Falling in Love with Sally

[el cid again]

For some reason, I reread my Francillon - author of the standard reference on Japanese aircraft -
about the Ki-21. I had always considered the plane obsolescent and was somewhat amazed it was
more popular than its successor, the more technically advanced Ki-49. But it is such a workhorse,
so many are available for so long, I had also always used it as at least "better than Japanese light
bombers."

Two things I learned which were new to me. First - the bomb load isn't four 250 kg bombs. I knew
this was not a standard pre war Japanese bomb when the plane was designed and I also knew that
1000 kg was the maximum load - not the normal one (which was 750 kg). That should have told me
- but it didn't - that the stanard weapon was a 50 kg bomb. Similarly, I knew the Ki-48 normally
carried 50 kg bombs - and that should have told me as well - but didn't. I had her fitted with 7 (normal)
or 10 (maximum) 100 kg bombs. But it was designed to carry 15 x 50 kg bombs - and could carry
20 of them - and like all Japanese bombers - in game terms - these represent normal and extended
range loads - with no actual "maximum load" in the allied sense at all: range was a consideration
and Japanese plane design traded range for payload. These loads never changed - and considerable
development of the plane over time didn't change the number or size of bombs carried. And the
crews like it. Compared to other Japanese bombers it was adequate - rather worth the risk of delivering
the load.

Second, although it was a change, it happened before the war - technically with the Ib - the Sally could
fit a 500 litre tank in the forward bomb bay - WITHOUT reducing the load. Replacing four of the 10 bombs
in that bay, the four bombs could be fitted externally if a maximum load of 20 was carried - and were not
needed if only 15 bombs were on a normal mission. This explains why the ferry range for the Sally was
not consistent with the normal range as a % - and I had long since gone over to considering operational
ranges the important ones - so my Sally could not transfer between distant points very well.

For RHS I had evolved the concept of a "bomb bay tank" - originally for Allied recon planes - and the unique
LB-30 - which fitted them for transport missions. While many bombers and transports could fit "ferry tanks"
only used for transfer missions, I reserved the concept for use by recon, bomber or transports that fitted them
for operational missions. A bomb bay tank is simply a drop tank - but I define it as "internal" instead of "external" -
and it works just like drop tanks do in game terms. Also - usually it is a different size than drop tanks of the
same air force are - being custom designed for a particular bomb bay. But I had no 500 litre version - so I had
to create one for Sally.

Testing (after far too much time in development) RHS in its third iteration for AE - technical problems forced me
to recast the thing twice already - revealed these two things transform Sally into a rather fabulous early bomber.
It has range - and in game mechanics terms - loads of 15 or 20 bombs are far more effective than stock's 4 -
or even my previous 7 or 10 - larger bombs. Re unarmored ships, land units, airfields or ports - and quite likely
industrial targets (which I don't bomb because I was trained it is illegal - never mind both my parents flew B-17s -
in wartime - my mother to train the gunners and bombradiers - using cameras to tell if they aimed properly) -
Sally is a remarkable plane. Much to be preferred to naval bombers - whose standard weapon FYI is a 60 kg
bomb (standardization was not popular with the Japanese until mid war). To be able to deliver this over longer
ranges than any other early Army bomber - and to be able to transfer to the full range of a Sally to change
bases - turns it into a bomber I want to build - and to use.

[el cid again]

Data in RHS is in terms of nautical miles - since hexes are defined in terms of them.
In RHS terms, the Sally has a maximum transfer range of 1860 nm, or no less than 47
hexes, when using drop tanks (or 1420 nm = 36 hexes without them). This results in
extended ranges of 1060 nm (or 810 sans tanks) for 14 (11) hexes, and in normal
ranges of 893 nm (682 sans tanks) for 12 (9) hexes. FYI - wether or not you use tanks
is a consideration: it appears that using them increases the logistical cost of the mission:
tanks cost supply points - rather more than bombs do (bombs - even huge loads - seem
to be virtually free - and supply costs for bombers seem normally too low). Nemo once
complained my Me-264, with vast internal tanks, cost too much to afford to use! [The
264 isn't in RHS AE - it was in a Japan enhansed version of old WITP RHS called Empire
of the Sun - not a historical mod]

