[GS] [1.0.0.290]Overflow missing bug.

[Wandering_Dandelion]

1. Consider a production 5/65 for an archer, and assume that chopping gives us 50 production.
2. Without Agoge policy card (+50% production towards archers), chopping changes the city production of this archer to 55/65. This is obvious.
3. Now if Agoge card is applied, chopping will complete the archer for sure but ZERO additional production for the next unit/building/district in GS. We think this is a bug.
4. Notice that in the previous RF expansion, this chopping process will give 50*(1+50%)-(65-5)=15 production for the next one. When Ed said they fixed the chopping overflow in GS, we thought it works as follows: 40 out of 50 chopping applies on the archer production. Those 40*(1+50%) (Agoge card) finish the current archer and the remaining 10 out of 50 production would be saved for the next. However, it is not 15 (RF), not even 10 (reasonable value) but 0.

In general, we define x as the remaining production (in the above example x=65-5=60), y as the chopping production (50 in this case) and a as the production amplified factor (50% towards archers in this case). We found that:
i. if y>x, chopping will complete the current production and y-x remaining. (which is also wierd, it should be y-x/(1+a) )
ii. if (1+a)*y>x>y (the above example), chopping will complete the current production and ZERO remaining.
iii. if x>(1+a)*y, chopping works normally that all (1+a)*y will apply on the current production and x-(1+a)*y more production is needed to finish it.

By the way, city production is also follow this BUG. That's mean that the higher the city production is, the more you lost(both a city with 40 production and 65 production overflow 0 when the train the boat with the +100% modifier).

That means that when we finish a project, the last part of this project before we finished will not be effect by any policy or UA. Due to this bug, even when you train same units in a city by chopping, the production that the trees provide are not the same. I think the overflow in GS is just as fantastic as it in RF, and the most scientific formula is that in CIV4.

Here are the details : https://forums.civfanatics.com/threads/is-chopping-overflow-overcorrected.642113/

 

[Browd]

May not be a bug, as you appear to be making an assumption about the manner in which the cogs from various sources should be applied, rather than how Firaxis chose to apply them. As I recall Ed's brief description of the fix, it was to remove (in this case) the impact of the Agoge bonus from overflow (unless the next item queued also qualified for the Agoge bonus).

In your example, without Agoge, there is no overflow, and with Agoge you generate overflow that is, per the fix, wiped away.

 

[Wandering_Dandelion]

May not be a bug, as you appear to be making an assumption about the manner in which the cogs from various sources should be applied, rather than how Firaxis chose to apply them. As I recall Ed's brief description of the fix, it was to remove (in this case) the impact of the Agoge bonus from overflow (unless the next item queued also qualified for the Agoge bonus).

In your example, without Agoge, there is no overflow, and with Agoge you generate overflow that is, per the fix, wiped away.

You may misunderstood something. See #31 at https://forums.civfanatics.com/threads/is-chopping-overflow-overcorrected.642113/.
In this example, I chopping for Saka in the same city and the same turn for three times, but the production they provided are different. Theoretically, all these these trees should provide 54*1.5=81 production for that I only trained Saka in this city. In fact, the first one provided 81, the second one provided 54, and the third one provided 65.
That's mean even the next item queued also qualified for the Agoge bonus, I'll also lose the overflow. By the way, that's not I chose to apply them, that's how Firaxis did in CIV4.

 

[Browd]

I didn't check the thread you referenced, just the example you gave above. The examples in that other thread are curious.