Notes on 1.069

[mystikx21]

Most UUs will need adjustment still, including obsolete tech where they differ from default (UU warriors and Legions are most obvious from first glance to be possibly distinct). The most obvious UU strength adjustments were the ranged knights (Keshik/Camels).

I saw three march promotions listed, one of which appeared identical to air repair. Not sure if both are active but that'd be confusing for the AI at least or it could stack.

Pikes/Spears/Vanguards do not appear to have 25% defence. Could be intentional but doesn't appear to be from unit proposal.

Still a chunk of AI priorities to reset. Will have notes on those to compare to Mitsho's diligent work later.

Constable/Police in Order tree up to +5 , Secularism tool tip is +2 to (all) specialists (Update: GK default appears to be what it does now, so this appears accurate)

AA Gun didn't update correctly. Appears to be 680 cost (I think 400 is right?) and 50 instead of 40. Does have ATG/SAM defence bonus. Unit class in .sql should be ANTI_AIRCRAFT_GUN, not ANTI_AIR_GUN

Note: I will update the specific units wikia armies page once the navies seem firmly laid out. Made notes though of non-UU land/air units and am updating my spreadsheet as I go. All units are more expensive owing to higher GEM production level. I suppose that can be recorded if it changed base cost (prior to 1.8x adjustment). Most units by the look of it have strength/ranged and base cost updates. I will use round numbers for simplicity.

One random note: Should landships be 5 moves? Seems like 4 is fine. If they're the WW1-interwar variety of armor they could be slower than the WW2-Korean variety (roughly pre-Gulf War style armor), and a -1 move shouldn't be a huge gameplay hindrance.

Update: Walls/etc appear to be -3 base. Note on patch was for -2. I prefer -1 with changes to pop/tech formula myself as the buildings *should* give substantial effect on both strength and HP.

 

[mystikx21]

Notes/issues in 1.0610

Building costs are increased roughly 1.8...except on Granary, Shrine, Stable, Walls, Barracks which are all still fixed at 55. Sql cost change on these buildings needs to be changed to 1.8 * 55 (100 roughly).

Spoiler :

Question for Thal: I'm not entirely sure how that's working as the various xml lists a lot of building cost adjustments that don't seem to be then getting multiplied by the 1.8 factor but are instead the base cost being used (mints, colosseums, gardens, etc). XML overrides sql change then?

I'm not seeing the Vanguard +25% defence or 20% camradarie except on Scouts. UNITCOMBAT_RECON tag in sql doesn't work. My guess at it is that it loads the scout first in sql because it already has this tag in GK default, then the new units are added in xml, and then the tag isn't working because they didn't exist when it was applied, or the tag wasn't applied yet as in the case of Paratroops.
Are walls etc intended to be -3 or -2? Looks like -3 right now.
Unnecessary iron requirement removal for catapults/trebuchets. Default GK.
There's a double mention for temples .2 / in the faith.xml for buildings. Is this supposed to be for Hagia Sophia instead of (likely useless) free temple?
There's no need to have null building on Notre Dame/Angkor as they don't give them by default, just in the previously modified version. (I have a habit of looking for these obsolete but mostly harmless codes from previous changes of this sort, policy xml files are strewn with them).
Notre Dame is listed as +2 Culture in wonders.xml and +8 culture in faith.xml (I assume +8 is correct). One can be deleted. It appears to give neither at the moment because GK Notre Dame doesn't have a culture yield to update and both points call for updates rather than inserts. Change to this:
<Row>
<BuildingType>BUILDING_NOTRE_DAME</BuildingType>
<YieldType>YIELD_CULTURE</YieldType>
<Yield>8</Yield>
</Row>
Machu Picchu is likewise double listed in wonders.xml and GEC_Data.xml. One gives and the other . Faith seems to be working, the other could be deleted. Or are both intended?
Mech Infantry is listed as 3 move and 4 move in the same file (GEA_Land).

Spoiler :

(Mod feel free to move the thread to bug reports now).

 

[mystikx21]

Three other catches.
Mobile Sams are missing an _ and not updating correctly.
Marines are UNITCLASS_MARINE (no 's').
Helicopters are UNITCLASS_HELICOPTER_GUNSHIP. (missing "gunship").

