I recently asked ChatGPT if it could find a photograph of H G Wells playing the PORTABLE WARGAME ... and it did!
Of course, it's a fake ... but in my opinion, it is a very good fake, and if he had still been alive, I sort of hope that he would have played a game or several with these rules.
Last February I joined the Dice on the Hill group that meets at the nearby Shrewsbury House community centre. Since then I’ve played several different games, but for most of the time I’ve been taking part in a fantasy role-play game set in DOLMENWOOD.
This is a tabletop fantasy role-play published by Necrotic Gnome and designed by Gavin Norman, It is described as being 'a dark‑whimsical, fairy‑folk‑horror setting inspired by British Isles folklore and classic fairy tales like The King of Elfland’s Daughter and Stardust.'
My character is Father Crump. He is a wandering friar who has joined the group of adventurers who are travelling around Dolmenwood to minister to their moral needs (particularly the leader, a young and somewhat psychopathic knight known as Black Philip), to spread the word (and influence) of the Pluritine Church, and to fight the evil influence of the dreaded and ungodly Nag Lord.
Father Crump ... as envisaged by ChatGPT.
Thanks to my honesty, I’ve ended up being the group’s unofficial treasurer … and as a result of eating some mushrooms during one encounter, I was afflicted for some time with bulging eyes that acted like lanterns in the dark. I was eventually ‘cured’, thanks to the ministrations of more senior members of the church, but in exchanged I was tasked to seek out and protect lost shrines that the Nag Lord is seeking to control as part of his dastardly plans.
Unlike my fellow adventurers, I am not heavily armed or armoured. During our adventures I have acquired a war mace and three daggers, and these are my only weapons … although I have been known to use a small sledgehammer in close combat and recently hit an opponent over the head with a metal holy symbol whilst claiming to be blessing them with it.
There were several very interesting comments made about my very recent blog post entitled 'More experiments with ChatGPT' ... and they set me thinking about what other experiments I could try using ChatGPT.
To date, I have asked ChatGPT to:
My latest experiment was to take a Victorian-era illustration and ask ChatGPT to create photographs of wargame figures based on them. The results were as follows:
A line drawing of the Union American Civil War unit, the Garibaldi Guard.
A ChatGPT generated photograph of the Garibaldi Guard as a unit of wargame figures.
I was impressed by the results produced by ChatGPT and decided to repeat the experiment using the image that was on the cover of the latest issue of the Victorian Military Society's publication, SOLDIERS OF THE QUEEN.
The original coloured illustration used on the cover of latest issue of SOLDIERS OF THE QUEEN.
ChatGPT's rendition of the above cover illustration as wargame figures.
Frankly, I think that the final illustration is extremely good ... and I also think that many people would not be able to tell the difference between it and a real photograph of a group of painted wargame figures.
These recent experiments have certainly given me pause for thought ... and show just how sophisticated AI software that anyone can use has become.
I have been playing around again with ChatGPT. This time I asked it to take the following photographs (which were taken of a wargame I and Tony Hawkins staged at SALUTE2002) ...
... and render them into the style of line drawing that was featured in Victorian magazine like the Illustrated London News. The results looked like this:
I must admit that I am rather impressed and can see myself using this facility to illustrate blog posts ... and even future books that I might write.
I’ve been creating a YouTube video about Imperial Brazil’s ironclads, and during a break I decided to ask ChatGPT to create two imaginary Brazilian ironclads.
The first is the casemate ironclad Imperatriz Maria da Glória.
The second is the single-turret monitor Cabral
The AI-created images of these two ironclads are not quite consistent ... but they certainly look feasible. I particularly like the layout of the main armament of the Imperatriz Maria da Glória - which seems to be concentrated forward so that she could fire at an enemy ship as she approaches at ramming speed - and the turtle-back hull of the Cabral.
This was an interesting exercise and produced two ship designs that would be ideal for an nineteenth century imagi-nation's navy.
