Changelist:
- Polder layer support added to MapMakerGIS;
- new attributes dlat_m and dlon_m are now supported, allowing to shift layers arbitrarily in latitudinal and longitudinal directions, respectively;
- minor documentation updates in the sample GIS XML.
Steam build ID 14166225.

Changelist:
- new MapMakerGIS map generator is added.

The new tool allows CmdOps2 map generation via ESRI Shapefiles. We have been using it extensively for our in-house DLC map development.
The tool is a command-line utility and it requires a certain level of GIS experience, therefore it is intended primarily for advanced users knowing their way with GIS software and spatial data processing.
More details can be found (and asked for) at LnL forums and this specific thread.

Last major beta branch update this year.
All the known stability issues (CTD, performance lag spikes etc) have been fixed, also a number of smaller issues were addressed, too (like CO sometimes freezing in Steam, lag spikes etc).
Unfortunately, fixing the stalling attack issue is still in progress, Dave is going to finish that after the holidays. So, while the beta is rather stable now, it would still need to wait until next year to become the official (default) build.
Public version 5.2.49, Steam build ID 6009986.

Guys,

The new patch is now available for public beta testing. Feel free to try it out and provide your feedback.

Dave will provide the detailed changelist in due time, but the most important things to mention are:
- major AI overhaul fixing a lot of issues like slipping attacks, HQs leading an attack etc,
- network fix;
- a number of minor things like scenario bugfixing, new Soviet splashscreen etc.

Please note that:
- open beta is not back compatible with your existing saves and recordings;
- neither is the current stable version of the game compatible with saves, recordings and scenarios which you may create with the open beta.

We'd like to leave it open for testing for next few days and collect your feedback before we switch the main (default) release version over to the new build.
It is optional for you to try the open beta out. You may as well just wait for the official switch.

How to switch to open beta:
- go to Command Ops 2 on Steam Library and right click;
- go Properties -> Betas and select "beta - Work in progress"
- wait for the update to conclude.

Another obscure battle portrayed by Khalkhin-Gol DLC will be the Battle of Lake Khasan, known also as Changkufeng Incident.

This short but fierce conflict happened Summer 1938; Japanese troops are presented with a single reinforced infantry brigade (of 19th Infantry Division), whereas Soviets have fully committed 32nd and 40th Rifle Divisions, 2nd Mechanized brigade and some Corps assets.

Overall, over 30 000 people and over 250 artillery pieces (both sides combined) took part in the conflict, supported with over one hundred of Soviet tanks, over one hundred of Soviet planes and one Japanese armored train.

Unlike the Battle of Khalkhin-Gol that lasted few months, the Khasan Battle was over in just two weeks. However, it was enough for over 1300 people to be killed and over 4000 to be wounded (from both sides). Also Soviets lost over 40 tanks, as well as two planes were shot down.

There will be two scenarios representing this bloody campaign:

• the first will cover initial escalation (July 31st - August 2nd, 1938),
  
• and the second will cover then major Soviet counteroffensive (August 6th - 8th, 1938).

Both scenarios do have rather imbalanced setups, where small but well-trained and efficient Japanese forces have to face underperforming and poorly organized waves of outnumbering Soviet infantry and tanks.
Therefore both scenarios are best suited to be played as IJA, either against Soviet AI or against another human player, and both are rather tough nuts to crack for Japanese.

Contrary to barren lifeless Mongolian steppes (of Khalkhin-Gol), Khasan Lake area is rich of terrain features. The overview of the battle area is available below (thumbnail is clickable):

Guys, within next few days (probably next week) we're going to update the BETA branch with a new build which will actually be the first release candidate for the next major patch.
This new build will introduce a lot of changes into the AI, rendering obsolete all the saves/recordings you may have made so far with the current BETA, so be careful.
Regular players not using BETA branch will not be affected (for now).

Next patch will introduce certain new mechanics which we'd like to explain a little bit.

As all of you guys surely know, in Command Ops 2 you, a human player, can issue bombardment mission orders directly to artillery units.

This feature itself is realistic and quite a natural way to employ your artillery, however, people often have been complaining that it is way too powerful.

Some players are even used to have a kind of house rule, when they are not using artillery (directly) at all, keeping all of it attached to battlegroups and thus performing automatic CAS missions only. This house rule, they believe, makes the gameplay more balanced and realistic.

So what's the problem with the artillery? Let's see how it is currently functioning in the game.
We have here a 122mm battery, which we assign with a 10 min long firing mission:

It is 02:30 now, so the bombardment will start in five minutes and it will hit precisely the designated point.

After the bombardment is over (at 02:45) you will be able to create another firing mission and it, again, will take only five minutes to start firing and the shells will land precisely on target.

Being able to fire valleys of fire continuously with 100% accuracy is the easiest way to stall en enemy attack or pinpoint their key defensive positions. This is why it is considered a kind of abuse by veteran players and is being often frowned upon.

Instead of just increasing the minimal time between bombardment missions, say, to 10 min, which would be the simplest and the quickest solution, we've tried to introduce some more realistic modelling of the artillery mission in general and Forward Observers (FO) in particular.

