Info on Next Patch

[PawelS]

As well as Pitboss (no word about PBEM though).

 

[Zechnophobe]

It may not exactly equal 1/3, but it's an awfully strange percentage to choose if you don't want to imply exactly 1/3. If it's purposefully less than 1/3, call it 30%.

The point, though, isn't the 1/3. It's that the programmers are obviously using integer math with values and fractions that aren't conducive to that system. I think most people expect rounding, not flooring.

3 * .33 = .99, which floored is 0, not 1. Even if the math is correct, it's either deliberately or unintentionally misleading. Personally, I haven't seen any compelling reason for it to be designed that way, which leads me to think it's a mistake.

The real best solution is to just allow for fractional Culture to two decimals. If 3 Culture +.33(3) = .99, that would be 100 times better, would fix rounding vs flooring problems across the board, and in general be more accurate.

Hmm, I feel like there was another game that I played that had these problems, and they fixed it the way I suggest. What was that game again?

Oh yeah Civilization 4.

 

[Lanthanoid]

I haven't read most of this thread, but I see people scrapping over whether beaker waste is important and whether it should be fixed or not.

I'm a software engineer, and for categorising bugs at my work we use two metrics: severity, and priority, measured on a 1 to 5 scale with severity 1, priority 1 being the most severe and important issues, and 5/5 being least severe (trivial) and least important issues.

Normally, very severe bugs go hand-in-hand with high priority - if it is really severe, like the programme crashes whenever you try to do a particular task, it needs to be fixed ASAP. On the other hand, sometimes there are bugs that aren't actually severe, but they still have high priority - maybe an important customer has decided (for whatever arcane reason) that this is a problem for them, and they want you to fix it ASAP, even if it is only a cosmetic issue that doesn't truly affect functionality.

Other things can affect priority too, like how difficult the bug will be to fix (maybe it needs major rework elsewhere, mabye it's a 1 line change) or how risky the fix is (complex system like AI, or simple XML change) or how often it occurs (once in a blue moon after the system has been running for 34 weeks non-stop, or every 2 minutes).

So, beaker overflow.
Severity: 4. Maybe 3.
Priority: 1.

If you use this framework to talk about bugs, you'll find it much easier for everyone to understand where everyone else is coming from.

 

[Buccaneer]

The problem isn't that tech is too fast: it's that everything else is too slow.

Really? You mean you would like to have buildings and wonders built at the same rate (comparatively) as researching techs?? Civ is not meant to be played at one sitting even though that seems to be a prevaling trend.

 

[Abegweit]

It may not exactly equal 1/3, but it's an awfully strange percentage to choose if you don't want to imply exactly 1/3. If it's purposefully less than 1/3, call it 30%.

The point, though, isn't the 1/3. It's that the programmers are obviously using integer math with values and fractions that aren't conducive to that system. I think most people expect rounding, not flooring.

3 * .33 = .99, which floored is 0, not 1. Even if the math is correct, it's either deliberately or unintentionally misleading. Personally, I haven't seen any compelling reason for it to be designed that way, which leads me to think it's a mistake.

Firaxis has always believed that the proper way to convert a floating point number to an integer is to floor it. It's not exactly a mistake. Tradition is probably a better way to describe their approach to math.

 

[Surgeon]

So, beaker overflow.
Severity: 4. Maybe 3.
Priority: 1.

If you use this framework to talk about bugs, you'll find it much easier for everyone to understand where everyone else is coming from.

Your mistake is assuming its a bug. It could just as easily be a design choice (perhaps a bad design choice but then people are arguing a whole lot of civ5's design is bad).

 

[atteSmythe]

Firaxis has always believed that the proper way to convert a floating point number to an integer is to floor it. It's not exactly a mistake. Tradition is probably a better way to describe their approach to math.

Regardless, it's a design mistake to choose a 33% bonus, then floor it, when dealing with small numbers.

