will not let my words to be manipulated into some shit, sir.
you didn't answere my question tough. do that instead of manipulating, or taking an "awe" on what you seen no-one else doing before.

18

Translations / Re: Polish translation

« on: September 30, 2013, 03:52:19 pm »

sprawdzę to wszystko w grze, patrząc na inne nazwy w odpowiadających kontekstach wątpię, żeby wystąpił problem z długościami stringów, odezwę się jeszcze dziś wieczorem na ten temat tutaj na forum.

19

Programming / Re: A bug in battlescape and a fix for it

« on: September 30, 2013, 03:38:50 pm »

https://www.youtube.com/watch?v=1Rqx8s2qKXM
he dont look like terrain noob ;P

nice material for people that don't know broodwar, but I don't get what you are relating to tbh. if you play on iccup then we can perhaps play some games - but be aware I haven't been playing for a really long long time, so I don't know if I will be able to match your skill nowadays.

If it will prove itself, it will become the default one

that is one of the first logical comments I have found on this forum, thank you for that. I encourage others, to do the same.

I will start from fixing the code around reaction fire as, to my surprise, it still isn't fixed, and the latest fixes were revoked, despite I've given pointers on why was that failing. I will post all the code on the forum with my commends for a review that everyone can take part in. Can you please tell me if the compilers used by the team support c++11 features ?

YET:

So, you're unhappy with the 2012 X-Com, you are unhappy with the current RNG OXC uses, and you seem to be extremely unhappy with the fact that your ideas don't seem receive the expected approval from the community.

extreme manipulation in these generalizations. let's apply a bit details so that your cheap marketing style wouldn't look as meaningful as they do where the information is insufficient.

So, you're unhappy with the 2012 X-Com

it's hard to say that I am happy or unhappy with it. I think it is a bad product - sure - because it's a bad realization of the original. it's overly simplified. I have commented the way it actually works, with the way it is supposed to work according to what's said about it.

what I am unhappy with is that bethesda has bought the company that was making a really neat remake of the original version and then closed the project, because their managers have expected a great game project to be closed as fast as their regular shit projects.

you are unhappy with the current RNG OXC uses

really? I wander how did you develop such an assumption. all I actually did was comparing what was written on the forums on the functionality and it's current implementation, and signalized a discrepancy, simply not accepting a "well, it's not a bug... it's a feature kind of bug, let's call it a bug-ture" response, which I don't agree as a right approach to solve problems.

and to note: changing the implementation of the RNG OXC uses was proposed by SupSuper, which I actually agree with ( so it worked in the different direction than you have implied it to be ) and I have only proposed an implementation which would be the mersenne twister, and I still keep my advice on that one without a slight difference : use mersenne twister, best if it's taken from the random.h from the new c++11 standard headers - there are two to chose, and very easy to use.

and you seem to be extremely unhappy with the fact that your ideas don't seem receive the expected approval from the community.

I'm not making my logic out of my emotions - I'm not a woman. I'd advise others do the same, it makes communication a lot easier. If you were reading the posts carefully, you'd notice one of the first posts on that forum I have posted have a comment, that I only PROPOSE and it is up to the development, because it is their product. I have my own private project, where I make all decisions, and I have my work projects (I have more than one under my supervision) where I make most decisions. believe me I have a lot of things to put my hands on, and very little time to catch some sleep ; )

20

Programming / Re: A bug in battlescape and a fix for it

« on: September 29, 2013, 11:39:22 pm »

you sound like a terran noob to me, if I am to be honest. having goliaths it is simply impossible to lose vs carriers unless the terran gets surprised by a swift switch, which is basically a consequence of "fruitful" thinking "I will so 200/200 3/3 him in one push lol ;3" which hundreds of terrans have for whatever reason, which basically conflicts with their cry about terran being the most impossible difficulty level race to play ever.

to keep your post from being a manipulation I will just add that interceptors that did not come back to the carrier weren't repaired, hence being easier to kill. they also cost quite a lot. I also find the protoss to be ultimately silly not to use an optimisation like this, and politely released the interceptors one by one when their armada is facing the whole army of the enemy. I recommend visiting the TL page. the thing with firing "outside the range" is that when the carrier sends interceptors it's as if a projectile was shot. the same thing happens if a dragoon shoots a vulture that runs out of its range - sometimes the hit happens almost at the distance twice the range. it even happens with zealots sometimes, making an impression of "ranged zealots" when the marine dies from the second hit when he is a bit away.

