Is this a bug or WAD, follow TF order cause both two TFs remains on station

[Nomad]

ORIGINAL: Ian R

I'm not convinced it's a bug. If by "normal" settings, you mean threat tolerance normal, is it possible the lacklustre Admiral Crace is failing a die roll or something?

What failure of a die roll would keep him from moving away from the enemy? And why will he move on his own when he is relived of his follow command?

[tolsdorff]

I did some testing as well and came across a weird result, indicating a possible bug. I am using the true WITPAE TF Numbering, not the TF-names created by the OP. *edit: In all cases the SCTF will be following the Air combat TF.

Notice that the Australian cruiser TF 18 has a higher number than the Lexington TF 14. Rerunning the turn without changing anything will result in the Lexington TF not moving.

If you delete TF 18 (the Australian cruiser group) and replace it with the newly created TF 4 (the first available TF number, lower than the air combat TF number) with the same orders as the previous TF 18 -> follow TF 14, (the Lexington group), everything will turn out fine. All TF's will be safe.

If, however, you delete TF 18 and replace it with a number higher than TF 14. (again TF 18, or TF 27, the first available number after TF 18), both groups will not move and will get attacked.

The same trick goes for the Yorktown group! TF 16 is the air combat TF from the Yorktown, if you create a SCTF in this hex with a number lower than 16 (TF 4 for instance) and set it to follow the Yorktown group, than all ships are fine. If, however, the SCTF you created in the Yorktown hex has a higher number than 16. The Yorktown group will barely move and be attacked.

The last tests I did were (After some fiddling and creating of TF's):

Test 1 :
TF 4 (2 destroyers from the Yorktown group)in the Yorktown hex with TF 27 (air combat, Yorktown). TF 4 was following TF 27.
TF 14 (australian cruiser group) and TF 16 (air combat, Lexington). TF 14 was following TF 16.

Result : All ships in all TF's were a safe distance away.

Test 2 the other way around:
TF 14 (air combat Lexington) and TF 18 (australian cruiser group) TF 18 following TF 14
TF 4 (air combat Yorktown) and TF 16 (2 destroyers from the Yorktown group. TF 16 following TF 4.

Result : In the Lexington hex, nothing moved. A lot of damage. In the Yorktown Hex, everything moved 3 hexes and all ships were attacked. Yorktown left in sinking condition.

On the surface of it, in this small setting with a lot of unknown parameters :
If the higher numbered TF is following the lower numbered TF, problems result on the same day that the orders were given (that's all I tested, I never tested a day later).
If, everything else being equal, the lower numbered TF is following the higher numbered TF, things will go as expected.

It seems that TF orders inside the code are being processed starting from TF 1 and going up.
When the TF, that is being followed, check's if it's being followed it apparently needs some info or a setting from the TF that is following it, before it can process it's own actions appropriately. A setting or piece of info that the following TF hasn't processed or set yet, because of it's higher number.

Anyway, I might be wrong, but it is a testable theory.





edit: added that in all cases the SCTF is following the air combat TF.

[BBfanboy]

ORIGINAL: tolsdorff

I did some testing as well and came across a weird result, indicating a possible bug. I am using the true WITPAE TF Numbering, not the TF-names created by the OP. *edit: In all cases the SCTF will be following the Air combat TF.

Notice that the Australian cruiser TF 18 has a higher number than the Lexington TF 14. Rerunning the turn without changing anything will result in the Lexington TF not moving.

If you delete TF 18 (the Australian cruiser group) and replace it with the newly created TF 4 (the first available TF number, lower than the air combat TF number) with the same orders as the previous TF 18 -> follow TF 14, (the Lexington group), everything will turn out fine. All TF's will be safe.

If, however, you delete TF 18 and replace it with a number higher than TF 14. (again TF 18, or TF 27, the first available number after TF 18), both groups will not move and will get attacked.

The same trick goes for the Yorktown group! TF 16 is the air combat TF from the Yorktown, if you create a SCTF in this hex with a number lower than 16 (TF 4 for instance) and set it to follow the Yorktown group, than all ships are fine. If, however, the SCTF you created in the Yorktown hex has a higher number than 16. The Yorktown group will barely move and be attacked.

The last tests I did were (After some fiddling and creating of TF's):

Test 1 :
TF 4 (2 destroyers from the Yorktown group)in the Yorktown hex with TF 27 (air combat, Yorktown). TF 4 was following TF 27.
TF 14 (australian cruiser group) and TF 16 (air combat, Lexington). TF 14 was following TF 16.