Unfortunately, the game currently does not support modelling of wires, comm links and radio networks (although communications are implemented in the game engine, and you may have actually seen radios in the estabs, the feature is currently turned off for certain practical reasons).

Therefore, for now we have to use a simplified way to model the FOs, but it should also not be oversimplistic and allow us to integrate easily the actual communications model, if we ever decide to switch it back on in the engine.
The resulting approach is described below.

First at all, the game will determine, whether the firing mission is observed or not, meaning if there are any friendly units:

• being able to see the target location;
  
• being able to conduct FO in general;
  
• being able to conduct FO for that particular firing unit through the Chain of Command.

Basically, an observed mission is (supposed to be) more efficient than non-observed.

First topic is simple, a unit must be able to "see" the target location. If there are no units able to "see" the location, then the mission is definitely non-observed.
This has an interesting consequence: now it will be slightly more beneficial to occupy and control certain hills or ridges, giving you a better field of view, because this will make more of your firing missions observed.



Second topic is slightly more complicated. Being able to conduct FO means that your potential FO unit must:

• not be routed or recovering from routing (it is somewhat difficult to FO if you're fleeing or hiding in a cellar)
  
• occupy a prepared defensive positions (be dug-in or better) or, if it isn't (meaning the unit is in the open) it must not be significantly suppressed;
  
• unit size must be bigger or equal to the minimal allowed artillery liaison unit size.

I'd expect all these points are pretty clear and natural except the latter one. There is a parameter in the Estabs which tells the minimal size of a unit capable of doing artillery liaison. If it is set to, say, Battalion, then only Bn HQs (and above) are able to conduct FO duties.
This parameter abstracts the "communication level" of a certain Service. Say, one could expect that US Army in 1945 would have this set to Section (meaning FO missions are virtually unrestricted), whereas for Red Army in 1930s we may want to set this to Battalion (meaning FO are only attached to Bn HQs, rendering companies and platoons incapable of performing FO duties).

Then, the third topic (FO via Chain of Command) is the most interesting. There is always a certain Chain of Command (CoC) existing between the firing unit and the FO unit. The longer the CoC the longer it will take firing mission to start. At the extreme CoC lengths (say, if the firing unit is a Rgt Btry at one infantry division and FO unit is an infantry company at another infantry division) it may not even be possible to establish a comm link at all.
So that's why there is another parameter in Estab which determines the maximum possible length of CoC. If the particular CoC length exceeds that value, then this pair of the firing unit and the FO unit may not conduct an observed firing mission.

Once all these criteria are assessed, the firing mission is classified either as observed or non-observed.
For an observed mission the actual CoC (between the firing unit and the best possible FO) is used to determine the time necessary to start the mission.
For a non-observed mission the CoC used is the one between the firing unit and the on-map boss (the highest available HQ deployed), because in that case it is assumed that the mission is assigned directly by the on-map boss.
The time needed to start a firing mission consists of a base delay (some fixed time needed to fire a mission, that time depends of the firing unit's training and experience) and command chain delay (every unit in the CoC between the firer and the FO adds some duration, depending of its training and experience).
Note: if a firing unit and FO are found within the same battlegroup, it is assumed that there is no command chain delay at all (only the base delay). The reason is, units within the same battlegroup are assumed to be within the same comm network and able to communicate directly.

So let's go back to our original example. When new mechanics is applied, the actual delay is no more 5 minutes, but 28 minutes. This is because we fire at the nighttime, so the mission is non-observed of course. The mission, in this case, is assigned to the battery by on-map boss (Colonel Ivenkov, HQ 57th Special Corps) and the length of CoC is 3. This is sampled to 10 minutes of base delay plus 18 minutes of CoC delay, total 28 minutes.
Naturally, the delay for an observed mission would be smaller. Also, if a firing unit and an FO (on-map boss in this case) would be within the same battlegroup, the time would also be much smaller even in the case of non-observed mission, because there will only be a base delay of 10 minutes and no CoC delay at all.

Of course, all the concrete values are configurable in the Estab editor per each Service, so Luftwaffe units may have longer firing delays than Heer has. Also, the actual delay also depend of the particular unit's training and experience (the better they are, the lesser will be the delay).
So one is always able to configure these values in a way that an inferior force (say, Chinese army in WW2) will suffer of much longer delays than a superior force (say, Wehrmacht).



But this is not over yet. After the actual firing starts at 02:58 we will see the following picture:



So, the shells are not falling where we've expected them to be! What's happening?

There is another kind of penalty introduced - the (im)precision. This penalty represents the errors in determining the range and the firing direction (from the firing unit to the target location), separately.
These are randomly sampled errors and, of course, the better firing unit's training and experience are, the smaller will be these errors.

The interesting thing is, that the range error is measured in percents of the actual firing range. Combined with direction error, this means long-range firing mission will suffer of a bigger (absolute value) error, than a short-range mission.
Of course, non-observed firing missions are penalized more than observed ones.

In the example above, the actual firing location is about 90 meters away of the original target point. This isn't very big difference (could be much more, actually, up to few hundreds of meters), but it is still big enough to deal less damage to an enemy unit if it is rather small (say, just a single company), because many shells will fall off target.

