Creator AMA (RDC 2024) with Nick and Tian [September 19, 2024]

[Update] - 4:20PM PT

[Update] - 2:30PM PT


Hi Creators,

Nick (@skywise84) and Tian (@lowercasecreator) here, officially ready to kick off today’s AMA with you all. From 1:30-3:30 PM PT today, we’ll answer questions related to what we announced at RDC.

You can now start asking your questions below! Please stay on topic, be brief where possible, and keep to one question per comment. If you have questions outside of today’s topic, please refer to Creator Docs, the DevForum, or our Creator Roadmap.

Thank you and talk to you all soon!

Nick & Tian

66 Likes

This topic was automatically opened after 3 minutes.

How is it possible that even to this day, the engine is being held back by tons of useless projects, AI and the visions and wishes around automated, magic and one click solutions? We the people see and hear promise after promise about how the engine helps you build and create unique, creative, any and all sorts of experiences which is NOT the case and has never been the case. The engine itself often poses insane restrictions and limitations for anything which isn’t the standard simulator and/or simple low-end targeted mobile experience. Time and time again developers have to rely on very limited and restrictive optimizations, imaginary optimizations, the constant abuse of tricks/glitches/bugs to achieve a specific effect which 9.5/10 times will be half baked, underperforming and unusable in most situations compared to absolutely any other engine out there. How is it possible that even TODAY, you guys hold the same views around the engine as far back as you guys did nearly two decades ago? You guys have even promoted developers actually trying to push the bar (via showcases and sometimes even experiences) of the entire platform yet your response to them is to throw them more one click solutions which do nothing to actually serve anyone looking to do more than the bare minimum?

I have nothing against the existence of automatic systems and features. What i DO have an issue with is the fact that these automatic systems is all we actually get. What exactly are we supposed to achieve with those? Are we all supposed to simply fill your corporate pockets with the money of small children because that’s what the engine was built for? I’m not sure about you guys but relying solely on experiences like this will eventually just dry up the platform and crash it into a wall. No substance means no one cares, and why would they?

Every update directed towards the engine ALWAYS comes with one or more cons all for the sake of conforming to this magic future of the engine in which everything is automatic and magic. Most things we actually do get are often insanely limited and just do nothing to serve people the way they actually wish to be served. Even worse is the fact that some of these changes are SPECIFICALLY TIED to previous features often forcing developers to either conform with the new magic addition or sacrifice BOTH the new addition and a previously implemented feature just in an attempt to actually have their experience match their vision as closely as possible.

These days it would be absolutely absurd for any other engine like Unity, Unreal and many more to simply come out and say “Guys we believe in a simple, automatic and one click solution future. This will let us increase security, safety, user friendliness and overall usability by 100x!” without them going bankrupt the week after. You guys wish the engine to allow us to create realistic experiences, no? Not sure how to break it up to all of you but the kind of realism the engine allows us to create is lesser than early 2000’s AAA titles while somehow STILL yielding lesser performance than those AAA titles. What do you guys have to say about this?
I believe you guys along with most other people here have seen new showcases time and time again yet its pretty clear NONE of them will really be able to push past simple small 50x50 stud playable area’s due to how unperforming and limited the engine is. A full on experience with showcase levels of visuals are nearly impossible without simply either making tons of super specific sacrifices or just overall butchering the overall visual quality in hopes that it’ll maintain a “good enough” 60 FPS. I find it insane how the engine is even being constantly ran at like what, half resolution on mobile devices? How is this even possible considering every other major mobile game can run at full resolution, high framerates and WORLDS HIGHER visual fidelity than what roblox could ever wish to provide within reasonable performance metrics? What does roblox have that makes it so much more resources to run on mobile than any other game like Call of Duty: Mobile for example? Is it the engine just being poorly optimized? Is it the fact that developers can still barely do anything to actually provide high performance to these systems? Is it both? Or is it something else?

I find it absurd how so much of the company money is going towards AI models (which most of them will likely not even see the light of day) which are all so far hidden behind the argument that they will one day exist in the future. It’s even more insane how you guys claim to be really pushing the bar with these AI systems and models while being pretty clearly far behind what other AI companies can do. Open AI could give you their best AI models tomorrow and your variants of them would somehow STILL be far behind what OpenAI got.

So i must ask again, why is the engine taking a back seat for all these meaningless and baseless promises when developers all around wish to actually do something meaningful with it without having the engine stop them dead in their tracks early into their projects due to its countless limitations, missing industry standard features (not AI) and countless other issues which have YET to be addressed? I’m really not sure how you guys are expecting an engine built around limitations to really serve people who wish to actually do something with it in the future.

81 Likes

Hey guys! Thanks for doing another one of these. It’s always awesome to be able to ask the both of you questions directly.

