Meraki - Doing something with soul, creativity, or love. It’s when you put something of yourself into what you’re doing. It forms the core of my game design philosophy.

Wednesday, 25 June 2014

Super Glyph Quest Dev Diary 07

In which productivity takes a hit thanks to Pikmin 3* and the World Cup**.


As well as being on my mind, it's also almost always too tight to mention. For the last few weeks, I've been steadfastly not looking at my bank account. My head was firmly in the sand and I didn't want to see how long we had left. The other day, I finally plucked up the courage to take a peek an it wasn't as bad as I thought it was. Still pretty tight, but we've got a couple of months I reckon.

That is, of course, unless the Asian version goes crazy and makes us a stack of cash, in which case we can take our sweet time over this.

The Asian version ditches our 'shareware' model and goes for something a bit more F2P-ish. Also, in game ads which you can disable by purchasing one of the gold bundles. With that in mind, it's very interesting keeping an eye on Apple's current thoughts about making money on iOS. It feels like they're finally fed up with the swathes of cheap, knock-off, exploitative F2P crap that's out there and beginning a push back towards premium.

Okay, that paints a somewhat saintly slant on the situation. One of the main features that is now coming under scrutiny is the incentivised advert. That's something that the player can elect to watch in exchange for currency. As far as F2P goes, I quite like those things. The fact that they're entirely opt-in rather than rammed down your throat is pretty cool. In fact, I've taken advantage of that feature in games like Pocket Mine on a number of occasions. Apple would much rather that people bought in to the iAd system, as that's the one they make money from.

A lot of this is done under the guise of not confusing or unduly influencing the App Store - buying favourable reviews or otherwise trying to prevent people posting negative ones. An admirable intention, I'm sure you'll agree.

The thing that I like about this whole thing though is that even some of the heavy hitters might well have to re-think their approach and there may well be some space opening up at the top of the charts for something else to break through. Wouldn't that be exciting?

Project Management

There's a space above the fireplace in our living room that would normally be filled with a mirror or a picture or something. Whilst both of those would be nice, we're thinking more along the lines of getting a whiteboard instead. We would use it for task lists and stuff. In the meantime, we've also gone a bit old school.

Post It Notes.

With these Post It notes, we can take over the world!
We've begun laying out the story and game progression in Post It Note form. Each note corresponds to either a location, character, quest or conversation and is linked by a series of arrows that shows us the various dependancies. Normally, I'd use either a spreadsheet (but that's not so good for tech trees) or Visio (but I don't have a PC anymore) but this time it's back to Post Its. I had a brief look in town to see if I could find a nice big piece of foam board or mounting card or something, but to no avail. Then I remembered that I hadn't taken the box that the baby's crib came in to the dump, so I've cut the top off that and it's now our project board.

Now many evenings are spent poring over the board, occasionally re-positioning a Post It because we've had a better idea of where it should appear in the sequence or because it has just fallen on the floor.

Willow seems to like the bright colours too.


The cardboard sucks. The Post Its keep falling off. I think we're going to resort to the wall*** instead.

Quest Arcs

There's a main storyline running through the game, but there are also plenty of little side quests for the player to explore. With all of the various dependancies, it's quite easy for the distinction between Main and Side getting blurred on occasion. The board can occasionally resemble some kind of crazy maze with little arrows going everywhere to indicate things that get unlocked in a particular order. I can envisage a major re-organisation pretty soon.

But those things are fun anyway, so I'm not particularly bothered about that.

I might be getting a little bothered about the amount of story we're going to need to write for the game. Each encounter in the game can have a conversation and quest associated with it. The quests themselves can also have two conversations - one at the start and one at the end. As you'd imagine, this little lot adds up pretty quickly. Of course, it's not the case that every encounter needs all of the things, but we've certainly got a whole heap more text than we had for the first game.

There's a bit of a concern that we're drifting away from the drop-in and play, mechanic-driven gameplay of the first and getting a bit bogged down with content. But the gameplay remains the same as it was - plus tweaks to fix the issues that cropped up in the first one - so it should still be great fun. This time though you might care a bit more about the world in which everything is taking place. Get a bit more invested. That sort of thing.

In Other News

"Which began Glyph Quest - The Wizard"**** gets its Asian release this Thursday. We are all very excited by this.

* Which I have now finished, but subsequently found the leaderboards and score attack modes.
** Which has made me dislike Suarez even more.
*** You know nothing Jon Snow.
**** Google translate is a wonderful thing.

Monday, 9 June 2014

Super Glyph Quest Dev Diary 06

The days are beginning to blur. It's all starting to get a bit crunch-y. But that also means it's all beginning to come together, feature-wise.

Glyph Types

From left to right: Light, Air, Water, Metal, Dark, Earth, Fire, Primal
Leanne did us a lovely suite of the new Glyph Icons. The old elements are still there and the new ones are called Metal and Primal. For Metal, think things like blades and for Primal, think spirit animals. Kinda mini summons really.

Aesthetically, I really like them. They're a bit of a blend between our original set and the ones Chorus did for the Japanese version. We'll probably still tweak the colour balance throughout testing as we do find people trying to blend things like Air and Earth still.

