The cost to develop a new game?

[Trew]

Humour me here...

 

I was wondering how much it would cost to develop a new Lynx game. Imagine I came into some money and wanted to make a "retail" style game (something comparable in looks, length and gameplay). Print up boxes and do a short run of cartridges. The goal being to, at best, try to cover cost. So, as the money-man... in this hypothetical situation, I'd need all the development done by someone else etc etc. Let's also assume there was a concept/storyline and so on, we just needed the programming/practicals done.

 

Also, I am sure there are many challenges and "piece of string" things to consider. Let's just think top-level, and for the fun of it. I guess I am thinking along the lines of the reverse of what happens with homebrew, where someone chips away at a game, and we all sit here hungry for it, hoping that person has time to progress it. What if we planned it intentionally with an idea of cost and timeline and did it as a project.

 

Is that something that's even realistic?

[Willard]

As with any game the cost would depend on several factors. The amount and the quality of the art / music / sound fx in the game. Some genres require more resources than others.

 

Then the complexity of the game. Everything that happens in a game has to be programmed.

 

To put it in perspective we found a game called hotdog (platformer). It was made by two guys - one created the engine, the other did all the art & level scripting. It took him 18 months to do this by himself with an already established engine (used on games like Fat Bobby). It was an ambitious game that had a lot of variety (plays much like a console platformer of the time), and is of high quality, so it'd probably be one of the most demanding kind of games to produce.

 

Then there are games like the traffic jam game we were developing (based on the sliding tile game) that was pretty simple and done in spare time by the coder and me. With a long-term deadline or a smaller game (or both) it becomes more feasible. But you also run the risk of someone defecting because things get busy for them or their interests change.

 

The hardest part would probably be finding someone who would be willing to do the programming. You are not providing a reliable job offer, more like a one-off contract. Most people who are reasonably talented will have a full-time job with a real company that has a sustainable business model. So they have and want job security (they have to live!) and you're paying them to make a lynx games instead of do something else - like program an XBOX game (which allows them to gain experience developing for a viable platform which is also valuable). The fact that you're paying them to make a less technical game doesn't matter, the value of their time is still the same. And they may like seeing their contributions benefit a lot of people which they won't get with the Lynx.

 

Factor in that the Lynx has probably the smallest profit margin and "market size" / user base of any Atari home console and that complicates things further.

 

You could probably finagle something with enough money and a balanced approach in regard to game size or time horizon (or just an absurd amount of money), you might get the developer a little more amped up if the resources could be shared with another platform. Or if a hobbyist programmer liked your idea enough ... but that is always a risk because they may not trust you to follow through so in that case having the resources ready is an advantage. But having the resources ready is a double -edged sword because you might not find a programmer.

 

^ Just hypothetical of course, I think it's an interesting question but probably not very realistic unless you want to go through a lot of effort to lose a lot of money.

Edited March 21, 2015 by Willard

[GadgetUK]

It varies on the game type, complexity, sound, music and gfx. For Alien I spent approx 21 days of around 7 hours each (roughly). 108 Stars spent a fair bit of time on gfx, and Der Luchs on music - no idea how long but I would guess half as much as the code, but that's just a guess.

[kskunk]

Source: I did some contract game programming when I was a teenager.

 

Here's my two cents on the 'piece of string'. I think it's affordable - a few thousand dollars - if you want a simple, derivative, low-fidelity, game.

 

Simple: A game that fits on one screen could be doable in a month. That's 160 hours at (local programmer's wage). Sophisticated scrolling adventures can take several man years, which ought to be into the six figure range no matter where you find your coders.

 

Derivative: A lot of game programming time goes into making a game enjoyable. Just having a list of original ideas and a story is no help. The programmer has to go through several cycles trying ideas and mechanics, tuning them until they're fun, and giving up on the ones that can't be fun, starting over with fresh ideas. If you can just copy an existing game exactly, you can avoid this highly risky and time consuming step. Most "original" games simply fail at this stage. You could lose your entire investment and never find the fun.

Low-fidelity: 8-bit programming is a skill modern programmers don't have. Anything fancy on the Lynx requires fresh experience with bare-metal 6502 assembly - 25 years ago, it wasn't too bad to find that skill, today it's a needle in a haystack. If you aim for very low complexity, it will help. I heard a story of someone trying to develop a Genesis title, but his contract programmer's 15 years of modern programming experience was useless in an environment with 1/1000th the performance. The budget was squandered, and the project (probably?) abandoned as a result.

 

I've done a couple of hobby projects, and I do them because it's fun to watch my vision come alive - my dream come true. When it's somebody else's vision, that joy just isn't there. So I'd probably charge the same as my job. It's the same thing. Deadlines, changing requirements, dissatisfied customers, etc. So, I don't think you could get a discount because it's the Lynx.

 

- KS

[7800fan]

Anywhere from free if you learn to code yourself and compile one. You would need to find someone with a flash cart to test your game on a real Lynx as emulator may not reveal obscure bugs.