While answering a question during his Q&A at RDC, Dave brought up the possibility of Luau being used as a shader language in the future. Allowing developers to author shaders using Luau is something that’s been requested time and time again, year after year after year. The last “official” statement we have regarding it is from @zeuxcg’s 2016 devforum post where he basically said “this is probably not happening, ever.” This RDC marks the first instance (to my knowledge) that Dave has ever publicly acknowledged this feature as a possibility and so there’s this renewed sense of hope that it may happen at some point.

Has there been any additional effort, research, investigation, or development put into “LSL” internally at Roblox? If so, is there any information you can share on it at this time?

27 Likes

My questions are about the recently announced Shopify x Roblox integration from the keynotes at RDC

Is there plans to post a roadmap soon for this integration to keep creators posted?

I’m getting ready to launch my first physical products with some amazing partners soon and the topic of ecommerce on Roblox is one I’m super interested in connecting with relevant teams on.

18 Likes

Hey there! Are there any plans to bring back “Best friends” feature to the platform alongside the friends 1000 friends slots?

Also, would we be able to customize permissions to only allow best friends to add us in team create, join us etc.

23 Likes

Hey! Can you separate your questions into two separate threads?

10 Likes

I was really excited about the “Text To Speech” API coming out soon, and noticed that the voices sounded very robotic in the preview. Is this subject to change? Also would we get options of different voice types / accents to choose from?

Also, I was wondering if unlike plugins, if we would still have the option to have experiences set as a Robux amount to enter as well as a real currency fee? I was saddened when I learned that plugins can ONLY cost real money now, and there was no option to pay in Robux.

15 Likes

When can we expect to see UGC creators get more of a fair revenue split, similar to what game developers see? Right now, Roblox still takes over 90% of UGC creator revenue, making it impossible for most creators to make meaningful earnings off of their creations.

20 Likes

With the upcoming party system having a feature that looks like a VC chat room system in friend chat, would we potentially see the ability to voice call players in the same format? Like push a “call” button and a player can accept or deny?

9 Likes

There’s a bunch of people here that want their questions answered – please don’t take up so much reading space with such a long rant-ey post. This is not a good faith reply and it’s disrespectful to other forum posters who have questions.

59 Likes

When will the clouds instance get an update? It’s been in phase 1 forever.

21 Likes

While I have this opportunity, I’d like to address the following:

In My Longer Response to "Discovery on Roblox: Past, Present, and Future Vision", I point out the flaws in Roblox’s current discovery system while detailing how Filters & Tags would improve it.

Filters & Tags is an industry standard when it comes to content discovery. Why is it not on Roblox? Is the core value of “tak[ing] the long view” the hold up? I fear this is a particular decision by someone at Roblox to stray away from transparency and user action in favor of ‘Semantic Search’, which I deem to be a flawed vision.

Imagine Filtering by Price (hopefully robux or USD), Server Size, Genre, etc. Not just one sort made by Roblox. That’s the system for 10 percent of the gaming market. That’s the system for a billion users.

13 Likes

Will Android be seeing Future lighting and Depth of Field relatively soon?

(~ a sober user stuck with mobile rendering)

11 Likes

What does the future of UGC look like on the platform with respect to game compatibility? You mentioned new scaling controls, but how robust will this really be given how inventive creators are? Is there any tech ready to talk about at this point?

Right now, there seems to be an unstoppable problem with people bypassing avatar accessories and clothing to the point where games are now starting to simply outright turn off UGC. This is the new “cat and mouse” problem. We can imagine systems such as “separated categories for standardized versus outlandish UGC” + “multiple avatars” + “games choosing the type of UGC they allow”, but what direction does Roblox plan to pursue? Will we only be getting stronger scaling and hitbox controls, or are there other ideas currently in motion?

With giant spongebobs in the sky, it’s hard to imagine one generic avatar scaling solution will be enough.

16 Likes

When will we be able to upload dynamic heads to groups?

6 Likes

Hey there! Always like seeing these.

Can you tell us a bit about the new voxel lighting grid coming next year and occlusion culling, which is scheduled still for this year?

15 Likes

My other question about the Shopify x Roblox integration is below

** When do you guys plan to post information regarding payment splits? (or any other associated fees?) **

I imagine there is a lot of parties that all want a cut of each sale (Shopify, Roblox, Appstores, etc) so I’m curious how this dilutes the creator’s margins.

9 Likes

Hey there, is there any plan to add a new API for planar/screen space reflection or proper cubemap reflections? The current one we have for indoor environment mapping is very outdated and visually looks very inaccurate.

9 Likes

Last q&a, you guys said PBR was coming soon for UGC accessories. It’s been a while and we still haven’t received any updates. Is there any timeline for that?

9 Likes