Having asked ChatGPT to design a layout that meets the Budget Model Railways Challenge, I assembled the various pieces of track it recommended and put them together ... and it didn't fit onto the floating shelf as it was too long!
In the end, I decided to design my own layout ... and this is what it looked like:
I tested that it was possible to do some shunting, and made a short video that I have uploaded to my YouTube channel. It can be seen here.
The next step is to paint the floating shelf and then fix the track to it.
Whilst playing around with ChatGPT, I decided to ask it to design me a layout to meet this challenge. I phrased my request as follows:
Please design me a model railway end-to-end layout that will fit on a 60cm x 23.5cm board and that uses Kato Unitrack. It should have at least one set of points and at least one siding. Please list the Kato track pieces I will need to make the layout.
In reply, it gave me the following answer:
Here is a compact end-to-end model railway layout designed specifically to fit on a 60cm x 23.5cm board using Kato Unitrack (N Scale). It features a main line, one turnout (point), and a siding, ideal for shunting small trains or scenic dioramas.
Layout Overview
- Type: End-to-end
- Dimensions: 60cm x 23.5cm (board)
- Scale: N Scale
- System: Kato Unitrack
- Features:
- One main line
- One siding (via a turnout)
- Space for scenic elements or small structures
Required Kato Unitrack pieces
- 2 x 20-000 Straight track 248mm
- 1 x 20-020 Straight track 124mm
- 1 x 20-045 Turnout #4 left (or right) ~124 branch
- 1 x 20-040 or 20-041 Curve R481-15° or R282-45° (optional, if curving siding)
- 1 x 20-010 Straight track 62mm (for siding)
- 1 x 24-850 Bumper track (buffer stop)
ChatGPT's suggested track plan.
Total Length Calculation
- Main line:
- 2 x 248mm = 496mm
- 1 x 124mm (with turnout) = 124mm
- Total: ~620mm (layout allows small overhang for board edge buffer)
- Siding:
- Branch from turnout: 62mm (or longer, depending on available space)
- Buffer at end (62mm)
- Curve optional depending on shape/orientation
Notes
- If you want to save a few cm, replace one of the 248mm straights with a 186mm (20-090).
- You can mirror the turnout (use #4 right instead of left) depending on siding direction.
- Power routing can be done via Kato 24-818 feeder track (substitute one straight if needed).
- A small engine and 1-2 wagons would suit this micro-layout.
This is very helpful, and although I will probably use the list of track as a starting point, I expect that my final layout will be somewaht different.
I have been continuing to experiment with ChatGPT's image creation tools. Having used it to create individual images of myself as a Russian World War II general, three different German World War II officers, and as a Viking chieftain, I decided to see if I could use them to create battles scenes based on a photograph of a wargame. I chose a photograph that was taken during a wargame I fought during a session at my local game group.
I asked ChatGPT to turn this photograph into an image of a battle scene, and it created the following image.
I them asked it to remove the helmets from the Russian troops and replace them with fur hats.
I them decided to try something a bit off-the-wall and I asked ChatGPT to produce a version of the image in the style of a Hergé Tintin cartoon ... and the result looked like this:
I rather like the latter and can see myself using this sort of image in appropriate blog posts.
I have been playing around with ChatGPT’s ability to create images based on photographs.
The first was of a Russian World War II general …
… who I have named General Georgi Provgorat. (For some years I was the Masonic Province of Hertfordshire’s Provincial Grand Orator, which was abbreviated to ProvGOrat … which sounded Russian to me!)
The next three were German World War II soldiers. The first is a middle-ranking officer, …
… the second was a general, …
… and the third was an member of the Mountain troops.
Finally, there is a rendition of me as a Viking chieftain.
I was rather pleased with the end results and hope to use this function again soon.
I have been playing around with ChatGPT again, this time to create images of Sultan Chor Darrai and the Byzantine general Demis Roussos. The results look liked this:
The Seljuk general, Sultan Chor Darrai.