Phat Lewt

Loot glyphs are back in. They work just the same as before - you have a chance of them appearing and, when you collect them, they'll either be a nifty item or cold, hard cash. This time though, we're keeping tabs on them a bit better and capping the maximum number that can appear on the board. This number is something you can increase by Upgrading your Luck stat.


Oh yes, Upgrades are in. Remember that whole levelling up thing I was talking about before? Well, it has finally gone from sitting in it's own little scene in to the main game itself. You can access it from your room in the Inn. So the way character progression now works is this:
  1. Kill monsters.
  2. Get XP.
  3. Level up at prescribed XP thresholds.
  4. Gain upgrade tokens for each level you gain.
  5. Spend upgrade tokens on upgrades.
  6. Kill monsters, but more effectively this time.
The Upgrade Screen itself is functional, but requires some serious art love. The layout could use a tweak or two as well to better make use of the themes. We've broadly split Upgrades in to three different themes - Offence, Defence and Utility*. 


Offence is where you'll find things that help you do more damage - from upping base damage on a per-element basis to increasing the chance of a critical hit or the amount of damage you do from a chain. You can even throw in extra glyphs to add a bit of Aftertouch.


Defence is where you'll find things that help you stay alive for longer - normally reducing damage you take or increasing your base health. It can also help you resist harmful status effects, so things like those Goblin Shaman may start to lose their sting.


This is where all those other neat features that can't fit into straight up attack or defend fit in. These are all to do with things like the aforementioned Luck or gaining XP at a faster rate. There's also one that increases the material drop rate of monsters. "Materials?" I hear you say? Why yes. Materials.


Oh ho! Crafting. Okay, so I haven't written this bit yet, but the idea is that you can harvest all sorts of icky items from dead creatures and turn them in to really useful bits of kit. New robes or weapons and the like.

There's a part of me that's a little concerned by the Crafting bit - if nothing else, it's just more work. It should add a bit of longevity to the game as players might want to grind to gather material for cool pieces of kit. That doesn't mean to say that we need Crafting per-see - ultimately, it's topologically similar to just awarding kit for killing enough creatures, so that's our fall-back position if it ends up too much work.

It's always a good idea to have the odd contingency plan or two. In fact, we'll often spec out two or three ways of implementing something and list them in order of difficulty. If we can't do it the first way for whatever reason**, we'll look at doing it the next way and so on.


Young Parker dropped by last week on a bike ride. He started talking about servers and accounts and all sorts. I nodded and smiled. This is still Plan A - piggybacking Game Centre and it's user IDs and friend lists and stuff. Need to look up the Google equivalent. We're probably going to stop short of cross-platform stuff - that is to say that iOS people will only be able to summon other iOS people and likewise for Android - otherwise we'll have to set up our own user accounts system and that's just a complete pain in the backside waiting to happen.

As a precursor to this (slightly terrifying) prospect of taking your friends' characters on quests with you, I had to hook up henchmen to the customisation stuff. It's still not finished but now you can swap yourself out for a randomly generated test mage. Next, I need to hook up the gameplay stuff so that these henchmen can have stats based on their Gear and Upgrades.

Plan B probably means we don't do the online stuff. Instead, we'll have a local repository of generated characters for you to pick from. In fact, we'll need Plan B anyway for such times as you can't connect to the ol' interwebs.

Either way, I still need to write some kind of 'recruitment' screen, where you select which character you want to take on a quest with you.

Narrative and References

In keeping with the higher narrative content of this game, we've got some really interesting things in the pipeline. Collaborations. In jokes. References. That sort of thing. I don't want to ruin any surprises (or, indeed, jump the gun) so I'll probably stop talking about that stuff now.


Actually, that's something to mention. Nothing dates a work quicker than throwing in a pop-culture reference or two. Pop-culture moves very fast and memes get left behind even faster***. It's generally a good idea to avoid references like that. But we don't mind. This isn't games-as-a-service. We don't expect you to be playing Super Glyph Quest in a year's time. What we'd like you to do is play it now and have fun doing so. We don't have the resources for Big or Polished so we have to go for Funny. The easiest Funny to pull off appears to be pop-culture references and in jokes. So we're loading up on them.

Along those lines, Leanne found a weird thing on Reddit**** the other day. There was a thread somewhere that accused Glyph Quest of being a rip off of Dark Souls because of the Summoning Stone item. I'm not sure the guys got far enough to summon the Sun Knight*****. Thank you internet. There are many, many references to many, many things in the first one and it's only going to get better in the next.

Well, better for us. Worse for Shin who's going to have a hell of a job translating all of those references to ensure they're still funny on the other side of the world. At least this time the text is all in a single file and not embedded in the deepest, darkest recesses of the code.

Little Things

I noticed an interesting button in Unity the other day. It lets me toggle between different aspect ratios for iDevices. Man, that would have been a good thing to know about when we were struggling with image sizing for the original game.

It's much nicer writing things the second time around. You have a much better chance of getting rid of the annoying bits. Now scrolling through things like the shop screen won't select items whenever you let go.

Burn, familiar goblin chap
Macs are pretty cool for things like screen grabs or video capture. I wanted to send a short video to Shin, showing him the Sunstrike effect and it was a work of seconds to capture a video of the relevant part of the screen. In fact, how about a screenshot of the game as it currently stands?

Willow now fits in her Creeper baby grow. It's still a bit big, but hey. Next up: the "Little Short For A Stormtrooper" one.