I've been playing Civ since the original, so I understand the history. And I'm a developer myself, so I understand integer math. Using straight, truncating integer math with values that are expected to be this small is the wrong choice. It's misleading, and has no place in a UI-facing environment like this. Any non-technical person, and I daresay most technical people, are going to expect rounding. It's how almost every daily interaction with numbers is handled. Especially with a value like 33%. Whoever decided to *.33 instead of /3 should take a serious moment to reflect on what they've done (assuming the situation is as described, which I've not personally tried to recreate)

Edit: I realize that this is a tough argument to make, because man, when you look at it, there's technically nothing wrong with it. What's wrong with it is twofold: 1) requiring this sort of technical expertise of your users without any mention of how fractions are treated in the documentation (that I've seen), and 2) making everything else so streamlined, and information so limited, that assumptions like this are natural. Everything about the game screams accessibility - the lion's share of making an accessible ui/program is anticipating your users' expectations

 

[Abegweit]

Regardless, it's a design mistake to choose a 33% bonus, then floor it, when dealing with small numbers.

I've been playing Civ since the original, so I understand the history. And I'm a developer myself, so I understand integer math. Using straight, truncating integer math with values that are expected to be this small is the wrong choice. It's misleading, and has no place in a UI-facing environment like this. Any non-technical person, and I daresay most technical people, are going to expect rounding. It's how almost every daily interaction with numbers is handled. Especially with a value like 33%. Whoever decided to *.33 instead of /3 should take a serious moment to reflect on what they've done (assuming the situation is as described, which I've not personally tried to recreate)

I'm a developer too, and one who is old enough to remember when rounding would cost 30 or 40 machine clocks (at 1.something MHz) even when you had an FPU, which was actually quite uncommon. My comment was tongue in cheek but I really do think that Firaxis is still living in the days of the 286.

I remember one thread back when Civ4 was young where Sirian, who is maybe half my age, was ranting about how costly it is to work in floating point. In truth floating point has been cheaper than integer since long before he ever joined the business and he has never worked in a world where floating point was the least bit costly but that didn't stop the ranting. He quite obviously learned this nonsense from his co-workers.

There is nothing wrong with approximating 1/3 by 33% if that works best in a human-readable XML file. Just kindly realise that it takes far more time to read the file than to interpret it properly. This is 2010. Take the billionth (or less) of second required to add 0.5 so that 3 * .33 rounds to 1 instead of to 0.

Edit, to clarify: back in the day we really spent a lot of time considering the trade-offs between accuracy and speed. Today this is just stupid. With a modern CPU, there are almost no gains to be had by sacrificing accuracy. So do the job right.

 

[ohioastronomy]

Really? You mean you would like to have buildings and wonders built at the same rate (comparatively) as researching techs?? Civ is not meant to be played at one sitting even though that seems to be a prevaling trend.

The basic problem with the game is that things take too long to do - and that the peaceful routes (like space) are much slower than the military route. So, yes, I don't think that tech needs slowing down - buildings, etc. need speeding up. Just slowing the tech down will make non-conquest games agonizing.

 

[Jaycob]

Your mistake is assuming its a bug. It could just as easily be a design choice (perhaps a bad design choice but then people are arguing a whole lot of civ5's design is bad).

Let's see.
Culture applies overflow, production applies overflow, golden age happiness applies overflow, etc etc... what on earth makes you believe the frustrating lack of research overflow is intended?

 

[Lanthanoid]

Let's see.
Culture applies overflow, production applies overflow, golden age happiness applies overflow, etc etc... what on earth makes you believe the frustrating lack of research overflow is intended?

Exactly. This is either a requirements or specification lack, or an outright bug.

I don't see how anyone could think that they specifically chose research alone to not apply overflow to. In fact, out of all of the things not to apply overflow to, it is probably the worst one to lack it.