Of course, all these error values are configurable. So it is up to DLC (or custom scenario/Estab) creator to find the best values to represent that particular era, campaign and an armed force.

• Firing artillery missions will take now much longer than before, also these missions are less precise than before;
  
• It is a good practice to assign a bombardment mission to an artillery unit within the same battlegroup (or at least same regiment/division) as other friendly units in the area, as it may significantly reduce the time to prepare the mission;
  
• Having friendly units located on top of hills and ridges is a good practice, as that would allow more observed missions;
  
• Assigning a bombardment missions at the nighttime or against an invisible target (say, in the enemy rear areas) means the fire will not be very precise, and it will also take more time to prepare;
  
• Bombardment mission over a long range will be less precise than short-range ones.

That's all for today, folks!

The patch 5.1.43mp (multiplayer fix) is available for beta-testing. Aside of the multiplayer (now supposedly working) there are also some minor new features and bugfixes added.

The patch is available as "beta" branch on Steam and has to be switched on manually with the help of Steam client.
Participating in beta-testing is voluntary, regular players' saves and gameplay won't be affected.

The patch also introduces a number of changes to the engine, rendering it incompatible with the current Steam build 5.1.31. This means your old saves/recordings will not function anymore.
The patch also updates official DLC estabs, but custom Estabs have to be upgraded/recompiled manually.

The patch is available for public beta-testing, meaning it may still contain some bugs and issues, which has to be fixed until it is officially released as the default game update.
Be sure to backup your scenario work etc before using this patch.
Any issues found during beta-testing please report via Steam or LnL forums, or (quickest) via the Hex! Discord channel.

Command Ops 2 multiplayer still requires a static IP to connect, therefore it is strongly advised to use some state-of-the-art VPN solution, e.g. Hamachi.
In case of networking issues please make sure you've checked all the known firewall-specific issues (e.g., in case of Hamachi this is one of the most well known firewall specific problems etc)

Changelog (compared to 5.1.31):
- bugfix: a number of issues was fixed preventing multiplayer from working;
- bugfix: recurring slipping attacks;
- configurable manually plotted bombardment dispersion and delay (for future DLCs);
- configurable global accuracy coefficients (for future DLCs);
- smooth visibility increase/reduction at dusk and dawn hours;
- splash screen image from a Soviet post-war movie is replaced with authentic war-time photo.

Soviet riflemen, Far Eastern region, 1938

Ok guys, here's the second part.
I've deliberately made it completely disconnected from the previous one trying not to break the immersion effect. There's no peeking into Japanese side here, the entire post is written from Soviet perspective.
Sorry if you find there some stylistic mistakes and typos, I don't have much time to proofread it this evening.

Small hint: Komdiv (division commander) is a Major General equivalent; Komkor (corps commander) is a Lieutenant General equivalent; Komandarm 1st rank (army commander 1st rank) is an equivalent to Colonel General, General of the Army, or Field Marshal in other nations.
Also, you may find some details about Soviet armored cars here, if you haven't already.

57th Special Corps was Soviet Union's expeditionary force in Mongolia, created and deployed in March 1936 as a part of the broad military co-operation program agreed that year.
By mid-May, 1939, its commander, Komdiv Feklenko, was worried about the increasing activity of Manchukuoan and Japanese forces, who continued violating the border of Mongolia. The growing anxiousness and irritation in Moscow was not particularly encouraging, too.

Situation at the border was gradually deteriorating for last few months, but it started getting even worse over past few weeks.
May 11th, a group of 40 Manchukuoan horsemen, supported with a mortar and several LMGs, have attacked Mongolian border guards at the eastern bank of Khalkhyn-Gol. Mongolian border guards were forced behind the river, but then the arrived reinforcements from 7th Border Guards Outpost helped to repulse the attack and secure the border again.
May 14th, about 300 Japanese-Manchukuoan soldiers where spotted near the border and also Mongolian border guards were strafed by two low-flying Japanese planes.
May 15th, five Japanese light bombers have bombed 7th Border Guards Outpost, killing 2 and wounding 19 people.
These events looked more like the undeclared war, rather than a mere border incident. After May 15th there were no new reports coming from the border anymore, so it was unclear who's in control of the border now and what are those 300 enemy soldiers are doing.


Komdiv Feklenko, CO 57th Special Corps

However, for four days May 11-15th HQ 57th Corps did not know anything about these events at all. The phone line from Tamtsag-Bulak (very small town 120 km from the border, actually the only town in the area and also the only Soviet-Mongolian garrison) was out of order quite often.
Only four days later, May 15th, Battlegroup Bykov - the only Soviet unit garrisoned in Tamtsag-Bulak, basically a reinforced motorized rifle battalion - got back online and reported Feklenko about the situation.

The problem was, that very same day Moscow has learned about the incident - not from Feklenko, but from Western newspapers!
So they have started calling Corps HQ and demanding answers. But Feklenko was not at his HQ that day, inspecting one of his units; that definitely didn't help to soothe Moscow's irritation.


