February 13, 2012 5:00 PM | John Polson
PSN developers will have a leg-up when porting existing code and assets to the PlayStation Vita, shares Chris Harvey of Drinkbox Studios. He further suggests that current generation devs won't have to sacrifice much to get their game up and running, describing the Vita as a "mini-PS3."
Harvey's team is currently working on Tales from Space: Mutant Blobs Attack!!!, one of 25 titles scheduled for the Vita's North American launch. This is far from the team's first console effort, having been certified to develop across leading gaming platforms and having worked for studios such as Vicarious Visions and Electronic Arts. Drinkbox last year went solo with its PSN release of About A Blob, with the help of Sony's Pub Fund. The team's Mutant Blobs Attack follow up, however, is without Pub Fund assistance.
In the following interview, Harvey shares Drinkbox's experience with porting onto the Vita, its development as compared to other platforms, and the benefits found in using the Vita GPU's Tile Based Deferred Rendering (TBDR). Harvey also shares what he thinks could make more successful North American and European launches.
PS Vita has a lot of user input possibilities. What can you share about integrating these with your game?
Our main objective with the inputs has been to find control updates and additions that enhance core gameplay. Our general process for experimentation has been:
Step 1 - Brainstorm control change ideas, implement draft versions of our favorite ideas.
Step 2 - Watch people try the game once control changes have gone in. Think about the results, go back to Step 1.
Mutant Blobs Attack still plays like a platformer, with thumbstick-and-button-based core movement controls, but we've been finding that touch controls in particular can be worked into the gameplay in pretty interesting ways. We're also still experimenting.
You've ported onto many different platforms. How do other platforms compare to working on Vita?
Managing shaders and VRAM is similar to the PS3, although simpler. Because the CPU is a symmetric multi-core device, the threading process is similar to the PC or Xbox 360. The basic porting took about 2 weeks. At that point the original game was playable on the device, although it ran too slowly. Since then, we've probably spent another 4-plus weeks on performance. We've had to get a lot more aggressive with how much threading we do, like threading the input-device reading.
Fortunately, these changes have benefited the engine for all the other platforms. The Vita's API hasn't changed too much since we started, and the documentation was been pretty good from the start, so I think the porting process was on the easy side. Keep in mind that our engine had already been ported a couple of times, so we were in a good position to port to Vita. If you haven't ported your engine before, then it might not be quite as easy.
What benefits has the GPU's Tile Based Deferred Rendering (TBDR) offered you?
Most noticeably anti-aliasing is inexpensive, similar to the Xbox 360. But like the X360 you can't render part of a scene, do some other rendering, and then come back to finish on the previous render target. Fortunately our render pipeline already accounted for this since the engine runs on X360. If your engine hasn't been designed around this setup then the switchover can be a pain, but fast anti-aliasing is certainly worth it.
How much, if at all, easier has Vita development been compared to your PSN/PS3 experience?
Much easier. I think Sony has made a real effort to improve their development pipeline and simplify the hardware. While setting up a PS3 for the first time can be a bit of a pain, the PC development environment is much better on the Vita, and is similar to the X360 approach. The GPU and CPU profiling tools are coming along nicely and are already very useful.
Overall, it's a much cleaner process. As with any pre-release hardware, there are always problems to deal with, but compared to other pre-release devices we've worked on I think it stacks up quite well.
What have you learned about porting About A Blob's PSN code and assets to Vita?
As I mentioned, the porting was pretty straightforward. Most of the port time was spent on graphics, and then later on thread-syncing and thread-core assignment.
What I dislike most about porting to a new device is when you have to make a lot of device-specific changes to the engine to compensate for a very unique hardware-architecture or hardware feature set. For a small team like ours, doing port work that only benefits one device is not very desirable. With the Vita there wasn't much of that, and many of the changes we made simply benefited the engine overall.
Regarding assets, the Vita's display resolution is roughly half of 1080p, and its video memory is about half of what's on the PS3, so everything works out nicely if all your textures are half the size on Vita vs PS3. This fact made our art process really easy, because you can just skip mipmap 0 when loading any texture and use all the remaining mips on the Vita, thereby using less than half the texture memory. That approach isn't going to work for every texture, but for most of our content it's actually been fine, so most of the content is still authored as it was for PS3.
What else would you tell developers about working with Vita?
If I didn't know much yet about developing on Vita, one question I would ask is, "What will I have to sacrifice anything to get my PC/PS3/X360 game running on the system?"
Because the GPU allows for shadow maps and post FX, multiple render buffers, etc. it has felt a lot like a mini-PS3 to us rather than a completely different mobile device. There are occasions where we've had to hold back a bit in terms of quantity of data being pushed through (e.g. vert counts), but we haven't really had to make any significant feature cuts. But of course that really depends on the kind of game you're working on and its scope.
By the way, what do you think about the Japanese sales so far?
Regarding Japanese sales, obviously it's been disappointing so far. The 3DS also had a rocky start, but seems to have turned around based on a price drop and the big-name games that came out this fall. The big-name game for Vita launch is Uncharted, which feels very North American, so I wonder if Sony are missing a killer-app for the Japanese audience. PSP seems to be pretty big in Japan right now, apparently on the basis of Monster Hunter (which I've personally never played). I think that if the price were a little lower and Monster Hunter had been available on day-one things might have gone better.
What do you think is needed for stronger North American and European launches?
The first thing needed is general awareness. I haven't seen a lot of advertisement where I live downtown, but I hope it's going to ramp up soon. I remember that when the PS2 launched, even though the Dreamcast was already out and had a bunch of great games the anticipation for the PS2 was so high that it didn't make any difference. To a certain degree perception is reality, so I hope the launch marketing is able to drum up excitement.
The second thing is that Sony needs to make a clear argument for why people with the iPhone or 3DS still need to get a Vita. I think they can make a strong case here, since the Vita's power, screen size, and dual analog sticks make it possible to have great action games on the system. I don't own Street Fighter IV for my 3DS because, fairly or unfairly, it just doesn't seem to me like the experience would be great. I will, on the other hand, probably pick up Marvel vs. Capcom 3 for Vita.
Third, of course, is the games. The Vita may be a great piece of hardware, but in order to get someone like me to buy it when I have all the other systems already, I have to feel like it has multiple games I can't afford to miss. Uncharted for Vita is a good start, but if there was an original God of War available, and Resistance Vita and Little Big Planet Vita were launching on day-one, I think it would be really hard for me to tell myself to wait before buying it.
Obviously, what I'm suggesting here isn't easy, but I think Sony is in a good position to deliver on this in the next year or so. It seems like over the PS3 generation they've built up a stable of really solid first-party studios that put out consistently great games, so like Nintendo they don't have to rely as much on third-party studios to help to make their system strong.