 

[smellymummy]

it could just well be that it was taken out because it sped up research more than it is (and imo i find it fast)

either way, there's mods available that will restore overflow, so either the devs will rebalance tech speed, properly add tech overflow, or none of these and just use custom changes to have it for yourself

 

[Carl5872]

Okprobably a stupid question, but is this patch already released or planned to be released?

Another problem is that there are very few diplomatic options available other than war, there is no espionage or sabotage at the moment, no tech stealing either, if the AI becomes angry it only really has one recourse.

With a few more diplomatic options the game could become more indepth, where a peaceful Civ may have prefered to spy or sabotage on an enemy before, now they too only have war as their way to show displeasure, it leads to the AI leaders all acting in a similar way.

Yea great idea! I miss sabatoging from CIV 4. Every time Mansa Musa stole a wonder off me but I was too busy to DOW him, I would send a spy to poision his water supply then exclaim IN my best Toy Story Woody voice: "SOMEONE'S POISONED THE WATERING HOLE!!!"

The diplo patch has to be linked with improved victory conditions. Right now the AI seems to be programmed so that only domination means something. They are useless when it comes to any other condition. (And not that swift with domination, to be honest.) However it is fixed, improved victory conditions will greatly help the AI: Monty wants to win by Domination; he should be far more difficult to reason with than Gandhi (Mr. Culture) or Washington (Mr. Apollo Program) or Harun (Mr. Diplomacy.) The Civ IV system should be a model to be transcended, not copied.

An extra thought for a future expansion:

I do not mind opaque diplomacy. It would allow the reintroduction of espionage. I do not mean that annoying garbage from Civ IV (poison the water, sabatoge the workship, disrupt the library production) but real espionage:
-get part of the enemy map
-get a (partial) glance at the enemy tech tree
-get an idea of the enemy relationships
-get a glance at enemy social policies
-get a glance at enemy cities
-steal a tech
-spies are invisible: make open borders mean something
-maybe even get insight into the enemy's plans
Real espionage: information gathering. The modifiers from Civ IV were silly. Take some risk, use some effort to get the diplo info. Instead of espionage points, missions can cost money; the more you invest, the greater the chance for real rewards. If you fail it might mean war.

Another that might be interesting is to sabatoge a resource. Like have a spy sabatoge an iron mine and it may reduce that mine from 6 iron to say 2 iron, or eliminate the resource all together. THat would be a huge incentive to actually use it. In civIV it was such a pain in the ass to send a spy to break a mine on iron, which his workers would just rebuild. And when you were caught, most of the time it didnt matter much. It would also be a huge incentive to DOW someone for trying that on you.

To veer away from the pressing issue of fishing-boat realism...


I like that now when a civ asks a player to go to war, the player can request 10 turns to prepare. However, I wish that the player could decline war, but offer help in other ways. A player should be able to offer units, gold, or open borders to help with the other civ's war efforts.


For example:

Assume that Washington pops up and asks whether you'll go to war.

You can respond in 1 of 5 ways, 1) Yes, 2) No, how could you? 3) Not interested, 4) Give me 10 turns to prepare, or 5) No, but how might I help in your war efforts?

Assume you select option 5: No, but how can I help?

Washington could answer in one or more of the following ways: 1) Never mind, I'm mad at you, 2) Thanks for offering, but you can't help me, 3) Give me X gold, 4) Give me X gold per turn, 5) Provide me X number of units within 10 turns, 6) Guarantee my troops safe passage for the duration of my war against X, 7) Close your borders with X for the duration of my war against X


I really hoped that C5 would take diplomacy to the next level....I want the experience of BARGAINING with other civs.....I think this additional diplomatic option would be great....what'cha think?

I like this idea!

the game is playable for me, but that doesn't hold true for everyone. What we have is these assinine generalizations being thrown about. It may be playable for me and it may be playable for you but for someone else it may not be playable. In my opinion, both sides are guilty of using these all encompassing absolutes.

