Blog for Adam Beaumont at GarageGames.com

Where did all the time go?

Adam Beaumont — 2007-07-16T03:30:18+00:00

Wow - a year and a half (almost) since my last post about being almost there and Roboherd is only just getting out of the door. I guess the bad thing is its taken so long, but the good thing is at least it's done. Yep, all zipped up and ready to go - Roboherd should be seeing the light of day fairly soon now, once we get everything signed, sealed, and delivered. It's been an epic development cycle to say the least (I think 3-4 years on and off (mostly off)) but I guess perserverance pays off.

Having got that out of the way we're now starting to take a look at what we can do next - we have a number of ideas, ranging from a 2d platformer to a couple of multiplayer on line games. I think the 2d one would be the easiest but we'll probably end up going online - I have a sports sim that I'd like to do but we've got to compare it against the other ideas and get some kind of consensus about which direction we should go in.

The main goal though is to reduce development time. Part of that we think will come from being more focussed around design and communicating the game vision, part of it will come from getting more people on board (well 1 or 2 more I think). We really suffered on Roboherd in not having enough people to keep things moving and deliver assets quickly enough which meant that things got really bogged down or it was difficult at times to see the final vision. The plan is to try and get some more skills on board on the back of Roboherd but I guess we will have to see how well that works out.

I'm also interested to find out how much we can leverage the work we've put into Roboherd - not really from making cash on selling the game but more in terms of what we can get out of it in terms of reputation, increased exposure, more interest from other developers, etc. Interesting to see.

Roboherd on Mac?

Adam Beaumont — 2006-04-23T01:37:29+00:00

Feedback from our latest demo release of Roboherd has been ok and looking at the online scores it seems like some people have been able to get the hang of things and post some pretty good times. We're currently figuring out how much more work we should do on it and if we can realistically get it to a level where we can publish it or if we should finish it up, stick a working version on our website and move on to the next project.

One of things I'd like to do is get a version up and running on Mac - things I have are:

- Heavily modded version of v1.2 source code for PC (VC6)
- iMac with intel inside
- XCode

How hard is it going to be for me to get my PC source code set up on Mac? Anyone have any experience of this - I am gonna try and just copy it over and see if I can bodge it to work but tbh I am expecting there to be tons of issues. If anyone can give a helping hand I'd appreciate it - otherwise just wish me luck !

Roboherd Demo

Adam Beaumont — 2006-04-05T14:05:32+00:00

Well we've finally got something together for the Roboherd demo - it's not quite as far along as we would have liked (some of the menu screens still need some work) but instead of delay another few weeks I decided to stick to the deadline we set ourselves and get something released.

You can find the installer at www.roboherd.com/downloads/Roboherd.exe - I just put it together with NSIS and I am pretty sure it works but if you have any issues then let me know (there could well be a few teething problems).

There are 10 levels in this release, we have about 25 done so far in total but are looking to get that number somewhere north of 50 I guess before we can go live. The webstats should be working (at www.roboherd.com) so you can set up a username and see your best times online - by default they will just get recorded under demouser.

As always any feedback appreciated - still looking for someone to help out on level building so if you're interested then let us know.

Roboherd update

Adam Beaumont — 2006-03-25T22:29:20+00:00

Roboherd progress has been pretty good this month - we're up to about 20 levels now and have totally revamped the graphics (both 2d and 3d). Everything is looking much more polished and you can almost believe that people would pay money to play this thing! There's still a long way to go but we can see the light at the end of the tunnel.

Currently just working on getting the next demo release sorted out - there are a couple of loose ends to sort out but once they are done we'll put up a dev snapshot and a link to the download. We are looking to get some dedicated testers on board as most of the levels have only really had a few playthroughs - needs a dedicated effort to track down all the bugs and iron out any wrinkles. If you're interested and have the time then let us know.

I also bought a new imac this week to start building a mac version on - so far haven't done much with it but I have an XCode book and am optimistic I should be able to get everything up and running. Not sure yet whether I should port my existing code or take 1.4 and start adding stuff into that, or take TSE and go from there. I think I will probably have to start from a fresh build and add stuff back in but hopefully that shouldn't be too difficult - will keep you updated on progress.

Next Roboherd release coming soon...

Adam Beaumont — 2006-02-26T07:56:14+00:00

Is it just me or does the new blog page not work in firefox? All the text boxes had been squashed off the page so I couldn't enter anything and have therefore had to revert to IE - weird!

Anyway - progress this month on Roboherd has gone pretty well. Music is in game, controls have been tweaked even more and a bit of work has gone into the menu system (although there is still a bit more to do there). I've even got round to building a couple of new levels - well about 10 in fact. I came up with a way of prototyping levels and its going ok - it's pretty easy to do and the results are playable. I guess it remains to be seen if we'll need to go back and redo them all but even if we do the ideas are down there for someone to follow. The web stats at www.roboherd.com have all been updated too with the new level layout so if we can get some people playing after this release then all their best scores should be getting captured (and displayed) online.

