Ren'Py What is your workflow on creating VN with Renpy?

[hkproduction]

Hi everyone,

I'm working on a VN. It's going to be my first time making one so I'm wondering how to get things going on creating content.

Here is my current workflow for creating 1 chapter of content:
1. Write all the script into .rpy file. The script contains narration, dialogue, and maybe some background/scene if the CG already existed.
2. Open Renpy, do the first round of proof-reading and test play to find bugs and fix awkward sentences
3. List out the scenes that need to be rendered, based the script
4. Open DaZ/Studio Neo and start rendering, export the image, and paste into the images folder
5. Update the script with newly created CG
6. Do the final round of playtesting + check if needed. Fine-tuning some transitions between scenes.
7. Build and upload...

How is your workflow? Did you render first then write the script, or opposite? Do you draw a draft version of the rendering on paper first before opening DaZ/StudioNeo?

Any ideas are welcomed. Jjust want to know how people are working, and what can I learn from to improve my workflow.

 

[hkproduction]

I'm also curious how a team can work together to create CG.

As a writer, how do you describe the scene and give it to the designer/animation? Do you just write it down, draw a draft version on paper, or include examples/references from other art?

 

[mickydoo]

Some people write first, some ppl render first. However you do it be prepared to change your script on the fly, when you see renders you always think of something else, or that scene should have this dialogue in it etc.

We all have some sort of note taking, I work a lot off the top of my head, knowing where I want to go and want to achieve, others will get more detailed in their approach.

I have at least 200 renders and 8 animations ready for my next chapter that have not had any dialogue written for them, I know what scenes I want and what happens in them, I write to the renders.

There is no wrong or right way to do the above, whatever is best for your brain.

You can test play it a 1000 times someone will still break it and find the tiniest typo

Can't help you with teamwork, I am my team

 

[8InchFloppyDick]

Workflow... very much an each to their own thing... but mine looks a little like this:

0. get up bright and early, eager to make progress with your VN
1. stare at screen and check the weekend's football results between the bouts of staring
2. ponder the myriad plotholes in your utterly derivative story
3. stare at screen some more
4. consider if your MC really should be an orphan going to a top college in a big city on a scholarship
5. get up and wage war on the fridge
6. come back victorious and do several pointless test renders of a character that isn't even in the cast of your VN
7. have dinner while watching inane DIY TV-shows on some .ru pirate stream
8. wallow in self-loathing for again not making any real progress on the VN you started 3 years ago
9. call it a day, go to sleep and restart at step 0 in the morning

I find step 8 to be especially productive

 

[8InchFloppyDick]

Oregon1898 - sorry to make you sad... so in the spirit of fairness and all that, here's what really happens.

1. I find that my VN is plot first, renders/gfx later.

That said, I do trawl the three main websites that provide us with an endless stream of props, environments and models. Sometimes a new asset will help me get my writing over dead points and other blocks. I've changed my story a couple of times simply because a particular asset turned up and I wanted to use it in my story. I've also changed the models of some of my characters on the way.

2. I do all my writing in a plain text editor without any Ren'Py markup. Geany/VSCode/Vim/EMACS - it doesn't matter. I don't use anything that does 'formatting'.

3. I keep a separate directory with plain ASCII-files full of notes and ideas.

Some are relevant some aren't, Though I'm sure some will be worked into my story somehow. (a very some-rich sentence that...)

4. I take my plot points and turn it into prose.

As I write, I take care to be aware of the length of my sentences so that in Ren'Py they'll fit nicely on at most 2 lines of the display box. I also proofread and edit my text while it's still outside of Ren'Py. Finally, I add the filenames of the renders that I think I'll need to illustrate my story, each filename on its own line. Each file gets a a descriptive name such as: bakery_roxana_smiling.jpeg

Note: the renders don't exist yet. The .duf files don't exist yet. It's just filenames at this point.

5. I 'scaffold' my project in preparation for Ren'Py mark-up.

First I make a 'definitions.rpy' which holds various defines and the character names etc.. Then I make a 'main.rpy' script that setups a few useful functions and has the 'start' label and a few other general things like the splashscreen end the end notes etc. Other than that, each episode in my project has its own *.rpy file. Images also go into per-episode directories.

6. When I have my writing done and scaffolded, I open Daz and wish I hadn't gotten the VN-writing bug.

I don't feel the need to do much, if any, sketching and stuff beforehand. I just do what needs to be done to get the renders out of Daz. I tend to render at low resolution 1280x549 (21:9) and at 250-300 samples. 500 samples max for those odd super close-ups of skin and eyes. Render quality OFF, Firefly filter ON.Then I pull them through the nvidia AI denoiser. I don't have fancy-enough hardware to do higher resolution and huge amounts of samples. I also use Scene Optimiser for just about everything.

I do quite a bit of kitbashing and for the most part I'll either do my own posing or customise off-the-shelf poses. I keep lighting as simple as possible. I don't bother with Daz animations. Partly because of speed issues, mainly because there's xvideos.com.

7. After review, the denoised images get copied into the relevant game/images/episode_XX directory.

8. I do the little bits of code needed to make things 'work'. all quite trivial as my VN is more story than mini-games and sandboxxe.

In summary my main workflow things would be:

1. use ASCII text and write to a 'happy end' outside of Ren'Py first
2. create descriptive filenames for render assets as you write
3. structure your *.rpy files into smaller files
4. use the predefined filenames in your text to keep track of what needs to be rendered in Daz
5. favour the use of a denoiser over a high render sample-rate

And that's about it. For me.

Hope this helps in some small way.

 

[Playstorepers]

Rule 0: I can't stress this enough:
DON'T EXPECT TO MAKE A LIVING OUT OF IT.
Seriously. I've seen too many v0.1 games, which immediately get abandoned, once they realize, that it takes a lot of effort to make a few pennies, let alone a living out of a game. Nobody's gonna suck your dick, because you make a game. It will most likely take years, before you get a certain amount attention, if even so.

Rule 1: Make a masterfile (A spreadsheet e.g.), which has all your ideas. Separate them into loose ideas and ideas, you want to implement. Make sure the masterfile can create a minimum viable product, then start making the game, while slowly adding to the masterfile.

Rule 2: Make sure to always keep your curiosity up. Perserverance, discipline and curiosity will carry you through your workflow. Nothing else. Neither creativity, money or intelligence. Just curiosity, perserverance and discipline.

Rule 3: Make sure to keep your workflow balanced, which is to say, work on everything, you can/want and make sure to single out what you can't do to hire or ask experts.

From there on out it's just the three things above: Discipline, curiosity, perseverance.

You can add a few pointers and tricks here and there to help you organize and structure your workload/reduce future work, but you'll probably have to revise shit anyways, despite a million preparations. Which is not to say, that you should disregard prepwork/structured work.

TLDR: Even the most structured person doesn't have ONE workflow. It's always just a small guideline/roadmap.

 

[PTSdev]

Workflow... very much an each to their own thing... but mine looks a little like this:

0. get up bright and early, eager to make progress with your VN
1. stare at screen and check the weekend's football results between the bouts of staring
2. ponder the myriad plotholes in your utterly derivative story
3. stare at screen some more
4. consider if your MC really should be an orphan going to a top college in a big city on a scholarship
5. get up and wage war on the fridge
6. come back victorious and do several pointless test renders of a character that isn't even in the cast of your VN
7. have dinner while watching inane DIY TV-shows on some .ru pirate stream
8. wallow in self-loathing for again not making any real progress on the VN you started 3 years ago
9. call it a day, go to sleep and restart at step 0 in the morning

I find step 8 to be especially productive

Great post. And so true.