This is what leads to trolling; this entire vicious cycle of generalization - exception-criticism-counter-generalization-exception-criticism. Every time one makes a generalization it is a challenge, not a statement of opinion because unless you are a deity, you will most likely not know if something is universally true. So folks, stop trying to overcompensate for your opinion, by making it sound universal.

Thus can we please keep the patch thread from being a another loveboi/hateboi flamefest please?

Rat

loooouuuuuudddd nooooiiiiissssseeeessss!!!!!!!

Moderator Action: Quintuple post merged. Please use the 'Edit' button in future.

 

[blind biker]

In my previous life (before returning to actual science) I worked as a SW engineer myself, and one factor that affected priority was also how easily a bug could be fixed. If it's trivial to fix (and test) then it would get a higher priority. the Erlenmayer Flask overflow (not a beaker ) seems trivial to fix and regression-test, so it should be high priority. IMHO.

I haven't read most of this thread, but I see people scrapping over whether beaker waste is important and whether it should be fixed or not.

I'm a software engineer, and for categorising bugs at my work we use two metrics: severity, and priority, measured on a 1 to 5 scale with severity 1, priority 1 being the most severe and important issues, and 5/5 being least severe (trivial) and least important issues.

Normally, very severe bugs go hand-in-hand with high priority - if it is really severe, like the programme crashes whenever you try to do a particular task, it needs to be fixed ASAP. On the other hand, sometimes there are bugs that aren't actually severe, but they still have high priority - maybe an important customer has decided (for whatever arcane reason) that this is a problem for them, and they want you to fix it ASAP, even if it is only a cosmetic issue that doesn't truly affect functionality.

Other things can affect priority too, like how difficult the bug will be to fix (maybe it needs major rework elsewhere, mabye it's a 1 line change) or how risky the fix is (complex system like AI, or simple XML change) or how often it occurs (once in a blue moon after the system has been running for 34 weeks non-stop, or every 2 minutes).

So, beaker overflow.
Severity: 4. Maybe 3.
Priority: 1.

If you use this framework to talk about bugs, you'll find it much easier for everyone to understand where everyone else is coming from.

 

[JBConquests]

I haven't read most of this thread, but I see people scrapping over whether beaker waste is important and whether it should be fixed or not.

I'm a software engineer, and for categorising bugs at my work we use two metrics: severity, and priority, measured on a 1 to 5 scale with severity 1, priority 1 being the most severe and important issues, and 5/5 being least severe (trivial) and least important issues.

Normally, very severe bugs go hand-in-hand with high priority - if it is really severe, like the programme crashes whenever you try to do a particular task, it needs to be fixed ASAP. On the other hand, sometimes there are bugs that aren't actually severe, but they still have high priority - maybe an important customer has decided (for whatever arcane reason) that this is a problem for them, and they want you to fix it ASAP, even if it is only a cosmetic issue that doesn't truly affect functionality.

Other things can affect priority too, like how difficult the bug will be to fix (maybe it needs major rework elsewhere, mabye it's a 1 line change) or how risky the fix is (complex system like AI, or simple XML change) or how often it occurs (once in a blue moon after the system has been running for 34 weeks non-stop, or every 2 minutes).

So, beaker overflow.
Severity: 4. Maybe 3.
Priority: 1.

If you use this framework to talk about bugs, you'll find it much easier for everyone to understand where everyone else is coming from.

We use this exact system where I work also. Hmm, wonder if we work at the same place... Or perhaps this is just a really common system.

It is a good system and it can be fun to debate the settings.

I would give it:
Severity 4
Priority 3

Which would mean very minor issue that should be fixed but we wouldn't have to fix it.

Where I work it is either classified as a Bug or as a change request. (new feature) We would have to have someone from Firaxis chime in and tell us which it is.

 

[PieceOfMind]

Okprobably a stupid question, but is this patch already released or planned to be released?

It is planned for release. As for when, we don't know, but probably some time this year.

