Game Design, Programming and running a one-man games business…

Medieval internet infrastructure woes

Posted on April 15, 2017

I’ve been stuck with the same web host for a billion years, and a home internet connection thats probably older than my parents (that last bit is probably true). Its time to upgrade the shit out of this.

Back in the dark days, I registered my awful company name (positech.co.uk) and then went on a silly spree where I would register domain names for my games, like kudos, democracy, gratuitous blah blah… Those days are gone. I didn’t bother getting a new domain name for Production Line. Nobody cares these days, and why should they. Plus when you move server hosts, domain names are a PAIN. Even now, after my switch-over, I’m only 95% sure everything will work at the end of the month when the old server gets switched off…

The new server (again, a dedicated one, not a VPS), is very slightly more expensive than the old ones, but it hosts multiple sites, this blog, forums, and some backend stuff for the minor web integration of some of my games (high scores, upgrade checks, challenge-sharing etc). Under $250/month for all that seems a good deal to me. The best bit is that with a dedicated server you aren’t at the whim of anybody else, you can reboot it whenever you like, install whatever you like. its freedom. The host change means better tech support, and less worry, and for under $3k a year. For an established indie selling lots of games, its not a major expense by any stretch.

I wish things were so simple for my own home internet.

Thats on a GOOD day, and thats when its connected. The 7.54 down is not bad actually. I can stream TV (but not in HD I assume, my eyesights crap enough not to be 100% sure), the 0.57Mbps up is depressing. This means that uploading a 300MB game installer (like big pharma) can take ages. Assume 3 builds (3 platforms) and 3 destinations (humble, BMT, GoG), thats 9 x 300MB. It takes me about 2 days (I don’t leave it overnight). What a drag. There are games I don’t download during free weekends because it would actually take the whole free weekend to grab them. And this is with my new, spangly business ISP that has truly unlimited bandwidth. When I was paying £45/month I had a 110GB/Month cap.

I know right… its literally medieval.

Why is my internet so bad? We have FTTC (fiber to the cabinet) but the nearest cabinet is further than the exchange, so ADSL is actually faster (madness). Because I live in a field, BT (the monopoly on internet here) refuses to lay fiber to the village unless we pay (at least) £40,000. That is not going to happen.

Can you see the nearest fiber cabinet?…. me neither.

The only solution is what they call a ‘leased line’. This is effectively a dedicated piece of fiber from my office to wherever the best exchange to connect it to will be. This is expensive. This is very expensive. This is almost ‘send your kids to private school’ expensive. And the net result would be 20Mbps down, 20Mbps up.

If, when they do a survey, they can do it in under 3 months and with no additional install fee above the £600 already quoted, I’ll probably do it (BTW I’d be locked in for 3 years). On the one hand its insanely pricey. On the other, I’ve lived here 7 years and there has been NO change in internet speeds here. Also totally reliable, uncontended and symmetrical upload/download does sound appealing. 24/7/365 support also sounds good. As a business cost, its slightly more affordable due to being an expense against tax, and frankly, being able to stream on twitch, to upload HD 60FPS videos easily, and generally live in the modern age is something thats kind of non-optional for an indie game dev in 2017.

Of course the real story is the ripoff prices of BT openreach, effectively a private monopoly that charges a fortune to rent a piece of plastic cabling after its installed. I’m all for hefty install fees, but surely once the cable is in the ground, its in the fucking ground. What am I paying for here? defense against mole attacks? c’mon. We get water through pipes in the ground too, but I don’t need to sell a vital organ each month to rent the pipe.

Oh well…you have to speculate to accumulate. Now I need to save up for botox and moisteurizer as HD streams of me pimping my games will show off just how old and uncool I really am.

 

 

Switching car body styles. in production line.

Posted on April 7, 2017

Something that is much requested, and that I am also keen on (and currently working on), is different body styles for Production Line. Currently there is just one body style: the sedan. It has optional spoiler and aerial, and a bunch of colors, but otherwise thats it. Obviously we need SUVs and compacts and sports cars and pick-up trucks and many more (I held a poll, and SUV and compact won).

Deciding that new bodystyles need to go in, and will go in, is easy. The tricky thing is how to implement it within the game.

I’ve mused on this for a while now. My initial plan was to keep things simple. Forcing the player to have completely separate production lines for each vehicle might make sense in the real world, but I think it would be pretty annoying for the player. On the other hand, some players may want to do precisely this, and feel that not doing so is ‘cheating’. We all know that factories tend to produce just 1 model, and that the same robot/person does not screw the same headlight onto an SUV or a compact car.

Actually the main problem is that if we *do* have different production lines, that also means we start needing new resources (if we follow that train of thought). So suddenly we don’t have sunroof, we have sedan sunroof, compact sunroof, SUV sunroof…etc, and the complexity starts getting ridiculous. As a result, I want to have a compromise situation where there is reason for the player to organize his production lines into separate bodystyles, but not enough realism to force them to worry about 6 different types of trunk or 6 different headlight types. There is complexity that leads to interesting and fun gameplay, and then there is just annoying busywork… I *could* allow component re-use across body styles but never allow lines to cross or meet, but then does it become more like tetris and less like a business sim? and do I now need a lot more export slots? Plus the routing code would become even more like spaghetti.

So I think I’m going to say that a ‘sunroof’ is a ‘sunroof’, and it can be fitted to any car. Obviously the final visual effect ‘on the car’ will be different, but I’m not going to force the player to micromanage different types. Obviously this isn’t totally realistic, but then sexy elf boob-armor and men carrying guns six times their weight isn’t either so throw me a damn bone here ok?