Result : All ships in all TF's were a safe distance away.

Test 2 the other way around:
TF 14 (air combat Lexington) and TF 18 (australian cruiser group) TF 18 following TF 14
TF 4 (air combat Yorktown) and TF 16 (2 destroyers from the Yorktown group. TF 16 following TF 4.

Result : In the Lexington hex, nothing moved. A lot of damage. In the Yorktown Hex, everything moved 3 hexes and all ships were attacked. Yorktown left in sinking condition.

On the surface of it, in this small setting with a lot of unknown parameters :
If the higher numbered TF is following the lower numbered TF, problems result on the same day that the orders were given (that's all I tested, I never tested a day later).
If, everything else being equal, the lower numbered TF is following the higher numbered TF, things will go as expected.

It seems that TF orders inside the code are being processed starting from TF 1 and going up.
When the TF, that is being followed, check's if it's being followed it apparently needs some info or a setting from the TF that is following it, before it can process it's own actions appropriately. A setting or piece of info that the following TF hasn't processed or set yet, because of it's higher number.

Anyway, I might be wrong, but it is a testable theory.





edit: added that in all cases the SCTF is following the air combat TF.

Yes, the AI handles the movement in order of TF#. This is not a new discovery - I've advocated it for years as part of the discussion of follow failures from daisy-chained TFs. When someone mentioned daisy-chained follow near the start of this thread I assumed everyone knew about the TF# being part of the explanation so I did not rehash that. But I forget that this issue hasn't come up in a while and newer players may not know those discussions from a couple of years ago. Sorry for not posting on that detail which led to your testing for the same result!

[Nomad]

I guess I never have run into this situation. I work to get my ACTFs with as low a number as I can then they often follow a SCTF with a higher number.

This really was not a daisy-chain of TFs, only one TF following another.

[Ian R]

I have had a similar problem where I have an amphib command TF (AGC plus ASW escorts) with everything following it to an atoll assault - lower (by which I mean lesser) numbered TFs following the AGC don't always do what they are supposed to during the unload phases.

After a little searching, I found this from many years ago:

ORIGINAL: Don Bowen

ORIGINAL: Miller

ORIGINAL: Bradley7735

I've read that you need to have the transports follow the surface fleets. Or, you have to try and create a surface fleet with a low task force number. One of those two options help to have the surface fleet engage first, instead of the transport task force.

Surely there is a way of programming the game so that the TF number is not the deciding factor[&:] And it still does not explain why neither SCTF even got a shot at the Allied ships[:@]

Unless the player gives some other instruction, the TFs do indeed move in the order created. Follow allows the player to control the order of move and the spacing between the TFs.

https://www.matrixgames.com/forums/tm.a ... s&#2536147

A bit more searching turned this up, from 2013:

ORIGINAL: Don Bowen


The options here are pretty straight forward and, I think, very logical.

Remain On Station, No Reaction.
Orders the TF to stay somewhere, with no options. TF will remain "on station" until forced to retire by low fuel/ammo, accumulated damage, defeat by enemy, or orders by player. This was the only option in original WITP.

Remain On Station, Reaction ordered
Orders are generally contradictory but might be useful in some circumstances. TF will remain on station until something causes it to react (or any of the things above). Once it reacts it will retire. Sort of like a picket.

Patrol Zone(s), No Reaction
TF will patrol the specified hexes until forced to retire (damage, fuel, etc. - as above).

Patrol Zone(s), Reaction ordered
TF will patrol as above, will react to enemy as required. After reaction will either retire or resume patrol, depending on (you guessed it) fuel, ammo, damage.


A separate item is Reaction and follow. Follow is mutually exclusive with Patrol so there is no interaction there.

Follow with no Reaction
Just Follows, ignores enemy if possible and just tags along.

Follow with Reaction Ordered
Follows but will react to enemy when conditions are met. After reaction will return to follow.


Glad to know I'm a Guru. I'll bring that up next cocktail party.


https://www.matrixgames.com/forums/tm.a ... s&#3446832

Then I found a discussion between Alfred and WITPQS from 2015 which indicated Mr Bowen might have decided the issue of whether reaction cancels a follow, or visa-versa, might be more complicated than it first appeared.... [To be continued]

[Ian R]

The answer here, may as BBFB said, to do with TF reactions canceling your best laid plans, including, in this case, the auto reaction of CV TFs.... and there can be multiple such reactions. Noting that Aubrey Fitch has a good enough aggression rating that he might have passed a die roll where Black Jack Fletcher wouldn't, a CVTF doesn't always do what you tell it when other CVTFs are nearby, and blaming the unspectacular Mr Crace may have been premature on my part.