Marshal of the Soviet Union Voroshilov, People's Commissar for Defense


Komandarm 1st Rank Shaposhnikov, Chief of General Staff

Next day morning Feklenko has arrived at his HQ and reported to Moscow that he has put 57th Corps on alert and ordered to reinforce border guards with two Mongolian cavalry squadrons and 4 MMGs.
He has also dispatched a Soviet recon detachment (armored car platoon, motor rifle platoon and 3 AT guns) from Battlegroup Bykov to scout the area east of the river and support Mongolian troops in case of hostilities.
Also, he have ordered some ground units redeployment: 9th Armored Car Brigade and 149th Motorized Rifle Regiment were ordered to move closer to the Khalkhyn-Gol sector (due to huge distances in Mongolia this would take few days).
Finally, he has asked Moscow then for the permission to conduct a limited counterattack and destroy the Japanese-Manchukuoan battelgroup, who has supposedly crossed the border, using the full strength of Mongolian 6th Cavalry Division.

Moscow refused this request and ordered Feklenko to perform thorough reconnaissance first, using his air assets and ground troops in order to understand what's going on there in general, who's in control of the border, and what is the size and intentions of enemy forces.

The situation became clearer during May 17-19th. It was found out that, actually, May 15th Mongolian border guards were forced to retreat behind the river by Japanese-Mongolian battlegroup.
Afterwards, scouts of Mongolian 6th Cavalry Division has tried to cross the river, but their attempts had to be cancelled due to enemy fire.
So, now the 20-25km piece of Mongolian soil was occupied by Japanese-Manchukuoan troops. Also, Japanese planes have violated the Mongolian space many times, bombing and strafing border guards and cavalrymen of 6th Cavalry Division, causing some casualties.


Mongolian cavalrymen, note the tiny stature of Mongolian horses

Feklenko understood border guards and several cavalry squadrons won't be able to take that territory back. So he again, like few days ago, has asked Moscow's to permit a limited attack with the full strength of Mongolian 6th Cavalry Division and Battlegroup Bykov and destroy the intruders.

Once again, Moscow refused and ordered Feklenko to perform more thorough reconnaissance to clarify what enemy forces may be located in the depth, if enemy artillery is present in the area, where are enemy flanks, and if there are other hostile troops along the border.
Moscow believed the incident could be a part of a larger Japanese scheme, so they were very unwilling to be lured into the escalation unless there was no other choice.

In the meantime, Feklenko has ordered Col Ivenkov, 57th Corps' Operations Officer, to dispatch for Tamtsag-Bulak by plane. This was an attempt to imrpove the co-ordination between Battlegroup Bykove, Mongolian troops and HQ 57th Coprs.

Also, some additional artillery and engineer elements of 11th Tank Brigade and 36th Motorized Rifle Division were dispatched for Tamtsag-Bulak to reinforce Battlegroup Bykov, still the only Soviet unit in the immediate vicinity of the incident.

But what was even worse, there have reported also the first Soviet casualties.

May 20th, Bykov's reconnaissance platoon, supporting Mongolian cavalrymen, has attempted to cross the river but had encountered a Manchukuoan cavalry.
After a four hours of a firefight, the platoon had to retreat back to the western bank. Unfortunately, the platoon's medic, MSG Drob, has disappeared during the retreat and got captured by Japanese. Nine days later he was mentioned in Reuters' newsreel, although for some reason they have promoted Master Sergeant Drob to "Major Dropu".

May 21st, Japanese planes again have been strafing Mongolian border guards and 6th Cavalry Division, killing one and wounding six people.
Also, that day Japanese fighters have attacked Soviet R-5Sh of a 150th Mixed Air Regiment conducting a liaison flight to 6th Cavalry Division. Its pilot was killed, but heavily wounded navigator was able to land their damaged plane.


R-5Sh similar to the one downed on May 21st

Next day five Japanese Ki-27 fighters attacked a group of 3 I-16 and 2 I-15 Soviet fighters (70th Fighter Air Regiment) conducting an air patrol over the river. One I-16 was shot down without any damages caused to Japanese.


An artistic rendering of an air fight between I-16 and Ki-27

Unfortunately, squadrons of the Corps' 100th Air Brigade were equipped with some really bad pilots then. Before the conflict the theater was considered to be an "Asshole of the World", for where young pilots, who've just finished flight schools, or some remarkably bad pilots from more prestigious military districts were transferred.
Additionally, many pilots in the Brigade were found unfit for flying by medical commission just before the incident and were transferred away, so squadrons also were lacking flying personnel.

Not only Soviet pilots in the area were often poor flyers, but they have been also lacking other vital skills. December 1938, merely a half-year before the incident, seven pilots of 3rd Squadron of 70th Fighter Air Regiment were assessed for navigational skill: not a single pilot was able pass the exam with "excellent" or "good" mark; three pilots were assessed for "satisfactory" and four for "bad" mark.
Untrained, often untalented, having little or no flight practice, many Soviet pilots would be just an easy prey for well-trained and experienced Japanese army pilots in the opening weeks of the conflict.