It's definitely starting to look good as a game now and the more levels I get built I think the easier it is for people to see how the game is supposed to work. Then there is the question of is it fun but tbh I have been developing it for soooo long and I still like playing it so there must be something there (for at least one person).

So hopefully we should be in line for a new release pretty soon - just waiting for some assets to be finished up before putting them in game. I guess if they don't make it out of the workshop soon then I'll just do the release anyway.

No screenshots but hopefully there should be a devsnapshot (and another release) coming soon!

Roboherd keeps on movin

Adam Beaumont — 2006-02-03T22:39:13+00:00

In spite of all the moving around the world, having a baby, and a day job that takes up about 12 hours a day Roboherd is still progressing! That's the good news - the bad news is of course that progress is slow.

Since we last did a release we've done a fair bit of tinkering based on the feedback we got from people who played the game. The control system is pretty much overhauled now so that you can play with mouse alone although if u do use the keyboard you can probably finish levels more quickly so theres a nice learning curve built in that will mean pro players can get more out of it if they put in the effort to learn. Vanilla has done a ton of stuff on the look of menu screens and in game fonts etc so its all starting to feel a lot more polished. There are still some things that need finishing off though and the better everything else looks then the worse those things look in comparison!

We also have some music which is cool although i'm having a small problem with balancing the relative volume levels of sound effects and music. Once I get some of the other things off the list I'm going to work on the audio in a bit more detail.

I'm currently recoding the level system - previously each .mis file contained 6 levels but we're scrapping that and now each .mis file contains 1 level. Load times are quick enough for this to be ok and it means that it's easier to manage all the data in the .mis file. It affects quite a lot of areas though so I'm having to work through all the impact - I pretty much only have the web stats left and then it should be done. All this reorg is basically to make building levels easier - we still don't have very many and we still don't have anyone else on board to help out so i figure the easier we can make it to build them the better !

We'll keep plugging away to try and get this thing finished - look out for a new demo soon!

Wednesday Dec 7 8:11

Adam Beaumont — 2005-12-07T08:11:45+00:00

Quite an eventful month since I last posted - my wife and I had a baby girl so the whole world feels like it's been turned upside down, what with night feeds and nappy changes. Anyway, development on Roboherd continues...

We've done a bit more with the site and the forums are now up and running and being used (by the dev team at least) - currently we have one public and a couple of dev forums. There are only 5 people signed up so we're not really being overrun with a need to add any more! I've posted up a link to the Roboherd guide to mapping on the general forum - this is the first draft of what I hope will be the repository for all things required to make a roboherd map work. At the moment its just a list of things that need to go into the .mis file but eventually it will have details on all the map objects and how they can be used and linked. If you get round to reading it then let me know if you have any feedback.

Feedback from the last demo we released highlighted an issue with our control machanism - for some people it was fine but the general consensus was that for noobs or none fps players a two handed system was going to be too complicated. Solving this control problem was a bit of an issue - an overhead view was suggested but we felt that was going to change the whole nature of the game and turn it into something that might almost as well be 2d instead of 3d. So we thought about it for what seemed like ages with no success and then finally a breakthrough ! We have now come up with something that keeps the 3d look but also means you can play with just your mouse (although there may be a few other keys you can press if you get stuck). We're calling it 'chariot mode' because you basically just get dragged along behind the robo you're focusing on.

So far it seems to work really well and actually makes the game easier to play (which must be a good sign!). We're putting together some HUD improvements and then we'll be ready for a new release - it might come out as a dev snapshot but if not then watch out for another plan or maybe even something on the website www.roboherd.com

We're still having trouble getting anyone else to join our team - so far we've had a few close misses from people wanting to help out on the mapping side but nothing done as yet. We've still got things we can work on in the meantime to make the game better and hopefully each new release should increase profile and interest so for the moment we aren't panicking and are willing to see what happens.

After we get this release out of the way I am seriously considering stopping new development and trying to port the code over to Mac to see how easy it would be to get it compiled up there. Obviously this could be a huge undertaking but I've been thinking about doing it for a while and if we have a demo that's playable I think getting some feedback from the mac community might be a good idea.

Finally I had that mid-project thing happen to me the other day where I came up with a *really great game idea* - so far I have resisted the temptation to do anything more than idly think about it as there is nothing worse than stopping what you are doing now to chase after another 'even better idea'. I read somewhere that in this situation if it really is a good idea then it will still be there when you finish the project you are already working on so don't get distracted. Hopefully we can stick the course with Roboherd - it's slowly starting to come together so actually getting it finished is becoming something that is more and more believable.

Saturday Nov 12 13:51

Adam Beaumont — 2005-11-12T13:51:15+00:00

Things have been quite slow this week on Roboherd. We're trying to increase the content in the game which basically means getting some more levels done. Last couple of weeks I've done quite a bit of coding building new objects that the robos have to avoid or interact with, now its time to try and get them integrated into some levels and released for people to play.

