As most of you know, I've taken a slight break from working on my games. But I haven't stopped thinking about how cool they could become with a little more editing, a little more focus, and better optimized scripts. At some point, I'd like to return to them and finish the three I've started.
The original game, The Adventures of Powerstick Man, still holds my brain captive, as I often wonder how adventurous it could become without stumbling into the realm of overwhelm.
It's already a big game, with an involved story, plenty of side missions, and a basket of Easter eggs.
But I wonder if it could get even bigger...
I've certainly spent much of my OHR time trying.
Now, one thing I often do between updates is change the title screen. It's a weird thing to do, but I find that each update is so much larger than the last that an upgrade warrants it.
I've already created several iterations, each one to mark its transition to a bigger and more interesting experience.
Powerstick Man Title Comparison (Art).jpg (338.16 KiB) Viewed 2356 times
Still, I wonder if it could get even bigger...
and cooler...
and more bigger cooler...
Powerstick Man Reality (Title Mockup).jpg (1.44 MiB) Viewed 2356 times
Nah, probably not.
But I've been having too much fun with AI art and mockups lately not to dream of the bigger and cooler, so I thought I'd share an image of a future that could become reality someday, especially if/when the OHR's graphics overhauls are finished.
That future, unfortunately, is not today...
But maybe someday.
Anyway, work will resume on this game one of these days. My first order of business is to rewrite the scripts so that they're easier to modify. My second order of business is to remove all references to trademarks and ripped music. Some work on the former has already been done. But writing novels has slowed me down a bit.
The third order of business, of course, is to pick up where I left off in 2016. No promises about when that might happen though.
That said, stay tuned for more updates...
When they happen. Assuming they happen.
In the meantime, if anyone wants to play the 2016 edition, just ask. It won't look anything like the above screenshot.
Wow, just yesterday I was thinking to myself that it's too bad I'll never see another Powerstick Man update! While this isn't really one it is, at least, an update of what I might hope for.
What are those graphical features you're waiting on?
To be honest I always thought that the titlescreens let the game down a bit, while they certainly improved the importation wasn't great and they're not up to the quality of your later tiles/maps or even (if may risk relying on hazy memory of) your battle backdrops. But drawing tiles and drawing backdrops are completely different styles of art. I didn't know what to suggest. A different style? Well, wow, this is different. It's great, but then you'd have to make the rest of the game match :P
For fun I tried shrinking that generated titlescreen down to 320x200 (actually 320x226) and 256 colours and I'm surprised that it still looks photorealistic. Wow, I didn't think generative AI was going to be that much use at 320x200... but of course if you'd wanted pixel art rather than photorealism I think you're still out of luck.
My second order of business is to remove all references to trademarks and ripped music
Well, Pepsi... ummm ... I'm not sure which is the taller order, that, or rewriting the scripts. Actually I don't know why you would want to rewrite "the scripts" in general; I hope you mean rewrite a few of them, and leave all the existing cutscene alone. But this is honestly a question I ask myself, "how can I design script multitasking (or whatever other feature) so it could be used in even Powerstick Man without rewriting anything?"
Honestly, I haven't checked in on the latest updates since the last stable version, so I'm not sure if the graphics updates are what I'm looking for yet. Last I checked, battle backgrounds were still limited to 320x200. Has that changed? I also wanted to go full color instead of 256 color palettes. Maybe that's all finished, but when I last tried to import a better version of Entrepreneur's title screen (the one I posted in the Entrepreneur thread a couple of years ago), it looked ugly and disappointing. I'd really like to just redo the whole game in a larger resolution.
And, of course, there are the fonts. I count font design as part of the graphics update, since they assist the aesthetic, and importing fonts from my system and changing their size as needed to fit custom-sized textboxes is preferable to designing them from scratch inside the engine while still adhering to the old 40-character line limit. Not sure if that's been updated since I last opened the engine, but it hadn't been when I'd last used it.