We have a lot of cake in the house. This is probably a Bad Thing.

* LoL and WoW have a lot to answer for...
** Normally a lack of resources.
*** Except for things involving cats. It would appear the internet has a long way to go before we get fed up with them.
**** Hardly news, I know.
***** \o/ Praise the Sun! \o/

Tuesday, 27 May 2014

Super Glyph Quest Dev Diary 05

After an interesting trip up to Birmingham to introduce Willow to her great-grandparents and her great-great(!)-grandmother, it's back to sunny Worthing and work.

Character Customisation

This style requires a pillow
This style requires a lot of effort
After going back and forth on this for a while, I think I've decided to get over myself and just implement the damn thing. You buy clothes. You wear them. They can affect your stats. Simples. There might be clothes that you like the look of but are not very efficient against the types of monsters you're going to meet on your current quest, but hey.

I've written a compound sprite system to allow us to layer in the various clothing options. Not just clothing either - it will also be used for things like hairstyles and weapons. It sounds like a simple thing but, as is so often the way, starts getting a bit more complex when you think about it a little more. The main issue is making sure everything looks right and things overlap other things in the correct way. That means that robes should be drawn in front of the character's body...

... apart from the robes that you might want flowing out behind or if you needed to do the back end of a cowl or big collar. So robes are split in to two bits - front and back. The player will only ever choose a single item and everything else happens under the hood. The same is true for hair - there are times when we want it drawn in front of the robes and bits of it that might need to be drawn behind.

Japanese Feedback

Whilst the boys at Chorus Worldwide were doing the conversion, they tweaked a few things and added the odd feature or two. Some of these bits and bobs are going to find their way into Super Glyph Quest. Their version of the characters feature a couple of frames in addition to the standard pose - one for when the player attacks and another for when they take damage. It all adds to the feedback presented to the player.

It does somewhat increase the art demands on Leanne as we now have to have multiple versions of each item of Gear to cope with each pose, but it's totally worth it. Actually, we're skimping a bit by some judicious code that tilts the idle sprite when it gets hit, meaning that all we'll need to do is adjust the underlying body sprite and make them blink or wince or something.

Hang on! Where's the baby? NOBODY MOVE!
There's also an attack frame and the character leaps forward a bit. It's a lot more dynamic and even though it's a pretty simple thing, really adds to the feel of the game. Again, a simple task if one that has extra layers of complexity the further you dig in to it. Different sprite for robes? Check. Wait - two different sprites for robes given that robes themselves consist of a front and back sprite? Check. Different sprite for weapon to make it look like the character is actually pointing it? Check. Wait - different position for the weapon sprite because the character's hand has moved as part of the robe 'animation'? Okay, now male and female versions for each...

Sticking with the art, the new character designs are also taking a leaf out of the Japanese version's book. That is to say they're a weeny bit more brightly coloured than they were. In all of this, it was very exciting seeing Leanne back in front of her computer.

Work Hours

There are two* things guaranteed to mess up working from home. Daytime TV and a baby. We have both. Cake Boss is a problem. It makes us want cake**. All day. I've also got a soft spot for Fast and Loud. Pointless isn't a problem though - we have that on series link.

The rest of the time is spent looking after*** Willow.

Whilst I can normally get on with some stuff during the day, it's only after she's gone to bed and we've had dinner that we can really get down to business. We tend to get quite productive but unfortunately, it's usually around 2am.

I can see why new parents - dads in particular - might be eager to return to the office, but I can't stress enough how privileged I feel to have been here these past 4 months. Willow has changed so much in that time and, if I were back in the peace and quiet of an office, I'd have missed it all. Sure, it's not easy - we're getting it done though. We're shooting for a submission in two months as that's when the money runs out. Well, assuming, of course, that the Japanese version doesn't go all Angry Birds on us in the meantime.

Items & Shopping

The Items are back in****. You can buy them from the shop and use them in combat. They use the ever-so versatile Attack system, although for the more specialised items there's going to have to be a layer of special case stuff added. I'll probably end up adding it to the Status Effect stuff and just giving it a single turn's duration.

Remember what I was saying last time about having possibly unsuitable-for-mobile game systems in there? Well, we're thinking about that again with the shops. It would be quite simple for us to hook up multiple shops that sold different items. It makes narrative sense - here's the Hat of Cold Resistance, weaved in the frozen north and only available from the Viking Village shop, for example. Or the exotic Robe of the Sands that can only be bought from a faraway land to the south.

In other games, this would be fine. It would also encourage exploration and travel (such as the game allows) to find these places. But in a mobile game for this increasingly lazy bunch of modern gamers, I reckon most developers would tell you it was madness to fragment your storefront in such a way. Of course, that might only apply to those selling you items for actual money, but I'd hope it was okay for in-game items.


Given that we've got two extra elements this time around, the board can get a bit crowded. This means it can be trickier to make longer Chains although that definitely becomes easier when you unlock Combo spells and utilise the features of the new spell parser.

Still, it got me thinking that maybe we'd got a bit carried away with the extra elements and they'd be more trouble than they were worth, so after I'd hooked up the Attunement***** bit, I wrote an elemental bias thing. It works alongside the Attunement to weight the randomness of the glyphs getting swapped in whenever a spell is cast. Elements used in a previous spell are given higher priorities and so are more likely to appear, therefore giving the player a better chance of maintaining a Chain. It still retains the 'at least one swapped glyph will match a neighbour' feature of the original, but the others will be decided according to this weighting.

