Today, we're going to take a look at a slew of C Dance tips you need to know and prompts that really blew up. But, more than just providing you with the prompt, which of course, you know, I will do, we're also going to excavate why these prompts work with the idea that you can come up with your own unique spins for your projects. Also, we're going to head back over to Martini, the new AI studio platform that we checked out a little while back and you guys really liked it. I will say up front that is a sponsor segment, but trust me, it's not salesy and they've got a feature on there that I have not seen anywhere else. All right, in the meantime, let's get dancing. Kicking off with our C Dance tips and prompts, I do want to note that I'm going to try to credit as much of these prompts as I can, but of course, by nature of prompt shares, they tend to get spread around and modified quite a bit. They sort of enter up like a public domain remix vibe, but again, I do want to shout out at least where I discovered them. First up, we have the viral snap stop time prompt from uh well, friend of the channel, Chris First. Uh this one I have definitely seen a lot of modifications on. It's a really good prompt and works extremely well, especially for like, you know, if you're in the ad game. So, digging into that prompt and you can feel free to either like screenshot this now or um if you don't feel like going through the hassle of OCR-ing it and all of that stuff, uh I will publish it in uh this week's upcoming newsletter. Uh the interesting parts about this prompt, obviously it does it is using a reference image here of Chris as a character. Uh we're going to touch on that a little bit more. I have found that that seems to be lightening up a little bit lately. We sort of have the like standard prompt uh format here for C Dance uh calling out an Arri Alexa Mini 35 mm lens and then moving into timed uh shots here. Now, the interesting part here for me at least is the uh at the snap, a subtle spherical shockwave burst from his fingertips. So, that is something that we can definitely pull from this prompt and use in other places as well. Um yeah, I guess C Dance does a pretty good job with these spherical shockwaves. And of course, now that you have the prompt, you should modify it to your own needs. Uh for example, as I did here and I'm not about to use the Yankees. Sorry, Chris. Perfect. >> >> It's funny, you can tell that I actually manually modified that prompt considering that there's like one guy rocking all of his Yankees gear here. So, clearly I missed that later in the prompt. Some garbled text over here or maybe not. Maybe uh Yiannis Dirk is uh but, you know, a rookie season player that is showing some promise. This next one I have been enjoying quite a bit. It utilizes uh C Dreams first frame last frame feature, so not Omni reference. It comes to us via Framer and is just simply show me what happens in between. Use multiple camera angles. Yeah, you're not going to need to write that one down. As we can see here, Framer provided two images, one of a princess She's not really stuck in that tower. She could totally make that jump. Uh and then them, you know, camping afterwards. And then the resulting video um actually has our knight climbing up to rescue her. Again, lady, you you really could have just, you know, hopped off and you would have been fine. What I like most about this is how well C Dance ends up kind of surprising you. So, I got this image I Midjourney generated of a rather buxom lass hanging out in a pirate tavern. I just noticed like, does she have a scar down here? There's a story there. And then for our second frame, I had these three crusty pirates from another generation uh decided to again, put them together, show me what happens in between. >> >> As it turns out, she's a siren. Look at all those pirates just completely captivated by her. Yeah, they are all going to end up at the bottom of the ocean. For another example, I ended up taking this image and this image, both of which were referenced off of weapons cuz I I was I really liked weapons. So, I was trying to kind of aim for something in that visual look. Um and then again, running these two, just show me what happens in between. So, yeah, not too shabby. It does end up creating a narrative bridge between those two shots. The emergent storytelling stuff of all of these AI models, I mean, I find just endlessly fascinating. The other thing that I did want to mention here is that you will notice there's a lot of faces in these generations. It does seem to me that C Dance is kind of lightening up a little bit on the face filters. I am definitely getting a lot of like realistic faces coming through. I mean, you're still going to get blocked and pushed back if you're, you know, again, trying to put Tom Cruise and Brad Pitt in there. Don't do that. Next up, we have a really fun prompt from Coda. This gives us that like sort of morning