And, I still want more than 160 tiles per set for more highly detailed scenes (and more diverse shop signs). But I accept multiple layers as a feasible workaround in the meantime.
But that's what I'd still like in a graphics update based on my last use (I believe I last used hnotvir(sp)).
That said, for kicks, I used the same prompt with the new image as a base to see what I could get for pixel art. While the results were spotty, they could be reworked to great effect. This one's my favorite:
Pixel_Art_a_blonde_male_superhero_in_white_spandex_adorned_wit_3.jpg (397.03 KiB) Viewed 2347 times
I'd probably upload all of my existing backdrops into the art generator, give them more specific prompts under a realistic model, then re-convert them back into pixels for better imaging and resolution. I'll try it when I get home from work (or later this week, since my vacation starts tomorrow night) and see how it plays out. I'll post my results here if they're worth sharing.
I actually opened POWERXE to double check the resolution and was a bit surprised it was 320x200 (since I'm so used to looking at the maps in higher resolutions in the map editor) but of course it's 320x200, and I forgot the battles were limited to 320x200 anyway. That limit is now raised but if you go much higher (not sure exactly how high) sprites will start to appear/disappear before reaching the edges of the screen: the animations need rewriting.
True colour backdrops is something I'm planning to add in the very near term, I was even considering sneaking it into the release, since it's not a huge project. Better (including non-8x8 and variable-width) fonts and text also really soon since I've done so much work on it already which I want to finish (after release -- what's missing is an editor and file format for fonts), but updating everything across the engine to make full use of them, most notably removing the textbox size limits, will be plenty of additional work. You had a custom textbox system? I've actually forgotten.
Larger tilesets would be one of my top 10 priorities but possibly the one on which I've done the least work.
---
That pixel art image is a very different style from what I got by scaling down your previous one, which remained photorealistic. Pretty neat. However it's the wrong resolution, it's actually too low! I guess you have no control over that?
I'm really curious to know what kind of results you get by prompting with the old backdrops. Especially since the existing Powerstick Man backdrops are very typical for OHRRPGCE games.
If you look closely that image isn't actually pixel art, although the generator has done a very good job at creating consistently sized pixels -- better than I'd see before! It seems close enough that I expect you could trivially convert it to an actual low-resolution image by downsizing it with "point" sampling.
Yeah, the game is still set for 320x200, but scaling it to 640x400 with native resolution (not scaled as the current behavior handles it, I think), or even somewhere in between (480x300) would be preferable, as computer screens keep getting bigger, and pixels become more pixelated. I'd also, quite frankly, like to rewrite all my games with an updated widescreen perspective. I was in the process of doing that for Entrepreneur, but I can never remember what the right ratio is. I have it in my journal. Whichever size I've chosen for Entrepreneur is the one I'd likely choose for this one, too. Hold on...
448x280
That's what I'd written down as the perfect resolution for Entrepreneur. Not sure if I'd feel the same about Powerstick Man, but I can't think of a reason why I wouldn't. For Entrepreneur, that resolution allows for the shop windows to display outside the indoor map screens. Powerstick Man has different needs, of course, but not necessarily. Given the display windows I've been using for battle previews, larger horizontal screen space may still be important. At any rate, it fits modern screen shapes better, so I'm interested in scrapping the 320x200 resolution across all my games. When fonts are more customizable, I'd probably increase the native size to allow for sharper graphics. Not sure yet. It would be a lot of work.
But I don't do that already because the textboxes get really tiny if I rebuild the game at a higher resolution. Text anything in the OHR needs 320x200 as the native resolution to scale larger on modern monitors last I checked, and it really should adapt to whatever resolution is selected natively. That's why custom text is important to me. It's more important for Entrepreneur, since that's a business adventure that needs to convey a lot of statistical information in user-friendly ways. But Powerstick Man is a text-heavy game, so making the text easier to see at higher resolutions is important.
