Master of Magic - Conquest Victory, Single Opponent, Impossible difficulty.

It's always been a dream of mine to submit something to SDA, and this is the first thing I've ever felt was worthy.

So, a little background:

I've been staring at TravelDemon's Master of Magic run on SDA for years, somewhat frustrated.  It's an insane run, well executed, and Lionheart is a seriously overpowered spell.  The Lizard folk are definitely the right choice in terms of race, definitely the best (and only) call to go all in on eleven spell books and Mana. 

But something was nagging me every time I watched it.  It felt very... safe in terms of choices.  I also thought I remembered beating MoM in just under a minute with all Death spell books back in the late 90s.  I didn't think much of it though, as I knew full well how brutal the RNG was for map generation.  I'd never be up for trying it myself, or so I’d thought…

Six months ago, after 15 years of sitting on the sidelines and merely watching speed runs, I decided to start recording my own runs.  And I knew, eventually, I'd have to try Master of Magic.

About a month ago, I started doing some testing on the basics in earnest: Map generation, CPU behavior, etc.  Took about a week, but I learned that about 1 out of 20 maps had about the same or better opponent starting location as TravelDemon's run, and there was always the chance that the opponent would spawn in Myrror, which would ultimately waste time and end the run.  With two starting units, 8 directions to explore, and the possibility that hostile neutral towns would spawn on your landmass and attack you, ending the run early, it was pretty clear the odds were stacking up against a better run.  At the end of about 300 test runs using the same strategy as TravelDemon, only five of those runs were successful in killing the CPU, and only one was under 5 minutes.

Something had to change and change radically.  Was my memory incorrect?  Was that "run" from so long ago just luck?

The How and Why:

After briefly exploring other options, I settled on Death domain spell books.  This is because the Death domain has two early spells comparable to Lionheart in strength - Wraiths and Shadow Demons.

Wraiths are expensive though, 100 MP more than two Lionhearts and 175 MP more than Shadow Demons.  Both fly. Both do decent damage, but Shadow Demons have a ranged attack, which tends to be quicker to animate.  Most of the Wraiths usefulness comes from creating undead units, but since we want to win in a single battle, and undead spawn after battles, that isn't terribly useful. 

Shadow Demons also have two major properties that make them one of the strongest early summons in the game.  The ability to Plane Shift and Regeneration.  Plane Shifting allows them to travel to Myrror to scout for the opponent’s starting town, which allows for a backup scouting option.  If you have your Lizards scout the same direction as the Shadow Demons on the overworld, you get two chances to find your opponent.  Regeneration on the other hand, is both useful and contains a beneficial glitch in the last 1.31 patch.  When an individual Shadow Demon in a unit of Shadow Demons is killed, it can regenerate on the next combat turn, and if it does, every figure gains another health point that stacks.  So the more individuals that are killed and allowed to regenerate, the harder the Shadow demons are to kill over all.  Combine this with their weapon immunity, and you get a shockingly powerful melee and ranged attacker.

Shadow Demons were the clear choice, but they severely limited my choices in terms of opponent wizards I could successfully beat on Impossible.  Wizards with Life, Chaos and Nature Domains could all be problems, and Sorcery could also be an issue depending on the Wizard's randomly generated properties.  Too many of any particular domain of spell books could mean high level summons early game that could easily kill my Shadow Demons.

To mitigate this, I changed up the settings slightly:

Land size small is the only way to try to prevent the possibility of neutral towns ending my run.  Having this set to larger landmasses would potentially allow for more chances for the opponent to spawn closer, but it also expands my Wizard's vulnerability exponentially.

Magic set to weak to force the opponent Wizard to expand early and leave their starting town less protected.  Certainly, riskier for aggressive Wizards that might send units after my Demons directly and waste precious time, but it makes more scenarios winnable. I'd be resetting on bad RNG anyways.

My choice of Wizard is the very same Wizard TravelDemon defeated in their run, and the name is an inside joke (Marwan Kenzari is a very pretty man). 

A full complement of Death spell books is a must to gain access to Shadow Demons early so I can cast from turn 1, and the cost reductions that comes with doing this.

And of course, the Lizard folk for their water walking and early scouting ability.

Now on to the run:

After a deep breath and a second or two to take in the initial starting map, I clicked away the town name to start the run.

There's not much land around me and I'm surrounded by water on a peninsula, so I assume by the shape of things that there's landmass below me to the left or right.  Left has been good to me in the past, so I try it. I move my starting Lizard folk over a screen down and left.

I start my Shadow Demon cast, and then I go to the Magic Menu. 