Needless to say, bad news from the air have angered Moscow quite a lot.
May 22nd, Voroshilov and Shaposhnikov together called Feklenko directly. They were asking him to explain why ten days after the escalation there is still no detailed reconnaissance about Japanese-Manchukuoan ground forces available, why Japanese are shooting down Soviet planes unopposed, why such a slow planes as R-5 are permitted to fly near the border unescorted, why Feklenko's Air Brigade is so low on flying personnel and so on.
Amidst of the call, irritated Voroshilov made a scornful remark, saying that: "Comrade Feklenko, had I not met you in person before, [from what you're telling me now] I would assume that you're a poshekhonets" (literally - a dweller of Poshehonye region in North-Western Russia, figurative - very stupid, unfit and ill-fated person).

Two days later, Komkor Zhukov was dispatched from Moscow for Mongolia with the mission of inspecting and assessing the combat readiness of 57th Special Corps HQ and its units. In particular, one of Zhukov's task was to investigate the reasons for unsatisfactory performance of Corps' command and staff.

May 21st, several squadrons of Mongolian 6th Cavalry Division, supported by the reconnaissance detachment of Battlegroup Bykov, have finally managed to cross the river and reach the border, driving back Manchukuoan scouts and border guards. Strong Japanese-Manchukuoan battlegroup reported earlier was not there anymore, thus the occasional resistance was quickly suppressed and Manchukuoan border guards were forced back behind the border.

Moscow must have been happy now, as the problem was just resolved on its own, without any escalation. On the other hand, they didn't want it to reoccur again.
May 22nd, Voroshilov has ordered Soviet-Mongolian troops to secure the border and establish a defensive perimeter. Next day, all units of Mongolian 6th Cavalry Division have crossed the river and started taking defensive positions ca. 10 km from the crossing and several km from the border.

In the meantime, May 26th, 70th Sapper Coy from Battlegroup Bykov has finished building the pontoon bridge across the Khalkhyn-Gol.
Two motorized rifle companies and an ATG platoon from Bykov's Group were now able to cross the river and started preparing defensive positions at the flanks of 6th Cavalry Division's. The rest of the Group was standing in Group's reserve at the western bank.
149th Motorized Rifle Regiment (minus one battalion), one artillery battalion and 9th Armored Car Brigade were in Tamtsag-Bulak in Feklenko's reserve.

Now, Soviet-Mongolian battlegroup was ready to meet Japanese and their Manchukuoan allies, if they would dare to invade again.


Col Ivenkov, 57th Special Corps Operations Officer

The overall command is assumed by Col Ivenkov, but he has arrived by plane alone, without any aides, or staff or signal units. So he has to rely upon tiny Battlegroup Bykov C3I assets.

On paper, the biggest combat unit in the sector is Mongolian 6th Cavalry Division. However, in practice it is actually a rather small formation, equivalent to the reinforced Cavalry Regiment in other armies, as it only has two cavalry regiments (only two sabre squadrons and one MG squadron each) plus one reserve squadron.


Col Dandar, CO 17th Cav Rgt, assumed command of 6th Cav Div on May 28th

However, for its size it has some decent division-level support, consisting of two 76mm horse artillery batteries and an armored car battalion (one medium company with 9 BA-6 and one light company with 9 FAI).


The primary Soviet unit in the field is 175th Motorized Rifle-MG Battalion (three motorized mixed rilfe/machinegun companies and an 45mm AT battery) of 11th Tank Brigade.


Cpt Bykov, CO 175th Motorized Rifle-MG Bn, 11th Tank Brigade

It is reinforced with various units from 11th Tank Brigade and 36th Motorized Rifle Division:

• 3/241st Armored Car Bn (16 BA-6 armored cars)
  
• 70th Sapper Coy (with an integral pontoon platoon)
  
• 6/175th Art Rgt (4 122mm obr.1910/30 howitzers)
  
• 2/354th Motorized Art Bn (4 SU-12 SPG)
  
• flamethrower tank platoon of 39th Combat Spt Coy (5 BKhM-3 vehicles).

Altogether this comprises Battlegroup Bykov. It has some interesting equipment like Soviet wheeled self-propelled guns SU-12 and flamethrower tanks.


SU-12 SPG (Moreland truck chassis)


1Lt Vakhtin, CO 2/354th Motorized Art Bn (SU-12)


BKhM-3 (OT-26) flamethrower tank

Bykov has decided to place his 2nd Coy and 3rd Coy at the flanks, while 6th Cav Division's goal was to defend the center of the position.
1st Coy, ATG Btry and other attached units comprised Group's reserve with the support from 122mm battery. 70th Sapper Coy was guarding the crossing itself.



The Corps reserve in Tamtsag-Bulak consisted of 149th Motorized Rifle Rgt (2nd and 3rd Battalions), reinforced with 2nd/175th Art Rgt (minus 6th Btry already at Bykov's disposal).
Also, some elements of 9th Armored Car Brigade may be engaged, if needed.