The risk here is that the weighting is so high it's possible for the player to fill the entire board with a single element or that it's so low that it doesn't make a blind bit of difference. Thankfully, this is just a couple of numbers in the balancing file but we're still going to have to keep a close eye on it.

Leanne remains sceptical. Willow doesn't have an opinion on it one way or the other.

* Three, if you're single or the missus works elsewhere.
** A problem I have solved thanks to a judicious trip to the local cake shop.
*** Read: staring at until she goes to bed. Actually, after she's done that there's still plenty of good staring time to be had. So cute!
**** Well, the system is in place. Now it sits there, screaming for content, like so much of the rest of the project.
***** A player can adjust his Attunement to a particular element, making that element more likely to appear as well as being more effective (damaging) when it's used. This can happen either though levelling up or wearing the right Gear.

Friday, 9 May 2014

Super Glyph Quest Dev Diary 04

No luck at the poker so it looks like I'd better make a go of this game development lark after all. At least I'm getting much better at eating solid food, even if my jaw looks like someone has set about it with a sledgehammer. Also, we had a very pleasant weekend subjecting a couple of friends to more board games than they would otherwise have been comfortable with.

Narrative & Characters

It is our intention to give the game a much stronger narrative than it had before. Given that it had almost no narrative, that shouldn't be too hard. We've got an idea for an overarching plot that loosely ties everything to the first game but it can also be constructed in quite a modular way. To that end, everything will be a series of side-quests and little adventures, all building up to an epic climax. Leastways, that's the aim.

One of the first steps was to start creating a bunch of characters. Our process for this has been pretty simple - find one of our friends that we want to include in the game, assign them some kind of relevant character trait and come up with a reason for them to be there. Whilst it won't mean much to the user-base at large, it gives us a great starting point and hopefully shows some measure of thanks to our friends for their support. I guess it's the sort of thing you might expect to find in a Kickstarter reward tier*.

Anyway, we now have a pad with a whole bunch of characters and some key story points written down. I've managed to write a rudimentary conversation system which will be used to communicate with these characters as well as give out quests and the like.

As for the player characters, we're currently still discussing the level of customisation we're going to allow the player to have. At the very least we want different robes, hats and staves.

Aesthetics & Gameplay

This customisation stuff raises an interesting point - the link between aesthetics and gameplay. It's something that often bakes my noodle. It makes sense that the reason the player would pick a particular item of clothing is based on one of two criteria. Either they like the way it looks or they want the gameplay bonuses conferred by it.

The problem arises when they actively want one aspect without the other. That is to say, here's a mighty Hat Of Badassery that has awesome stats but looks like someone curled one out on your head. Actually, it doesn't have to be all that bad but because aesthetics are so subjective, it can get quite emotive. You may end up with your avatar in the game being dressed in stuff that you just don't like the look of but you need to wear them because of the perceived gameplay bonuses you're getting.

We are by no means the first developers to run in to this. This scenario can also be used as a great illustration for the issue of seemingly easy solutions causing even more problems or contrivances later on down the line. The simplest solution is to just not care - make things look interesting and give them stats accordingly. But that's when you can end up with the above scenario and have players uncomfortable with their own avatar.

Or we go the other way - clothes are purely aesthetic and there's some other aspect of non-visible equipment (like Rings for example) that apply gameplay bonuses. Many MMOs these days utilise some kind of display system where you actually get to equip your character twice - once for a hidden layer that confers the stat bonuses and a second, visible layer that displays what you actually want your character to look like.

Personally, I think we'll probably just go for the aesthetic clothes + gameplay rings approach.

How Much Game?

Another of the issues we're having with SGQ is the fact that we're both gamers. This may seem a little odd, but hear me out.

The problem is that we're falling back on established tropes and doing things the way we expect them to be done in other games. It's habit really. For example, the way quests are handled. This is an RPG. Therefore you go somewhere** and chat to someone*** and they give you a quest****. Then you navigate to said quest, kick the crap out of a bunch of monsters, harvest whatever phat lewt is required and return to the quest giver for your payment. This is simply The Way It Has Always Been Done and an obvious starting point.

The problem arises when you begin to realise that we're making this game on mobile and people who play mobile games aren't really "gamers" as we would understand them. They have the attention span of a fruit fly and unless they can simply pound a single green button to move everything along, they find things too complicated and lose interest. At least that's what any modern mobile developer's analytics will tell them.

A typical quest log
Our initial design revolved around Inns, Quest Logs and Maps. As with the original, these would all be separate scenes in Unity. To be fair, the Old Way could use a little streamlining, so we eventually came up with something a little more fluid. The eureka moment was when we realised we could use something akin to the tech tree stuff to populate the world with quests and therefore author the entire game. That has now become the plan.

The player now has a global Map to scroll around. It is filled with interesting Locations - like the Town or the Goblin Hills or whatever - and each of these Locations has a number of Encounters in them. These Encounters can be anything from a Quest to a Character to talk to. Each Location is unlocked according to a set of criteria dictated by what the player has already done. Normally this just means the player will have to have had a particular conversation (say from the quest giver), already completed a particular quest or is in possession of a particular Item.

