Thread by @SebAaltonen

Recently a book was released about Nokia's N-Gage handheld game console. Back then Nokia was one of the most valuable companies in EU. They spent lots of money to conquer the gaming market.

I was the lead programmer in their flagship game.

Thread...

hs.fi/taide/art-2000…

The book realistically covers how crazy the spending of the money was. Nokia spent massive amount of money into marketing. They wanted to look like a gaming company. Big stands at GDC and other places. Massive launch parties, lots of press invited, etc, etc.

Their first problem was the hardware itself. N-Gage was identical hardware to their camera phone, except the camera was gone. Display was still vertical and very small. There was no GPU and no floating point unit in the CPU either.

The only SDK we got was just two functions: One to get the start memory address for display memory for direct write. Symbian gfx APIs were dead slow. The second one was for reading multiple key presses so that we could read diagonal (up+down) on dpad. That's it.

So the hardware wasn't really suited for gaming. Competitors all had GPUs back then. Nintendo DS also lacked FPU. PSP didn't, and it had a great GPU too. And competitors had bigger screens in landscape orientation. Much better designed hardware for gaming.

The first N-Gage version had MMC memory card slot under the battery. You had to take off the back lid and take off the battery to change your game. They didn't even bother doing hardware modification to add hotswap memory card slot. It was just a default phone hardware.

Their lead game Pathway to Glory (our game) shipped with full network capabilities. Nokia bought SNAP online library from SEGA. SEGA just quit their own console business and was licensing their online tech to GameCube/PS2 game developers. Nokia bought it fully.

This was actually a very good move from Nokia. SEGA knew what they were doing and they got their "Xbox Live" library just in time. It worked great. It was fun to be the first ones to play online games on a handheld device. It was magical...

But it was way too early. GPRS latency was almost 1 second (in average case). Unlimited GRPS data plan was +50€/month extra on your phone bill in Finland. Only rich people could afford that. And the phone didn't even have an internet browser.

I wrote our client side network code. We were among the first one to ship rollback netcode. This kind of netcode has become popular in fighting games recently. This was required to hide the 1 second network latency. We also designed the game to hide the network latency very well.

I wish they could have spent some of that massive marketing and PR budget to design a proper gaming device with GPU and FPU and a bigger landscape screen. SEGA's network tech was nice and they had big market share. Could have been a different story.

As said in the book, Nokia's launch schedule for the device was overly optimistic. It was just a minor HW modification as said above, so they only gave game devs one year to develop big games on the platform. They wanted same quality as Nintendo's own games back then.

Their early hardware and firmware had bugs and we had to debug these bugs to them, which wasted a lot of time. This added stress on top of the super tight schedule. Back in the day they didn't support OTA update, so there was a lot of stress to get firmware ready for launch.

So we were crunching a lot. At one stage we worked over a month in a row without any weekends. We ended up shipping the game and got 90+ Metascore too. So it was fine. But I was so burned out that I didn't participate in the sequel. I started writing our Sony PSP tech instead.

That was my personal Nokia N-Gage story.

Nokia hired the best game producers and marketing talent from Microsoft and others and succeeded on that front. But they didn't spent enough (if any) effort on the gaming device hardware. Their HW devision didn't seem to be on board.

Just wanted to add some tidbits about the tech in the Pathway to Glory game. As I said above, we had deterministic rollback netcode. We had terrain rendering tech that was similar to Comanche and Outcast (voxel heightfield tracer)...

The pathfinding algorithm was basically a 2d version of the modern Recast algorithm. We rasterized surroundings with software triangle rasterizer for A*, Back then we did this again for different radius units, as distance fields were not very well known back then.

We had a very early version of dynamic virtual texturing too. It was projected from above to the voxel terrain. Our whole terrain was textured with decals, just like modern VT terrains. We rendered the decals to the sides of projected VT offscreen texture when the camera moved.

We basically reused my old terrain tech for the Xbox 360 Trials games. Trials Evolution (Metascore 90) was almost the first 100% virtual textured game to ship. ID-software Rage shipped one month earlier.