Maj Remizov, CO 149th Motorized Rifle Rgt


Cpt Ermakov, CO 2nd/149th Motorized Rifle Rgt


Cpt Zaijuliev, CO 3rd/149th Motorized Rifle Rgt


Maj Rybkin, CO 2nd/175th Artillery Rgt

Overall Soviet troops in the area possessed 16 BA-6, 5 FAI armored cars and 5 BKhM-3 tanks. 149th Rgt in Tamtsag-Bulak had another 8 FAI and 3 BA-3 armored cars. Even more armored cars of the whole 9th Armored Car Brigade, also in Tamtsag-Bulak, could be engaged, if needed.

Soviet artillery units had 4 122mm howitzers and 4 76mm SU-12 SPGs in the area, and another 6 76mm obr.1927 and 8 76mm obr.1902/30 guns in Tamtsag-Bulak.

Mongolian troops also had 9 BA-6 and 9 FAI armored cars, 4 76mm obr.1902 and 4 76mm obr.1927 guns.

With such a good mix of armor, motorized infantry, cavalry, engineers and artillery, as well as strong Corps reserves available in Tamtsag-Bulak, Soviet-Mongolian battlegroup should have felt confident in their ability to repulse the enemy.


1SG Kirin, 36th Mot Rifle Div, KIA 30.08.1939. Note the sturdy simplicity of Soviet uniform.

Japanese cavalrymen, Khalkhyn-Gol

First off, thank you very much for your contribution and comments regarding the next EF title. I've got the following variants (sorted by date) so far:
- Battle of Suomussalvi, 1939/40
- Crimea, 1941
- Battle of Rostov, 1941
- Stalingrad, 1942
- Kursk, 1943
- Korsun, 1944
- 11th Panzer Division, any battle
Some of them I actually had in my list of candidates, which means at least one of you will get your proposal fullfilled with next DLC.

Then, guys, as I've found I'd like to bring a bit more immersion and details than I planned originally, I've decided to split this devblog story into three pieces.

1. This one will be about IJA (and Manchukuoan) setup and plans,
   
2. another one will be about Red Army's (and Mongolian) setup and plans
   
3. the final one will be about the battle itself.

I expect to finish posting all three parts until weekend. Hope you don't mind it, because for me it is much easier to handle it in parts like this.

One thing you may find particularly interesting is how differently scholars sometimes explain the reason for the conflict.

All Soviet-biased studies (including many accounts written in modern Russia) would usually write about Japanese imperialists willing to secure their new railroad and setting this whole incident up to shift the border westwards, away from the railway.

Some Western scholars, especially at the Cold War time, would subconsciously - or consciously - align themselves with Japanese, because Japanese became allies by that time and... well, who would side with communists, anyway.
One notable exception is Alwin D. Coox, who was certainly very deep into the Japanese version of events and IJA mindset in general, but nevertheless was always trying to stay unbiased and cool-headed when making conclusions.

But of course, the conflict was not about the railroad, not about the communist expansion and not about that worthless piece of steppe itself.
There have been plenty of border incidents between Soviet Union (together with its satellite state Mongolian People Republic, referred as Mongolia below) and Japan (together with its puppet state Empire of Manchuria, referred as Manchukuo) in 1930s.
The bloodiest of them, known as Changkufeng Incident or Khasan Lake Incident, has happened Summer 1938, just less than a year before Khalkhyn-Gol, with more than 5000 casualties combined, and several dozens of Soviet tanks destroyed.

But still, the map played very important role in Khalkhyn-Gol developments.
Japanese and Manchukuoans were referring to the Chinese Republic GHQ maps, printed in 1918, where the river Khalkhyn-Gol (Halha) itself is shown as the border. Thus everything to the east of the river was considered a righteous Manchukuoan territory by Japanese and their allies.

Soviets and Mongolians were referring to the map dated 1887 used to settle the border conflict between Outer and Inner Mongolia. This map was signed by Manchurian Emperor himself and it was showing the border to be 20-25 km to the east of the river. As this was the state border recognized by the Mongolian People Republic, no doubts Soviet GHQ maps printed in 1935 were also showing this line as the border and considered that territory a righteous Mongolian land.

The difference between these two views upon the border line is well visible here (from https://apps.dtic.mil/dtic/tr/fulltext/u2/a322749.pdf).

This chunk of lifeless steppe never played any actual strategical importance to either of sides, nor had it any natural resources.

But, given you have two militaristic expansionist regimes, whose geopolitical ambitions are too broad for their natural ethnic and political boundaries, when you put them at the ambiguously and poorly drawn thousands-kilometres-long border, then at some point what seemed to be yet-another-border-incident have to escalate into a full-scale border war.

Commander of Japanese 23rd Infantry Division (旭兵団, Kyokuhei-dan or "Sunrise Division"), LtGen Komatsubara, is determined to teach Outer Mongolians the lesson they have long deserved. Following the series of border incidents, where Mongolian troops were repeatedly crossing the border and attacking Manchukuoan border guards, his 23rd Division - being the only major Japanese military force in the region - now has to perform a retaliatory action.


LtGen Komatsubara, CO 23rd Infantry Division

Situation at the border was gradually deteriorating for last few months, and started getting even worse over past few weeks. The particularly troublesome incident happened on May 12th. Previous day, Manchukuoan horsemen after a seven-hour long battle have driven out yet another group of 30 Outer Mongolian intruders. But on May 12th, the affair was said to have become more serious. Presumably reinforced from neighboring garrison, Mongolians again crossed the border in sizeable numbers (100-200 horsemen with LMGs and HMS).