I was rather impressed with the end results, and will probably use this facility again to produce more images of myself (and others) in various roles.The Byzantine general, Demis Roussos.
I recently received an email from Ian Dury about his experiments using ChatGPT:
I have started to use ChatGPT to flesh out some of the lower-level details for the campaign and just asked it to produce a scenario for a battle. The results are attached and I thought that they might be of interest. I thought they were pretty good.
This is the instruction I gave to the software:
- Prepare a fictional scenario for a multi-player wargame set in the Kohat district of the North West Frontier of India in 1881.
- At a minimum, the scenario should include:
- Introductory context explaining why the scenario is occurring and the events that triggered it.
- Details of the forces involved (one or more Pathan tribes, British Indian army troops).
- An outline map for the scenario.
- Initial briefings for each player taking command of a force.
- Victory conditions.
- Random events (both positive and negative) that could impact any or all of the forces involved.
- The scenario should be based around an outlying British fort that has been damaged during an earthquake.
He attached a copy of the document that ChatGPT produced … and when I read it, I was impressed. By setting out very specific parameters, Ian ensured that ChatGPT produced results that made a lot of sense and did not include anything that was off piste. The final document is produced below.
Wargame Scenario: The Siege of Fort Lockhart (Kohat, 1881)
Introduction & Context
In the early hours of March 2, 1881, a violent earthquake struck the Kohat region on the North-West Frontier of India. Among the structures affected was Fort Lockhart, a small but vital British outpost guarding the caravan routes and serving as a forward base against hostile tribes. The earthquake severely damaged the outer walls, creating breaches and causing a partial collapse of the south bastion.
Seizing this opportunity, the Orakzai and Afridi tribes have mobilised, gathering warriors for an assault before British reinforcements can arrive. Meanwhile, the British garrison must hold out until relief forces fight their way through the treacherous mountain passes.
The stage is set for a desperate struggle between the determined Pathan tribes and the beleaguered defenders of the British Empire.
Forces Involved
British Garrison at Fort Lockhart
Tribal Forces
British Relief Force (Optional, Arriving Mid-Scenario)
Outline Map & Terrain Details
The game master should provide a basic topographical map featuring:
Initial Briefings
British Garrison (Major Edward Harrington)
Tribal Commanders (Malik Zarin Khan & Gul Rahman)
Random Events Table (Roll 1d6 Each Turn)
As I mentioned in a recent blog post, I have been using ChatGPT to create a set of Portable 3 x 3 Chaco War wargame rules. They needed a minor amount of tweaking to improve the clarity of the rules but what follows is probably 90% ChatGPT and 10% me.
A 1933 map of Paraguay. The disputed area - the Gran Chaco - is shown shaded in pink.
Portable Chaco War Wargame
A fast-play grid-based wargame representing the difficult jungle and desert warfare of the conflict.
1. Game Setup
2. Turn Sequence
3. Units & Special Rules
| Unit | Move (in grid areas) | Weapon Range (in grid areas) | Attack Value | Defence Value | Special Rules |
|---|---|---|---|---|---|
| Bolivian Infantry | 1 | 2 | 2 | 3 | Well equipped, but suffers if moving through jungle. D6 die sore of 3, 4, 5, or 6 needed to move through jungle. |
| Paraguayan Infantry | 1 | 2 | 2 | 3 | Better adapted to the terrain; gains +1 Defence in jungle. |
| Machine Guns | 1 | 3 | 4 | 3 | May not move and fire in the same turn. |
| Mortars | 1 | 3 | 4 | 3 | May not move and fire in the same turn. Can fire over obstacles, but have limited ammunition (D6 die mortar rounds per gun). |
| Cavalry | 2 | 1 | 2 | 2 | Can move after attacking (this represents their ability to hit-and-run), |
| Artillery | 1 | 4 | 3 | 2 | May not move and fire in the same turn. Can fire over obstacles, but have limited ammunition (D6 die artillery rounds per gun). |
| Air Support | - | - | 3 | - | May be used once per battle. May only target non-jungle grid areas. |
| Fortifications | - | - | - | +2 | Provides cover bonus. |
| Commanders | 2 | - | 1 | 1 | Boosts morale by adding +1 to morale of any unit in an adjacent grid area.. |
4. Combat
5. Special Features
6. Campaign & Scenarios
It is interesting to note that ChatGPT has taken into account the terrain found in ther Gran Chaco and the differences between that Bolivian and Paraguayan infantry. The former were mainly drawn from the Bolivian Altiplano (which is a high semi-arid and arid plateau with a cool and humid climate) and the latter were mainly Guarani who were used to living in the semi-arid, broadleaf jungle and scrub-covered lowland terrain found in the Gran Chaco.
