Fixing casemate first shot accuracy

The issue:
Casemates have a reduced accuracy for their first shot when they need to turn towards their target, because their hull still rotates while the gun is already pointed towards the target. This means they shoot with moving multipliers (reduced accuracy and higher scatter), giving them a miss more frequently.

This topic surfaced shortly again in the JP4 thread, and I had a bit of time to test it. Unfortunately, I tested it within a mod where I played around with target sizes (e.g. I gave the T34 a target size of 80, T70 around 60-ish or so). Interestingly, the unmodded JP4 still missed about half of its shots (I counted about 20), which is a hint that moving modifiers are applied after capping the accuracy at 1 and the fix as described below should still work as intended.

The idea:
If the unit is still rotating when shooting, we must make sure it stops before the shot. The easiest way is probably to delay the shot enough until the unit is fully rotated. For example, the JP4 rotates at 20° per second, and the unit is allowed to shoot when the target is 5° off the center. This means we need to delay the shot by 5/20 = 0.25 seconds.

I then increased the ready aim time (this is only applied when a new squad/vehicle is targeted) from the standard 0.125 seconds to 0.26 seconds (slightly above 0.25 to make sure it works) and I did not have any accuracy issues anymore. I did not test for an awful lot of shots, but the first shot was suddenly reliable.
Since MMX brought up the issue of tracking I then took a T34 as normal and a vet3 T70 as a super fast target and circled around the JP4. I didn't see any tracking issues. As long as the JP4 kept up with its own rotation, it was able to shoot. Are there any other issues with this fix, or what was the reason it was not implemented?

Sorry for the tags @Sander and MMX. You guys seem to have either tested it yourselves or talked with people trying to fix it, so I'd be interested what exactly went wrong at the time so that the fix was not implemented.

It's well known that rotation means moving. You can even test it with AA HT. If it's rotating, it won't fire the main cannon.

It's well known that rotation means moving. You can even test it with AA HT. If it's rotating, it won't fire the main cannon.

Yes, but what does it have to do with my suggestion?

This seems like quite am elegant fix, with the only caveat being that casemate TDs are not underperforming and are in fact very strong

That's an elegant fix! Could probably revert the SU-76's moving accuracy buff with this.

Great work Hannibal, I'm surprised just increasing aim time would have fixed the problem wish this had been implemented in a patch sooner.

One question, with the increase to ready aim time does that slow down follow up shots?

Great work Hannibal, I'm surprised just increasing aim time would have fixed the problem wish this had been implemented in a patch sooner.

One question, with the increase to ready aim time does that slow down follow up shots?

edit: unsure, I thought yes but maybe it only applies to when you have to reacquire targets or new targets.

Great work Hannibal, I'm surprised just increasing aim time would have fixed the problem wish this had been implemented in a patch sooner.

One question, with the increase to ready aim time does that slow down follow up shots?

There's two fire time delays.
Fire aim time is applied each time the target "aims", so every shot. Ready aim time is only applied once a unit aims at a new enemy squad, so for the very first shot (afaik, ready aim time is not even applied if an infantry soldier picks a new target of the same enemy squad because his current target got killed).
For this reason I picked the ready aim time, since I only wanted to fix the first shot.

I am still sightly surprised that this accuracy issue does not happen when circling the casemate. At least it did not for me when I circled the JP4 at around range 60 and range 20. Maybe there are some special ranges in between where rotation and stop-shooting do not fully match up, leading to the same issue. In this case, fire aim time should be increased (and reload decreased accordingly).

On the topic for this thread, if the whole "casemate tanks are inaccurate while rotating" thing is repeated in coh3, I'd like for this fix to be implemented first and foremost, but I also would like a "facing" button for all units. Just a button that you press and left click somewhere to get the unit to look in that direction.

I understand you can somewhat do this with the move command but with a "facing" button you don't have to be as completely perfect with your micro. Sometimes when I'm trying to get an AT gun to face a tank I click a little behind or to the side and it causes the AT gun to move. Same for the MG.

EDIT: It doesn't have to be on the grid or anything.