All of this happens in a single scene but we're going to load up others for the Shoppe, Spellbook and Bestiary. Either way, the Map will form the hub of the game and the act of scrolling around and exploring these Locations is pretty tactile and very suited to touchscreen devices.

We're also big fans of the way Dark Souls does it's NPC stuff. You encounter NPCs out in the wild in various locations depending on your progress through the game and where you last met them. Sometimes they move on to another location deeper in to the game and other times they return to the hub for you to interact with whenever you return home from your adventures.

Henchmen & Summons

Another cool new feature for SGQ is the Henchmen system. This allows for the player to bring other characters in to combat instead of them. I guess it's a bit like Pokemon where you swap out to another member of your party. In reality, we'll probably just restrict it to a single Henchman that you bring on any given Quest with you. The system is also going to allow us to re-work the Summons. That is to say that you'll actually be able to summon these creatures to fight for you rather than just have them appear for a single attack.

But the really cool thing will be the Player Henchmen. Your player data would be stored on a server somewhere and your friends would be able to access it and take you on Quests as their Henchman. This gives the game a more connected and social aspect that, certainly in mobile gaming, can be very beneficial indeed. 

As an idea, it's very exciting. There are some practicalities to consider though, not least of which is not knowing how to construct and then maintain a server infrastructure as well as accounts and friends lists. I'm sure there's a load of Game Center stuff that could handle it, but it's a big ask. It also rather relies on some serious levels of player customisation which, as we've already explained, is a lot of work - certainly on the art side. The more customisation we have though, the more players can express themselves, but all of that is seriously undervalued if there isn't a public forum for them to show their creations off to others.

* Hmm. Not a bad idea now that I think about it. Do you want to be in the game? Well, we're open to offers...
** Almost always an Inn.
*** Mysterious stranger.
**** Bring me n Withered Ogre Genitalia. And don't ask questions.

Tuesday, 29 April 2014

Making a God Game

I read Eurogamer's article on the problems that Godus is having recently and it got me thinking. Firstly, how accurate is the Eurogamer verdict? In the interests of full disclosure, I have not played Godus at all - no early access or sneak peaks or anything. Everything I'm about to write is based purely on speculation and a small insight in to the way that Peter used to work. Take that as you wish - many might recommend a pinch of salt - but the point remains: If the items the Eurogamer article highlight are true, then it seems like the game is in trouble. Why?

Populous Reboot

Very much WIP
That's how Godus was sold to me and I'm very happy about it. Glenn and I used to talk about remaking Populous many times but we just never got around to it. As one of the original creators, Peter has every right to give it a go and even from a very early stage, the stuff I was hearing about it was very encouraging. To me, it sounded like he was going about it the right way.

To establish what I mean by that, we need to delve back in history. I was at school when Populous was released. I only had a lowly Spectrum, so the whole 16 bit thing was very new to me. My friend Scott had an Atari ST though and pretty soon we were bunking off school* and sneaking off to his to play Populous. I  loved it. It was unlike any game I'd played before.

Pretty soon, I was at Bullfrog and working on a sequel. For Populous 2 we didn't mess with the formula too much. It was still very much a Populous game. In fact, looking at what became of Populous: The Beginning, I'd say it was the last of them. It wasn't perfect by any means - too many effects could become a little overwhelming and a nightmare to balance. We innovated a little in Pop 2 but we didn't mess with the core principles. For me these are Landscape Modification and Indirect Control. P:TB had gone too far down the RTS route and recruiting wild men just wasn't as compelling as carving out the landscape.

Landscape Modification

Populous had it easy. It was made in a time before everything had to be realistic. Everything was an abstract representation. A single person sprite could be anything from a single dude to a whole army. Sweeping hills and valleys were constructed from low resolution building blocks. Godus certainly ticks the abstract box. Personally, I love the style, but it's missing one thing from the original: It was all digital - a house took up a square, a slope was always a particular angle**, a click was a click.

That last point I feel is something quite relevant. Each click would raise or lower the land by a prescribed amount. There was no ambiguity about it. It was a very simple system that could then be exploited. The most efficient method of rapidly raising a settle-able swathe of land from the sea was to employ the two-up/one-down approach. Each sequence like this would result in a much larger square of land than had you raised each point individually. Knowing where to perform these little combos was all part of the skill of the game (so you could get better at it) and was a very compelling thing to do (so you wanted to do it). As you can imagine, this is quite a potent mix in terms of gameplay.

I'm not sure smearing the land flat has quite the same appeal - certainly not on anything other than a touchscreen device - but this does bring us to a very valid point in the Eurogamer piece. Your entire aim is to flatten the world. Turn it all in to one sprawling metropolis. Whilst the same is true in the original, it occurs to me that there are two key differences, one of which was the raise / lower land interface. Then again, because of the compelling nature of said interface, we used to see an 'issue' where people couldn't walk past a screen showing hilly Populous landscape without stopping to flatten it all out. It was like digital bubble wrap.

None of this is to say that you can't do analogue landscape sculpting well. In fact, why not take a look at something like Glenn's Topia for a decent example*** or Eric Chahi's From Dust.  But if the aim remains the same (to flatten the land) and the only real tool at your disposal is a smearing, flattening one (as opposed to a combination of raise and lower), then you might end up with something quite dull at the end.

