Some of you remember my complaint about lack of sales of my first game... and also because this situation was getting me out of motivation to continue with indie dev.
Recently I got a great motivation: in my soccer management forum got a msg from Mark Vaughan in person! :p

For those of you that don't know who is, well, he is the Lead Programmer of the award winning Championship Manager series :D

For sure receiving a congrats msg from him was one of the biggest incentive I could receive!!! ;) even if sales are still under 10/month at least I got a moral motivation now... :D

So... guess I can go back writing more games :p

As many people confirmed in other threads, "GO ON!". Remember Steve Pavlina saying many indies succeed in the fourth, fifth or sixth trials or even more. You are at a far point beyond that. Yesterday I checked download.com and saw something like 3900 downloads only in two weeks. I think you have catched the word of mouth situation. Support USM in every aspect; the web site, advertisement, new features...

ergas

;)

Yes got lot of downloads in effect :p
Need really to get back to improving the game... those damn 3d cutscenes take forever to make :(

Originally posted by Jack_Norton
those damn 3d cutscenes take forever to make :(

I agree that 3d cutscenes are important, but I also think that they are not vital. Are 3d scenes real-time rendered? What do you exactly mean with that? Remember "The Manager" game long ago, having animated goals. The goals were obtained by combination of about 5*5*5 different scenes to get different impressions of goals. Is that what you mean? If so, I still do not think that it is vital, but very impressive indeed. I did not spend enough time on USM therefore I cannot really give you a decent feedback, but in a general aspect, I think in manager games, the most important thing is "having the gamer to feel that the strategy that s/he has determined works right when applied at the right time and at the right case." Let the gamer feel s/he is doing a creative thing. A simple example: at the 88th minute you are loosing the match, but during the game you give clues that some tactics do not work and the gamer is responsible to figure out why, then, s/he finds the solution like activating a certain tactic such as cross run at long balls, and gets a goal scored. Now, the gamer feels that the goal is a product of his creativity. This might be hard to obtain, but this is what most manager gamers look for I think.

Another idea is to insert funny features. Like you transfer "Maradona", but from his reputation you take the risk of having him taking drugs before the match. The match day comes and Maradona is drunk, and the manager understands that he is drunk late in the second half, and replaces with another player. Just an example that was, but the idea is finding original features that makes the game fun. All are hard to implement and hard to extend the application, but could be more beneficial and cost effective than 3d.

Lastly, another true statement for most types of games: more the number of variables, less the game play, and slower the game tempo. Again from "The Manager" example, there are only 3 features of a player in that game: fitness, skill and form. The Manager is a fast fast game, you can finish a league session in only two hours. This is not a good solution for giant manager games because manager gamers like to spend weeks on a league session, but for a game like USM less variables could be better.

Nowadays I do not have much time to play USM, but when I can spare time, I will try to give you more specific feedback.

ergas

Congratulations Jack!! That's excellent news! I'm glad your game is doing OK and with that kind of qualified feedback :)

The cutscenes are pre-rendered, and yes, I plan to make those similar to "The Manager" (great old game! :))

I agree with you that aren't essential: however I think they can add "value" to the casual gamer. I think that most player just saw the nice screens, but not a single shot of the playfield, and I want to give them at least some :P

About the "less stats"... I put lot of stats in USM, that are 14 for each player, but for each player role can be reduced to just 4.
Example, for an attacker the most relevant skills are: speed, shoot, dribbling, heading.

So basically you can just ignore all the other and choose the player based on those stats. But for the soccer fanatic who like CM4, you have also the other 11 skills/stats to look at :D

Finally, making the user see that his decision influences the matche is a great thing, probably the main focus of every soccer management game.
However is one of the most HARD thing to do: you can't even think that if a player does the right move he will for sure win (because it's not like that in the real world too!).

I'll try to improve that thing too, but for sure it is really hard, even the commercial games like CM4 haven't fully implemented that I believe...

Originally posted by Jack_Norton

About the "less stats"... I put lot of stats in USM, that are 14 for each player, but for each player role can be reduced to just 4.
Example, for an attacker the most relevant skills are: speed, shoot, dribbling, heading.


That's great! If I were still a university student, probably now I would be playing USM.

ergas

Jack, well done, USM is an excellent game and should in time do very well.