As posters here may have read before, I am fond of the saying "always the farmer, never the dog". I remain unconvinced that there is any bug here, and reading this thread:
https://www.matrixgames.com/forums/tm.a ... w%2Corders
has possibly confirmed my skepticism.

Please read Alfred's post #14. The key paragraphs are 4, 6, & 8 -especially the last line of 8.

The answer to Tcao's initial question: I think there is a distinct possibility this is WAD, and WAD is complicated. The IJN CVTF is TF1 in this scenario. So, according to Mr Bowen it moves first, and comes sniffing around your guys. If you rerun the turn, and Aubrey Fitch, sniffing the IJN CVTF, passed the reaction die roll that pulled him back towards them. Crace just followed orders.

[tolsdorff]

ORIGINAL: BBfanboy

....


Yes, the AI handles the movement in order of TF#. This is not a new discovery - I've advocated it for years as part of the discussion of follow failures from daisy-chained TFs. When someone mentioned daisy-chained follow near the start of this thread I assumed everyone knew about the TF# being part of the explanation so I did not rehash that. But I forget that this issue hasn't come up in a while and newer players may not know those discussions from a couple of years ago. Sorry for not posting on that detail which led to your testing for the same result!

Ah, ok, makes sense you already knew that! Don't worry about it. I don't mind wasting time testing things myself. It's how I learn best.

[HansBolter]

ORIGINAL: tolsdorff

ORIGINAL: BBfanboy

....


Yes, the AI handles the movement in order of TF#. This is not a new discovery - I've advocated it for years as part of the discussion of follow failures from daisy-chained TFs. When someone mentioned daisy-chained follow near the start of this thread I assumed everyone knew about the TF# being part of the explanation so I did not rehash that. But I forget that this issue hasn't come up in a while and newer players may not know those discussions from a couple of years ago. Sorry for not posting on that detail which led to your testing for the same result!

Ah, ok, makes sense you already knew that! Don't worry about it. I don't mind wasting time testing things myself. It's how I learn best.

While apparently not common knowledge, this info is not hard to find.

All you have to do is read one of the umpteenth number of posts of mine regarding the need to create minesweeping TFs FIRST at Manila on December 8th, 1941.

What the OP has experienced is something I don't.

I learned a long time ago not to daisy chain follow orders.

I constantly work to lower the TF number of my air and surface combat TFs.

The result of the above action typically means my Death Star leading TF has a very low TF number and the same is true of any Surface Combat TF that may be leading.

Where things can get dicey for me is when I need the gaggle to follow a Minesweeping TF.

It can take a lot of disbanding and creating TFs to get an MSW TF with a low number to do the leading.

Establishing standardized methodologies for manipulating the TF number movement sequence mechanism goes a long way toward establishing standardized performance.

[tolsdorff]

ORIGINAL: Ian R

The answer here, may as BBFB said, to do with TF reactions canceling your best laid plans
....
The answer to Tcao's initial question: I think there is a distinct possibility this is WAD, and WAD is complicated. The IJN CVTF is TF1 in this scenario. So, according to Mr Bowen it moves first, and comes sniffing around your guys. If you rerun the turn, and Aubrey Fitch, sniffing the IJN CVTF, passed the reaction die roll that pulled him back towards them. Crace just followed orders.

As always excellent info by Alfred. How is he doing, does anyone know?
I never really understood any of the reaction mechanics that well and in the interesting link you provided, there are 100's of possible reasons why the Lexington group could have stayed where it was.

What still puzzles me though : if the outcome of this particular setup is determined by a reaction roll of a leader, why do apparently all TF's consistently reach safety when a lesser numbered TF follows the higher numbered one, and vice versa consistently meet disaster when a higher numbered TF follows the lesser numbered one?
I would assume every commander has to pass the same die rolls in both setups and the other reaction-checks still need to be made as well, right?
When the Lexington TF and the Australian cruiser TF combine to form 1 larger Air Combat TF, with the same commander and unchanged overall settings, it will also consistently reach safety. Again, every commander has to pass the same die rolls as before and all the other reaction-checks still need to be passed as well.


