Ancient Frog (App Store link) is a colourful and vibrant puzzle game for the iPhone and iPod Touch with James Brown as the one-man team behind it. Now, James shares his valuable experience with the development process of Ancient Frog and the adventures of selling a game on the highly competitive App Store. In this interview, among other answers, he discusses porting the game, the difficulty in pricing an app, and regrets on handling the perceived price-drop syndrome applications face.
No DPad: For those unfamiliar with Ancient Frog, can you give us an idea of how the idea of Ancient Frog began and what this game is about?
James Brown: Ancient Frog is a puzzle game. You have a small grid of points, and an articulated frog which you have to move from its start position to an end position. The constraint is that it must always have at least three feet resting on grid points. It’s a very simple mechanic, but it gives rise to a surprisingly wide range of puzzles.
The idea arose from a few different interests, all of which had been kicking around in my mind for a long time before they came together in one idea. The closest ancestor to Frog would be the plank puzzle – a game where you have to cross a grid of points using planks of different lengths. I’d got really into an online version of that several years ago, and sort of idly thought it would be cool to be able to play it on my phone.
Later, I worked on a game which required a huge amount of character animation. This was all done by hiring a large team of animators to sit there churning out thousands of individual hand-crafted animations on a production line basis. It set me thinking about ways of automating the process, systems for creating realistic animation automatically. Doing so for that game would have been a very difficult undertaking, but it did start me thinking about games where the animation requirements were limited to something achievable by procedural systems.
The third piece of the puzzle was again sparked by a problem I encountered working in the industry. The company I was at had started work on a Tomb Raider clone. Part of this involved your character climbing a rock face, and to accomplish this we were going to define a cloud of points on the rock geometry indicating to the animation system where the character could find a foot- or hand- hold.
The initial concept which brought these factors together was a game where you had to guide a gecko up a wall to reach its dinner. You’d control it with a joystick, and it was more about walking to an area where there was enough room and enough purchase to make a turn, then walking to the next goal. This turned out to be a bit too loose to make a decent puzzle from, so I reduced the number of possible foot holds, and made foot movement an explicit act (dragging with the mouse) rather than a procedural response to the inputs. The gecko became a frog because having legs of different lengths would open up the puzzle space some more. After that, it was a matter of repeatedly prototyping and testing, playing with the parameters until I had a combination of frog skeleton and grid size that worked. It had to allow enough variety to make puzzles in the narrow realm between ‘always trivial’ and ‘always impossible’.
The end result is a puzzle game which is rather looser than the plank puzzle – on later levels it would be hard to plan the entire route in your head – but tighter than the original drive-a-gecko game. As you work your way through the levels, you learn what patterns to be looking for, and can aim to get your frog in a position to bridge them.
ND: You mentioned on your blog that Ancient Frog started out being developed for the Mac and PC as Frog. With Frog already running on PC, what was the transition to the iPhone like? What was the most difficult issue you faced?
JB: In fact the very first version ran on my mobile phone – at the time, a Windows Mobile device. It always seemed the sort of game that I’d want to play as a bus-stop time-filler rather than something to sit down in front of. But there are so many different hardware configurations, most of them rather underpowered, so I decided to hit the PC first and convert back down if it looked like it was going to work as a game.
Because I run a PC as my main desktop system, and a Mac as my laptop, I necessarily write my games to be cross-platform (not least so I can keep working on them wherever I happen to be). Once you have code that works on two systems, adding another turns out to be fairly easy.
The one problem I encountered actually turned out to my benefit. I was using SDL (an open-source cross-platform layer) to handle my platform-specific code. I just needed to wait for that to become available on the iPhone and I could port the Frog code across. But this was back in the days when Apple had an NDA in place preventing anyone from sharing or discussing code, so nothing was happening on the SDL front. Eventually I decided that enough was enough, and removed it from my code altogether, instead writing my own interface layer. It was less work than I had anticipated, and has the advantage that I’ve now removed an external dependency, and have more control over the low level side of things.
The hardest issue I faced was simply making the game work on a small screen.
The resolution of the iPhone is pretty decent in terms of pixels, but there’s a physical scale limit on interactive items imposed by the size of the human finger.
This set an absolute limit on the size of grid I could fit on the screen (scrolling would have been clumsy), and at the new scale, a lot of my puzzles didn’t fit and had to be discarded.
The interface also had to change, removing as much clutter as possible. As is so often the case, working to tight constraints considerably improved the design. Folding all of the buttons and onscreen information into the daisy made the layout cleaner, simpler and easier to understand.
