Forest Chop in both Quick and Normal Speed games is 3 turns?

[Sun Tzu Wu]

Forest Chop in both Quick and Normal Speed games is 3 turns?

I recently completed a few Ancient Era Quick/Normal/Epic BtS 3.19 games using BUFFY 3.19.001. I noticed that Forest Chops were taking the same number of turns (3) in Quick speed games as they were taking in Normal speed games.

I recently played a Modern Era Warlords games where Forest Chops were only 1 turn without running Serfdom. So, it seems that Forest Chops in Ancient Era Quick BtS games should be taking no more than two turns?

BtS 3.19 Ancient Era turns to Chop a Forest at different game Speeds:

• 3 Quick
• 3 Normal
• 5 Epic
• 9 Marathon

I've personally verified the numbers for Quick, Normal and Epic in BtS 3.19 games; the number for Marathon games is from 3.17 (I assume # turns for a Marathon Forest Chop hasn't changed for 3.19).

It just seems that the no. of turns to Chop a Forest at Quick Speed ought to be two turns rather than three. Could this be a bug of some sort?

Sun Tzu Wu

 

[TheMeInTeam]

Bug possibly, but most would call it a rounding error (a problem that just never seems to find its way out of civ ).

 

[Sun Tzu Wu]

Bug possibly, but most would call it a rounding error (a problem that just never seems to find its way out of civ ).

TheMeInTeam, thanks for your kind and frank answer, but I respectfully disagree 100% that this can be dismissed as a rounding error ...

That's a whopper of a rounding error! That's like saying x/2 = x/3 for x > 0!

Does Civ4 really have rounding errors this blatantly wrong?

I've seen Ancient Era Quick Settlers that cost 65 Hammers compared to Ancient Era Normal Settlers that cost 100 Hammers, whereas one would expect 66 Hammers = 330 Quick turns / 500 Normal turns * 100 Hammers. This is what I'd call rounding error that shouldn't even exist. I'd show the software engineer that did this the error of his ways; he obviously did the computation wrong, because the correct answer is a perfectly even 66 Hammers. There's no reasonable explanation of why it should be 65 Hammers instead.

Where and what was the quality assurance department doing when this bug, err..., rounding error got loose.

I really can't image what formula based on maximum turns of each speed would generate the following values:

• 3 Quick
• 3 Normal
• 5 Epic
• 9 Marathon

• 3 = 330 / 110
• 3 = 500 / 166.67
• 5 = 750 / 150
• 9 = 1500 / 166.67

Sorry, I can't seem to come up with a Worker Forest Chop divisor that works for all three speeds. Something between 150 and 166.67 seems plausible, but 110 is just plainly too far away from the other two values.

However, when one assumes that a Forest Chop for Quick speed should be 2 turns as opposed to 3 turns, the numbers are most closely aligned:

• 2 = 330 / 165
• 3 = 500 / 166.67
• 5 = 750 / 150
• 9 = 1500 / 166.67

Now, the divisor 150 almost works:

• 2 = 330 / 150
• 3 = 500 / 150
• 5 = 750 / 150
• 10 = 1500 / 150

I just can't imagine what they were thinking; one Epic turn should equal two Marathon turns. Therefore, a Marathon forest chop should be 10 turns rather than 9 turns to be linearly equivalent to a 5 turn Epic forest chop.

So, I'd conclude there's a bug in the way the number of Forest Chops is computed for either Epic speed or Marathon speed as well.

Sun Tzu Wu

 

[WastinTime]

So, I'd conclude there's a bug in the way the number of Forest Chops is computed for either Epic speed or Marathon speed as well.

That's an easy one. Normal = 3 so Maraton(3x) = 9 and
Epic (1.5x) = 4.5 rounded up is 5.

Still doesn't explain why quick isn't 2, but certainly your conclusion about a bug in epic or marathon is incorrect.

 

[kcd_swede]

TheMeInTeam, thanks for your kind and frank answer, but I respectfully disagree 100% that this can be dismissed as a rounding error ...

That's a whopper of a rounding error! That's like saying x/2 = x/3 for x > 0!

Does Civ4 really have rounding errors this blatantly wrong?