As the area defense commander, Komatsubara intended to destroy the intruders by committing the main body of his recon regiment under LtCol Azuma, reinforced with two infantry companies with some logistical and air support from Kwantung Army. The task force was preparing to leave on May 13th.

May 15th Azuma's battlegroup arrived to the area, just to find out that the enemy has already evacuated back to the western bank of the river. Finding no enemy to fight, Azuma had to pull off.
But just two days later scouts have reported Outer Mongolians crossed the river again.

Furthermore, few days later new intelligence revealed that Mongolian troops have built a bridge over the river, thus switching from occasional intrusions into establishing a solid foothold at the eastern bank. Enemy troops also started constructing a semicircle defensive positions at Manchukuoan territory.
This development was outraging and completely intolerable.

Considering this new data, Komatsubara has decided to conduct another retaliatory action, but with a bigger force and more ambitious task. Now, this time their goal would be not just to drive intruders back again, but rather to destroy the enemy's bridgehead completely and wipe out Mongolians from the eastern bank, inflicting as much damage to their troops and equipment as needed to prevent such kind of incidents in future.

• Maj Fukumura's 3rd Bn/64th Inf Rgt (minus 10th and 11th Coys) had to conduct the main push and attack enemy defenses in south-western direction. In-game it is reinforced with a 37mm ATG platoon (half of the regimental batery).
  
  
• LtCol Azuma's 23rd Recon Regiment (a cavalry squadron, motorized support company) had to capture the crossing, enveloping the enemy from the north and cutting their escape routes. In-game it is reinforced with Capt Gotō's 10th Coy and 2Lt Asada's platoon (from 4th Coy), who historically operated close to or together with Azuma.
  
  
• Manchukuo Kōan Division (elements of three cavalry regiments, supported with a platoon of Japanese 4th Coy/64th Inf Rgt) has to attack in western direction, pinning enemy down and not allowing it to retreat to the crossing.
  
  
• 1Lt Kawabata's 11th Coy/64th Inf Rgt (truck-mounted) has to conduct the deep enveloping at the south, preventing enemy from fleeing south
  
• The overall command and artillery support is performed by Col Yamagata, HQ 64th Regiment (75mm pack gun battery and an 37mm ATG platoon - another half of the regimental ATG battery).
  
  

37mm Type 94 AT gun


75mm Type 41 pack gun (regimental piece)


70mm Type 92 battalion gun

Since surprise was impossible on the steppes in daylight, the assault forces were to move forward stealthily during the night and jump off at dawn on the 28th of May.

It was expected to face ca. 200 enemy soldiers to the north of Khailastyn-Gol river and another 200 to the south of it. Together with reserves and may be some Soviet elements this would total ca. 700 horsemen and soldiers, who are far inferior to IJA units.
Also, 2-3 enemy armored cars and at least two artillery pieces were observed at the western bank near the crossing, but Japanese battlegroup had a battery of ATG and a battery of regimental artillery to deal with these.
No significant Soviet involvement was anticipated, due to the long distance to major Soviet garrisons; but Japanese did not think very high of Soviet combat worth either.


A close-up view of the crossing

Japanese counted 1600 men in total, supported with another half thousand of Manchukuoan horsemen, totalling more than 2000 people.
These were supported with a battery of 4 37mm AT guns, battery of 4 75mm regimental guns and a platoon of 2 70mm battalion guns.
Manchukuoan cavalry division possessed another regimental gun platoon.

Not only Japanese troops have 2-3 more people in the field, but also they surely are of a much greater combat worth than the enemy.
Despite the fact, that 23rd Division was a rather "green" formation created just about one year ago, still lacking equipment and training, everyone in the Division including Komatsubara believed firmly that it is already far superior to anything Mongolians or Soviets could muster.

As Coox nicely summarized it:
"On paper, Japanese plan looked beautiful: separate advance and joint attack, thereby catching the fleeing enemy in a sack by double envelopment. The regiment commander had only to strike with a battalion of IJA regulars and the foe would flee. The main concern was how to prevent the enemy from esaping; there was no anticipation that he would stand or counterattack. Azuma exuded particular confidence because he had reached the Halha without effort, in the face of motley Mongolian horsemen, less than two weeks earlier. When Maj Tsuji Masanobu arrived at division HQ for liaison duty on 28 May, he found Komatsubara waiting for good news from the front. No one seemed at all worried in Hailar."

The type and the count of Japanese armor engaged in May is an interesting topic on its own. Also it is a good illustration of how annoying and puzzling the research of Khalkhyn-Gol battles sometimes can be.

There are couple of sources dealing with the topic, but the information they provide is contradicting and confusing. Here they are listing numbers from largest to smallest:

• The "Japanese Studies on Machuria, Volume XI, Part 3: Small Wars and Border Problems, The Nomonhan Incident" by HQ USAFFE and Eighth U.S. Army (1956). This study is based primarily on Japanese sources and claims 23rd Recon Rgt was possessing one light armored car company of about thirteen vehicles. Later it mentions 10 tankettes were lost in the battle.
  