ND: Besides the benefit of the iPhone having a fixed resolution, what other advantages have you found with developing Ancient Frog for the iPhone and iPod Touch?
JB: The fact that it’s a single hardware implementation is the best thing about it from a development point of view. This surely won’t last—doubtless Apple already has variations in its labs with different sizes of screen, different processors, different GPUs. But for the moment, if you target the iPhone you don’t have to worry about either aiming high and losing older platforms, or aiming low and not being able to do all you want. You also know that once you’ve tried it out on the phone in your hand, it’s going to play exactly the same for everybody else.
So it has all the advantages of a console, but without the restrictive publishing model.
ND: You have commented that the effect of having Ancient Frog featured on the App Store front page was a ten-time increase in sales. Has being featured given a sustained increase in sales, or would you say being featured only serves as a temporary sales boost?
JB: It’s quite hard to tell definitively because there are so many factors that get overlaid – reviews appearing, word of mouth spreading, and the various different charts and ‘what’s hot’ sections in the various different territories. But certainly the day it made the US app store front page, sales jumped straight up by (more or less) a factor of ten. They then stayed pretty steady there for the week it was featured, after which there was a step function back down again.
It didn’t go all the way down to where it started – it came off the front page, but remained in ‘what’s hot’, and has been gradually blinking out of the easy-to-find places on the app store. It’s still only a few clicks down from the front page, and that’s presumably keeping sales above the level they were before featuring. But I still haven’t really got a handle on how well it’s going to sell over the long term. You don’t get the sort of stats from Apple that you do if you’re selling directly from your own website, so there’s a lot of guesswork involved in analyzing sales performance.
ND: Ancient Frog launched with a price tag of $4.99 in February and has not gone on sale. How did you decide on the original price? Have you considered putting Ancient Frog on sale to see the effect it may have? In retrospect, would you launch at a different price?
JB: I believe it makes more sense to put the price up than down when sales tail off.
It’s a good few weeks of idle distraction for the price of a cup of tea and a biscuit – once you’re no longer reaching the impulse-buy audience, there’s less mileage in paring the cost right down.
One thing I regret is not explicitly saying “special introductory price” in the blurb about it from the start. I think a lot of people were expecting a price drop, and holding off buying until then. I wouldn’t rule out a temporary sale at some point – perhaps after 6 months or a year – but I don’t really have the marketing clout to make the sort of song and dance about special offers that they need to be effective.
Pricing is a notoriously tricky subject on the App Store. There’s such a focus on 99 cent applications, and the simple fact is that most of them don’t make their money back. If you’re one of the big studios you can afford to price low and give a big enough marketing push to sell the volume to make a profit. Otherwise it’s a matter of knocking something out quickly enough that you can afford to sell at 99 cents.
Ancient Frog was a very labour intensive game. It took a couple of weeks to get a stick-figure frog that you could drag around, but it took months to make the movement fluid and naturalistic, to make it so that you couldn’t drag it into positions where it would lock or hit an awkward pose, to find the sweet spot for snapping to pegs, snapping to the finger, and so on. That’s on top of the time spent designing levels, ensuring that there really is a solution with no short cut, that the solution is possible to find by pattern matching rather than trial and error. Then there was tramping around in various lovely outdoor locations shooting textures, or gathering materials to shoot in the studio (most of the artwork comes from my own photographs, with only a couple of stock images). It’s a truism that the last 10% of the work takes 90% of the time. Every little polish effect – every tweaking of the speed of the bounce of a menu transition – takes a disproportionate amount of the development time.
The 99 cent version of Frog would have been the stick-figure version.
In the end, I looked around at games that I thought were in the same league in terms of production values, and it seemed to me that there was a tier at the $4.99 mark that might be sustainable. That’s a lot lower than I expected to be selling for when I started work on the game, although the convenience of paying through the App Store helps make up for it a bit in volume. It remains to be seen whether it makes its money back at this level.
ND: Do you have any plans for future iPhone games?
JB: I’m taking a bit of a break first, updating Ancient Frog as I get feedback from players, and trying to get the word out about it. I’m mulling over ideas for new iPhone games, but it’s a matter of coming up with something simple enough and quick enough that I can still make a living selling it for under $5. Financially the most sensible thing for me to do would be to finish the Mac / PC / Windows Mobile / Android versions of Ancient Frog so I’m spreading the development costs among all of them, but really I’d be happier moving on to the next shiny new thing.
Thank you, James! Your time is truly appreciated. I’m sure developers and gamers alike have enjoyed a behind-the-scenes look at what happens with a game on the App Store.
Ancient Frog comes with 100 levels and looks beautiful; we here at No DPad highly recommend it!