I could name an actual bug, when at times sunken colonies can hit a siege tank in siege mode despite it is outranged in reality, and that is due to the whole attack being done in 3 parts in the engine.

I wouldn't call sc2 that is a grid based game a FIXING 'thing'. making terran able to outproduce the zerg, when the zerg is on more bases, making the zerg the least mobile race ( excluding mutalisks ).

ending the off-topic : I fail to understand how a broodwar terran cry, or bringing up blizzards mistakes addresses the problem. bugs exist while development, and only a few categories still remain bugs after the game is shipped. only the obvious things. those are things like crashes, denials of part of functionality ( such as being unable to finish the game, or advance a certain branch of the plott ) or documented behaviour that should be guaranteed ( such as when your manual sais A and your game does B ) or when the case just looks like shit - for example in one of the versions of ufo I've been plaing on an amiga where there was no checks of the caps of soldier stats, and after one mission my soldiers time units extended above 255, the unsigned char was overflown, and the soldier had 1 time unit making all of his attacks cost 0 time units giving him the capability of destroy everything that was before him if you only have given him a laser weapon

and the reason for that is trivial and very pragmatic - you actually have to prove something is a bug. how will you prove something is wrong if it is undocumentet?

21

Open Feedback / Re: Promotions

« on: September 27, 2013, 03:55:34 pm »

nice option

nice option? THAT would be bloody ingenious! you sir should have been involved more in designing the product.

22

Programming / Re: A bug in battlescape and a fix for it

« on: September 27, 2013, 03:53:26 pm »

well I've seen and played the 2012 ufo, and I think it's a bad product. now, there is a substantial difference between bug abusing, and cheating. if you change the save - yeah, you've cheated. but when you are utilising the existence of the bug you do what is perfectly legal in the world of the game. its like pushing units through minerals in starcraft broodwar, or pushing away stasised units on the ramp with your workers in that game. it was legally used in professional tournaments, because it IS a feature of the game.

and as it is with 2012ufo, this isn't actually working like that. i didn't play too much of 2012ufo, because i get no fun from it and deem it to be shit and nothing more, but my coworker did play a lot, and he used the "load and retry" tactics, and with the same set of moves he got different results in the end - it sometimes took up to 20 tries, but it DID NOT work as designed - there is a bug, and anyone can abuse it ; )



I'm not pushing on this solution - i've just signaled that it is there, and what are it's consequences it is up to you guys to make decissions on your product. I do sometimes abuse this bug in critical moments, because it is a part of the game.

When you have a truly random number generator

you NEVER have one. no random generator created and working in a deterministic environment ( on a processor ) will EVER be trully random. All software random number generators are fully deterministic. to get a trully randomized sequence you need special hardware - usually this is a (quite big) box, with radioactive material and a geiger counter inside. those things and nothing more can provide you with a trully random sequence of bits.

exploit every weakness of the system

really? how about mind controlling mutons, and giving them laser pistols to artificially boost training of your soldiers? this is also bug using - as far as I remember the ufopedia remakrs on mind controll sais that you will not be able to access the alien's inventory.

another bug abusing scenario is making the alien drop his weapon - he will never pick it up which is simply stupid but it is a bug that facilitates the game so everyone accepts it without a word of protest

another example:
time relativity is not conserved. you can have 10 guys, one spots an alien and spends half of his time units on that (hence logically for the half of the turn there was no knowledge of the alien standing where he was spotted), then you take all the rest of the 9 guys, shot out all their time units ( hence, not preserving the parallelity of the actions in a turn ), and then use the other half of the first guys time units, so that he isn't reaction shot by the alien.

there are numerous things like these, and none of them can be called cheating - bug is still a feature of the game and describes the world's game. no more, no less than that.

there are numerous "tricks" like those. abusing the weaknesses of the enemy and using features of the game to get better results is a winning strategy and will  be used.

Bottom line:
Every system has a weakness, but if I have to choose between these two, I would happily go with the current implementation.

23

Open Feedback / Re: Promotions

« on: September 26, 2013, 11:21:26 am »

good skills

well it was one of the lesser skilled soldiers - my 8 best are on an avenger. he had 71 accuracy and 45 reactions which is weak vs the weakest guy in the transport who has something like 74 accuracy and 84 reactions the guy in the base couldn't possibly have his secondary skills increased, because he didn't use his primary skills. that soldier was also inferior in terms of psi.

and as it is to the extended promotions reasoning by one of my predecessors - I wouldn't give a captain or colonel to a guy who has been in the field once, while there's an open war between two worlds, and the first squad has like 30 missions and killer stats compared to him. who would have followed such a captain or colonel ? I wouldn't.