I attempt a trick I call the 12-wand here.  Normally you can't get more than 11 Mana Points from changing your priorities in the Magic menu at the beginning of the game, but if you have mana crystals around your starting town and you click very precisely at the top of your Mana gauge, you can set an extra mana point that would not normally be available.  Since I can't see in a 5 square radius around my town to check for mana crystals without wasting time, I risk the extra second or two it takes, as it could potentially save two turns in casting the Shadow Demons. 

It doesn't sound like much, but those two turns could mean the difference between an extra unit or a really nasty summon my opponent could conjure up. I unfortunately don't get the trick, but it only takes about a second, so no major time lost.

Next turn, I choose the closest clickable spell to research once the screen loads, and quickly move to the next few turns.  My lizard folk get lucky and stop a space away from my opponent’s main town!  I get the Wizard introduction screen… it's Kali and the yellow eyes of the gargoyles on either side of the viewing mirror tell me she isn't immediately aggressive towards me.  Good, though I must admit I'm totally floored! 

I know that Kali usually has a 5/5 Sorcery and Death split in her spell books and the Artificers Retort, which means her likelihood to have any hard to deal with summons is fairly low.  She also tends to focus on expansion and magic, also good! She tends to be very aggressive by default, so I beat the average there.  She isn't the best in the game since Sorcery allows her have some annoying in battle spells that waste time, but she's beatable. 

Well, usually beatable I should say.  You see, on impossible, there's always the possibility that your opponents have tossed out the usual spell books in favor of some completely random assortment of awfulness.  I'd lost several promising runs just as good as this to the "Impossible factor" already.

And the rub is, I thought I saw a unit she's not supposed to have for a split second as my Lizards were moving in.  Looking back on the actual video, I don't think there was much evidence of anything, but I was convinced I saw Sprites.  I pause for a few seconds, options racing through my mind.

Do I move my lizards out of the safety of their watery locale and around her town to get a better look and make sure she's expanding?  Do I take a second or three to look at her town?  Do I just patrol and watch?  Any of those options could trigger aggression and get me caught up in a time-wasting battle or two, ending my run.

And even beyond those immediate decisions, should I move in with both the Lizards and the shadow demons for safety, possibly wasting 10 or more seconds in unneeded battle movements?

After a few seconds of deliberations, I decide to take the risky path.  I'll patrol, and I'll send the demons in alone.

I don't see much while waiting for the demons other than a single unit group move INTO her town.  That's not what we want.  I'm nearly regretting my decisions.

The demons are finally summoned and I send them in post haste.  As they get to her landmass, I see units moving to take a neutral town!  I HAVE A CHANCE!

My movements with the Shadow Demons also trigger some angry words of warning from Kali, which wastes a little time, but any other movements to mitigate this would likely have wasted far more, and she's not overly aggressive yet.

As the Shadow Demons move in, sure enough I see something odd right outside of town: Sprites.  A Nature summon with a magical ranged attack?  This could be bad.  If Kali has managed to summon 4 sprites in town, I'm toast.  It's good that the ones I can see are right outside of town, let's take advantage of that while we can.

I move in for the kill, holding my breath.  The battle screen loads... 3 Sprites! Even with the usual nature magic protection spells we're good. This is easily winnable.  My first volley kills off a pack of sprites, then the second and finally with a few more the third, but they do more damage than the average and keep my Shadow Demons at 3 figures.  I have only a few volleys left to kill the melee unit with, and I fail to wipe them out.  The Demons must walk it which wastes about 2 seconds, but at least she's finished.

The end screen confirms she's defeated!  Time at about 50.58, about a 28 second improvement over TravelDemon’s run.

During the course of the run about 10 seconds could have been saved between my own decisions and map independent RNG, and it is possible to get slightly better map RNG than I did in terms of opponent Wizard and starting towns.  I could have used hotkeys in several places as well, which might have saved a second or two, but that can be risky in certain situations, as it tends to create some interface lag that can cause costly miss-clicks. 

It’s highly unlikely that all these moving pieces will align though, as this run alone was the product of about 300 testing runs with TravelDemon’s strategy, around 100 or so testing attempts with other magic domains and at least about 400 attempts with my current Shadow Demon strategy.

Well, anyways, in closing:

I’d like to thank my friend Fogel for introducing me to Master of Magic and the wonderful world of 4x Micropose games back in 1995, Tenka and any others I might have missed popping into my twitch channel attempts to watch what must have been the most boring series of speed run attempts ever streamed, and finally my wife Neebs for putting up with my constant clicking and cursing at digital wizards for the past few weeks.  I owe Neebs a ton of Marvel vs Capcom matches at the very least!

I'd also like to thank TravelDemon for the (probably) hours of enjoyment watching that run, and the inspiration to try it myself.  And finally, a thanks to anyone who's been a part of maintaining SDA over the years.  I've probably watched everything posted here from 2004 to 2012, so thanks for all the years of enjoyment and inspiration.

Krayzar