Dinosaursssssss

Of course! The last couple have been a lot of fun

Instead of making a new thread in banter I'm just gonna bump this one!!! Official Unofficial Idle Thumbs GDC Meetup 2018 Pagan Idol (thumbs) 375 Bush St (~1 mile from Moscone, 10-15 min walk) 6pm, Friday March 23

alright, callin' it: Official Unofficial Idle Thumbs GDC Meetup 2018 Pagan Idol (thumbs) 375 Bush St (~1 mile from Moscone, 10-15 min walk) 6pm, Friday March 23 I made a reservation for one of the little huts that seats 8, we'll see how that plays out. Apologies to folks who can't make it, know that I still love you, event planning sucks <3

I just realized a pal of mine in SFs art show is opening Tuesday night so I should probably go to that instead of drinking out of a ********* Alright here's some thoughts Saturday night (17th) I wanna do the tiki places that aren't downtown, I think @coaxmetal and/or @jennegatron will join, and that's probably the only time @zerofiftyone will really be able to hang with Jenna since she can't get into the city during the week. Though obv ya'll are adults and don't necessarily need me to organize your hangouts or whatever. I think for the Official Unofficial GDC Idle Thumbs Community Meetup we should do Friday (23rd) or Saturday (24th) to get out of the way of real industry meetups and stuff. I also would be down for a casual lunch thing on Mon/Tues/Weds cause it'd be cool for people to connect names/faces early on and see what people are up to the rest of the week. Like just stake a space in Yerba Buena for an hour and get take out from the mall.

The only thing about Tonga Room is it's ludicrously, prohibitively expensive, so it definitely shouldn't be the primary (or only) meetup.

oh yeah I...have been meaning to put something together. The past couple years we've done Saturday night (which'd be the 24th this year), but a good chunk of folks will be gone before then...maybe a pre-GDC meetup (Sunday night?) or mid-week (tuesday??) makes more sense. There is literally no way to encompass everyone so idk what the best time is. Also yes, I am planning some tiki...adventures. Saturday before I wanna get out of the city a bit and hit up Trader Vics and maybe Forbidden Island, during the week I would like to get to Tonga Room and Smugglers and maybe Pagan Idol. I was gonna just let those happen but maybe it makes sense for me to just declare that I'm doing Tonga Room on Tuesday.

NAILED IT

Elvaq has pointed out that someone (presumably a non-Thumb) found this game on Steam and wrote a nice comment about it on the forums. http://steamcommunity.com/app/779150/discussions/0/1620600279667227280/ It's perfect; I could not have asked for a better first post about this game. It was all worth it to experience this.

IF YOU DO BUY THIS GAME (I WILL ASSUME IT IS AN ACCIDENT) THEN PLEASE AT LEAST HAVE THE COURTESY TO NOT PLAY IT

that starting line spline setup is fresh as hell!! Very happy to see some fresh Tiny Tires content in 2018

I know some of you wishlisted this game, so I want to be very clear: IT IS A BAD VIDEO GAME DO NOT BUY THIS GAME UNDER ANY CIRCUMSTANCES

Thanks all for playing! One quick thing I wanna note is that enemies should respawn every 2 hours of in-game time; you can open the log with...F3 I think to see what the simulation is doing. It'll show when a tick happens, and enemies go on patrol, respawn, etc. I also want to reiterate that no one should play this game. Anyway, I tried to hook up a damage zone to the anomalies so they'd damage you if you were standing in them; my goal was to get something closer to STALKER, where you'd have a geiger counter type device that beeped as you got close to them, and could be activated to dispel the anomalies, but I ran out of time. Also I don't think I tested if the damage volume was working so they may do nothing. I also put an (also untested) healing zone near the guitar lady. Those both went in super last minute, and I ended up just giving you a ton of health because at the time I thought I wouldn't have time to get in any way to heal. Shooting is a bit weird because I'm sort of mashing two things together...UFPS fires raycasts from the camera center forward, and Tactical AI has bullets with physics and visible tracers that fire from the gun muzzles. I believe where I ended up was I have both happening for the players (enemies all just use the latter system) because all the guns I had in the plugin were set up as raycast only, but I wanted tracers visible. The physics based bullets had inconsistent hit detection, and the raycast ones were just wildly inaccurate at range, so as a result you're firing tracers that aren't going where you're actually shooting, and the actual shots you're taking are really inaccurate, so basically all the feedback is useless and it just feels awful. UFPS also didn't (I don't think) support what I think of as the typical firing model, where crouching/standing still makes bullets more accurate, and the longer you fire, the less accurate they are. The accuracy seemed to be consistently bad and didn't reward taking the time to line up shots. If/when I come back to this, I'm just going to remove the raycast based stuff, use exclusively bullets with rigidbodies, and probably rewrite the recoil/accuracy model.