I haven’t blogged for the past few days for a couple of reasons.
Firstly, I’ve been suffering from a general lassitude that seems to have crept up on me. I have no idea why, other than it’s the last hurrah of the flu-like bug I caught after Christmas, coupled with rather over indulging myself on my birthday.
Secondly, I’ve been learning how to use two new-to-me apps, ChatGPT and Discord. I’ve already mentioned ChatGPT on my blog, but what I’ve been doing is experimenting with its use as a means to create period specific versions of the PORTABLE WARGAME rules (e.g. the Chaco War), potential scenarios, and related campaign Matrix Games. So far, the results have been variable because I’ve had to learn how much and what type of information I need to input in order for it to generate meaningful results.
Discord is used by the local gaming group I have joined to organise its activities and as a means for players of particular games to communicate with one another. Having begun to explore its potential, I have begun to realise that so far that represents only a fraction of what it can do, and that it can be a viable alternative means of playing wargames online … something that I’ve tended to use Zoom and FaceTime to do in the past.
I may be old and worn out, but I hope that I’m not too old and knackered to learn - and use - new things.
As part of my investigation into the use of ChatGPT in wargaming …
… I asked it to design me a set of PORTABLE WARGAME rules for the period 1880 to 1910, and what follows is the result.
Portable Wargame Rules: Colonial & Early 20th Century Warfare (1880 to 1910)
A simple, fast-playing, and portable wargame system for small-scale battles.
1. GAME OVERVIEW
This set of rules is designed for quick and engaging wargames set between 1880 and 1910, covering conflicts like the Anglo-Zulu War, Mahdist War, Spanish-American War, Russo-Japanese War, and early colonial skirmishes.
The rules prioritize playability while maintaining historical flavour.
Scale & Setup
2. UNIT TYPES & STATS
Each unit has the following attributes:
| Unit Type | SP | Firepower | Range | Movement | Notes |
|---|---|---|---|---|---|
| Regular Infantry | 4 SPs | 1D6 | 2 grid areas | 2 grid areas | Core unit, steady fire. |
| Native Infantry | 3 SPs | 1D6-1 | 1 grid area | 3 grid areas | Fast but less powerful. |
| Cavalry | 3 SPs | 1D6-1 | 1 grid area | 4 grid areas | Charge bonus. |
| Artillery | 3 SPs | 1D6+1 | 4 grid areas | 1 grid area | Indirect fire, powerful at range. |
| Machine Guns | 3 SPs | 2D6 | 3 grid areas | 1 grid area | Devastating firepower but immobile. |
| Commanders | - | - | - | 2 grid areas | Boosts morale & activation. |
3. TURN SEQUENCE
Each turn consists of the following phases:
A. Initiative Phase
B. Activation Phase
C. Movement Rules
4. COMBAT RULES
A. Shooting
B. Melee Combat
C. Morale Checks
5. VICTORY CONDITIONS
6. OPTIONAL RULES
A. Command & Control
B. Special Weapons
7. EXAMPLE SCENARIO
Colonial Defence Situation: A British column is ambushed by a large force of native warriors. The British must hold their ground until reinforcements arrive.