Note also that the actual range of the bomber on missions is TWICE the operational range -
it must fly both ways. For that reason, while I use extended range = 57% of transfer range,
and normal range = 48% - the ranges that result are nominally 28 (22) hexes with tanks,
and 24 (18) without them - twice what you can fly. Range with tanks is also not the same
% without them - you get only 50% extended and 41 % normal vice the standard 57% and
48%. It is less efficient - as it should be. Regardless - the resultant ranges are quite useful
operationally, in game terms - and the Sally now stands ahead of the Ki-48 both in terms of
load and in terms of range - explaining why it was preferred.

Its failing, of course, is lack of protection. But that is almost universal with early Japanese planes.
It is more robust than a Japanese light bomber, and has more defensive firepower.

[Commander Stormwolf]

japanese were in desperate need of a heavy 4E or high speed 2E

once 1943 came around.. the IJAAF Bmr corps was useless against the
masses of P-47s and P-38s in new guinea.

[Commander Stormwolf]

needed more of this sort.. [8D]

4E , airfield size required = 0 [8D]

[Dili]

Without the weights it is all fairytales.
It is the weights that shows what is possible and what is not.

[Shark7]

Hmmm, I wonder what effect changing the bomb load from 4 250kg to the 50kg or 100kg would do in game terms? It would certainly give you more room for error (that is a greater chance of something hitting) than the current load does.

[DuckofTindalos]

It would result in more damage. A lot more. It's been tested many, many times since back in the days of stock.

[Dili]

That is why it is advisable to normalize bombs with bomb packs if possible. Say 100kg bombs are normalized in double pack=200kg with penetration of a 100kg but with soft damage of a 250kg or a bit more. 50kg are normalised in a 4x pack. With more soft damage than 250kg and 2x100kg but penetration of 50kg.

[bigred]

What exactly did the japs use as bombload w/ the sally in real life?

[Shark7]

May have to give it a try for myself in a test scenario, just to see what exactly happens. [;)]

[el cid again]

ORIGINAL: Commander Stormwolf


japanese were in desperate need of a heavy 4E or high speed 2E

once 1943 came around.. the IJAAF Bmr corps was useless against the
masses of P-47s and P-38s in new guinea.

Well - eventually they got there. It was long classified how the joint Army/Navy
raids on B-29 bases using the Ki-67 really was messing us up. They would
attack when we were about to take off - bombers fueled and armed on the runways -
and it as almost impossible to miss. Once one went up, it set off others. Small
raids had grossly disproportionate effects. So we decided to go for Iwo Jima -
a base they needed to refuel in route. These were long flights - more than 24
hours - and they arrived at night - when figher opposition was negligable. They knew
when to attack (lacking satellite imagery we would use today) by watching our
weather planes - we snooped the target precisely the same number of days before the
raid every time. And we took off at the same hour each time - so we were predictable.

There was another fast two engine bomber as well - but it was navy only.

I found a report last month that the one four engine bomber design they completed - its
factory and tooling were destroyed in a B-29 raid so it never made it into service -
was so highly regarded it was not destroyed - as is usually reported Allegedly the
plane was "found" not long ago at a development facility. I am going to attempt to see it
and find the reports allegedly used in our designs since. I worked on B-1 and don't remember
anything of the sort! We will see.

[el cid again]

ORIGINAL: bigred

What exactly did the japs use as bombload w/ the sally in real life?

I just told you - quoting Francillon

20 bombs was designed in and never changed as the max load = 1000 kg - 20 x 50 kg

15 bombs was the standard load = 750 kg - 15 x 50 kg

When the 500 litre bomb bay tank was fitted, it took up the space of 4 x 50kg bombs -
so on a standard mission - which didn't carry 5 - no change whatever.
When on a max load mission, all 20 were STILL carried - just 4 were external -
under the wings. Also according to Francillon.