 

[dlk]

All I wanted to see (and didn't) was the fix for the "screen freezing" problem when it moves from city to city later in the game on huge worlds. A lot of people have this problem and I quit playing a couple of weeks ago until a patch comes along and will fix it. This is a MAJOR prob that needs to be addressed. This is a GREAT game but it doesn't allow play on all type worlds. Dave

 

[Paradigne]

All I wanted to see (and didn't) was the fix for the "screen freezing" problem when it moves from city to city later in the game on huge worlds. A lot of people have this problem and I quit playing a couple of weeks ago until a patch comes along and will fix it. This is a MAJOR prob that needs to be addressed. This is a GREAT game but it doesn't allow play on all type worlds. Dave

I have the same problem. It's not a city issue, I was play testing huge map, 2 AI and myself. It didn't even make it to late game, more like renaissance.

 

[Putmalk]

Update!

AI

* Worker AI improvements.
* Update to tactical AI pillaging code. Additionally, always check to make sure it’s not trying to pillage in an enemy dominance zone.
* AI victory emphasis improvements (more efficient end-game when focusing on Science and Diplo victories).
* AI should colonize other continents regularly.
* AI will emphasize production of an Ocean going explorer unit when the time comes.
* Adjust Napoleon to make him more likely to go for culture.
* More aggressive second wave expansion (mostly off shore) after initial empire building and consolidation has occurred.
* Optimization when finding routes (pathfinder improvement).
* Multiple tweaks and bug fixes.
* AI will now build ranged and mobile units more in line with the flavor settings.
* Multiple defensive AI tweaks.
* Never use ranged units to provide flanking bonuses. (Added 11/18)
* Improve AI use of protected bombard attacks (melee in front, ranged in the rear). (Added 11/18)
* Worker priority adjustments (prioritize pillaged tiles, etc.). (Added 11/18)
* Further pathfinder optimization. (Added 11/18)


GAMEPLAY

* Cities heal more quickly.
* Only allow one upgrade per unit from a goody hut.
* Science building track adjustments (cost, specialist slots, GP Points, etc). (Added 11/18)
* Amount of damage caused during naval combat increased. (Added 11/18)
* Melee horse units combat value lowered, and now receive a penalty when attacking cities. (Added 11/18)
* Lowered bonuses received from Maritime city-states. (Added 11/18)
* Removed maintenance from defensive buildings. (Added 11/18)
* Multiple unit upgrade track adjustments. Most (but not all) units now have a full upgrade path from start to finish. (Added 11/18)
* Open terrain penalty lowered. (Added 11/18)
* Policies must be selected the turn they are earned. (Added 11/18)
* Promotions must be selected the turn they are earned. If it’s as a result of combat, then the beginning of the next turn. (Added 11/18)


UI

* Tweaked the single-player score list to hide the civs of unmet ai players.
* Added game option to disable automated workers from removing features. (Added 11/18)


DIPLO

* AI's attitude towards you is now visible in the diplo screen and diplo drop-down.
* Added info tooltip for an AI leader's mood. Lists things that are making an AI player happy/upset.
* New diplo system: Declaration of Friendship (public declaration with diplomatic repercussions).
* New diplo system: Denounce (public declaration with diplomatic repercussions).
* New custom leader responses (Serious Expansion Warning, Aggressive Military, Luxury Exchange, Borders Exchange, Gift Request).


MODDING

* Parent category counts now include counts of child categories.
* Selecting/deselecting a category now automatically selects/deselects it's children and its parent.
* Tweaked category name truncation to better fit names.
* Hide categories w/ no children and a count of 0.
* Added support for fallback languages (if mod is not translated, fall-back to English so text keys are not showing).


MISC

* Fixed save format which causes saves to increase the memory footprint of the game drastically when loading frequently over the course of the game.

 

[UWHabs]

Don't like forcing the picking of promos/policies right away. Otherwise sounds like they really listened to a lot of things from the board (nerfing horsemen, open field penalty, maritime cs)