The second difference may seem a little odd: You only saw a bit of the land at any one time. It sat in the middle of a book, surrounded by icons and other interesting (though largely pointless) features. Of course, this was mainly down to the technical restrictions - there's only so many things a poor ST or Amiga can draw before it gives up the ghost - but it did mean that at any given time, there was still something interesting to look at even if the whole thing was flat. This was illustrated further when the PC version of Pop 2 developed a full screen mode. Sure, you could see much further and you actually gained a gameplay advantage but it did result in this issue of turning the entire world into a giant car park.

Indirect Control

The second core pillar in a God Game's design is that of Indirect Control.

I feel this is crucial. A God Game is quite a voyeuristic experience. The instant you start ordering individual units around, you're a general in an RTS. Populous was about coaxing the population. Guiding it. Steering it. Never telling it outright. This created a level of indirection, certainly, but one that could be exploited by a skilful player.

If you're going to give orders, give them to the whole population, not individuals. A god doesn't concern himself with the life of an individual.

The closest the game came to giving the player direct control over individuals was the Papal Magnet - a device that could be used to attract every (keyword - "every") walker on your side to a single point on the map. This could be used to micro manage and get people to migrate to new areas of the map or invade the enemy's territory, but there was a skill to it and it was a very broad brush stroke. Also, time spent doing this meant that your population wasn't growing as fast as it could be as they would be ignoring settle-able land in favour of marching towards their fate.

From what I've heard about the "leashing" thing, I'm a bit concerned that this is a step too far towards RTS.

Living With Bumps

Why does everything need to be flat? Isn't that just a legacy hangover from the original, born out of the block-based technicalities and a decent abstraction for what constitutes 'buildable' land?

I would suggest that the landscape actually offers plenty of opportunity. Instead of 'flat' being the only resource, there are plenty of other things to be found in the landscape. Actually, use of the word "resource" might hang a few people up. By that I'm not just talking about veins of Tiberium or Vespene gas pockets - places where specific buildings can be built - although that would certainly open up some interesting strategy, especially if the person responsible for placing those resources was you with a godly power.

Instead, I'm talking about the landscape topography itself. I think the 22 Cans guys are already thinking about things like altitude affecting the buildings, but that just means that you might have plateaus instead. Still a "step up" from completely flat, but there's more that can be done.

What about delta altitude? That is to say certain buildings that require the landscape around them to have a difference between the highest point and the lowest. Say, for example, mines or quarries that can only be built into the side of hills or cliffs. If there were more in-game resources (okay, I'm back to the ones you expect now) that were produced by different buildings instead of just mana, a player would have to have areas devoted to each. Kinda Sim City-esque although instead of discrete zoning you'd employ landscape sculpting. Big flat bits would provide food and allow for population growth. Hilly bits would provide resources and allow for population strength or even technology. This sort of thing could be used to gate other advances or resources or even the rate at which the player acquires powers.

The nice thing about this is that it plays in to the stepped nature of the landscape. The delta value can be expressed in terms of digital levels rather than something more fuzzy and analogue.

Then, even simpler than that, you've got things like trees. Throwing down forests will brighten up the landscape - even the flat bits. These could be used for all sorts - slowing movement, defence, resources, kindling for fire-based effects...

And now I'm getting carried away.


When I look at something like Godus, I see it as a toolbox. A designer's plaything. There's a landscape and an AI system for the population. With these basics, you could make anything from a simple RTS (Total War) to an arcade shooter (Cannon Fodder) to, well, a god game.

A Knight. Quite a late addition to the game.
I hear people complaining that the game isn't fun and to them I say "so what?". It's not ready. I guess this is a problem when you open yourself up to things like Early Access. The people you're opening it up to (the general public) are not game developers. They're consumers. They expect a finished product regardless of whatever disclaimers you make them sign ahead of time.

Every game I ever worked with Peter on wasn't fun. At least, it wasn't fun until quite late in development. Then there'd be a two week period of insanity, during which the game would suddenly leap forward in playability. It happened every time and I see no reason to expect anything else this time out.

Joe consumer wants a shelving unit. Only he's gone to Ikea and come home with some boxes of raw materials.

And he can't find that allen key in the box.

* Don't do it kids - it's not cool. Well, okay, it's cool but not particularly helpful in the long term.
** Born out of a rendering restriction.
*** In fact, why not take a look at just about anything Glenn has ever written. Chances are it's got a sculpted heightfield in it.

Sunday, 27 April 2014

Super Glyph Quest Dev Diary 03

Must get Leanne to do a new logo...

Busy week next week. Not to do with SGQ - the GUKPT rolls in to Brighton and I might be thinking of a cheeky tournament or two. But then I've got that long-awaited hospital appointment on Wednesday to sort out my wisdom tooth at last. That writes off Wednesday and, most likely, Thursday as well. On the one hand, I'm really looking forward to a pain-free existence and the ability to fit food in my mouth again. On the other, I'm not entirely sure I can afford to lose that much wisdom... or money, for that matter.

With all of that in mind, it's been quite a productive couple of days.

Main Loop

Remember all that stuff about everything Just Working? Well, after a light bit of fettling, it does! The main game loop spawns the board then throws waves of monsters at you based on the current quest data. The multiple monsters thing works fine and everything gets stepped through in a neat fashion. It's easy, this coding lark.

