(see with artists, but indeed sword < spear < pike)

Let's make a comparison of current maximum melee attack ranges:

                 citizen  champion  hero
female           4.0
infantry  sword  2.0      5.0       5.0
cavalry   sword  6.0      6.0       6.0
infantry  spear  4.0      5.0       5.0
cavalry   spear  6.0      5.0       6.0
infantry  pike   8.0      8.0       8.0
battering ram             6.5

Can you update the patch to just the range changes?
Did you test with large armies? Did the battles have different outcomes (i.e. do we need to rebalance attack stats too)?

Out of curiosity, how have you chosen these specific ranges? E.g. why 3.5 and not 3 or 4?

In D988#54458, @Nescio wrote:

Out of curiosity, how have you chosen these specific ranges? E.g. why 3.5 and not 3 or 4?

I tried to fit best the animations and keep proportions between ranges (i.e. pikemen double other soldiers units).
This doesn't want to be a balance patch but a correct attack-animation matching fix.
Any better fitting value is welcome.
Also I looked at sieges but, since they interact with buildings, it seemed more complex and envolved other stuff that aren't units. A clear example is any sieges attacking roman Fortress where the ram body interesects the building body, thus the whole tree ( the ram'head) enters in the texture even while idle.

I remind you that the patch is intended to be an estetic patch only. Any gameplay change is unwanted thus I am still open for suggestions.

In D988#54967, @temple wrote:

Making champions and heroes consistent with citizen-soldiers is good.
The new values are sword 3.5, spear 3.5, pike 7. But it seems to me like spears should have a longer range than that? It looks like 5 is good.

I tried 4.5 for spears and apparently it was good too ( for hellenic spears at least ) as there wasn't any significant change from the esthetic point of view. Still i agree that spears should have longer range than swords. While higher range suits good for hellenic spearsmen, other type of spearsmen ( i.e. celtic ) are equiped with shorter spears and 3.5 seemed to be a good compromise. Still 4.5 seems good.

The new values for cav are sword 4.5, spear 5. But here I think they should have the same range as their infantry counterparts -- their weapons aren't any longer. You can use the actor viewer in atlas to compare. So I'd do sword cav 3.5, spear cav 5.

This seems reasonable.

Rams and elephant are siege, hard to judge. But women maybe 3, dogs 2, what do you think?

Sure, I'll have a look at women and dogs animations too. Sieges and elephants are hard to judge as they interact with buildings which seems to have smaller obstruction than their textures size. Matter of fact rams intersect roman fortress while attacking and workers intersect with gaul fortress while repairing/building.

Sword infantry max range was changed in rP16285 for balance, but if their range remains smaller than spear then there shouldn't be much of a difference in gameplay. Otherwise doesn't look like ranges have been adjusted. There's D981 which changes how range is used, so all the range values might have to be revisited in the future.

What really changes is just the first hit an unit can perform when he reaches the enemy unit. I didn't really notice if the hit is performed when the unit reaches the opponent instantly or if it waits attack rate time before the first hit.
Lower range unit will get closer to perform his hits anyway, too long range for short range animation/weapon looks not natural in game as they can't even reach each other ( the clear example are champions with the same max range ).

I think you should use absolute values not relative, since this is dependent on the animations. It's just coincidence that pike range is twice sword range.
You adjusted champion and hero capture range (which should stay at 4m). They don't inherit the melee attack so you have to change the max ranges all individually.
You adjusted women's slaughter range instead of melee range. But on that note, it looks like everyone's slaughter should be much smaller too, say 2m.

Maybe @Stan has an opinion on the numbers?