I've seen Ancient Era Quick Settlers that cost 65 Hammers compared to Ancient Era Normal Settlers that cost 100 Hammers, whereas one would expect 66 Hammers = 330 Quick turns / 500 Normal turns * 100 Hammers. This is what I'd call rounding error that shouldn't even exist. I'd show the software engineer that did this the error of his ways; he obviously did the computation wrong, because the correct answer is a perfectly even 66 Hammers. There's no reasonable explanation of why it should be 65 Hammers instead.

Where and what was the quality assurance department doing when this bug, err..., rounding error got loose.

I really can't image what formula based on maximum turns of each speed would generate the following values:

• 3 Quick
• 3 Normal
• 5 Epic
• 9 Marathon

• 3 = 330 / 110
• 3 = 500 / 166.67
• 5 = 750 / 150
• 9 = 1500 / 166.67

Sorry, I can't seem to come up with a Worker Forest Chop divisor that works for all three speeds. Something between 150 and 166.67 seems plausible, but 110 is just plainly too far away from the other two values.

However, when one assumes that a Forest Chop for Quick speed should be 2 turns as opposed to 3 turns, the numbers are most closely aligned:

• 2 = 330 / 165
• 3 = 500 / 166.67
• 5 = 750 / 150
• 9 = 1500 / 166.67

Now, the divisor 150 almost works:

• 2 = 330 / 150
• 3 = 500 / 150
• 5 = 750 / 150
• 10 = 1500 / 150

I just can't imagine what they were thinking; one Epic turn should equal two Marathon turns. Therefore, a Marathon forest chop should be 10 turns rather than 9 turns to be linearly equivalent to a 5 turn Epic forest chop.

So, I'd conclude there's a bug in the way the number of Forest Chops is computed for either Epic speed or Marathon speed as well.

Sun Tzu Wu

Is this new to patch v3.19 or BUFFY, or has it always been so (in v.3.17 and Hof 3.17.001)?

If its always this way, I am not too concerned... its like troop movement taking lots more years... just another handicap for Quick and everyone is on the same rules. But if its new, then I see it giving advantage to those games in older versions of HoF, perhaps making them not comparable anymore.

 

[ZPV]

It's always been this way, I believe. Presumably it's deliberate, because quick is primarily used for multiplayer.

edit: looking at the xml, it is a rounding error
Forest chop cost is 300 (workers provide 100/turn).
The quick speed multiplier is .67
Thus the forest chop cost on quick is 201. After two turns, only 200 are provided to the chop, so it takes a 3rd turn to complete the chop.

 

[Indiansmoke]

It's always been this way, I believe. Presumably it's deliberate, because quick is primarily used for multiplayer.

Exacly. Doing chops 2 turns would change the balance of the early game alot. For better or for worst I am not sure.

Also this rounding error makes quick speed the only speed that you gain a turn if you improve instead of chop first. So for example a forested hill if you improve straight is one less turn than if you chop and then improve.

 

[Sun Tzu Wu]

It's always been this way, I believe. Presumably it's deliberate, because quick is primarily used for multiplayer.

Can anyone confirm that a Forest Chop has always been 3 turns for Quick speed? I thought it used to be 2 turns for earlier versions of BtS, but I have only 3.19 working now (so I can't verify it).

I doubt very much that game designers wanted a Forest Chop to be 3 turns for both Quick and Normal games, especially when its 5 turns for Epic and 9 turns for Marathon.

Why would a multiplayer game need a 3 turn Forest Chop rather than a 2 turn Forest Chop for Quick speed?

edit: looking at the xml, it is a rounding error
Forest chop cost is 300 (workers provide 100/turn).
The quick speed multiplier is .67
Thus the forest chop cost on quick is 201. After two turns, only 200 are provided to the chop, so it takes a 3rd turn to complete the chop.

In my opinion, the quick speed multiplier should be .66 = 330/500 (Maximum Quick turns in a game/Maximum Normal turns in a game), thus the forest chop would cost 198 and complete in 2 turns as should be the case.

It amazes me how much resolution was allowed 100 units = 1 Worker Normal turn, yet the number of turns for a Forest Chop in both Quick and Normal speed is 3 turns, when the amount for Quick speed should obviously be 2 turns and not 3.

