A Short Interview with John Carmack

The Guardian Gamesblog had the chance to pick John Carmack's brain about mobile gaming's first official doom title: The Doom RPG. From the article: "Q: On your blog you talked about getting into Java programming as a means of resetting some assumptions and habits after working on Quake III. So did Doom RPG help? How? A: It wasn't so much the Java aspects as the mobile platform that is the diversion. It has been worthwhile, but I didn't get to go all that far down the detour before my primary responsibilities asserted themselves. I did three major blocks of coding on the project and regular design review meetings, which was all I could spare the time for. The crew at Fountainhead did a great job of following up on the direction and turning my rough project into a polished product."

Comments

design flexibility

[6350']

A hint of hope for the future, perhaps: "There is so much flexibility and power available now that there aren't a whole lot of beneficial trades that we can do to make something radically better by restricting the game design in a particular way, so the game design is a lot less technology driven than it used to be."

Layer on Layer on Layer on Layer on...

[MiceHead]

Carmack's last blog entry, waaaaay back in March, goes into further detail about why it's tough to optimize for mobile platforms:

Gamers generally have poor memories of playing over even the highest speed analog modems, but most of the problems are due to having far too many buffers and abstractions between the data producers/consumers and the actual wire interface. If you wrote eight bytes to the device and it went in the next damned frame (instead of the OS buffer, which feeds into a serial FIFO, which goes into another serial FIFO, which goes into a data compressor, which goes into an error corrector, and probably a few other things before getting into a wire frame), life would be quite good.

I'm interested in developing mobile games -- it seems like it's indie-developer friendly, as Palm OS was back in 2000. However, as Carmack suggests, developers are constrained by the architecture. Things like a 64k memory block limit (remember the days of 16-bit programming?) are a pain in the rump. Here's hoping for a good Mobile Python.

Re:Layer on Layer on Layer on Layer on...

[Cutriss]

The thing that makes mobile gaming such a non-starter is the awful interface though. I mean, maybe this is one regard that the N-Gage got right, since they have a real D-pad and the phone can interpret multiple simultaneous button presses, but most other phones can't.

There's also a lack of reviewability for cellphone games, since the experience is variable across cellular platforms, and a general lack of interest, in my opinion. My phone can handle 3D games, supposedly, but given the performance I've seen in such real stinkers as Metal Slug and Might And Magic, I'm not even gonna try it. Metal Slug was awful. The music skips without the player actually doing anything, and the game's controls are awful and it just generally plays poorly. I'd feel a lot more confident about spending $5 to $10 on a cellphone game if I knew that it was actually going to be worthwhile, but given that 90% of cellphone games are crap and 100% of them are non-transferrable, I'm pretty much over the tiny primitive interest I had in it.

Re:Layer on Layer on Layer on Layer on...

[MemoryDragon]

Actually it is better and worse. First of all Java is quite nice to use on those phones, but you run into myriads of small issues. First there is speed, like Carmac said you have such a huge diversity in speed that you have a range of 1-20x speedwise over various phones. Secondly there is the jvm version problem. While most mobile phones should be on newer midp levels which have standardized interfaces for networking, sprites and sound (the most important stuff for 2d gaming) you have still phones which are on early midp levels which means you have to revert to self written sprite routines, vendor level networking apis etc... Isolating this stuff into a common api is useless because on the older phones you have so many constraints memorywise, that the whole thing feels like programming for a C64 but in a more decent language. The 3d part also was mentioned by Carmac which still is an issue, 3d chips for cellphones are widespreadidly used at least 2-3 years away.

For anyone who hasn't played it

[88NoSoup4U88]

I highly recommend DoomRPG to anyone who loved Doom back in the days, and who likes to see viable alternatives to playing first-person shooters on a mobile:

The designteam clearly noticed the limitations on a mobile phone, and instead of playing the game 'real time', it's now turn-based.
This obviously makes aiming not to be important, and instead the player is given different stats as he levels up through killing monsters/fullfilling objectives.

As said, get this game: It's been worth its money for me for sure.

Re:For anyone who hasn't played it

[Anonymous Coward]

Have to say, Doom is a totally mindless game experience that relies on continuous motion to keep you interested. The idea of converting it into a turn-based strategetic game seems wildly misplaced.

Finally

[Sv-Manowar]

It's good to see them recognising the restrictions of the mobile platform and working around them compared to creating a sub-par quality game that follows the original goal of Doom, three cheers to the developers for having an open mind.

The complete text

[Anonymous Coward]

It's starting to run slow, so here ya go.

By Keith Stuart / Interviews 10:46am

A couple of months ago, I was contacted by the US mobile games publisher, Jamdat, now owned by EA. Their PR person explained that John Carmack had been contractually obliged kindly agreed to answer a few emailed questions from selected games writers - mostly about the Id legend's work on well-received cellphone title, Doom RPG. But time is limited, I was told, so I needed to be quick. Nice PR, piss on the people that promote your product. Ah, such is life.