My compromise back in the land of realism is this:

Because allowing components to be interchangeable, and therefore allowing fitting slots to be interchangeable between body styles *could* lead to the weird sight of a jumble of different cars all trundling along a production line, which feels *wrong*, I’ve decided to allow this… BUT…. *drumroll* …. have a ‘re-tooling’ penalty in time whenever the bodystyle changes for a slot. In other words, when you weld the doors on an SUV, and the next car to rock-up to your slot is a sedan, you have to wait X seconds (probably some multiple of the normal work time) to realign your welding widgets, or whatever. This *allows* for bodystyle changes and thus mixed-production, but penalizes excessive swapping.

I think that’s quite a neat compromise, especially because I can then allow you to research ‘rapid re-tooling’ to lessen the impact. I actually already have tech named ‘retooling’, which I can rename to ‘asset recovery’ as that’s more accurately what it does.

Anyway, that is my current plan. When multiple body styles are researched, the player will be able to select from a drop down list at the start of the line which body style each new vehicle uses, and that carries through its entire life. Obviously it will need more tweaking at the business side of things, and also I would liked to find a way to ensure that SUVs cost more (for example), but I think its best to get multiple car styles in there sooner rather than later as its clearly something people want to see in their factories.

I do have plans for proper design studios and clay car models, but they will likely come later, if people want them.

In other news I’m working on early access readiness for steam. GoG will come too. I have to get a video named (its in progress) and I’m getting proper music too. Busy busy busy. it looks like new car types are likely to be the last new thing (apart from polish and bug fixes) that goes in before we go into early access.

 

Production Line 1.15 released. Feeling really good about it

Posted on April 4, 2017

I admit that I didn’t expect so many thousands of people to pre-order Production Line. The target I set for ‘I should be on early access when this many people are playing’ has already been hit, and I don’t have the video done yet, or the steam info ready for early access. I’ve been caught a bit by surprise.

This latest build changes the way engines are built, changes (again!) the financial modelling, adjusts a lot of numbers, puts in proper achievement art and text, adds a simple bit of tutorial stuff fixes a whole bunch of small issues, and adds some new research tech too. I feel really good about this build, its definitely the most balanced, fun and playable so far by a long way. If you are already playing the game you should get an update popup, but either way, just re-download the installer and you’ll get the new version. You can trigger an order email re-send from here.

The near term will see me getting new car model types into the game (coming soon, my artist dude rob is already working on them) and some proper music (are you sick of that loop yet? :D), and then the immediate order of business after that is to get the correct steam and GoG integration done so that I can get into Early Access on both platforms. (I know GoG call it something else). The longer term will see me thinking about marketing, competitors, a better tutorial, mod support, multiple language support, more achievements, and an expansion of the tech tree in all sorts of directions. I’d also like to get steam workshop support in there, obviously trading card support, and a bunch of character and polish.

I haven’t really done much to promote the game other than my weekly youtube videos which are fun to do and fairly popular (by my paltry standards). I’ve spent a few hundred dollars on reddit and facebook to boost the game a bit, but nothing serious. I’ve partly relied on it having a bit of a bargain price (that is likely to go to $13 next week, and probably $15 for early access), but that wont be my strategy forever. I’m considering getting a booth for EGX in the UK in September (when I assume the game will still be in Early Access) to help get the word out a bit more. I haven’t made much of a push to contact youtubers, but once I hit early access I will start to do that a bit too.

Anyway, the most important thing by far is that I’m really enjoying working on it, its gets better every week, and so far the community have been really helpful and understanding when I break stuff or mess things up :D. If you haven’t already grabbed the game, its still $12 for a few more days:

Complete 1.15 changelist:

[version 1.15]
1) [Bug] Fixed crash bug when upgrading power plants.
2) [Bug] Fixed memory leak caused by the save game dialog. (Big one :D).
3) [Feature] Engine system redesign. Flywheel no longer used, engine is now a powertrain, new research, new slots for new engine production line.
4) [Feature] Stop-start engine is a new researchable upgrade. This is the first upgrade applying to a component you can manufacture.
5) [Bug] Fixed issues where changing the requirements for a supply stockpile would cause overlaps and confusion.
6) [Bug] Fixed routing for production line slots so resource requirements changes frees up the right room for newly required components.
7) [Performance] Vastly sped up drawing and creation of the car stock dialog when number of cars in stock is very high.
8) [Feature] Load game dialog now auto-sizes where possible and will not load invalid save games.
9) [Feature] Changes to economic model. Design screen now automatically matches market value unless you override. feature price tab removed.
10) [Feature] Basic tutorial pop-up windows added.
11) [Bug] Some achievements that did not display progress now do.
12) [Feature] 2 new researchable factory upgrades increase the refunds from demolishing equipment.
13) [GUI] Refunds now have a green rising value to distinguish them from expenditure.
14) [GUI] Supply stockpiles now also show their delivery point as a colored tile like normal stockpiles in resource-placing mode.
15) [Balance] Major re-pricing of upgrades and the market-values of features. Most upgrades to slots doubled in prices. Many features are now less valuable.
16) [Graphics] Some production slots now have work-pits next to the conveyor belt.
17) [GUI] Achievements now have final graphics and better text.

Production Line video blog #21 thinking about engines.

Posted on March 26, 2017

I talk a little about it in this video, but will blog about it properly soon. basically I want a separate system for putting together engines, but do not want to overcomplicate matters, or for that matter…under complicate them. Interested to know peoples thoughts on this. Also I remain excited about new car body designs, they can’t come soon enough :D. I should blog about promoting and marketing PL at some point too…its definitely something I have not done much work on yet.