Rounding error may be the cause, but it's still a bug to have a Forest Chop take the same no. of turns in both Quick and Normal speed.

Sun Tzu Wu

 

[Sun Tzu Wu]

That's an easy one. Normal = 3 so Maraton(3x) = 9 and
Epic (1.5x) = 4.5 rounded up is 5.

Still doesn't explain why quick isn't 2, but certainly your conclusion about a bug in epic or marathon is incorrect.

I came to the wrong conclusion based on incorrect rounding. I assumed truncation of fractions (rounding all fractions down). However, as you suggested and based on ZPV's input, we can see that all "fractions" are rounded up for this computation. For example, 201 hundredths is rounded up to 300 hundredths = 3 turns.

Actually, after 2 turns, there is 1/100 of a turn left to complete the Forest Chop, thus it takes 3 turns at Quick speed. Entirely bogus in my opinion. Unfortunately, there is no such thing as Worker turn overflow where the remaining 99/100 of a turn is applied to the next Worker action. That 99/100 of a turn is always lost at Quick speed.

EDIT: Some of the excess 99/100th of a turn can be utilized when the Forest Chop is part of an improvement, but this only win-win when one needs the improvement and needs it when it is completed.

A forest chop being 3 turns for both Quick and Normal speed is definitely a bug, regardless of how it happens to occur. That it is the result of some bizarre rounding error is no excuse; it's still a bug.

Sun Tzu Wu

 

[Sun Tzu Wu]

Is this new to patch v3.19 or BUFFY, or has it always been so (in v.3.17 and Hof 3.17.001)?

I hope someone that can still run BtS 3.13 or 3.17 could verify this one way or another?

If its always this way, I am not too concerned... its like troop movement taking lots more years... just another handicap for Quick and everyone is on the same rules. But if its new, then I see it giving advantage to those games in older versions of HoF, perhaps making them not comparable anymore.

This is definitely a valid concern.

However, I'd prefer that a bug were fixed and thus makes winning a Game a little easier or a little harder rather than leave the bug to keep a temporal level playing field.

Sun Tzu Wu

 

[WastinTime]

A forest chop being 3 turns for both Quick and Normal speed is definitely a bug, regardless of how it happens to occur. That it is the result of some bizarre rounding error is no excuse; it's still a bug.

Totally agree. not sure they'll fix it however.

 

[Hsinchu]

Wow... This dilemma reminds me of a contractor's dilemma, most comparable to a certain transportation service department here in California...

If I have 3 guys working on a piece of freeway and the job is scheduled for 100 man hours, that's 96 hours covered in 4 days. On the fifth day, I only have 4 man hours left but 3 guys that are getting paid for the whole day because it's a government job. Therefore, is it right for me to let the 3 guys goof off for up to 20 extra man hours over the course of the previous 4 days since the job is going to take as long anyways? The answer is to leave that piece of road alone and have fewer traffic jams and put these guys to work painting houses or something more useful than wasting government cheese.

Based on that logic, quick speed may have a HUGE rounding issue, but rather than try to change the game, the strategy has got to be different. I know playing on Marathon/Epic is a very different strategy given that militarily the number of turns to "rescue build" against an invader is significantly different and time scales for upgrade versus replacement change significantly. Quick therefore has a similar timber. Would I go for pre-chopping Pyramids in a quick game? Probably not.

This means to me when you are talking about HOFtacular finishes, quick cannot be compared to normal/epic/marathon. It's not apples to apples. I think it makes things more interesting in allowing for greater exercise of mental agility in order to adapt to game speed.

 

[ZPV]

Can anyone confirm that a Forest Chop has always been 3 turns for Quick speed? I thought it used to be 2 turns for earlier versions of BtS, but I have only 3.19 working now (so I can't verify it).

I doubt very much that game designers wanted a Forest Chop to be 3 turns for both Quick and Normal games, especially when its 5 turns for Epic and 9 turns for Marathon.

Why would a multiplayer game need a 3 turn Forest Chop rather than a 2 turn Forest Chop for Quick speed?

I've checked v3.17 and 1.74, and it is 3 turns for both of them.