The only things I noticed were the lack of real names:(, I know there is the problem of licencing etc but I hope you can sort something out.

Also am I missing something, when I played the game I couldn't see how to actualy 'watch' the match progress. I clicked next match and the result was shown, is it possible to view a running commentary on the game?, cos that for me is the most exciting part of Championship Manager. Even something simple like just the latest score and a clock would be fine.

Anyway thumbs up, the presentation is top notch, if you've done that by yourself then that's fantastic.

Cheers,

Oli Norwell
Rockingham Software Ltd - creators of 'Michael Vaughan's Cricket Manager'

Hello and thanks! :)
Yes, I've done all by myself, and sometimes I wonder how I was able to continue doing it while keeping a full daytime job (thank God now I switched to full-time webdesigner :))

I had strong motivation, that's for sure. I hope that in the long run my efforts will pay off.

About the match view question, in the first release (that is v1.0) I wasn't able to include it yet.
Next one should have text commentary like CM series, and also some 3d (pre-rendered) cutscenes (damn those are hard to make!!).

Jack, that really is excellent news. I've been lurking here for a while, following your progress, and that kind of response is...well...inspiring.

Seriously, seriously impressed!

Well thanks again :p

I'm getting more "congratulations" email/msg than sales, but I am happy anyway ;)

The game itself started as a "way to understand better Blitz3d". Well, from that beginning it went really so far. I think I pushed Blitz3d capabilities at max., and the fun fact is that I didn't ever used any 3D function.
Yes I could have made it in Blitzplus or even Blitz2d! :D

I've read that you got the blitzplus version. Well I can only suggest you one thing: if you plan to write a complex game like mine, DON'T use Blitz languages! :))

They're perfect for simple games, but for games like USM is way better to use C/C++ ... !!

Hi Jack,

What sort of goals have you laid out for yourself? Did you plan out your sales goals for the next n months or are you just going with the flow?

I can only suggest you one thing: if you plan to write a complex game like mine, DON'T use Blitz languages!


Just out of interest, what in particular did you find frustrating?

My goals:
I wanted to go fulltime indie. So I switched from full-time to part-time with my current job (webdesigner). I have now 2 days more to dedicate to my indie projects :p

Since this was my first attempt, I had a really small goal in mind (at least from what I've read here and around internet): to sell 100 copies of my game in TOTAL (that could even take 1 year).
That's about 2500$ (since I sell it at 24.95$, but I started discounting it at 19.99$).

The 17th of october will be 1 month since the game is out.
If things continue this way (of course they can go better...worse I don't think! hehe), in about 10 months I'll be able to reach my small goal... so I guess that for niche games (like soccer management), no matter how good your product can be, a conversion rate of 1% is clearly impossible to reach (at least this is my experience... wish everyone more luck!).

So now I'll make some changes/improvement to the game to try to get a better conversion (now is something like 1 sale every 2750 downloads...!!!).

To answer the question about frustration:
- the IDE is terrible... when you have over 15,000 lines of code with lot of variables, struct, etc is hard to keep track of everything
- the language limitations. I couldn't have ARRAY inside a TYPE. Tried that but the game instantly crashed or went out of memory.
example:

Type gi_t
Field league;league reference
Field home_s[10];goals scored
Field away_s[10]
End Type

the two arrays home_s and away_s are of 10, if I make them larger (say 20-25) the game run out of memory or crashes :p
that's because I had also defined:
Global cm.gi_t[NUM_GIOR]

where NUM_GIOR = 1800

- but the worse thing of all was DEBUGGING. Terrible!!! every time I had to change a thing, I had to restart from beginning!!!
Now with Visual C, I can just break, change pieces of code, hit continue, and in 90% of cases I can continue to debug the game without ever had to restart... I think that of the 350+ hours spent on the project, at least 100-125 I have WASTED with the debug thing :mad:

Fair enough. As far as ide/debugging goes, I guess you don't miss it if you've never had it... the last time I did a lot of programming was in the late 80s/early 90s so these days everything seems so easy by comparison - I'm currently maintaining and debugging a game of over 20,000 lines and rising without any problems. But I may be missing out on some nifty features that I've just never been exposed to!