For AI pixel backdrops, the results will almost certainly be disproportionate to the engine as long as generations remain limited to 512x512 or higher. None of the native AI resolutions match scale with 320x200, but I might be able to prompt a specific pixel depth and crop the edges. Not sure about that, either. But I wouldn't just prompt a backdrop and run with it. I'd use the Image2Image feature to clone an existing backdrop and create a new realistic model based on that image. Then I'd run the result through other models or prompts until I found a look that works for the game. I'm thinking of doing the same for characters (heroes and enemies) and de-rezzing them back into pixels. I just bought Aseprite on Humble Bundle yesterday, so I'd clean up my graphics in there. And since AI speeds up the process, if I can get one enemy model to look cleaner and sharper, I may create variants to make enemy layouts more interesting, as the same enemy may have three or more "poses" instead of the same one over and over. This could also make groups of the same enemy look more appealing.
Thinking out loud, of course. Not sure how any of this would look in practice. I'd like to develop a mockup before my vacation ends. If it works, I may start picking at both games again in 2024.
And if production takes off at a healthy clip again, I may start using some of my new tools and apps to inform everyone of progress. Wouldn't a proper newsletter about progress updates and links to exclusive downloads and maybe even a roadmap sound nice? Yeah, I can do all of that now. I still need to set up pages on my website for these games, though. Been slacking on that since my web host is shared, and I feel the performance hit every time I fix a period.
But the options are plentiful.
Oh, and one of these new apps is a chatbot that I can use to train my own custom information. I've built a few to answer questions about my books and website, but I'll likely do the same for my games at some point. Having trouble beating the first boss? Ask the chatbot how to win. If I can, I also want to train it on the plotscript dictionary and some of my scripts so that I can write scripts faster. Again, I'm not sure how well that'll work, but I think it's possible. It's something I'd like to try, at least. It may produce ugly results, but it may also produce solid ones.
I still haven't devoted much time to game design since my last post, but I have been toying with some ideas for possible improvements to my games, including this one, thanks to new tools coming out that can make production a bit more efficient.
For example, running my previous image through an image analyzer has allowed me to grab a specific text prompt that can better duplicate the idea I had for that sample title screen image above with the means to fine-tune it into a base image that I can better work with.
After some minor edits to the generated pixel image, I was able to create a more sufficient backdrop that can serve as the new official title screen.
Powerstick Man Title Concept 1.jpg (34.54 KiB) Viewed 1985 times
Granted, this is still a concept only and may not end up being anything like the title screen I settle on (or represent the final version of this image, as the insignia is still warped in the wrong direction). But it gives me something better to work with than what I've done with the original screens from earlier editions.
I've done the same thing with one of my battle backdrops.
Instead of this original Extended Edition version of the junkyard battle backdrop:
junkyard base final.jpg (48.44 KiB) Viewed 1985 times
I'm considering using something like this.
Forgotten Junkyard AI Option 1.jpg (36.11 KiB) Viewed 1985 times
Again, these are ideas, but I think they'll help improve the look of the game, even if they aren't fully mine anymore. If nothing else, they'll give me a stronger foundation from which I can better customize the scenes.
But yeah, they're just ideas. Nothing is set in stone here.
But it might be enough to get me back to work on the game sooner than later.
Music News
In other news, I've finally managed to create a soundtrack that will allow me to finish the long-unfinished music side quest.
In case you've forgotten, I'd set up a series of musical quests that were supposed to culminate in a concert sequence where Powerstick Man (the player) can watch the island's resident rock star, Cherry Chicklet, perform. Problem was that I had no way of creating original songs without spending a lot of money I've never had on real musical talent.
Thanks to Udio, I now have something I can use. In fact, I've now got two full albums of somethings I can use with a third in development.
So, this should also lead to a huge elevation of the game's soundtrack soon.
I've recently put together a trailer for my email newsletter featuring a sample of songs from her first concert, some of which will likely appear in the game at some point. I still have some editing to do before the songs are finalized, but they've got strong legs.