manic feeling uh and then, you know, followed by the trip to work and then uh after that, we arrive at work, go through our work day and then of course, you know, come back, pass out, go to bed and then loops back. If there's one thing I'm definitely a sucker for, it is time loops. Groundhog Day, Edge of Tomorrow, uh Looper, I love them all. So, extracting that prompt, which is pretty long, we are once again using an image reference for our subject uh with wardrobe changes throughout as well. Uh and then in Coda's prompt style, um you know, we have everything from the color logic being hyperreal pop look, uh an ultrarealistic style and um and then just I mean, I think it goes to show just how much you can cram into a C Dance prompt. There's a lot of information in here. What I think is pretty interesting about this one is that it actually does have sound effects tied to each shot as well. So, taking this whole prompt and again, giving it a shot of my own, I wanted to try out using this character who we took from a illustrated image into like a live-action image in GPT Image 2. We looked at that last video. So, I took that cinematic output and then had Image 2 create a 9 by 16 character reference sheet for her. Um we did lose some details in uh you know, the switch over from her outfit into like business attire. Although, I will say that um you know, she seems to be missing a button there. She might be getting a call from HR pretty soon. That's it. I mean, maybe she is HR. >> >> So, yeah, I mean, pretty solid generation. Although, that said, I something that I definitely want to note is that that was not a one shot. So, this is the generation that you actually saw, uh but that was actually one, two, three, four, five. Five other generations to get there uh and then one that actually failed for some reason. Um yeah, you'll actually Uh the other thing that I do want to note, too, is that um you'll note that some of these are actually in Chinese. Some platforms will have a character limit on like the prompt amount and Coda's prompt was longer than 3,500 characters as it is on Runway. So, all you have to do then is just like take the prompt and bring it over to Claude, ChatGPT, or Gemini, or whatever and just say, "Translate this into Chinese." And uh at this point, you are actually far under the character limit. I mean, Chinese is more efficient language. Uh I think the important part here though is that, you know, I can run this here and for whatever reason, our Image 2 character passes no problem. Rounding out with a prompt that brings me a lot of joy. This one is actually a lot going on with this shot. There's actually a great little like mini interview documentary thing with uh the cinematographer Don Burgess uh that I'll have linked down below, but it actually was achieved uh by using a blue screen. So, Plasmo decided that was a call to arms and decided to see if you could achieve that shot in C Dreams 2.0. I mean, it's not perfect, but I mean, it gets pretty close and definitely gets the overall idea of it as well. I think the really interesting thing in Plasmo's prompt here is um the core illusion. So, that's actually, you know, triggering to C Dance like this is a trick shot. Um we also do have of course, like the speed ramp down here, frame rate ramps from 24 to 48 and then into slow motion. Um so, taking this prompt and testing it out myself, so yeah, you can definitely get it. Though, I do want to note again, like last time, that ended up taking one, two, three, four, five, six, seven, eight generations uh before I got there. A lot of times what I would end up with is something like this. So, a lot of clothes, but not quite right. So, again, I think that's just important to stress that with any of these prompts, it might take you a couple of rolls before you get the thing that you're looking for. Kudos to Plasmo on that one. I do think that that is a pretty fun and interesting challenge, you know, trying to take, you know, the famous cinematic shots and seeing if you can recreate them in C Dance 2.0. All right, moving on. Next up is our look at Martini's latest update. This one is really useful. Plus, you get to hop the waitlist. You know what sounds good right about now? A Martini. Hey, it's 5:00 somewhere. Or maybe a better idea if you're not looking to take a nap in an hour, uh we can check in with Martini, the AI studio platform I took a look at a few months ago that you guys really liked. Quick note, Martini is sponsoring this segment, but just like last time, this is not a hard sell. Uh this is me just showing off the impressive updates they've been working on. I've been pretty impressed. I think you're going to be impressed as well. So, the standout feature that really grabbed everyone's attention last time was the step into set feature where we could take just this is like an image that I generated up in Midjourney. Uh if you click over here, you can hit step into set and this will