Actually for real, I strongly discourage all of you from buying / wishlisting / playing this game on steam, or telling anyone about it. Also the store page initially failed review because "there are screenshots which contain non-gameplay overlays." This is half true; I had a screenshot of the main menu, but I turned all the UI elements off except the logo. For reference, this is the screenshot. A quick look at the store page shows a promo image with some obviously badly scaled assets, and a low resolution trailer that isn't cropped correctly. And yet, this screenshot is why it fails.

please like and subscribe http://store.steampowered.com/app/779150/

The ice came out looking great, can you clarify what you mean by "blending textures/colors based on normals"? It's not really clear to me what's going on from the images. Also, how'd you like ASE?

Did you like how the guitar amped up a bit right then too?? Definitely intentional and not just a great coincidence (though it probably cuts a couple seconds too early) Anyway another product review from me is that Windows Movie Maker is garbage. I don't get why they never bothered making something to compete with Garage Band or (in any serious capacity) iMovie. Really feels like a pretty easy win to me, especially since with the Surfaces they've been going after a more artsy folk. Shrug. edit: it was incredibly boring and I hated it but I got all my various store page images cropped and stuff. Couple days for Steam to review it, 2 weeks on the store listed as 'Coming Soon' and it'll be out. Here's some of the fun that awaits you if you wanna upload a game to Steam: Generating store images at the following completely insane resolutions: 460 x 215 231 x 87 467 x 181 616 x 353 Generating community page images at the following resolutions: 444 x 208 184 x 69 Generating a 32x32 .ICO icon file. Uploading your game via the SteamPipe Build Uploader. This app, which looks like it was made in WinForms 15 years ago, is the easiest, sleekest way of uploading your game to what is currently the biggest storefront in games, made by a company that makes more money than every user on this forum combined:

alright actually turns out they wisely don't let you instantly put a game up for sale, so I have to wrap up making assets and wait two weeks. I forgot how much of a pain in the ass this is, I don't have all the promo images and stuff I need. So, one more day... I DID cut this surprisingly passable trailer though. And yes I know it's cropped weird, my OBS settings were not ideal and I'm definitely not gonna spend the time to redo it.

Extending asset store plugins has been a problem for me pretty consistently...I think it's worthwhile to get a working version of something in relatively cheaply, but for any core system like 'the shooting' or 'all the AI' I feel like you're basically guaranteed to come across situations where your design will be at odds with SOME portion of the plugin? Professionally I've also run into a ton of situations where plugins were clearly never intended for use in a large scale game, and have either performance or source control issues (or both). In terms of ease of use / commenting, the plugins I used for this were well made and had pretty solid docs. I only really dug into the AI and UFPS and both were commented, not amazingly, but...enough (with maybe the exception of the inventory stuff in UFPS). Not perfect, and Tactical AI in particular I think assumed you would spend the time poking through their samples to set things up rather than just reading the quickstart. Also for all the difficulty it gave me, setting up blend trees and AI that works at all on arbitrary characters like it does is pretty impressive. Now that I'm really thinking about it, the worst were the 3D assets...no LODs, blank Animator components left on prefabs, high poly physics colliders. One had a sample scene with a bunch of walls without back faces, so you'd bump into invisible walls all the time. That one (it's the brick warehousey stuff) also had buildings built to a weird scale, so a large warehouse was about 3m tall. Another didn't have a single prefab, just a folder of fbx's. And I don't think either of those had any metallic maps or atlassing. If I was going to continue working on this I would probably delete the cityscape, set all the prefabs up properly, and start over because it's a disaster right now. Alright anyway I'm gonna quit blabbin and go cut a trailer and put this thing on Steam.