The idea was simple: the main game loop steps through a series of phases - everything from spawning the player or monsters to processing their input and effects. There are a series of delays built in, otherwise the whole thing just flies by. Then there are other bits that have to wait for other things to finish, like Effects. I had it all working with some hard coded delays and no player or monster input. It was only when I tried to implement those latter features that it all started going a bit pear shaped.

It was mainly to do with damage resolution. Whenever a Dude (either a player or a monster) takes damaged, they go in to another state - where you'll see them wobble a bit - for a while until finally applying the damage. To prevent the main game loop moving on when it shouldn't, I made it so that all currently active Dudes have to be in an idle state. It keeps everything nice and neat.

Attacks and Effects

I pulled across the Effect System from GQ, then proceeded to gut the thing. The basics are fine - if it ain't broke and all that - but I've pulled out the bits that did damage and status effects and written a new Attack class to handle all of that. Effects are now just that - effects - and the thing that does the damage is the Attack. Much more elegant.

Integrating all of this wasn't a seamless process and there was a spot of negotiation with the main game loop to keep everything swimming along. I also implemented an early-out type affair so that it doesn't always have to wait for the effect to finish completely. This means that stuff like smoke can fade out in its own good time without hampering the gameplay.

Attacks can apply damage to a target as well as confer health to the attacker. In this way they can be used for everything from simple damage to healing to lifesteal type affairs. The original had a system whereby an Effect could do damage several times, for things like rapid fire. For the time being I've abandoned that and just got the Attack applying the damage in one lump at the end. It makes life easier and means that the bug with the flying numbers* shouldn't rear it's ugly head again.

Currently, I've got placeholder Attacks and Effects in for all of the spells and I'll hook the actual content up as we go.

Numbers Game

Actually, along those lines, it's probably a good idea to talk about how we come up with the gameplay content for things like this. The answer is simple.


Spreadsheets are your friend. We've got a doc that houses tons of information. Everything from the various spells we want and their combinations to a neat little table that is filled with formulae for health and damage at various levels.

Firstly, I should state that I'm totally against a completely mathematical approach to game design. To me, it's still an artform. A craft. The idea of game design as science is abhorrent to me. That said, getting some solid formulae down makes a great starting point. Then you play the game and see what it actually feels like. From that point on, you tweak based on your gut and playtesting feedback.

This table enables me to set up things like player health vs the amount of damage a monster of a particular level should do and how much DPS the player will need to pump out to be able to reach the end of the quest intact. If the player ends up with negative health, they will fail the level... or be forced to use healing. What I'm after is some kind of curve that starts off with a very high value (plenty of remaining health) in the early levels but actually goes negative later on in the game, requiring some decent strategy from the player to progress. That works just fine for the first few levels but it gets really complicated when you start introducing Status Effects.

Status Effects

Status Effects refer to things like Poison or Petrification - things that affect a Dude over a number of turns. They are applied as part of an Attack and they work similar to the way they did in the first game only, as you'd expect, in a slightly more elegant fashion. Status Effects can stack and there's a dedicated phase in the main game loop for processing them. Because any Dude can have multiple Status Effects active on them at any one time, this phase needs to loop through each of them. Again, looking at the old code, this got out of hand really quick and was very messy. The new one? Not so much.

Some Status Effects react differently depending or not on whether they've been applied to the player or a monster. For example, Blindness gives a monster a percentage chance of missing the target** whilst it flips the player's glyph board in to question marks. In real terms, this reduces the player's effectiveness in combat. Because it doesn't easily change into a DPS value, assigning a particular value to that in the spreadsheet becomes a bit trickier. Twelvety three. Plus or minus the square root of elephant.

Stat Multipliers

As a precursor to the player being able to equip Gear that will affect his stats, I've put a basic system in place to allow for stats to be altered on the fly. This has also allowed me to have monster variations. You know how Puzzle & Dragons occasionally features "Strong" variants of a particular monster? Well, something along those lines. Except it's not just "Strong" - we've come up with a number of different adjectives. "Tough" monsters, for example, have extra health. "Vicious" ones do more damage and so on. We might even implement "Formidable" ones that have extra health and do more damage. I know, right?

We could go the other way as well - have "Weak" or "Puny" versions. The sky's the limit here. I suppose we're not just limited to physical attributes either. We could have a "Cunning" variant that's more likely to perform attacks with those damned Status Effects in it. Or a "Cautious" one that's more likely to try to heal itself - I know how people love that***. Imagine a Cunning Harpy or a Cautious Troll, for example...


One of the least glamorous jobs is that of text and localisation. As befits the original's thown-together approach, the text was all over the shop. Sometimes it would use the name of the prefab. Other times there would be a string as part of the class. Occasionally it would be in an array somewhere in the code. This meant that when we handed the project over to Chorus to do our Japanese version, they had a hell of a job on their hands.

To make it easier the next time, all of SGQ text is pulled from a single file. It means a bit more work at the beginning but should really pay dividends in the long run and I know a couple of Japanese people who will be very happy with that****. Of course, we knew this is something that totally should have been done anyway with the first one, but we really weren't thinking that far ahead and I was happy to sweep it under the rug.

