HOME | DD

Published: 2011-02-13 09:41:44 +0000 UTC; Views: 12209; Favourites: 75; Downloads: 255
Redirect to original
Description
Starcraft 2: Terran Science FacilityModeled & Textured by Phill Gonzales
The Terran Science Facility was on my wishlist of Terran buildings I wanted to personally see return in Starcraft 2. For a short time it was on the multiplayer tech tree in Starcraft 2 until it was eventually replaced by other buildings such as the Ghost Academy and the Fusion Core.
During the Wings of Liberty single player campaign, it briefly returned as the buildings that needed to be destroyed during the dropship level. Unfortunately, after the artwork was made the map was revised and the Science Facilities were replaced by the Data Core building artwork.
Thankfully the Science Facility finally returned and even unlocks the secret mission in the single player campaign, it's in the bottom right of the "Media Blitz" campaign map.
I was very inspired by the work of the Project Revolution mod team when I made this model. Project Revolution created a Starcraft 1 mod in Warcraft 3, and alot of the artwork they made is pretty impressive. It was very interesting seeing how they defined some of the shapes from the vague Starcraft 1 sprite artwork. When the Science Facility was finally built in 3d, I took the lead from many of their shapes while updating the artwork.
[link]
© 2010 Blizzard Entertainment, Inc. All rights reserved.
Related content
Comments: 12
Randroid7 [2015-10-16 17:10:26 +0000 UTC]
Want access to the secret intel without getting massive mechs to blow this beautiful building up? Get a giant screwdriver and screw off it's lid of a roof! :3
👍: 0 ⏩: 2
LordDarkstarr In reply to Randroid7 [2016-03-03 07:46:14 +0000 UTC]
No, wait--that's where you plug in your iPod! See, it's a functional dock with awesome sound and cool light-up effects! But wait, there's more! If you order now, we'll even throw in a week's supply of vespene absolutely free!
👍: 0 ⏩: 0
soundsnc [2011-07-22 16:47:57 +0000 UTC]
It really means a lot to myself and my team mates to have developers of such a prestigious game give shout outs like this.
👍: 0 ⏩: 1
Boarguts In reply to soundsnc [2011-07-26 03:51:34 +0000 UTC]
It's well deserved praise! Don't stop kicking ass!!!
👍: 0 ⏩: 0
soundsnc [2011-07-22 16:35:58 +0000 UTC]
Hahaha...
Thanks again for the kind words man, I was the art lead on the Project Revolution team. We redid most of our models several times. Terran, while not perfect, was our closest replica race by time we ceased development.
👍: 0 ⏩: 1
Boarguts In reply to soundsnc [2011-07-26 03:52:59 +0000 UTC]
Out of curiosity, what is the turnaround time for the custom art you're working on? Also, is the model iteration based on changing game design or just finding the right look for the visuals?
👍: 0 ⏩: 1
Boarguts In reply to Boarguts [2011-07-26 03:56:01 +0000 UTC]
Also, about game design changes I guess i'm more specifically asking about the process of translating Starcraft into the 3d engine used for War3. We essentially did the same thing when building SC2 and it was a unique challenge to make those sprites carry over into 3d and still have the same impact...
👍: 0 ⏩: 1
soundsnc In reply to Boarguts [2011-07-26 21:38:38 +0000 UTC]
From a game play perspective, we had team members who's only job was to test things in SC1 and see how it reacts, record data and bring it back to our programmers.
We'd do interesting tests such as placing 50 marines spread in a specific distance of units, shooting a reaver scarab analyzing the results and replicating it.
Other aspects such as unit acceleration and movement speed we pulled directly from Starcraft 1's data.
Some of the challenges were the unique properties of some of Starcraft 1's damage and splash effects such as the firebat cone. It was really weird how our tests showed us the firebat cone worked. It would sort of bounce and fly back a few pixels to the sides.
We had huge challenges when translating to WC3 primarily due to WC3's hard coded limitations. Starcraft 1 for instance had an advanced build menu while WC3 did not. We would have to use the chaos ability to swap the builder unit for another one (each had their own separate build menus) in order to get a basic and advanced build menu. Only issue was that unit swapping via JASS is slow, especially in multiplayer so hotkeys worked sub-optimally.