These are based on songs I wrote for the novels but until recently have never been able to hear. Once I got the means to hear them, I wrote the rest of the songs for each album, and now I'm thinking of ways to integrate them into the game. I'm happy with the results so far.
Anyway, if you'd like to see the trailer, here you go. It also features the first song, "A Legend Defunct," in its entirety. The games in the video are rhythm games Melody Escape and Melody Escape 2.
That's all for now. Sorry I don't have much else to share. Life's been overwhelming, but these resources should make it easier to get my ideas back on track.
Let me know if you like the direction this is heading in.
And please let me know if you like the title screen. I'm tempted to use the same methods to reproduce all of the game's existing cut scenes into more palatable images.
P.S. I don't have plans to redo the walking sprites. There are way too many of them and too many animations attached to them. They will likely remain sized at 20x20 (unless we get larger tilesets that can support higher definition maps...then I might consider a sprite upgrade).
For the title screen (Powerstick Man Title Concept 1.jpg) the white text at the bottom is hard to read in front of the white suit, maybe change the text to a dimmer color or put a black outline around it so it doesn't blend into the background.
Other than that looks cool. Are you AI generating the pixelated images themselves or are you generating standard images and then manually editing them after scaling them down to 200p?
No, I'm generating them directly as pixelated images to build the canvas, based on concepts I've already developed from earlier handcrafted work. I'll edit them in an editor once I'm ready to refine them.
So, for example, with the junkyard image, I ran the original backdrop into an image analyzer powered by OpenAI, Gemini, and Claude, grabbed a prompt that best explained the contents of the image, rewrote any part that was clearly misunderstood, added additional instructions that I thought would be helpful, then fed the new prompt into Leonardo AI with the original backdrop as a line and depth reference set to "medium" influence. Of the four images it produced, I decided the one I posted below the original was the best representation of what I wanted for the scene.
Then I imported the produced image into Affinity Photo and set it on a 320x200 canvas and scaled and moved it to fit. Odds are high I'll redo this officially under a 640x400 canvas.
What I'll likely do next is duplicate the original image into new layers and inpaint or stamp over the major graphic flaws, like the ditch running across the scene, for example. I may also attempt re-mapping the colors to the game's palette, but for now I think it's better just to import with dithering. I'm not as interested in being meticulous with the look and feel of the backdrops as I might've been when I first started the Extended Edition project. There comes a point when I have to measure time against value, and value against ability. If this process gives me a sufficient backdrop, then I'll call it done. But if I find the process of matching the new image with the game palette is quick and easy, then I'll apply that to the finished image.
Regarding the title screen, yeah, I'll do all of that. I still have several things to fix before I'll say I'm happy with it.
This one is less of a game update and more of a lore update:
In my continuous quest to improve my games' impressions and build up the characters that comprise the stories that make up the games through the use of innovative technologies that mask the fact that I have very little artistic talent, I have created a concept music video for one of Cherry Chicklet's later songs (from the fourth album, made to coincide with the sixth year storyline from the novels).
Part of this experiment was simply to see if I could do it. The other part was to test the technologies that would allow me to do it.
But the end result, while flawed and unlikely final, is still leaps above whatever I thought was possible for a character belonging to the Powerstick Man universe. And I thought I'd share it in case it inspires anyone to find new ways to bring their own characters to new dimensions of storytelling.
So, what is this?
The song is called "Timeless," and it's the title track to Cherry Chicklet's fourth album, which coincides with the events that launch the novel series's second half. (For reference, the game will cover most of the series's first half, if I ever see it to completion.)
The song itself may not actually become part of the game, so this video is more of a demonstration of what's possible, while also contributing to the story's post-game (and the novels' sixth year) lore. But I thought it was cool enough to share, in case you all are interested in new ideas.
This isn't to say none of Cherry Chicklet's songs will make it into the game. Some likely will, especially the early ones from the Caffeinated soundtrack. But this one in particular may not be needed.
Then again, maybe it will. I don't know yet.
But it's still worth listening to and watching here.