allow you to well, like literally step inside the set. There are limitations um to this this is like essentially Gaussian splat base so you can walk outside of it but the idea here is essentially you, you know, kind of find shots that you like. An update that they've recently made is that all you have to do is hit the capture button now and then that will um yeah, capture the image for you. That'll load up here. Um and then you know, previously the idea was is that we could take a shot like this and um and then add a character into it. They've also added the bonus of like, you know, cuz with the Gaussian splatting obviously it kind of gets a little fuzzy at times. So, what this is doing now every time you hit capture is that it's actually kind of running like a like a secret Nano Banana kind of cleanup on it as well. So, that's pretty handy. This is where things get crazy though. So, what we have here is a Midjourney generated like location image and then two characters that I've been sort of workshopping a little bit. This is Rex and uh Pink Punky. Uh they are um stupid names. Very stupid names but you know what? It's okay. They don't actually exist. From there, I ended up utilizing Nano Banana Pro to comp them into the image as you normally would. is where things get interesting uh because if we do a step into set with this image, you'll see that our characters actually now appear in our set. Uh this particular set is a little bit on the extra crunchy side just because there's like so much detail in here. Uh again, we can change the lens length if we want to. Um so, this gives us some placement where they are. And the characters are assignable as well. So, red I already did this but red uh is obviously Rex and green is Punky. So, if we go through and um you know, just find a shot like here, let's just say uh and we hit capture, you'll see that we have our pre-render over here that's already started. After our characters have been tagged, you'll note down here that it's actually already built out not only an a face reference but actually uh turnarounds as well based off of your character input. So, you don't even need to create turnarounds for your characters. You can actually just, you know, hand it right over to step into set. Um so, that gives it the ability essentially that when we come into uh this pre-render which all we have to do after this is hit render, you can see that this gives us uh essentially maximum control uh in terms of where we want to place our characters. It's really really impressive. More than that, we actually do have, you know, 3D character control as well. Now, I do want to point out that this is not like full it's not a full 3D package or anything. Like we're not going to be able to do pose control here or anything like that. So, this is more about placing your characters within the scene. Um you do have some rotation as well here. So, if you want to turn a character this way um and then maybe we'll get Rex and we'll turn him as well. Um so, yeah, there's there's a lot of control that you can uh in terms of placement once again. It's good because it's not overloading you with a whole lot. It's it's not poser, you know what I mean? It's not it's not a 3D software. Um and if you're kind of like a little nervous about this um this character stuff and using the 3D controls, it's really uh it's really pretty easy. If you just kind of play around with it for a few minutes, I'm sure you're going to do just fine. And after just a few minutes of just kind of screwing around, we end up with this six-shot sequence here. Uh and you know, beyond the precision and control that you have with this and you know, consistency as well of like the location and the characters is the fact that like a shot like this with the negative space on this side here, uh that would be again very difficult to prompt for. Like Nano Banana does not want to do a shot like this. Uh even this one this is when we were playing around with our character rotations. Again, very difficult to prompt for. And look, I will say that the recent advances in Omni models which are fantastic will do a lot of this work for you. Uh but you know, also they're choosing the shots for you whereas this method uh allows, you know, returns control to you. Before moving on to the other big feature that uh Martini has recently released, I did want to point out uh this is another project that I was playing around with earlier today. This is our uh space cargo crew. We've used them a number of times on the channel here before uh Lyra and Griff. Um so, I've already turned this one into a set. What I do want to make note of is that um say with like Lyra here, you'll see if we if we double-click on her, it populates. We have our turnarounds here but you can So, that's her original source image. Um you can replace each one of these if you want. So, if you're not happy with these and you want to replace them back with like your original