We had many fairly clever solutions when using the WC3 engine but in the end we were met by things that simply could not be solved. We had a sick carrier system working but all those interceptor move orders being issued would lead the WC3 engine to queue movement orders. So when a player had 12 carriers attacking a base and the enemy tried to grab his units and counter...they wouldn't move...because the interceptors had WC3 bottlenecked on move orders. I could go on for ages with the amount of challenges we faced (such as the engine containing no ability to do cloak pixel ripple warp effects) but this message is getting long.
I'm sure you updated one of your existing engines when developing SC2, but we had no access to hard code, merely JASS solutions unfortunately (slow). We actually had our own launcher written so that we could intercept some storm.dll calls. This was useful for adding things such as unit wire frames to the engine.
From an art design and turn-around perspective. We'd often cheat in certain ways. We came to notice that some things were only meant to be seen from a certain angle. So we would billboard features of a model to always face the camera. For example, we decided to not allow buildings to turn when flying just because we felt that our sc1 designed buildings looked rather poor on the backsides since they saw less attention. Our timelines were dependent on the artist. Being a modding team we never had concrete deadlines. People did things because they loved what they were doing. Some artists took a month, others would take a day. Generally I'd say a week to two weeks. The biggest challenge was the harsh polygon limitations. Most of our models had to be around 500 triangles. Some ended up higher, but not by much. We had no ability to use normal mapping or specular mapping techniques to emulate some of the more dynamic effects we found to be in the BW sprites. In the later phases of our development, we had refined our art style by developing high poly models and baking all of the details directly to the diffuse map (ao, spec etc). It was static lighting, but it was the best we could do.
In the end, I love that you guys changed some of the game design up, even if it was a solution for some of the difficult translation problems (trust me, I understand). Starcraft 2 was meant to be a new game anyways so as always, new game design and balance is welcomed. I certainly wouldn't be disappointing if Blizzard were to release a "Starcraft 2: Broodwar" edition though. With the experience and development time you guys have put in at this point, I think it would be a lot easier, even keeping the art style as is and going for the gameplay aspect.
👍: 0 ⏩: 1
Boarguts In reply to soundsnc [2011-08-17 04:14:03 +0000 UTC]
This was one of the most informative comments I've had, and sorry for not responding sooner but i'm looking forward to the tech art team reading this and feeling good about some of the innovations they've developed over the War 3 engine.
To pick a quick topic though, I'm looking forward to the execution your team provides on art assets now that the polygon limit has been raised so dramatically from the War3 engine specs. Our own developers often had concern over increasing polycounts when they modeled despite the engines ability to handle more mesh information... The liberties of the new spec are so generous that it feels like robbery to put 5000 polygons into something as numerous as a Zealot but the engine handles it gracefully. I'm interested to hear your workarounds and philosophies regarding alpha maps and draw-calls in the SC2 engine. I've found that modeling in detail rather than using an alpha map has been more beneficial just for the sake of adapting to the processing power of the engine. Other circumstances such as unit population have also had a bigger impact than the War3 constraints for low poly models regardless of how many are on screen at a given time.
It will be interesting to say the least. your team is the most ambitions project we've seen so far and i can honestly say we're very curious to see how a outside development team adapts and utilizes the engine.
Once again, sorry for getting back to such an informative comment so late, but it was great to get so much insight and as I've said, I'm a huge fan of your previous efforts and looking forward to your seeing what new things you'll be accomplishing in the future!
👍: 0 ⏩: 0
smurfbizkit [2011-02-14 18:04:47 +0000 UTC]
The Project Revolution comment is awesome, its great when we see dev's inspired by modders (and actually admitting to it). It reminds me of when I visited EALA for a community summit for 'Command and Conquer 3'. They had this wall filled with inspiration and reference art, everything from GI Joe to old C&C concepts...but also right in the middle 3-4 printed shots of models from a (eventually killed by microsoft) halo-to-rts mod my friend had made. The overall GDI style ended up having a similar look to his take on halo-inspired designs.
👍: 0 ⏩: 1
Boarguts In reply to smurfbizkit [2011-02-14 20:01:06 +0000 UTC]
I have to give the Project Revolution guys major props, they were going through all the same pains we were in updating the artwork for a 3d engine. The whole reason I got into games was because of the mod community, and I agree that it's totally inspiring to see independent artists rocking out making their own games.
👍: 0 ⏩: 0