[el cid again]

ORIGINAL: Dili

That is why it is advisable to normalize bombs with bomb packs if possible. Say 100kg bombs are normalized in double pack=200kg with penetration of a 100kg but with soft damage of a 250kg or a bit more. 50kg are normalised in a 4x pack. With more soft damage than 250kg and 2x100kg but penetration of 50kg.

Can you say that again in plain English? What is a "bomb pack" - a fictional bomb statistically modeled?

From early days - although I don't remember seeing any plane use em - UV/WITP/AE always had cluster bombs as devices. Those are triple bombs in one device - although one can quibble with the math used to define them. It is those devices I put on B-25G - famous for them - and on some B-29s. Except I did define them in a statistically consistent way with other bombs. A 100 pound ICB = about 3x 33 pound bombs; a 400 pound ICB = about 3x 133 pound bombs. So what the effect would be if three small bombs is used to define one bigger one. Is that the sort of thing you mean?

[treespider]

ORIGINAL: el cid again

ORIGINAL: Dili

That is why it is advisable to normalize bombs with bomb packs if possible. Say 100kg bombs are normalized in double pack=200kg with penetration of a 100kg but with soft damage of a 250kg or a bit more. 50kg are normalised in a 4x pack. With more soft damage than 250kg and 2x100kg but penetration of 50kg.

Can you say that again in plain English? What is a "bomb pack" - a fictional bomb statistically modeled?

From early days - although I don't remember seeing any plane use em - UV/WITP/AE always had cluster bombs as devices. Those are triple bombs in one device - although one can quibble with the math used to define them. It is those devices I put on B-25G - famous for them - and on some B-29s. Except I did define them in a statistically consistent way with other bombs. A 100 pound ICB = about 3x 33 pound bombs; a 400 pound ICB = about 3x 133 pound bombs. So what the effect would be if three small bombs is used to define one bigger one. Is that the sort of thing you mean?

What Dili was trying to say instead of giving your Sally 20 x 50 kg devices...

...create a new device ... for example...the(5x50kg Bomb)Bomb which represents a stack of 50Kg bombs... with similar albeit different stats than a single 250 kg bomb...ala... lower values vs. hard and slightly higher values vs. soft.

Then equip your Sally with 4x(5x50kg Bomb) devices.

That way Sally still only gets 4 bites at the apple instead of 20.

[Dili]

Thanks treespider.

[el cid again]

ORIGINAL: treespider

ORIGINAL: el cid again

ORIGINAL: Dili

That is why it is advisable to normalize bombs with bomb packs if possible. Say 100kg bombs are normalized in double pack=200kg with penetration of a 100kg but with soft damage of a 250kg or a bit more. 50kg are normalised in a 4x pack. With more soft damage than 250kg and 2x100kg but penetration of 50kg.

Can you say that again in plain English? What is a "bomb pack" - a fictional bomb statistically modeled?

From early days - although I don't remember seeing any plane use em - UV/WITP/AE always had cluster bombs as devices. Those are triple bombs in one device - although one can quibble with the math used to define them. It is those devices I put on B-25G - famous for them - and on some B-29s. Except I did define them in a statistically consistent way with other bombs. A 100 pound ICB = about 3x 33 pound bombs; a 400 pound ICB = about 3x 133 pound bombs. So what the effect would be if three small bombs is used to define one bigger one. Is that the sort of thing you mean?

What Dili was trying to say instead of giving your Sally 20 x 50 kg devices...

...create a new device ... for example...the(5x50kg Bomb)Bomb which represents a stack of 50Kg bombs... with similar albeit different stats than a single 250 kg bomb...ala... lower values vs. hard and slightly higher values vs. soft.

Then equip your Sally with 4x(5x50kg Bomb) devices.

That way Sally still only gets 4 bites at the apple instead of 20.

Also thanks.