character, you totally can. So, that's pretty handy. Um so, you'll see here that uh again, uh some more shots that I was playing around with. Um some are rendering right now, some are pre-renders. The thing is that I didn't actually come up with all of these shots. Uh I actually ended up using Martini's new multiplayer feature. Now, I think most folks who watch this channel do tend to work solo but I will say as this technology improves and our ambitions get larger, teaming up and collaborating is going to be important part of bringing these projects to life. So, I ended up calling up Ko from Martini uh the CEO and we just kind of jammed on this board for a little while. You can see Ko's uh camera right there. So, we're looking at my camera view right now but we can actually watch his as well. Um so, and you know, the whole time that we're sort of choosing our shots and whatnot, we're both, you know, obviously on a voice call together as well. Again, I know most of us tend to work solo but I'm definitely meeting more and more people who are working in small studios or teaming up for collab projects or forming filmmaking duos and I do encourage you to do the same. Filmmaking has always been a collaborative art form and it opens you up to different ideas and per- spectives from like other humans, not just an LLM saying, "That's a great idea." Kind of going back to what I was saying earlier about the Omni models, we are reaching a point where the models are very good at making choices for you and there is nothing wrong with that. It's amazing in fact for speed, iteration and capturing lightning in a bottle but also remember that it's important to connect and collaborate with other people. Uh you know, obviously they're they might not gas you up like Claude will but ultimately through collaboration, I mean, you're always going to end up creating something more unique than a chain of AI agents are going to. All right, getting off my soapbox now and getting back to Martini. Uh like I said, last time I showcased Martini um you guys responded extremely well. In fact, actually I think it kind of like surprised them uh how many people ended up hopping on board. Um so, Martini is still in waitlist but not for you guys. So, you can just hop on with no waitlist if you use the link down below. Um plus they have expanded the first month of credits uh if you click on that link. So, if you sign up for the pro account at $30 a month, that's going to include an extra $20 of credits in the first month. And then if you sign up for the studio plan at $150 a month, that includes $200 in first month credits. Yeah, Mar- Martini's a very well-thought-out platform in all honesty and uh you know, I'd spoke with Ko earlier today and uh they they have some more exciting stuff coming up soon. So, I'm sure I'll be covering that as well. Uh in the meantime, if you want to check it out, link is down below. So, that's it for today. Uh definitely let me know if you guys enjoyed the sort of prompts and tips uh segment in the beginning there. That is something that I think uh is fairly valuable but you guys definitely let me know if you want to see more of that. Until next time, I thank you for watching. My name is Tim.
Here's some insane Seedance 2.0 prompts & tips you need to know! The Prompts are broken down so you can steal them, modify them, and build your own. From snap-stop time shockwaves to first-frame/last-frame storytelling, time-loop sequences, and an invisible VFX shot from the 1997 film Contact — this one's a tactical breakdown, not just a prompt dump. Plus we head back over to Martini for a feature I haven't seen anywhere else (and yes, you can skip the waitlist 👇). 🍸 Skip the Martini waitlist → https://www.martini.film/theoretically-media And if you use the link: $30 plan → $20 bonus credits ($50 total for the first month) $150 plan → $50 bonus credits ($200 total for the first month) 📩 Get this week's newsletter (with all the prompts copy-paste ready) → https://theoreticallymedia.beehiiv.com/ 🎬 Don Burgess on the Contact mirror shot →https://youtu.be/HQRu9cz5L9E?si=B47-tPxdbKUS8NjT CHAPTERS 00:00 Intro — Why These Seedance Prompts Are Going Viral 00:39 - Prompt Intro 00:56 Snap Stop-Time Prompt — 02:04 - My Stop Time Test 02:41 - First Frame / Last Frame Trick — Emergent Storytelling 03:19 - My Pirate Test 03:59 - Weapons Test 04:29 - Real Faces Appearing Now 04:59 - Day-In-The-Life Time Loop 05:59 - GPT Image 2 Character 06:31 - My Loop Test 06:51 - Rerolling 7:44 "Contact" Mirror 08:41 - Seedance Recreation 09:13 - My Contact Test 09:36 - How Many ReRolls? 10:08 Martini's New Studio Feature 10:48 Step Into Set Improvements 11:45 - Things get Wild here 12:00 - Step Into Set With Characters 12:45 - Character Turnarounds 13:22 - Character Space Controls 14:10 - Why This is So Good 14:55 - Turnaround Editing 15:48 - The Multiplayer Feature 16:19 - Why Multiplay? 17:08 Skip the Waitlist 18:04 Final Thoughts & What's Next