 

If you want to make PCB to sell game rather than offer the code for people with flash cart, that can run up about a few dollars a pop plus chip and other parts needed, you solder the parts on.

[Trew]

Some good info here, thanks guys.

 

I see that the Lynx community has so much talent that's still active, and while games are slow to come... they do come! That combined with the fact that the second hand/collector market is pretty heated right now (Double Dragon and Desert Strike are going for mad money, etc), it seems there's an appetite for the games still... and people that can make them. But the two worlds seem disconnected. Home brewers do their best, indie game companies are dividing their time as best they can etc etc.

 

It feel like all the parts are there (ability, ideas, demand), but not formerly organised. I wondered whether crowdfunding a game might mean we can either create something (even if it takes a year+), or... for it to pay to finish off some of the more promising ongoing games.

 

As Kskunk mentions... a derivative idea... "Straight Fighter 2" or "Fatal Kombat" or.. whatever... wouldn't that be great?

[Thorsten Günther]

Low-fidelity: 8-bit programming is a skill modern programmers don't have. Anything fancy on the Lynx requires fresh experience with bare-metal 6502 assembly - 25 years ago, it wasn't too bad to find that skill, today it's a needle in a haystack.

 

IBTD here: 6502 based machines such as the Atari 8-bit and Commodore 64 computers and even the VCS still attract several programmers today, and the resulting games show a lot of prowess.

 

The cost in money to write a program for the Lynx is negligible (buying one FlashCard to test on real hardware), but programming on the Lynx requires studying the graphics and sound hardware and preferrably already having the knowledge to write 6502 assembly, meaning the cost in man-hours may be quite high depending on the complexity of the project, making this platform infeasible for any but hobbyist projects (which - like Zaku - may very well have the quality of commercial ones).

 

Publishing is another issue, but this has been nearly perfected during the last years, usually Lynxman doing the card hardware and some other guys producing the package and manual (Zaku has a professionally made Lynx card, but the rest of them are PCBs with SMD part soldered to them, so these won't fit into the Lynx 1 very good).

[kskunk]

IBTD here: 6502 based machines such as the Atari 8-bit and Commodore 64 computers and even the VCS still attract several programmers today, and the resulting games show a lot of prowess.

 

Yes, you're right. I misspoke. I was trying to say that contract programmers, who you can easily hire full time, are rarely versed in 6502. There are plenty of 6502 programmers who do it in their free time, but they either have full time jobs or are retired. That's why these projects tend to take a long time. If you want to hire someone to do 6502 full time, I still think it would be rough. Maybe I'm wrong?

[Ninjabba]

Don't want to crush dreams here, but its very unrealistic to expect a "complex" game to be done in a year, even with full-time employment of 1 developer. Judging from past projects and personal experience, we're lucky to see simple puzzle games within a year of development.

 

Like Willard also mentioned, from a market perspective its not attractive to spend so much time on a device with many limitations as it requires a skillset that is rarely asked for as of today. I'm now advanced at Lynx development, which is something fun to talk about during drinks but not something that'll get me hired for the next job. If I could convert all the time I spent devving on the Lynx into Android development, I'd be a senior Android dev, which is a lot more valuable in 2015.

 

Also, whoever understands the platform to develop a full-featured game most likely has a well-paid job already (I know I have), and I doubt that one person is willing to pay a full year at the same rate for a Lynx game. Unless that person is filthy rich and is thinking to either buy a new sports car, say a Jaguar, or have a Lynx game developed. At that point anything is possible I guess..

 

The choices in life we make :')

 

Edit: just read about your crowdfunding idea. I been thinking about this myself, because indeed if I had some cash stashed aside to pay the bills, I probably could've cut down a year of development on my latest project. But by doing so, you'd have to have a proto-type to showcase you're actually doing something already. It also creates pressure to deliver a certain quality that isn't required for homebrew. I've been struggling with music/sfx for a whole year, and it was really making me depressed. Add expectations of on-time delivery on top of that and I might not sleep at night

Edited March 24, 2015 by Ninjabba

[+LS650]

To be fair, I'm surprised and thankful that anyone bothers to write homebrews for the old 8-bit systems. If you have the skillset to write a good retro game, why write something that will only sell perhaps 200 cartridges, when you can develop for tablet/smartphone and have a potential market of millions of people?

[Trew]

I guess my cunning plan to get people to come out, unite and decide to build and fund a new mega game didn't work... damnit.

[TailChao]

$150,000 upfront. Two year development time, full creative control.

That would be fine.

[Trew]

1000 buyers at $65 funded through kickstarter.... sorta kinda half-way there, right?

[108 Stars]

It's a fun idea to actually have a game developed under commercial viewpoint; it could perhaps be done when the profit expectations are zero and the team very small. A coder, a gfx guy and a musician, the latter two not being paid the entire project but just for a fixed price for a predefined set of work that the coder has to work with then.