The type/array problem seems strange though. I know blitz in its current form is a bit weak on data structures (I thought that would be a bit of a problem for a management game), and that arrays within types are currently a bit of a hack, but I just did a test and I was able to crate a type with 2 arrays in it, each dimmed to 200, and then created 10,000 instances of the type and it ran no problem. Well, the only problem was that it took about 10 seconds to initialise and another 10 seconds to clean up upon exit.

But I may be missing out on some nifty features that I've just never been exposed to!

Yes, I honestly believe that strategy/simulation games are quite more difficult than other kind of games. I made some adventures before in VB and that was more easy :p
The problem is that you have to make like a database (players, teams, matches, seasons, etc), with references between records of different tables... and well, doing that with a language that is NOT OOP is quite difficult :p

Well, the only problem was that it took about 10 seconds to initialise and another 10 seconds to clean up upon exit.
Yes, that thing was quite annoying (imagine the user having to wait so long every time he start the game...!).
But I also noticed that if I move/delete/update things in the arrays, the game CAN crash (I wasn't unable to determine if it was because of my bug or blitz one :p).

Well anyway I managed to find a compromise, but I'm already making some experiments with Visual C and I can say that is quite more powerful than Blitz3d.

As for the complexity, well I found a library (http://ptk.phelios.com) which is perfect for 2D games, basically I have passed from Blitzbasic to C in one week without big problems (because the library gives you lot of ready to use commands).

It is not for 3D, but it has cross-dev capabilities which is a invaluable thing for indie dev, I believe...!

Jack - thanks for explaining the problems you had with Blitz3D. BTW: What's your next project going to be?

Just from my point of view, I settled on BlitzPlus for a few reasons:

1) I have very little C++ experience. My head is alreay stuffed full of classic VB, Java, C# and the unstructured world of web programming. I simply don't have the time or motivation to get to grips with C++.

2) I know VB inside out and have heaps of experience with it. I figure I can use BlitzPlus to abstract the graphical functions and leave me to concentrate on implementing my game in a reasonably familiar language.

3) I think it's important to start small here. There's no point writing a huge multi-player, multi-platform game that will be in development for years and years. I want to crank out an fairly simple 2D game within a few months while gaining experience of the games development process. With my utilities I've been concentrating on features, and ease-of-use. With a game, though, I'm going to have to look into playability and addictiveness.

4) I have future plans for the Windows-GUI capabilities of BlitzPlus. There are a few packages I want to write that I'm reticent to do with my current skillset. Classic VB isn't much use due to the intallation issues and download size. DotNET is currently a big no-no because non-developers aren't going to download a 20mb framework just to run an app - no matter how good the app is.

I suppose this is a bit of a trade-off in the end because, as Jack says, debugging in any Blitz is tricky, but it will give me a bit of a head start with my game development. I'm going into this with my eyes open and feel confident about what I can do with it.

Just my 0.02p worth.

What's your next project going to be?
As for now I haven't yet started a new project, but I am experimenting with C and PTK.
I already have some ideas for a RPG, another simulation or a strategy game, but all depend on time and resources :)
Now I'll be much more careful when choosing the kind of game to do next: I must be SURE that I can complete it in a reasonable time and also that I have all resource needed.
For example I realized that making both graphic+programming require a lot of time, so I am still wondering if it would be better to have someone else doing the art... or vice-versa ;)

About the C++ thing, well, with PTK you don't HAVE to use C++ really, you can simply use C in a way that is really similar to basic syntax. If you take a look at examples on the website you can see what I'm talking about :)

But I understand your point of view, I too started with Blitz.
And if I was to make simpler games, or even 3d ones, well Blitz would be good, really good :p

And yes, you SHOULD start small. I didn't, but was a real pain :p
It's better to make a smaller but complete games than a big project but leave it unfinished or buggy (hey it's not the case of USM !!! hehe)

Ok, I'm sure I'm gonna be flammed for this but why ptk ?


Except for Mac support, there really isn't much to this library than many other free or cheap ones.Another thing is that audio requires BASS or FMOD, which means another 100 minimum on top of the library cost.

It also uses Open network thingy, which means you are bound to their license, its ok for now, but if they change it, you may be screwed.