Just a button that you press and left click somewhere to get the unit to look in that direction.

right click and drag works well enough

right click and drag works well enough

I mean, that's what he said: If you misclick a tiny bit, the unit wants to move those 2 meters, often costing a lot of time to spin around etc.

On the topic for this thread, if the whole "casemate tanks are inaccurate while rotating" thing is repeated in coh3, I'd like for this fix to be implemented first and foremost, but I also would like a "facing" button for all units. Just a button that you press and left click somewhere to get the unit to look in that direction.

I understand you can somewhat do this with the move command but with a "facing" button you don't have to be as completely perfect with your micro. Sometimes when I'm trying to get an AT gun to face a tank I click a little behind or to the side and it causes the AT gun to move. Same for the MG.

EDIT: It doesn't have to be on the grid or anything.

This would also be a nice improvement to at and mg teams. Units like the 17 pounder and pak 43 just reface and set up when given a move or attack move order so an extra ability should be possible.

That seems to be a very elegant fix to the moving accuracy problem, great work!

As to the potential tracking problem you've mentioned; I'm not sure what I did back then in terms of which values I tried to adjust (fire-aim or ready-aim time) and by how much, but IIRC sometimes the target would move out of the firing arc before the firing delay was finished, causing no shot to be fired. But anyway, since you've already tested this in your mod and things seem to work as intended, this was probably caused by something else.

Another thing that would be interesting to see is if this also affects moving scatter in the same way it does moving accuracy. I'd assume both are governed by the same rules as to what counts as moving, but, well, this is CoH2 after all so you never really know...

The issue:
This topic surfaced shortly again in the JP4 thread, and I had a bit of time to test it. Unfortunately, I tested it within a mod where I played around with target sizes (e.g. I gave the T34 a target size of 80, T70 around 60-ish or so). Interestingly, the unmodded JP4 still missed about half of its shots (I counted about 20), which is a hint that moving modifiers are applied after capping the accuracy at 1 and the fix as described below should still work as intended.

Now this is quite interesting as well. If it holds true, moving accuracy would have a much bigger impact than I previously thought. This would be especially apparent at short to medium distance where you'd normally have high enough base accuracy x target size to get to > 1 chance to hit even with the moving accuracy penalty applied. And if this applies to vehicles, does it translate to infantry combat as well?

And if this applies to vehicles, does it translate to infantry combat as well?

That's a really interesting thought- I assume it would follow the same rules, but the different accuracy vs RA values might change the behavior.
For instance, snipers have 2 accuracy, and never miss. The lowest RA in the game happens to be .5. This seems to indicate that if the weapon has above 1 accuracy, it isn't capped at 1.
My theory is that RA can only bring accuracy up to 1, not past it. That means only vehicles and high-RA infantry would be affected. Testing required, though.

That's a really interesting thought- I assume it would follow the same rules, but the different accuracy vs RA values might change the behavior.
For instance, snipers have 2 accuracy, and never miss. The lowest RA in the game happens to be .5. This seems to indicate that if the weapon has above 1 accuracy, it isn't capped at 1.
My theory is that RA can only bring accuracy up to 1, not past it. That means only vehicles and high-RA infantry would be affected. Testing required, though.

I think snipers and Osttruppen are the only units that can realistically generate >1 accuracy (apart from e.g. Grenadier K98 at vet0 and super close range). The sniper can't move though and Osttruppen need to be in cover. Since infantry RA is additionally <=1, I don't think there is much reason to really consider infantry engagements apart from theoretical knowledge. Infantry vs light vehicles is a different topic though, but this can easily be tested.

Quite some vehicles, especially TDs, as well as fights vs heavies in general should be affected though.

I agree that the accuracy cap effect you found is unlikely to have an impact on infantry vs. infantry fights. The sniper example might show that the 1 accuracy cap is based on RA, though. I'm curious what the results would be if the JP4's gun had high accuracy instead of having large targets. Just to know which modifiers are applied when.

Hmm, interesting. I agree the impact this could have on infantry combat is probably rather minor except for few fringe cases (iirc tactical advance can result in >1 chance to hit before moving accuracy penalties?).