I will take a peek on the code, perhaps there's a bug hidden there that I will notice.

24

Fan-Stuff / Re: Tactics to survive early games

« on: September 25, 2013, 11:19:36 pm »

I'm playing on genius (probably seem to be a newbie among all the guys playing superhuman), and the third UFO that came to earth was a sectoid battleship, that attacked my base I have even captured a commander, but he died because my alien containment was few days short to finish building...
I always was on low cash, because I think that banked money is wasted money. I spent as much, as I could if it didn't get me into loans.
what I've done is:
a) get some good soldiers -> I prefer reactions over fire accuracy, and that made me profit really well. I just hired some soldiers, fired everyone below 45 reactions, and hire some new guys. I quickly got some nice guys for the start with reactions above 50 each, and not bad shooting skill. I took 8 of my best soldiers to the skyranger, counting their value as 3*reactions + fire accuracy
b) took rifles for each of my guys except the person with the rocked launcher
c) got laser rifles equipped - this is almost all I actually needed so far throughout the whole game
d) the better armor you have, the easier it gets
e) I always used the least powerful weapons that do the job - you need to get more hits to kill the aliens, and have more chances for reaction fire, skills go up really fast. an exception is when my base is attacked, or when I attack an alien base.
f) I've made 2 additional bases quite quick, but I doubt any of them in the first month
g) at night you need tons of flares ( I had 3 per soldier )
h) grenades are a blessing. smoke grenades, high explosive and proximity grenades are really helpful, but I always used them as a last resort (except for smoke grenades, which are helpful if you don't want to get killed while getting out of the craft. when your reactions are nice ( for genius this is like 85 and more ) you don't need the smoke's anymore.
the rest I've done probably as everyone else. in my game it's july, and I have had a score like 13k.

25

Translations / Re: Translation tools

« on: September 25, 2013, 10:56:17 pm »

@SupSuper:
does this tool support handling debatable propositions in translations? I've just tried out the polish language version and some things are really off the original meaning by a lot. can I use the tool to propose a change, or the proposition should be done on the forums ?

26

Suggestions / Re: a thought from a c++ software developper

« on: September 25, 2013, 10:51:57 pm »

well that is nice I haven't had a chance to take a look on that part of the codebase yet. I'll check that out the first thing when I'm back on my feet with health stuff - now I can barely sit after work, because I'm sick.

what I like most in template metaprogramming is using SFINAE to make compilation time conditionally chosen classes based on specified types properties to generate safe inlined recurrent structures, so that you can make your more advanced templates always produce safe results. unfortunately this is very very rarely required. its fun to write these basic SFINAE structures on your own, but many useful cases are already implemented in boost.

27

Fan-Stuff / Re: Any pointers for learning C++?

« on: September 25, 2013, 10:36:35 pm »

@moriarity:
first of all, I'd like to thank you for your response.

what I've written here is related strictly to the post. the proposed way to learn to code in c++ is inferior, and leads nowhere near the basics that are actually required to call an individual a qualified c++ coder.

...complete re-design would actually help this project...

please take a re-read of my posts, and you will find yourself wrong on that.

the main coders of openXcom are not coders at all

a coder is not the same thing as a qualified coder. for example a qualified c++ coder easily handles tons of problems that simply couldn't ever occur in this project, thus which can't be learnt here.

But I do know that what you are doing right now is claiming that the code is all wrong

you "assume", you do not "know", and please quote anything in this post, or any other of my posts, that proves that statement
to add even more: I never said anything like this - the code does approximately what the users want, but its quality is really poor. (approximately, because there are still bugs, I've found some really quickly, tough one of them did not reproduce on other machines, but I still have one or two to post with saves)

and yeah, I can lie as well that the code is perfect quality, and its design is really intuitive if you think this will help out anything. I have never believed in the "don't worry timmy, you didn't lose. you're just the last winner" attitude -THIS does nothing but helps bad things corrupt into worse

reaction and psi attack logic is in the TileEngine class

and here I don't think you actually understand what stands behind these words. here's an projection of that situation to real life:

imagine you have a vacuum cleaner, and a blender. now imagine, that in order to use the blender you must bring the vacuum cleaner every time, open its dust container and that deep there's the button that turns the blender on.

now please tell me, how intuitive that is, and how a new user would be able to swiftly start using the blender without a manual

answering another point from what you've written:

complete re-design would actually help this project.

of course it wouldn't help this project, because a complete redesign basically means a rewrite, and it is a widely known fact that the deathrate of projects that get rewritten is tremendous.