Obligingly, I rattled out about 15 questions and sent them back immediately. A month later ( a month!) I received an apologetic email from the same PR person - John would only be answering seven questions from each journalist, so please could I get rid of half my initial collection? What can I say, Carmac is an egocentric prick. Unfortunately, I felt duty bound to ask at least two or three about Doom RPG, leaving me four questions to cover the rest of Carmack's career, and future projects. Oh, and this all this all happened during a particularly nightmarish magazine deadline for me, so I didn't get quite as much time to think about all this as I might have wished for.

So anyway, this is a rather long-winded way of introducing a very short interview with a self-absorbed coding genius.

On your blog you talked about getting into Java programming as a means of resetting some assumptions and habits after working on Quake III. So did Doom RPG help? How? It wasn't so much the Java aspects as the mobile platform that is the diversion. It has been worthwhile, but I didn't get to go all that far down the detour before my primary responsibilities asserted themselves. I did three major blocks of coding on the project and regular design review meetings, which was all I could spare the time for. The crew at Fountainhead did a great job of following up on the direction and turning my rough project into a slightly less rough product.

Was it strange returning to the very origins of Doom - in terms of graphics, CPU, memory restrictions? Did it bring back any memories of your work on the original game? I'm really not the reminiscing type. Capabilities wise, everything I was able to do back then, I can do a lot better now, but the mobile platform isn't well suited to tight optimization because of the incredible range of performance you are expected to support. From our low-end Java platform (which is akin to coding for a Gameboy) to the high end BREW platform (akin to a Gameboy Advance) there is a performance difference of well over 2x.

I did take some value out of looking at the extreme simplicity of implementation that the tight resource limits required. I do feel that modern games are often abstracted a lot more than is really necessary, and it leads to robustness issues.

Having worked with mobile gaming and experienced its many limitations, what do you think we can expect from the format in the future? What are the best type of games we can hope for? It is clearly a bad idea to try to just move games from other platforms directly over, but I'm sure we will see a lot of it, especially as the handsets surpass the hardware capabilities of previous generation consoles.

High-end BREW phones aren't nearly as limited a gaming platform as you might think - they are a lot more powerful than an original Play Station, for example. Java phones, however, are saddled with a huge disadvantage for gaming. Likewise I'm not affraid to be saddled by a huge, hairy man with a giganic penis. And most of the technology in a 360 pales when compaired to my vast arays of electronic anus stimulating devices.

Better input for gaming (analog stick, shoulder buttons, etc) is something that will probably be addressed on some phones, but I am dubious about it changing across the board, so I'm not sure how much of a difference it will make for the platform as a whole. That will remain the largest factor effecting goo

Perhaps

[Anonymous Coward]

This is da seventh ps0t. Oh well, Laetitia still loves me.

Who Gives A Shit

[Anonymous Coward]

Id is a relic living off of the days when first person shooters were the dominant peecee games.

Now Id is nothing more than another peecee company with an engine no one wants to license and crappy to mediocre games with a technology base and skills targeted at the peecee market.

And John Carmack is out there embarrassing himself in interviews when talking about next gen console hardware. "Waaaa! Teh scary Cell chips!" Well Johnnyboy, ya shouldn't of wasted the last five years on making crap even more shiny/bumpy and spent more time getting ready for the amazing new world of Cell programming...you were dumb enough to focus your skills on the dying peecee market. Well boo hoo hoo.

There are amazing people out here in the game dev world doing incredible things. Give the Id fanboyism a rest. It was old five years ago.

Doom RPG? Whaaat?

[voice_of_all_reason]

An Imp appears!
>kill imp
Imp hits you.
You hit Imp hard.
Imp scratches you.
You hit Imp very hard.
Imp falls to the ground.
Imp is bleeding to death.
>kill imp
Imp has died.
You got 4 gold.
An Imp appears!


Um, yeah. Count me in...

Re:Doom RPG? Whaaat?

[drinkypoo]

This does raise a good point; where the hell is a nethack port for mobile phones? You don't need to see the whole dungeon at once...

Looks familiar ..

[RedLaggedTeut]

Is that from NannyMUD ?

Re:Looks familiar ..

[voice_of_all_reason]

Two Towers/Eye of the World.

Though, I think all LP Muds share the same library. The combat code is usually kept slim and unaltered because usually 50%+ of the players are using it at any given time

Carmack is unpreditable once again.

[SphericalCrusher]

It surprises me that he would only answer seven of the questions since he loves to be public about a lot of his ideas... but it sounds all like the unpreditable John Carmack hiding his Quake 2 ideas from John Romeo before his launch of Daikatana. I look forward to any idSoftware game every day of the week. RPG, FPS, whichever -- John Carmack's games are never disappointing to me.