Pushing Engine Limits?

I’ve been playing SC for about 10 years now, and, er, I was wondering about the current engine limits. I am not sure if remastered uses the same engine as classic BW (probably does).

If so, it seems the limits can be pushed a bit, if the extended unit limits work as intended. This being said I’d like to see the possibility of-

  • Larger maps, larger than 256x256
  • Extended command limits (200+)
  • Larger player numbers (more than 8, perhaps 12)
  • A larger unit selection cap (more than 12)

Judging from the new content presented in Remastered this seems plausible. Thus I’d guess the only reasons it wouldn’t happen would be due to keeping the original feel of the game or simply lack of developer interest.

1 Like

Yes it uses the same old engine so everything you point in here is impossible today.

“Modernizing” and engine can be a very difficult task to cover because by doing so people tend to brake gameplay and also dynamics that the game and players have been using for years, its a very hard thing to accomplish “modernizing” an old engine without losing some of its features. We tend to think that behind every game there is a huge team of people working when in reality most of the time they are just humans with very specific goal and task set in terms of production, they have to be efficient overall.

There is a very obscure way of modernizing the engine, but it is both very expensive and hard to accomplish by regular team of Computer Scientist, its a Technic called “Cloud of Bits Re-engineering” which today is a term lost because the word “Cloud” now means server side hosting for virtualization software. This way of doing things has been used when companies want to migrate data from a very old server to a modern one but they are afraid the data will create bugs when runner on a new system or interface, i have seen this working on an old bank server in Colombia back in 2012 where they needed to migrate an old database from a bank that closed that was working on a very obscure Unix server implementation from back in the early 80`s, not only that, it was not only about the “data” but also the “services provided by contract” that the old clients where promised to have and if not delivered correctly they would break the contractual bond between the bank and the user, this “old” users where expecting for the new bank that acquired such accounts to make the smallest error to change to another bank, but the new bank was not going to let that happen so they called the Computer Science police.

So the Technic goes something like this, a very smart set of programmers create a virtualized environment of the old system, both hardware, firmware and software wise, and because it was such an old system they CLOCKED the times up so they could emulate transactions at higher speeds, at the same time a neural network was “learning” how all the bits of this virtualized system where being produced, set to registers, changed, saved in memory etc, so after a few weeks the amount of new “Bits changes” or new “bits configurations” that the neural network was seeing was around 98% (meaning that for this computer there was really nothing new to see), and thus is CLOUD OF BITS was born. A comprehensive massive set of bits which where the actual limit in the real of possibilities of any transactions made inside the old system.

With this Cloud of Bits now the migration started, the machines it was compiled into both a compiler and a real time Debugger, so when programmers where now creating a modern framework for how to use the data and migrate it to a modern x86 architecture they would find that when trying to compile small units of work they would get a very cryptic error of the type “Overflow for exception number…” and you could go to the Neural Machine documentation and check “oh that number represents how account are changed between bank deposits” and so on… i worked at that point and i can tell you it was the most annoying thing that i did in my life.

So after almost half a year of working with those “constrains” we compiled the new database GUI and web service and everything together, it worked !! but now it had to be tested with the Neural Network Cloud of BIt and it found a small percentage of errors so we had to do it back again until we hit close to that 98%…after almost a year and a half since they started (i only worked for 4 months) we finally did it. The interesting thing is that now the bank could also EXTEND the services provided to the old clients, meaning they could have modern usages for this while keeping the old contracts intact and only extending some clause (which is the main reason why the bank refused to just migrate the data, they where not going to renegotiate contracts). To put it in perspective if it wasn’t for our Computer Scientist team they would have preferred not to purchase the accounts at all, because it was too financially risky.

So this system was very crucial for the banks migration of and old system to the new one, at the end it worked well, for anyone reading this you have to understand that the bank system was much more complicated than a game engine, for instance and just as an eye opener, old clients in the bank where depending on dividing their account automatically in different wallets at a very high speed with different currency’s, while at the same time they where being managed by more than one user at the same time, trust me, it was a hell to do for anyone not using the Cloud of Bits paradigm.

To the point now it is possible to use the same Technic for the Starcraft Broodwar engine, creating a comprehensive Cloud of Bits neural network data base and then create all the tool to extend into those functionalities you mention, hell, this could also “theoretically” could be use in pair to some other sort of emulations that could make the NetCode and the code itself just tiny bit more efficient tho i dont expect this last thing to have any real meaningful product, maybe i am wrong and there is a HUGE improvement.

So yes it is possible without breaking the old gameplay, dynamics and aesthetics that Starcraft Broodwar has, is it easy? HELL NO, is it cheap? NOT AT ALL, is it necessary… i would say yes because at some point Blizzard would have to take all this game into a CLOUD service (and yes this time i mean Video Game cloud service) and by doing so they will have to modernize it quite a bit specially the NetCode.

1 Like

You have several good points. It’s amazing that they could potentially do it using the method you described working on (which, by the way, is quite nice).

Personally I feel the aesthetic of the game is lost for the most part with the new colors and the grossly spliced official ranked maps.

However, I do suppose that more drastic changes to the engine to allocate new space would be too drastic for the devs’ ideas, and they certainly have enough on their plate already.

Thanks for explaining all this stuff.

1 Like