Alright, game's submitted to the jam page! https://dinosaursssssss.itch.io/shadow-of-something Came out alright. I'll need to make a quick video and screens before it goes up on Steam so that'll probably happen tomorrow. For now I'll say that I think the stuff I bought laid down a pretty good foundation. As expected the rigging/animation working right took the most time; the Tactical AI plugin was a very solid first 90%, but getting the last 10% of the way was going to be a serious undertaking. The FPS plugin likewise had some great features that got me a long way, but I can't help feeling like what I paid for is a bunch of really minor stuff that doesn't contribute, and often interferes, with making interesting games. What I think I mean by that is there were a ton of great features in UFPS...weapon sway, recoil, inventory systems, muzzle flashes, etc, but if I had specific needs that didn't quite fit, implementation took longer because I was working in a codebase I only have a loose understanding of. And as a result it just didn't make sense to work on those features, but even though some of them were minor (bullet drop, leaning) I feel like it just drags the identity of the game closer to Average. And getting nice procedural weapon animations (they really are pretty good) probably isn't worth the trade off? Admittedly with more time I probably could have gotten that stuff in, and I have no idea how much I could have got done without these tools in a couple weeks, but it feels like I had to make some serious compromises. Anyway, when I started fiddling with the faction system and having AIs that patrolled areas, and actually started building the simulation the game really became interesting to me. I'd sort of like the scrap this and really rebuild the game around some of those systems. Gonna be pretty busy up through the spring, but post-GDC I'd like to come back to this and flesh out the simulation. Also the weather and environment plugins were rad and I'm gonna definitely use them again (Envirosky and Gaia)

Alright, I've now played Glitch Simulator. I'm happy to report that Shadow of Something (kinda runnin out of time to come up with a better name...) is a better game than Glitch Simulator. GS is arguably a little more feature rich in that it has a menu and settings so I have a little more to do. I actually might keep the name, unless I'm really inspired. Part of me would feel bad releasing for real with a Thumbs ep title, but I think it's kind of perfect? The format calls back to STALKER: Shadow of Chernobyl, and it is but a shadow of the greatness of that game. The light stupidity of the 'of Something' also serves the sorta half-assedness of the game as well.

Oh also I wasn't paying super duper close attention when I exported these animations, and I think I made the arms too long, which fucked up the auto rig a bit, resulting in this wacky wrist action: Probably not a big enough deal to bugfix. Also with the new way mixamo is set up I don't know if I could fix the rig anyway? I'd have to go back into Fuse, shorten the arms, re-export, and download/reimport everything. I think they changed the layout to reduce costs on themselves (I don't think you can save characters or custom fbxs you've uploaded anymore?) but it makes this kind of iterative work very time consuming.

Alright I figured out the save/load issues and persisting weapons stuff; it's really strange to me that it isn't a feature that's included. When people were asking how to do it on the UFPS forums the developer was like, 'oh interesting use case, ill have to see if its possible' and never replied again. Very weird. It was also actually a pain in the ass to implement; the code itself isn't bad, but the nomenclature in the plugin is sorta hard to follow. For example, a shotgun is an item that has an item bank of items of type shotgun shell. When I save, I have to record the contents of inventory.UnitBankInstances and inventory.InternalUnitBanks to get the weapon unit banks and ammo unit banks respectively. It makes sense in that it decouples weapons from ammunition, and lets you share ammo between types of weapons, but it was tough to dig through. In the end, for simplicity, I persist your ammo count but refill your current clip.

As we close in on the final here, there're a couple things I wanted to do to add that might not make it: Saving your equipment and the current faction status! And also add a new game/continue button up front to make that usable. Currently I'm trying to figure out some weird behaviors with UFPS's inventory system; it seems like if I pick up a gun, it doesn't add the ammo from that gun to my inventory, it adds a second gun with a full clip that I can't select. Along those lines, I also sorta wanted to make a Metro style bullet economy where the base pistol bullets are the standard currency and you can buy guns/bullets/whatever using them. I'd also like to actually play a Digital Homicide game (or the glitch simulator thing) and see if there's any other polish I ought to add in (menus maybe). I actually can't find their games for sale any more though; Jim Sterling may have destroyed them? I have about...a quarter of an anomaly system built that I'd like to wrap up; throwing bolts like in Stalker might actually be the most efficient solution. I wanted to do more of a geiger counter type device but I don't know if I'll have time (and I definitely don't have any art). There's no ending, but that's probably fine.

I added a rough pass at a faction simulation: Each faction has a strength rating. Every 4 hours (and at the start of the game) the SLife system spawns enemies until each faction has the same number as agents as their strength rating. During this tick, we also take 50% of their agents and randomly give them a new behavior to either patrol (walk between their own factions outposts) or search (walk between friendly and enemy outposts). Every time an agent dies, its faction loses .2 strength. Every time a faction captures an outpost, they gain 5 strength. Outposts are captured by counting the number of agents nearby and assigning the outpost to the faction with the most agents nearby. This is a lot of fun to noodle with, but I think the best thing I could do right now is fix damage numbers and tweak things so the game is actually playable.

I've been watching Futurama while I work, here's a gif I made