Some things I've been thinking about this week:

Demos - are they a good thing? I was thinking that we should try and put out a demo version of Roboherd every so often (couple of months?) just to show how things were going and to slowly build up a following of interested players. The game is definitely playable so people would get some reward for downloading. Is this a good idea? On the whole I think it is - its a good way of setting goals and measuring progress and it may generate some interest in the project (and get some help on board). The key thing for me is getting feedback from people when they play - that's where I see the real value of a demo. I've wondered about putting a button in the next release that will maybe give players an opportunity to answer some questions about playability and give feedback. I've already got some code that can post things to a mysql database on the web so with a bit of tweaking it may be able to collect feedback as well...

Mac Build - I am soooo close to buying a mac mini to see if I can compile Roboherd up on it for a Mac release next time round. I am naively assuming that the code will compile up without too many problems but I guess there is only one way to find out. I have a 15" powerbook already but its not really great as a dev environment - I tried using XCode to compile up a fresh version of 1.3 and there were a few things that needed sorting out. I assume that my version of the Roboherd code will have those problems and a few more. I saw a couple of vids from IGC though and mac seems to be a good way to go - by this time tomorrow I may be there.

Level building lite - we're a bit preoocupied with levels at the moment - one thing that struck me during the week was that there may be another way. We want to see beautifully crafted terrains and thoughtfully worked out puzzles and challenges, but until we get someone on board to help us out with that we are either stuck or we have to find a shortcut. Well, I came up with one. I decided that maybe I could build a level out of platforms - I could design them on graph paper, put them together easily even with my elementary f11 skillls, and use them to test out some ideas for the new objects I'd built - Metal World was born ! It's sort of like a Marble blast setting in that its platforms suspended in mid air - a bit of a tweak in the code and the robos can run around on them and your repellors work as well. I even came up with a level builder in Excel - you put letters in various cells, press a button, and a macro builds all the scripts to create the various objects and meshes for the level - sweet! It could be a good way for me to showcase some of the puzzle ideas I've had before someone else comes in and rebuilds them in beautiful 3d - kind of like level prototyping in a way.

Friday Oct 28 12:57

Adam Beaumont — 2005-10-28T12:57:16+00:00

Chasing shadows...

Things on Roboherd have been progressing this week, although not as quickly as we would have liked. I had some ideas over the weekends about some more level objects we could add in to the game and started coding a couple up. So far forcefields, switches and falling objects have made it in game but not without a struggle.

Getting a transparent forcefield stretched my modelling skills to the max (!) and I had a major headache with shadows - they just kept disappearing. After rewriting stuff, trying to use decals instead and a whole bunch of head scratching I realised that shadows don't get rendered if their casting object is culled from the scene. As my shadows were a long way away from the object casting them they were getting culled frequently. A quick rewrite and a bit of a change of perspective now means I have them where I want them. Unfortunately that took a bit of time so some of the other stuff I was thinking about is going to have to wait.

The good thing about this kind of game for me at the moment is that each object is a mini project in itself - you get to think of something cool, code it up and get it in game, then move on to the next thing. It's actually a really good way of developing, just building self contained little pieces of code and dropping them into the game world. Nothing takes too long to do, so you're always able to work on interesting stuff and keep the project moving forward. Obviously we still need a mapping person to put them all together in the right place but we're working on sorting that out.

Meanwhile as well as doing some new graphics Vanilla has updated the site (www.roboherd.com) - it's looking really good now and I've finally got the high scores hooked up to the right version of the game, so when people play it should be posting scores onto the website (firewall permitting). I'm probably going to have to clear all the scores down over the weekend as the levels have changed so much that some of the times up there are impossible with the existing layouts !

We've had quite a positive response to Roboherd over the last few weeks so it's definitely spurring us on to work harder on it. With a bit more effort we should have something that will give a better idea of how we see the game and where we think it can go - we're aiming to get another dev snapshot out there sometime soon.

Saturday Oct 15 3:27

Adam Beaumont — 2005-10-15T03:27:47+00:00

Things moving again...

Another week or so and another bit of progress. After releasing a version of Roboherd with the dev snapshot that didn't update the right online hiscore page (whoops!) I've re-uploaded something that now points to the official site. So from now on anyone downloading the demo and playing with stats enabled should be able to see their best times on the website. I also did a bit of work on the php to get more information out of the database. Every time I do a bit of php tho i keep seeing other reports and tweaks that would be cool to do as well (top ten scores, most played missions, most frequent player, longest games) so my todo list just keeps getting longer.

On the game side I've been adding in some features for new maps - we now have a force field object and a switch object. You can link the two together so that if a robo goes over a switch then the force field switches on(off). I've written it so that you can associate a switch with pretty much anything, it just needs to have some script to respond to the call from the switch. You set all this up in the mission so the idea is that it should be pretty flexible for when we actually get someone who wants to build us some maps ! I'm thinking about putting a page on the website together that shows all the objects we have and how to use them in a mission (sort of like a guide to mapping). Ideally I'd like to do some work on the in game mission editor (cos its not very good) but thats I think just too big a distraction and would take too long. Maybe someone else can come up with that and I can concentrate on Roboherd!

On the graphics side we're doing a bit of polishing so we might have some better screenshots to update the web with in a couple of days or so.