* Ooh, must remember to implement the flying numbers...
** A though just occurs to me. Now that I've split Attacks up, I can have the monster still Launch their attack but just not play the Hit if they've been Blinded rather than not play anything at all like the first game. Cool. I'll have to write that down too...
*** Actually, less "love" and more "hate".
**** One of their other bugbears was the lack of a centred board when you don't have enough glyphs to make a perfect hexagon shape - something I've also fixed for the new game. It's the little things.

Wednesday, 23 April 2014

Super Glyph Quest Dev Diary 02

Main Game Loop

An important area that is getting entirely re-written from scratch is the game's core loop. At the moment, I'm going through and trying to get the sequence right. This will move the main game through the various stages it needs to go through. When this works correctly, I can fill each bit out with the relevant module safe in the knowledge that it should all Just Work*.

This means that everything should be a little cleaner than it was in the first one and therefore easier to tweak and fix when it inevitably goes horribly wrong.

Multiple Enemies

One big change is that the system now supports multiple enemies at the same time. This opens up all sorts of neat stuff like being able to face a pack** of rats or wolves at once. It also means we can do stuff like having certain monsters summon others in to fight with them. Of course, just getting multiple monsters appearing seems simple enough but then you start going through the ramifications.

Suddenly, for example, you're going to need a targeting system to allow the player to choose which monster he wants to concentrate on. At the moment, it's just random. Also, you need to decide on things like how to process their turns. Do you do one monster per turn or give each monster a go? We've gone for the latter - it gives the hordes*** more weight and means that the player's ability is consistent rather than more effective as the number of enemies gets decreased.

Also, each monster needs a different place to spawn. In and of itself, this is a very simple problem to solve. Trivial in fact. But then we come to another issue - monsters won't be standing in the same place. This causes a problem for the effect system.

Knock On Effects

The Effects system was one of the bits that I was most pleased with from the original. It allowed me to make composite effects out of particle systems and scaling or rotating sprites. It was so versatile that it ended up getting used for just about every visual effect in the game and not just the ones during combat.

The problem was that each combat effect was a single entity that kinda relied on the two participants being in the right place. In the new game, this is going to have to change.

But that's not such a bad thing anyway. One of the criticisms of the original from some people was that the spells took too long. In the spirit of killing two birds with one stone, my plan is to split each effect in two. One part would be the 'casting' or 'launching' effect. This would be centered on the attacker. Then you'd have the 'impact' effect which would appear over the target. Nestled somewhere in the options screen would be a combat speed option that, on its fast setting, would simply skip the 'casting' part and show only the 'impact' effects. Should make things skip along a bit.

Of course, this alone doesn't fix things like projectile effects so I might need to add another effect component although I can still think of a work around which would just involve fading projectile elements out during the launch phase and having a new one fade in during impact. Your mind will fill in the blanks and there's no reason for you to see the entire projectile path.

The Design Process

Ever wondered what it must be like having two designers engaging in a "Wouldn't it be cool if..." discussion? Well, here's a paraphrased example for you:

AT - "What if we made it flip the entire board if you cast a null spell?"
LB - "I don't like it. What about the Resetta Stone?"
AT - "We'd get rid of the Resetta Stone."
LB - "It's too powerful then."
AT - "What if it reset your Chain?"
LB - "Then we could keep the Resetta Stone which would keep your Chain. What if it was something you could spec in to?"
AT - "I like that. I'd like to see several different schools of magic, each with their own play style."

There are a couple of important things to takeaway here.

Firstly, the discussion actively promoted a Devil's Advocate position. Far better for you to think of reasons why it's a bad idea now than to have a paying customer point it out later. Ask questions of your design. Try to make it fail. Then see if it can't be fixed or refined. Think of it a bit like science - come up with your theory then try to prove it wrong. If you can't prove it wrong, it must be right. An important thing here is that it's not personal. The person playing Devil's Advocate is trying to find something wrong with the idea, sure, but they're not doing it to be spiteful. It's due diligence.

Secondly, it went back and forth with all parties asking questions and suggesting possible solutions. This keeps everything nice and inclusive and makes the team (ie- both of us) feel involved and invested. Ultimately, someone has to make a call though and you should never design by committee. That's where the skill of the designer comes in - to sort the wheat from the chaff but balanced in such a way that you're not just throwing out ideas because they weren't yours but you're also not just trying to shoehorn every suggestion in there just to keep people happy. If the idea is bad, the idea is bad, but you have to be able to explain why it's bad. Just thinking it's bad for certain reasons is pretty rubbish too. Being able to prove that it is bad because of an example you can cite experience of - that's where you need to be.

Thirdly, jargon. We've got our own. It's organic and it comes from all over the place. Some of it is from established gaming tropes whilst some is from things like the original Glyph Quest. It's fine, so long as everyone who needs to be involved in the discussion understands exactly what it means. If you don't understand, you'd better ask someone to explain it. If someone asks you to explain it, you probably shouldn't take the piss out of them for not knowing****.

* A piece of coding terminology that is almost never true. Quite the opposite in fact. It's like when Murray Walker used to say "nothing can stop him now" during an F1 race or they say "well at least it's not raining" in a movie.
** Of three at the most. Not so much a pack, more a 'packette'.
*** Okay, three's hardly a horde either. Unless 'horde' is simply used as the collective noun for zombies...
**** L2P noob. Hey, I said "probably"...