Amended files below. I also put a "recon" tag in the sql.file on vanguards (like with the MG units), but I don't know if that will work yet. Testing momentarily.
Other file is a change for granaries et al to 100 cost and a change to Notre Dame.

 

[skodkim]

Thats impressive work. Lets hope Thal responds quickly!

\Skodkim

 

[[to_xp]Gekko]

thanx for the quickfix

 

[mystikx21]

One more. UNITCLASS_CROSSBOWMAN

Hard to spot all typos at once.

 

[mystikx21]

Fix for Vanguards did not work and seems to have broken the unit file loading after strength changes. I will keep playing with it this afternoon, but I'd rather have something that works out there.

This file is updated without it and should work correctly for paratroops and change unit costs and strengths for marines, Xbow, Helos, and SAMs.

Update: Tested and works (had to fix one more on helos that I goofed on).

 

[Thalassicus]

Thank you, I added a bunch of these fixes for Gem v1.6.11.

I keep old disabled code around in case I need it again in the future. It saves time in the long run. I clear the old stuff out a few months after disabling it.

Walls give 100 health and 2 in Gem, compared to 50 health and 5 in G&K. I would be okay with adding a strength point back later if we feel it's needed. Machu Picchu gave production in Vem, which I changed to faith in Gem. I was so happy when I got to add the faith bonus to that! The yields on Machu Picchu really represent this amazing city well. It's off-topic... but that's an exciting wonder!

The unit attribute problem for vanguards was because the promotion is added in gea_start, which runs before vanguards are created in gea_land. I moved the promotion-adding to gea_end to fix that.

 

[mystikx21]

I know it helps in case of revision later. I thought for myself it was easier to keep track of for things with a lot of changes (like when policies were updated) when there's fewer variables listed to know where things were in default GK instead of where they were 3-4 months ago pre-GK VEM. There's plenty in there from older than that but that's very nitpicked details.

Also probably nice to see/know when Firaxis adopted a change, somewhere along the line.

Was the double Temple related to Hagia Sophia? I preferred Machu being faith instead of production myself. Pleased to have it.

I still think -1 on walls was sufficient instead of -3 walls in favor of building something for defence instead of tech/pop changes. But the HP change might make -2 fine, yes. It's a bigger gap on the later buildings that only did +25 hit points and now do substantially more. Have to do some more actual play testing instead of hunting down the buggy bits to see how well that works.

 

[Thalassicus]

I recognize it's easier to sift through things without all the disabled stuff in there, and I'll do a pass of deleting stuff soon to help make that easier.

 

[mystikx21]

I recognize it's easier to sift through things without all the disabled stuff in there, and I'll do a pass of deleting stuff soon to help make that easier.

No hurry. The unit changes are now a completely new animal so there wasn't much ugliness to sift through. I don't usually have to go into the buildings and policies files to see what was done at this point and that's where the messy bits were mostly.

Probably can wait until/after you're adding the new wonders/buildings or changing the old ones. We still have leaders and some UU/UA balance to go after the navy, and a couple of those policies don't work quite right yet. Plus beliefs, any changes to the CS quests or espionage systems...

 

[Ahriman]

At ranged strength 33, cannons feel slightly too strong vs land units.

Remember that strength 20 -50% is much stronger than strength 10, because all the other modifiers still work on the base percent. So the terrain modifiers, great general modifier, heroic epic Morale modifier can get to canceling out the penalty, and so cannons are just blowing muskets away left and right.

I also think that the Heroic Epic morale promotion bonus of 20% is probably too high, particularly for siege units.

 

[Stalker0]

At ranged strength 33, cannons feel slightly too strong vs land units.

Remember that strength 20 -50% is much stronger than strength 10, because all the other modifiers still work on the base percent.

I was just thinking the same thing.

If the cannon gets 2 core promotions (40%) then overall its only at a -10%, which is 29.7. That's pretty strong against competitive land units.

 

[mystikx21]

Is that a cannon specific problem or does it impact artillery (and rockets) too for balance?

I think cats and trebs *should* be okay as is.

@Stalker, terrain promos should be reduced to 15%, so it would be at -20% still. But the general point and question still stands.

 

[Ahriman]

I don't know, I haven't tested those units lately. The problem gets worse the alter in the game you get because the number of % modifiers around increases, with more experienced units.

 

[mystikx21]

One other note. I'm not sure if it impacts design/balance or not.