It is hard to do any sensible testing in these small setups however, when the underlying mechanics are not known and pre-determined random die rolls are in play.

[Sardaukar]

Yea, the classic mistake in Manila is to create Minesweeping TF last/later than important TFs...when comes it's turn to move, all others already have had their time with mines in Bataan already... [:'(]

I think most of us have made that mistake...[8D]

[tolsdorff]

ORIGINAL: HansBolter

While apparently not common knowledge, this info is not hard to find.

All you have to do is read one of the umpteenth number of posts of mine regarding the need to create minesweeping TFs FIRST at Manila on December 8th, 1941.

What the OP has experienced is something I don't.

I learned a long time ago not to daisy chain follow orders.

I constantly work to lower the TF number of my air and surface combat TFs.

The result of the above action typically means my Death Star leading TF has a very low TF number and the same is true of any Surface Combat TF that may be leading.

Where things can get dicey for me is when I need the gaggle to follow a Minesweeping TF.

It can take a lot of disbanding and creating TFs to get an MSW TF with a low number to do the leading.

Establishing standardized methodologies for manipulating the TF number movement sequence mechanism goes a long way toward establishing standardized performance.

I never paid much attention to the entire 'low (or lesser(?) number TF discussions to be honest until today. Kind of skipped over it when it was mentioned somewhere.

Never quite realized how it could impact the game (WAD or not), as for instance in the context of the OP's problem or what you mentioned about Minesweeping TF's, either in Manila at the start of the war or leading a large collection of TF's through mine infestations.

Having read up on the topic today, everything about lesser numbered TF's and how they impact some aspects of the game, makes a bit more sense.

Establishing standardized methodologies for manipulating the TF number movement sequence mechanism goes a long way toward establishing standardized performance.

Yes exactly, that concept seems very sensible to me now as well.

[BBfanboy]

tolsdorff: What still puzzles me though : if the outcome of this particular setup is determined by a reaction roll of a leader, why do apparently all TF's consistently reach safety when a lesser numbered TF follows the higher numbered one, and vice versa consistently meet disaster when a higher numbered TF follows the lesser numbered one?

Well to get the TF react your ACTF must be aware of the enemy. If the Japanese just sprinted into that position and you did not have night search in range to detect it, and discern that it is also an ACTF, the react will not happen.

[Tcao]

Thanks all.
that's interesting, looks like the TF# sequence is the cause of this mess. I will keep that in mind in my future gameplay.

Here is a few more tests I mentioned before. I understand now it's better to have lesser number TF follow higher number one. That explain many of the tests results. (except one, pls see Save 31)



save 26

https://www.dropbox.com/s/3hcg3n7gudgo093/wpae026.pws?dl=0
order:
TF18 (CA Australia) follow TF 16 (CV Yorktown)
result
TF 18 remained at 101 138, TF 14 and TF 16 moved to safety

save 27
https://www.dropbox.com/s/6gsown6uwckjzmc/wpae027.pws?dl=0
order:
TF14 reaction range set to 0
result
TF14, TF18 remained at 101 138


Save 28
https://www.dropbox.com/s/nhfyj04huldz0ol/wpae028.pws?dl=0
order:
TF 18 follow TF15 (an AO Group)
result
TF18 remained at 101 138

Save 29
https://www.dropbox.com/s/0qj5bbcayez4jmm/wpae029.pws?dl=0
order:
TF 18 cancel follow order
result
everyone moved to safety

Save 30
https://www.dropbox.com/s/b3h539fp0y30xtf/wpae030.pws?dl=0
order:
TF 16 (CV Yorktown) follow TF 14 (CV Lexington), TF 18 cancel follow order
result
TF 16 moved one hex to 101 138, then TF 16 and TF 14 stayed at 101 138, TF 18 moved

save 31
https://www.dropbox.com/s/0dhz93nd3445k20/wpae031.pws?dl=0
order:
TF 14 (CV Lexington) follow TF 16 (CV Yorktown), TF 18 cancel follow. NOTE: this is a lesser number TF follow higher number one
result
TF 16 and TF 18 moved, but TF 14 remained at 101 138

Save 32
https://www.dropbox.com/s/79wkjubla60eqd0/wpae032.pws?dl=0
order:
TF 18 cancel follow order, TF 14 follow TF 18 instead
result
everyone moved