Home Creators Posts Import Register Favorites Logout
haven't archived this post yet. have a subscription? use the importer!

Poll

About Requests, Part 1 / Upcoming Plans

  • Option 1 7
  • Option 2 10
  • Option 3 5
  • Option 4 8
  • None of the options (let's see what comes up in Part 2) 32
  • 2024-12-22
  • 62 votes
{'title': 'About Requests, Part 1 / Upcoming Plans', 'choices': [{'text': 'Option 1', 'votes': 7}, {'text': 'Option 2', 'votes': 10}, {'text': 'Option 3', 'votes': 5}, {'text': 'Option 4', 'votes': 8}, {'text': "None of the options (let's see what comes up in Part 2)", 'votes': 32}], 'closes_at': None, 'created_at': datetime.datetime(2024, 12, 22, 16, 59, 6, tzinfo=datetime.timezone.utc), 'description': None, 'allow_multiple': False, 'total_votes': 62}

Content

Hey! This week, I was planning to post a big write-up with my ideas for the future of requests, as promised, to get your feedback. But since the post turned out really long—and because some options need extra demos/examples I didn’t have time to finish 😅—I decided it’d be better to split it into smaller parts.

First off, here’s the plan for the near future:

  • Regardless of which request option we choose, new requests will reopen next month.

  • Additionally, there will be a short break next month (around a week or so). During that time, the usual content won’t be posted, but I’m hoping to share a few examples of alternative content formats I’ve been working on but haven’t finished yet. I’ll definitely post something from that lineup during the break.

  • By the end of this month, I also plan to post some extra content with both new characters and some previously featured ones. These will include examples of a “request format” that I’ll explain more in Part 2, and I’ll also be using the newest version of my main image generation model to create this content.

Now, about requests: while thinking through the options that could work for me and be interesting for you, I realized that maybe the current system is also an option… just with some tweaks. Here’s the breakdown:

  • Option 1: Requests would periodically open and close. For example, requests open today, and once, say, 20 requests are submitted, the form closes until most of the current batch is completed.

  • Option 2: A downside of Option 1 is that requests would only go through for people who manage to submit them fast enough 😅, which might not feel great. But considering that [Max] isn’t too expensive and has other benefits like extra content, maybe it’s okay? I’m not sure. To make this fairer, though, here’s a variation: During an open request period, there’d be weekly polls for the requested characters. The ones with the most votes would be done as full requests, and those with fewer votes would still be done but in a “simpler” format (fewer images or missing some detailed specifications).

  • Option 3: Option 1 + an extra feature for guaranteed request completion.*

  • Option 4: Option 2 + an extra feature for guaranteed request completion.*

* - By “extra feature for guaranteed request completion” I mean either an additional tier with a higher cost (to avoid getting swamped with requests 😅) or maybe handling it through a different service—or both. You get the idea.

Okay, that still wasn’t very short 😅. I’ll post Part 2 next week, but in the meantime, feel free to share your thoughts on these options—whether we should rule anything out or add any of them to the poll for Part 2.

Thanks again to everyone who keeps supporting! I hope you’re still enjoying the content!

P.S. The full set with Nyotengu from the images at the beginning of this post will be ready and published tomorrow.

Files

Previews only

Comments

kahvipapu

First of all, congratulations on clearing the request queue :). I'd suggest the request format should be tailored for how much volume and overhead you want to handle. The slow lane/fast lane seems like a decent compromise. - Guaranteed tiers: in my experience if there's a subscriber limit and the price is low (and the quality is high, which it is :)) these tend to sell out very quickly thereby locking future users out, which isn't ideal. Haven't seen it done but I'd honestly consider setting a very high price and using it to control demand. - Polls: they tend to generate a high number of characters with a near 1-to-1 ratio of requestors to characters. It's workable but a lot of overhead. Overall if you're creating off of user requests there's no way to please everyone when demand exceeds supply.

euskyart

Hey! Thanks for sharing your thoughts! There’s definitely something to think about here :D

Pete

I think that it would be better if requests would take up some of your time and not all of it)) so maybe there should be a limit for them. Requests introduce a lot of new characters and are awesome in general but I think many were making requests just because they have such option not nessesary that they wouldn't be happy otherwise (that was my case). So maybe introducing different tiers for max content and for making requests would be good. I would be fine with not making requests as long as I get all the sets with all extra images)) And I do agree about the polls that very often different characters would have very simular number of votes which makes them less useful