I can see how that would reduce game execution time - the many devices being evaluated prolong combat resolution - as Big Red observed.
I do not grasp why one would want to that, however, assuming the game system is "right" in a statistical sense. If a soft value ought to be
by scheme x (stock = weight of weapon in pounds) or scheme y (mine - where for HE it is 2 x square root of weight in pounds, and fractions of
that for AP) - then each device ought to be evaluated separately. Same for penetration - if penetration ought to be by scheme x (stock) or scheme y (mine) - both yielding a given penetration in mm for any given bomb weight - than each separate hit ought to be evaluated to see if it penetrates. While I could build a model to simulate 5 such hits - that is a lot of work. In general, more bombs should matter more to less hard targets, and bigger bombs - and AP / SAP - ought to matter more to really hard ones. I added another kind - bw bombs - with zero penetration - they are made of ceremic material and break on impact - but high soft value - to simulate effect on people.

Still - now I understand his point - and it is nice to do that. Also thanks to both for taking the time to explain a point.

[dwg]

I can see how that would reduce game execution time - the many devices being evaluated prolong combat resolution - as Big Red observed. I do not grasp why one would want to that

It depends on how the game evaluates bombing. 20 individual bombing attacks have a different probability distribution to one attack of 20 bombs, and whether that matters depends on whether you are attacking a point or area target. Grouping the weapons provides a way of forcing the game to evaluate them as a single attack, though it then introduces the question of whether to model the damage of the group as equivalent to a single bomb (most appropriate for a point target like a ship), as equivalent to all of the bombs (most appropriate to an area target) or as something in between.

But without knowing how the game evaluates bombing attacks, I'd be hesitant to make changes.

[el cid again]

ORIGINAL: Shark7

Hmmm, I wonder what effect changing the bomb load from 4 250kg to the 50kg or 100kg would do in game terms? It would certainly give you more room for error (that is a greater chance of something hitting) than the current load does.

This is visible in RHS - and also IRL. In the days of gravity bombs, more, smaller bombs give you a much better chance of damage to a soft target.
More chances to hit the runway sort of thing. Which is why modern bombs often break into large numbers of tiny bomblets.

The time you want a big bomb is vs a hard target. In RHS I tell you the load of the plane in its model name. You may get 2 to 4 different versions of the very same "model" - armed with AP (or SAP), HE, or ICB weapons, or torpedoes, or ASW bombs (or air DC which amounts to the same thing). You can use the "wrong plane for the mission" with some chance of damage, but it more effective to choose the kind of load approprate for the kind of target.

I just got a report from a game run to completion in 1946 where vast armada's of B-29s were not effective in strategic bombing - not even incendiary bombing at night (which is worse than atomic bombing - the worst bomb raid in history was Tokyo - not Hiroshima or Nagasaki. And two raids in Europe were worse as well - Dresden and Hamburg). But small numbers of larger bombs did not work. My tests with ICB on B-29s - or even on B-25Gs - indicates devastating results.

[el cid again]

ORIGINAL: dwg

I can see how that would reduce game execution time - the many devices being evaluated prolong combat resolution - as Big Red observed. I do not grasp why one would want to that

It depends on how the game evaluates bombing. 20 individual bombing attacks have a different probability distribution to one attack of 20 bombs, and whether that matters depends on whether you are attacking a point or area target. Grouping the weapons provides a way of forcing the game to evaluate them as a single attack, though it then introduces the question of whether to model the damage of the group as equivalent to a single bomb (most appropriate for a point target like a ship), as equivalent to all of the bombs (most appropriate to an area target) or as something in between.

But without knowing how the game evaluates bombing attacks, I'd be hesitant to make changes.

I think this is a misunderstanding. One might replace many small bombs with one device - but if one were "honest" about it - one would calculate statistically what the effect of the many small ones would be and give that value to the one device. Otherwise - it isn't modeling the same weapon - but creating a new - and ficitonal - one. It is, however, true that area targets differ from point targets. And for them - you want many small bombs - to the extent today a big bomb may carry literally hundreds of bomblets!

[Shark7]

Well I tried the 20 x 50kg route...interesting results. Lots of ship hits, lots of airfield hits. Definately makes the IJA much, much more effective than stock...gonna carry it out a little farther in my experiment to see if it seems to be skewing it too badly.