The point about MP is, the game balance is different, and Quick speed is meant for MP play, with the rest primarily for SP. It just makes no sense to compare them.
At the start of an Ancient Era game in MP, the players will often chop every tile they can, to speed their early production and avoid being choked. (I'm no MP expert, but I do know a little bit. For the finer details, you'll have to find someone who regularly plays MP). One example that comes to mind is, can you imagine how overpowered Fast Workers would be. (+50% instead of +33% chop yield per worker turn, including movement)

In my opinion, the quick speed multiplier should be .66 = 330/500 (Maximum Quick turns in a game/Maximum Normal turns in a game), thus the forest chop would cost 198 and complete in 2 turns as should be the case.

Well, the game speed multiplier does not have to be defined by the number of turns. (it is a somewhat arbitrary number, used only for calculation of final score and time victory).
Instead it is defined explicitly by some generic multiplier, and 0.67 was chosen for reasons unknown to me.

 

[oyzar]

It should be noted that if you combine the chopping with some improvement you most often gain a turn compared to doing one after another, on quick that is.

 

[Sun Tzu Wu]

This means to me when you are talking about HOFtacular finishes, quick cannot be compared to normal/epic/marathon. It's not apples to apples. I think it makes things more interesting in allowing for greater exercise of mental agility in order to adapt to game speed.

For the HoF, none of the speeds can be compared to each other, even assuming they could be perfectly calibrated, which is probably impossible due to the complexity of the Game.

What I find so frustrating is the obviousness of this bug: A Marathon FC is 9t, a Epic FC is 5t and both a Normal and Quick FC are 3t; obviously Normal and Quick FC shouldn't both be 3t ...

One expects the different speeds to be calibrated in regards to such game details. If such an obvious bug slipped past programmer validations, play testers and quality assurance, it makes one wonder what other non-obvious bugs exist in the Civ4 series of Games.

Sun Tzu Wu

 

[Sun Tzu Wu]

I've checked v3.17 and 1.74, and it is 3 turns for both of them.

Thanks for the info! I'm a little surprised that 1.74 has this bug; probably Warlords has it too.

The point about MP is, the game balance is different, and Quick speed is meant for MP play, with the rest primarily for SP. It just makes no sense to compare them.
At the start of an Ancient Era game in MP, the players will often chop every tile they can, to speed their early production and avoid being choked. (I'm no MP expert, but I do know a little bit. For the finer details, you'll have to find someone who regularly plays MP).

I don't know anything about using Civ4 for MP, but I imagine that if it was really important, internal Game parameters could be changed for MP and the asset checks adjusted accordingly.

One example that comes to mind is, can you imagine how overpowered Fast Workers would be. (+50% instead of +33% chop yield per worker turn, including movement)

Fast Workers have only a +1 movement advantage over normal Workers. If you simply mean the Fast Worker can start Chopping a Forest (with no Hill) the same turn it moves onto it, I understand the point you're making. In such cases, MP Games could ban the Indian Empire or nerf the Fast Worker to be a normal Worker.

Well, the game speed multiplier does not have to be defined by the number of turns. (it is a somewhat arbitrary number, used only for calculation of final score and time victory).
Instead it is defined explicitly by some generic multiplier, and 0.67 was chosen for reasons unknown to me.

That is true, but Vanilla and Warlords would have been better had they used the same number of turns for each game speed as is used by Beyond the Sword.

In my opinion, calibrating the different speeds based on the number of turns is the only sane way to do it. At least that way players can see a logical basis for the turn speed coeffecient rather than some magical value like 0.67.

Sun Tzu Wu

 

[Sun Tzu Wu]

I've reported this bug to the Civ4 Bug Reports Forum:

Forest Chop in both Quick and Normal Speed games is 3 turns

Discussion can continue here or the bug thread above. I'm not sure which is more appropriate, but I'm more comfortable posting here ...

Sun Tzu Wu

 

[jesusin]

Totally agree. not sure they'll fix it however.

Agreed. It's a terrible bug. Not because of any comparison with other speed, where there are too many differences to get something balanced, but because of comparing different worker actions inside a Quick speed game. Chopping has been made incredibly expensive compared to any other worker action!

I think it should be easy to solve: just define the cost value of chopping as 290 instead of 300. Same for every other improvement.

By discounting a bit on every action but keeping the "round up" policy, better balance is achieved at all speeds.