You are necroing threads just to press your points.

I'm not. if you read carefully, you'd notice that I start being cynic when something "fuzzy" appears. I am not tolerant to lies, that is all. AND I really respect truth, such as you expressing your thoughts on my posts.

when I see a sentence "well, we didn't use polymorphism, because <something that makes sense as a justification>" then it's a valid decission that was made thoughtfully, but when I see "well, we didn't use polymorphism, because it doesn't line up with game development" then it's inavoidable, to recieve a proper comment to that.
false things are false, and insisting on them being true means either that someone is wrong, or is intentionally lying.

finally, if you think that speaking the truth is ill tempered, bad attitude or insulting, then please consider your honest words, as the same kind of temper, attitude or tone.

last comment:
learning materials need to be as close to being flawless as possible. there are tons of materials posted as "learning resources" that should be destroyed, because of their low quality - even if you take out all the crap that contains errors (in code samples or in the explanations of topics).

I have actually been teaching programming in java and c++ and managed to make people write their final projects on their own despite they were starting from different programming paradigms at best, so I know how this looks like. Still, I wouldn't call any of the people I've been teaching how to program in a certain language a  decent, or qualified person.

28

Fan-Stuff / Re: Soldier HiScore

« on: September 25, 2013, 06:28:41 pm »

poor shuji tanikawa... others have to be bullying him/her because of the difference in kills... : (

29

Fan-Stuff / Re: Any pointers for learning C++?

« on: September 25, 2013, 06:27:12 pm »

@MKSheppard:

take Stroustroup books thinking in c++ volume 1 and 2, read them and tweak the code around them. when you aren't sure about something ASK questions. when you are through with the books, understand mechanisms like dynamic and static polymorphism, templates, STL, are keen to pointers, references, understand the difference between them, understand static, const, be aware when you can and when you can't throw exceptions, when to use which cast operator, what is an interface, what is an abstract class, what is the difference between a struct and a class, when you should overload operators and when you shouldn't, what is the 'this' pointer for, and so on...

the next step is to understand the difference between object oriented programming and object based programming - here all the code I have seen so far is object based.

once you know all this, you know the "language" - meaning the important keywords and some simple features, and have a real chance to ask real questions. this is the right moment to start to dig into the code, make some changes and make a code review with someone really competent.

you will also need to lear a lot about the environment around the language, and know something on a little lower level than what the code gives you.

I highly discourage learning c++ based on the code of openXcom.

another thing is do not learn c++ using CLI, as long as the project doesn't contain interops between the two.

OpenXCom is designed in such a way that it is actually intuitive

nope - as a nice example I can give that reaction and psi attack logic is in the TileEngine class, where it should have never landed, as well as the issue with many places in the code that need to be modiffied in order to introduce a new psi attack type.

These days i feel confident calling myself a qualified C++ coder, and it's completely due to OpenXCom.

utterly brave. far too brave in my opinion.

one of the professors on my university told us very wise words on his lecture once "a coder isn't a person who gets the thing done. its the person, that gets the thing done in the RIGHT way".

30

Translations / Re: Polish translation

« on: September 25, 2013, 02:45:50 am »

things I've found utterly odd in the polish translation ( tried it out recently, up to that moment I was using original language version, I just wanted to see how good is the transaltion ).

hidden movement - nie widać ruchu -> why not : ukryty ruch  ?

firestorm - nawałnica -> really? how come the firestorm got related with a "blizzard".

cannon is stranslated as działko ( properly )

but heavy cannon and auto cannon 'ciężki karabin' and 'karabin maszynowy'
this is inconsistent and it doesn't sound good at all.. 'ciężkie działko' and 'działko automatyczne' or 'działko samopowtarzalne' are better in my opinon.

personal armour - pancerz lekki -> "pancerz osobisty"

terror ship - terrorownik -> this made my day, but I don't have a nice substitute, yet in polish it doesn't have to be a single word. it's the same as if you've made "terrorer" as the ship name in english. there is no such word in our language, and I doubt that was a good decission. the english name describes the purpose of that UFO, and so should the polish translation, so as a start I'd propose "transporter terrorystyczny" or "statek terrorystyczny" .

psiamp is an abbreviation of psionic amplifier, so perhaps "wzmacniacz psioniczny" would be a better choice here ?