If we consider that ptk then only offers you original :

KWindow
KGraphic
KMiscTools
KInput
KText
KIniReader

And taken that KWindow and KGrpahic and KInput are just used as wrappers, they don't even offer you some advanced stuff like native animation support or particle systems or whatver I don't see how it can be really worth the money (sorry to the author but this is just my oppinion). I would really like to know why you think it is worth it.

Another thing is that it uses OGL, while I have no problem with this, I think you should think twice (check Alien Flux OGL problems).


my 2 cents.

ps: not directly targetting ptk as being a bad library, it is just my oppinion that there are better cheap ones out there, and I would like to know why ptk instead. If the author doesn't like me saying this (I forgot who it was) let me know and I'll delete the post.

No flaming ;)

You're right, it's just a wrapper like other wrappers, you might as well use SDL or torque or allegro.

Put simply : with PTK I programmed 10 games in 10 months and am now full time indie.

I'd really like to hear about succesful stories with other libs.
pat.

Another thing is that it uses OGL, while I have no problem with this, I think you should think twice (check Alien Flux OGL problems).
I'll start with this :)
OpenGL is the only way (as far as I know!) to have cross-dev with pc/mac/linux :)

Neverwinter Nights, a big commercial title, uses OpenGL too :)

About the library itself: I had NO TIME to study DX, C++, or other complex/advanced things. Of course that is not the correct way or the best way to use C, but I think that the most important thing is to GET results :p

I see that patrox already replied: well, he is right, his library is nothing "magic", but simply let an ex-blitz user like me to make the big step from basic/constructionkit languages to pure C.

And also I get cross-dev capability :)

Sure there are SDL, Allegro, Torque and other things: but for someone like me who don't know well all the problems that may arise (and I am SURE there are) it is better to use a library who has been tested. As he said, he made already 10 games with it :p

I am not in any way affiliated with him, he's only a friend: but since now I must say that I had no problem using PTK, and even if I hadn't made a game yet, all looks very promising.

Said that, if you want to use 3D or particle system and so on, don't use PTK... you can try Torque instead ;)

I tried it, spent 250$ to buy the engine, then after 2-3 months I gave up without even being able to compile an example.... (I am a C/C++ geek, that's for sure!!! hehe)

The great thing about game development compared to software (for the web) is you really don't have to worry about version control since you don't access too many external files (or do you?).

Actually I'm sure you have external files that you reference, how do you handle version controls....do you ever have to handle multiple versions at the same time? what's the best strategy for this...

I agree Blitz3D was not the best choice for developing a football management game. However regarding the ide, I personally use Protean IDE (http://www.proteanide.co.uk ) which is a superb development environment that has saved me a ton of time.

By the way my own personal project written in Blitz3D has reached 10,000+ lines but is still easily manageable, but I guess it all depends on the type of game you make and how you structure your code.

Originally posted by Akura
[B]Except for Mac support, there really isn't much to this library than many other free or cheap ones.Another thing is that audio requires BASS or FMOD, which means another 100 minimum on top of the library cost.


And taken that KWindow and KGrpahic and KInput are just used as wrappers, they don't even offer you some advanced stuff like native animation support or particle systems or whatver I don't see how it can be really worth the money (sorry to the author but this is just my oppinion). I would really like to know why you think it is worth it.


Just added another wrapper around directsound, ( Mac version coming soon! )

->You don't need bass or fmod anymore.

pat.

Ah, a fellow Protean user ;) Really made my Blitz life easier. Makes it much easier to break a project into small chunks without an organisational headache. I don't think I would have been able to manage my current project without it.

I tried protean too, but crashed after 5 min of usage... :)

Well, anyway the greatest lack of Blitz is a good debugger with "break and continue", watches, etc capabilities...

People who like basic like languages may want to check out python. It's even easier than basic imho, and has really mature libraries and development tools. Like debuggers, profilers, ides, interactive interpreters etc.

Check out http://pygame.org/ for a python game library.

You can also use python for scripting game making tools like blender, poser, gamespace, the gimp etc. As well as lots of 3d engines(nebula, ogre, torque, crystal space, soya3d, irrlicht, openscene graph).


Have fun!
http://www.holepit.com/