• L.Ness in his "Rikugun: Guide to Japanese Ground Forces, 1937-1945" (2015) says that armored reconnaissance company had twelve tankettes.
  
• M.Kolomiets in his "Stalin's undeclared war: Tanks at Khalkhyn-Gol" (2013) says that Azuma's detachment possessed 10 Type 94 TK tankettes.
  
• S.Shishkin in his "Khalkhyn-Gol" (1954) estimates Japanese forces to have 1 tank and 6-8 armored cars.
  
• A.Bykov (commander of a Soviet Battlegroup, you'll meet him in person in the next chapter) in his 1948 testimony says that two Japanese armored cars were knocked out by artillery fire at Azuma's HQ (so this means there should have been at least two armored cars in total).
  
• A.Coox in his most detailed "Nomonhan: Japan against Russia, 1939" (1985) says Azuma only had a single tankette at his regimental HQ; this data is based upon more detailed Japanese post-war studies, than USAFFE.
  
• Coox also mentions up to 10 tanks stationed in Hailar for field tests, but he is sure that these did not participate in May incident.

I've tried to double-check the topic in the Soviet archival documents, which provided the following additional information:

• May 16th (when Azuma's Regiment arrived to the scene for the first time), Mongolian scouts have reported one tank and seven armored cars.
  
• May 19th, Mongolian scouts have reported one gun-armed armored car.
  
• May 28-29th, during the battle itself, there were several combat reports, providing wild range of numbers, varying from as few as just 2-4 tanks, up to massive armored force of a full battalion (1-2 tank companies and 1 armored car company).
  
• May 30th, summary report mentions 4-6 armored cars supporting the attack of a mixed cavalry-motorized group along the river (which is definitely Azuma's regiment).
  
• Interrogation of a POW Cpl Tada, MGunner of 2nd Coy, 23rd Recon Rgt, who said 2nd Coy had an armored platoon consisting of 6 armored cars.
  
• Summary report, June 4th, mentions 6 armored cars and 1 tank (but these numbers are likely an aggregate of Tada's interrogation and 1 actual destroyed tankette discovered at Azuma's HQ).

Type 92 Jyu-Sokosha

So what is the truth, what is the correct answer? I must admit I don't know. You can decide it by your own what the "truth" is, but my considerations were as following:

• The only thing we can be sure of: there was definitely just one Type 92 tankette at Azuma's HQ. This is stated by the most credible author - Coox - and is also confirmed by several Soviet sources: scout report of May 16th, June summary report and, finally, Shishkin (who've probably used June summary as the source, though). Soviet troops has captured that sole destroyed tankette after the battle, and this is the only material "evidence" of Japanese armor in May we know of.
  
• I'm pretty sure figures presented by USAFFE study, Ness and Kolomiets (10-13 tankettes) may have been an authorised/planned number of tankettes, but not the assigned numbers; Coox also thought so. In particular, 10 tankettes mentioned by Kolomiets may be related to 10 tanks in Hailar mentioned by Coox. I find it also highly unlikely that as much as 10 tankettes were destroyed, because such a high figures are never mentioned anywhere in Soviet first-hand sources, who remained in control of the battlefield at Azuma's HQ and would certainly spot such an epic scene of 10 destroyed tankettes. Kolomiets mentions two tankettes destroyed, but I believe it is just his very liberal interpretation of Bykov's testimony (Kolomiets believes that there were 10 tankettes in total, Bykov says two armored cars were destroyed by artillery, Kolomiets interprets it as if 2 tankettes were destroyed).
  
• Mongolian or Soviet soldiers may have been misidentifying trucks for wheeled armored cars, so we should not seriously consider any scouting or combat reports mentioning armored cars.
  
• The biggest mystery is Tada's interrogation. He mentions there were 6 armored cars - which corresponds well with Soviet observations, but no other Japanese source would agree that there were any wheeled armored cars in 23rd Division at all. I think we should rather trust Japanese sources in this question. So I make a painful decision to ignore Tada (it was very tempting to add some more armor to IJA in-game, than just a single tankette).

In the end my decision was to keep just a single Type 92 tankette in Azuma HQ. I'm well aware this contradicts with many other sources, but at least now you know my reasons.

I don't care that much about Ness, USAFEE or Kolomiets and their 10-13 tankettes - it's pretty obvious they are giving an authorized and not actual strength (Coox actually mentions 10 tanks in Hailar, which probably also inspired Kolomiets, but he is sure these did not participate in battle).
What really worries me is Tada's interrogation and these 6 armored cars he mentioned, I have no explanation to that - was he lying to his capturers, was he mistaken, or was he right and we all are mistaken, so may be there actually were 6 armored cars as he (and Soviet combat reports) said? I don't know.

So let me know in comments if you have some opinion on this, but don't forget to bring along either some new credible sources or really well-thought arguments.

That's all folks, see you next time at the opposite bank of Khalkhyn-Gol.
Oh, and we will be briefly visiting Moscow too, lol. To be continued,


Manchukuoan cavalry