But I noticed the excel sheet included with the mod has listed gunships at 65 in GK and they're only 60. This means that they come out as 55 and not 60 as appears to be intended.

 

[mystikx21]

Checking the math... From the base spread, cannons and artillery are the same as GK versus units. Both are slightly weaker now versus cities without a further siege promotion.

One obvious issue is that arques/muskets are weaker and I thought fortification was intended to be as well (perhaps not). A fortified Musket in GK is worth ~ 50 , while it's down to 44 now.

Are there any available promotions for siege units that would directly impact unit combat besides Hero Epic, the built-in volley against fortified, and GGs? Because with all of those included (and considering changes in the value of those promotions from GK), Cannons can get up to a max of 35% stronger attacks against units, and artillery gets 30% (compared to GK). That would be 53 for a hero/GG/3 terrain cannon against a fortified unit in that terrain (which as an arque is 44-52), and 43 against a non-fortified arque (which would be 32-40 ). I can see that's a problem, especially in flat terrain, but that's only +3 from what GK would have done.

Artillery would be 72 against fortified and 59 against non (again +3). That should be less obviously a problem against riflemen from the look of it, though they would be powerful against muskets one tier down. Fully updated rockets are actually weaker (104/120).

I suspect extra volley power (30%) and weaker targets are more obviously at fault than the +5% change from Hero Epic. Perhaps the extra volley power should be on a promotion rather than default as one option, and/or reduced to 20-25%?

Reducing slightly should be fine too. Say cannon at 30, artillery at 43, others stay?

 

[Ahriman]

Checking the math... From the base spread, cannons and artillery are the same as GK versus units.

What do you mean by base spread?

Are you taking into account this:

Remember that strength 20 -50% is much stronger than strength 10, because all the other modifiers still work on the base percent.

If they were strength 17 and are now strength 33 minus 50%, that is a big boost.

Are there any available promotions for siege units that would directly impact unit combat besides

Uh... the regular terrain promotions? Siege units still get those.

are more obviously at fault than the 5% from Hero Epic.

Heroic epic morale promotion gives 20%.

So my double open-terrain promoted cannons built in a Heroic epic city with a great general nearby are strength 33*(1 - 0.5 + 0.3 + 0.2 + 0.15) = 38 vs units. Against strength 30 newly named musketmen.

[I have to admit: the musketman for napoleonic-era guys and Riflemen for WW1 era is really a big flavor boost.]

 

[mystikx21]

They were 20 strength, so yes. That's what I looked at.

I'm using thal's spreadsheet to calculate. "Base spread" is what the unit would be using thal's default terrain modifiers for each. Adding a further terrain modifier to cannons helps another 15% in GEM, but in GK would have been another 25% +10% more from the existing two, so it doesn't add very much new power that wasn't already there in the game. What is present in GEM and isn't in GK is an anti-fortified bonus, and weaker mainline rifle units.

The Hero Epic is 15% in default. 5% is the difference that GEM adds. I could be fine with moving it back to 15 too but it isn't adding very much was the idea there. Basically +1.5 from what we would expect on cannons and +2.5 on artillery.

I included the terrain promotions in my calculations, just didn't list them. Sorry. The Muskets and arques would also get terrain advantages (sometimes) and we'd have to factor that in. On flat ground with a further -20% is where I could see an obvious problem, on hills it should be less close. Muskets fully upgraded are 48 unfortified (without Hero) however (vs 43 from the cannon with Hero). The bigger edge is against arques, which is admittedly where it should be as they're a tech line down.

Another way to approach this might be to slightly strengthen mainline rifles, or at least raise Muskets to 32?

 

[Ahriman]

Maybe I'm confused here, but it seems like we're still talking cross purposes.

It isn't just about whether the potential multipliers are slightly higher or not.

It is that, say, a 10% difference on a strength 20 unit is 2 strength points, while a 10% difference on a strength 33 units is 3.3 strength points.

So suppose there weren't any additional strength points available at all between G&K and GEM. Without any difference at all, if there are a cumulative 50% of modifiers available, in G&K those 45% meant it was strength 20*1.5 = 30, while in GEM is is strength 33*(1 - 0.5 + 0.5) = 33.

Basically: moving from a low strength model to a high strength with -50% model means doubling the marginal effect of other modifiers.