 

IIRC, a frame of animation costs about 10$ if you are cheap. Like me.

 

Also it would have to be someone who does not have a well-paid job; and finding someone who does not, but is good at ancient coding is probably tough. You won't find young people working with the hardware in their bedroom who are happy with rather low wages.

 

 

In the end though, I think chances are best to just keep encouraging developers to take greater challenges with each project. I have two dream projects where I am working on design documents, and hopefully, a homebrewer will feel the motivation to pick one or both up eventually. Things like Zaku, Hack, Slash, Loot and Wyvern Tales are pretty much commercial grade anyway. It's the fun of doing it that leads to great results even with little to no budget.

[+karri]

The thing is that if you have a crystal clear specs of what the program has to do then just coding it takes a week.

 

If you start from scratch writing a car-driving game it is probably done within a week.

 

But if you have no clue of what goes into the game you have to be creative. And that takes time.

 

Even a simple things like: Make the characters say funny things would tie you up for a month - at least.

[GadgetUK]

Absolutely! The creative part is the bit that takes 90% of the time. It's also common to spend 90% of the programming time on 10% of the code lol - those complex bits or bugs etc.

[Turbo Laser Lynx]

I liked Karris idea (a few years back) of having a kind of Lynx IDE that simplifies the development and in a way adds another "abstraction layer" as this would draw in more tinkerers and artists that would be able to create games, which obviously would result in more Lynx games for everyone. As discussed earlier in this thread programming anything more than small demos on the Lynx requires quite a high "engineering IQ" and deep understanding of assembler and/or C programming and the Lynx hardware. Personally I had a really great time creating minigames, pretty much only caring about the main game loop and creating som music with abc-music, but not being able to move beyond that (sadly) forced me to move on creating games in more modern environments (which is easy when you don't have to worry or understand the underlying technology as much). However, always keeping the lynx-game-creation-dream alive. ^^

[7800fan]

Or we could see if someone can make Batari-esque system for Lynx and let us code in BASIC equivalent.

[Shawn Jefferson]

On the Atari 8-bit there is a club called ABBUC out of Germany that holds annual software contests, with cash prizes. This seems to really help spur on homebrew developers and several great games and other programs have been created specifically for the contest.

 

Maybe a similar thing can be done here? We'd need someone "trusted" to keep the money (maybe Albert?), and have a well defined rule structure and voting system.

 

Just brainstorming:

- all games would have to be released in electronic (or physical) format to the "membership" (those that have paid into the prize fund)

- all rights remain with the author

- a public release can be done by the author if they so choose (after fulfilling rule #1 above.)

- the game cannot be released in any form prior to the contest

 

Things like that. The ABBUC rules may be a good starting place.

Edited March 25, 2015 by Shawn Jefferson

[+karri]

That is a nice idea. Personally I liked the 7DRL style contest with limited time to work on the game. This format is a great "sandbox" for testing some special feature of the Lynx. In "Shaken, not stirred" I wanted to test how to lighten up part of the screen. A kind of XOR cursor.

 

 

You can see that the background is lighter at the crosshair.

 

But the theme could be broader.

 

I also don't want people to code everything from scratch. Using pre-made chipper music should be ok as well as using complete engines like 3D maze drawing or 3D vector engine.

 

I was thinking for a while. Your entry can also be creative hacking of existing titles like new levels for an existing game or even enhanced modes to existing games. But of course you must tell what your starting point was so that the voters can give credit to your hack and not the entire game.

 

So the rules could be along the lines:

- please plan what you will do before the contest starts

- coding has to be done within one week starting UTC xxxx and ending UTC xxxx

- you may team up

- every entry needs a blog page with description of the game and screenshots. The blog page is your "announcement to the competition" and is put up before coding start. After the competition ends you update the blog. (If you are over-powered you may also update the blog instead of coding during the competition)

 

Just my idea of the rules...

 

PS. If I get my 512k + 2k cart off the ground then we could make a cart release for voting. A limited run made with "no profit". Most likely the entries will be small enough to squeeze them all in one cart. BLL and cc65 is ok at least. Perhaps other compilers as well.

Edited March 25, 2015 by karri

[Trew]

See! This is much more like it, we're gaining traction now!

 

Not sure how I could contribute, but my day job is writing... so if we wanted a pretty blog (with sub pages for each entry) managing? I do music production too, but never for chip/SID sorta stuff (just general MIDI). But if there was a comp cart, would poss be able to help with audio bits (others may be more qualified).

[Turbo Laser Lynx]

I read in one thread that there might be a midi to chipper implementation in the works, that would be too good to be true really. Or any way to convert midi to lynx music.

[obschan]

$150,000 upfront. Two year development time, full creative control.

That would be fine.

 

1 year and I give the creative control to who wants it !

 

II have two dream projects where I am working on design documents

 

Ok, now you'll have to make your own post to give more information !