So, we are currently looking very seriously at a console release for Defender's Quest II. Of course, PC is our top priority, and don't take this as an official statement of specific platform choices or anything just yet, we still have a lot of details to work out.
In any case, if we want to release DQII on consoles, wer'e almost certainly going to have to release the original DQI game there as well. I've taken special care to make sure the new DQII engine is backwards-compatible with the original game data, which has been very useful in testing basic functionality while we continue to work on DQII game content.
Of course, if we release DQI on consoles, we're going to be expected to support HD resolutions, and we don't have the usual "just scale it up" options available to truly low-res pixel-art games. The NES/SNES topped out around 240 vertical pixels, and the Playstation 1 was 480, if I recall correctly.
Defender's Quest has 600 vertical pixels, and all the art is designed around that. It's pixel art, but it's big pixel art, and it gives us some awkward problems to design around with regard to screen sizes.
Here's Shovel Knight at 427x240:
That easily upscales to various HD resolutions. At 480p you've already hit 2.0x, 720p gives you a nice clean 3.0x, and according to their article they decided to go all the way to 1080p with 4.5x.
Even though the pixel boundaries are technically blurred at that scale, it's barely even noticeable since you've got 4 full-scale screen pixels for each game pixel, and only a tiny smudge on the edge.
By contrast, here's Defender's Quest: Valley of the Forgotten at 800x600:
This doesn't give us a whole lot of options. At 600 vertical pixels, the next clean jump to a whole-number upscale is 2x and requires 1200 vertical pixels -- that's higher than 1080p! And if we settle for 1.8x, almost every on-screen pixel will be hideously smudged.
Our only option for doing a clean upscale would require us to show less of the actual screen then we normally do, which might work for the overworld, but simply isn't an option in battle. One choice is to just slap on some Black Bars, target 720p, and call it a day. But I wonder if there's another solution.
We had initially toyed with the idea of an HD re-release using all-new art like this:
But after doing mockups with that test art we figured it would be overkill, as we'd much rather spend our art dollars on our new game rather than touching up the old one.
I came up with another option is a bit of a middle-ground solution that could have a good bang-for-buck tradeoff. I've worked out a specific pipeline for upgrading the pixel art to HD, and I'm wondering if anyone has any feedback on the proposed method.
Here's a quick before-and-after comparison:
Here's the pipeline:
1. The original art looks like this:
2. Let's blow that up by 3x to get a better look:
3. We use the xBR algorithm (3x, unantialiased) on the original:
At this point we would hand the asset over to an artist for a little bit of manual cleanup. To demonstrate, I did a little cleanup myself. It's not a great job but it's just for demonstration's sake.
4. Clean up the asset manually:
5. Image that shows which pixels were changed during cleanup:
6. The game uses the asset at a maximum size of 2x the original resolution, so here's how it looks downscaled with bicubic filtering in its final ready-to-use form:
This would be the maximum size the asset appears at in-game, and would be downsampled at run-time from there depending on the user's target resolution. This lets us natively target 720p, 1080p, and anything in between without any major issues.
So the question is -- what do you think of the pixel-art --> HD pipeline I've just described? It would be much cheaper and faster than doing new art from scratch, but it only makes sense to do if enough people like the look.
And of course, if we did something like this we would keep the original pixel art with or without letterboxing, it'd be a setting to flip in the options menu.
Here's a few mockups of the overworld to compare:
To be clear, we're still moving forward full-steam on DQII, there's just a lot of overlap between what I'm doing on a daily basis with the DQII engine and what could be used for a DQI re-release.
DQII will only ship once BOTH the new engine is finished AND DQII's game content (story, characters, art, level design, balancing, testing, etc) is done. However, an HD upgrade for DQI could ship pretty much as soon as the new engine is finished as with a few small tweaks the content is already done.