8. FINAL NOTES
I find it interesting that ChatGPT has not just copied the text and mechanisms from the published versions of the PORTABLE WARGAME. For example, it has added Morale rules, and the Melee rules are very different from those I have written. It has also included ideas for Victory Conditions (something that I know some wargamer like) as well as some Optional Rules that have rather piqued my interest.
I was intending to play test Joseph Morschauser's FRONTIER rules in the near future, but I am now giving serious consideration to shelving that idea and play testing these rules instead.
Over the weekend I took part in several VCOW (Virtual Conference of Wargamers) sessions. These included:
Wake Island: A virtual battlefield tour of the Alamo of the Pacific
This presentation made extensive use of Google Earth images as well as contemporary and modern maps to tell the background to the United States occupation and defence of Wake Island. It was a fascinating insight to the little known event surrounding the Japanese attack and capture of the island group in the early days of the Pacific War.
Midway
The presenter of this session was injured in a car accident earlier that day (luckily, not seriously) but another attendee – who had seen the original presentation – stepped in and gave what turned out to be an excellent description of the Battle of Midway. There were lots of extremely interesting contributions in the subsequent discussion from the very knowledgeable attendees.
Matrix Game Simulation
This session looked at the way in which advances in Large Language Models (AI) – and ChatGPT in particular – is having an impact upon Matrix Games. After a short introduction, the presenter took the attendees through a Matrix Game about the growing US interest in acquiring Greenland. The results were very believable and went some way to convincing me that this technology has much to recommend it.
The presenter then used a Matrix Game about the Spanish Civil War (Viva la Muerta!) that I designed many years ago as an example of how quickly the text of a Matrix Game can be adapted for use with ChatGPT. This took a matter of minutes, and the presenter then ran through the first few turns of the game before asking the program to complete it. I was most impressed with the result ... and I became a firm convert to the concept of using such programs to enhance my tabletop wargaming.
As far as I am concerned, this was the standout session of VCOW, and it has convinced me to investigate the wargaming potential of such computer programs.
Article 5
Article 5 of the Treaty that set up NATO states:
The Parties agree that an armed attack against one or more of them in Europe or North America shall be considered an attack against them all and consequently they agree that, if such an armed attack occurs, each of them, in exercise of the right of individual or collective self-defence recognized by Article 51 of the Charter of the United Nations, will assist the Party or Parties so attacked by taking forthwith, individually and in concert with the other Parties, such action as it deems necessary, including the use of armed force, to restore and maintain the security of the North Atlantic area.
Any such armed attack and all measures taken as a result thereof shall immediately be reported to the Security Council. Such measures shall be terminated when the Security Council has taken the measures necessary to restore and maintain international peace and security.
This session saw the participants take on the roles of leading politicians in NATO member countries. They were told that an armed attack had been made on a NATO member on Christmas Eve, and were tasked with identifying what NATO’s response would be.
After a short briefing, the participants were split into four groups, and each group was asked to come up with four or five responses to the attack. These were then summarised by each group’s chair during a plenary session … and it was interesting to see the range of answers the participants came up with.
An Introduction To Cyberwarfare
Much has been written about the threat (or non-threat!) of cyberwarfare, and this lecture looked at several examples of this form of warfare. For example, the fact that the Ukrainians lost nearly 90% of their artillery in the fighting that took place in 2014 because the artillery observers were using a smartphone app to control their guns … and were unaware that the Russians had infected the app with malware that transmitted the GPS location of the observer and their guns to the Russians!
Other examples included the cyber attacks made by China on the Philippines after a confrontation in the South China Sea. Most of these involved DDoS (Distribute Denial-of-Service) attacks on Filipino government networks by groups based in China to which the Filipinos responded with the help of hacker groups and software companies.
Although my participation in VCOW was somewhat limited by it taking place over the same weekend as my birthday, those sessions I did take part in were all excellent, and I’ve come way with lots of ideas that I want to develop.
.JPG)
