Senario Migration to CMx2

[J Ruddy]

I thought of the individual light sources on my drive home. Dang! I wish I would've posted it & sounded intelligent for a change...!

As for the engine - I am dumbfounded.

You guys are nuts (in a good way) & I bow in your general direction.

I buy into #1 and #4 and by way of #4, #3. As for #2 optimizations - I am not so sure - polygons are polygons - but frankly, I don't give a damn because you guys (as usual) are doing the *right* thing - not the *thing other people do*

you are good enough, smart enough and Doggone it, people like you...

big smile here - I'm going to sleep tonight with visions of paraflares dancing in my head...

[Battlefront.com]

The thing about optimizations is that there are more ways to get the polygons to do stuff than you might think. Each way comes with its own pros and cons. If the game is tweaked out for FPS then it might not be possible to get the polies to do what you want simply because the code is geared towards speed rather than flexibility. Or something that seems like a good idea for RTS might not work for FPS. Again, all depends on what the designers had in mind as "optimal" when they coded stuff. An engine that assumes LOS to be no more than 500m probably took a lot of shortcuts and undoing them might be impractical.

Steve

[J Ruddy]

Just so you know where my worries and assumptions are coming from - I saw CMBO at a gaming convention (you know the kind of place where men are odiferous and women are nervous) and I didn't run out (log on) and buy it because the graphics were pretty far behind what was considered average at the time and I couldn't get past the graphics to give the game a chance.

I did buy CMBB and CMAK, although the graphics were still a bit weak compared to some of the games coming from the big name studios.

I understand there are CPU & GPU limitations to take into consideration and so on and I was a happy camper when CMBB ran on my 2nd machine's on board (shudder) shared memory video card.

I know at the time CMBO came out, the Graphics and Game engines that were publicly available were crappy and expensive.

However, I have played with a number of demo versions of professional engines in my own experiments and man some of them are pretty sweet. RTS or WEGO - it's all moving a lot of pixels, the big difference that I can see is that WEGO has an advantage when it comes to processing power - all the moves are precalculated, so you aren't hogging up CPU cycles during graphical processing.

But the thing that I may be out to lunch on is that if you're using DX3D or OpenGL 3D functions, most of the graphical processing takes place on the GPU on the video card, not the CPU...? So in fact the number of polygons in a RTS graphics system, like with the TOTAL WAR games, is limited not by how much spare processing power you have, but by the graphics card's processing power? So my question is why would RTS allow for significantly less detailed models than a WEGO system?

I don't know why I'm getting defensive here - it is a very good thing to have total control of the Graphics engine, if for no other reason then to have the code 'in house' for fixes and enhancements...! I just hope it is scalable to take advantage of graphics hardware advances so that you won't have to write a new engine yet again in 2008.

I also hope it is as good as I imagine it is going to be.

—Han Solo I don't know; I can imagine quite a bit."

[Sergei]

Steve,

so no heightmaps either? Dang... you know, the map editor could at least be a bit more ergonomical if you don't plan to support 3rd party software. Ideally, you could edit the map in the 3d preview mode.

Originally posted by Sergei:

It'd be nice if they'd allow the importation of heightmaps.

But somehow, I suspect Steve will be just as lukewarm to all these suggestions as he's been before. I mean, it'd only make scenario making a bit easier, but as 99% of players just play Quick Battles against the AI, it is more worthwhile to model stellar charts. :mad:

[Battlefront.com]

J Ruddy,

Remember, you are comparing apples to oranges. The better looking graphics engines are that way because they are optimized for things like highend hardware, limited distance shown, hardcoded max units, predictable polygon counts, predictable texture needs, simplistic gameplay, etc. etc. etc. They also look better because the game company's art departments are 4 or 5 times lager than all of Battlefront.com put together. Yup, when you spend $1,000,000 on art it does tend to make the game look better

John Carmack, the man behind DOOM and Quake, would have been hard pressed to make a significantly better looking CMBO with the specs we were looking to satisfy. CMBB and CMAK could have looked a lot better if we had rewritten the game engine for each (which is what the big companies do for each release).

Unfortunately, this is not possible to communicate to the average gamer because their understanding of what is possible stops only at what is displayed on the screen. They have little to no comprehension what it takes to make the screen display those images.

Having said that, a heck of a lot has changed over the last 8 years When we started we had 4MB of rather slow VRAM to work with, now we have 128MB and higher. 8 years ago processor and subsystem speeds were not all that different than they had been 9 and 10 years earlier. Since then several families of CPUs and subsystem components have been introduced with processor speeds now 10 times what the high end used to be back then. So on and so forth.

What this means is that the CM environment will not be as highly detailed and pretty as the games that have budgets that run into the millions of Dollars, but it will be darned close. And we don't think anybody could do a seriously better job.

Steve

[Battlefront.com]

Sergei,

Yeah, in a perfect world we would have import/export tools that interact with various map data APIs. Unfortunately, it just isn't in the cards for CMx2's first release or so. Same for a 3D realtime editor. Possible, for sure, but extremely time consuming to code, test, and tweak. It isn't even on the design doc list, as desirable as I know it to be.

Steve

[Sergei]

I don't really mean that, but just being able to import a grayscale bitmap.

[J Ruddy]

Thanks Steve - I'll shut up about this now.

(Dang, I'm starting to get that deep routed longing for a screenie again...)

[Claw]

I had no idea that a Scenario Migration question would morph into a star chart discussion.

Thanks, Steve, for your definative answer. Believe me when I say I understand data migration challanges between versions. Focus on what's important - the new version.

As for Star Charts in CMx2: instead of playing a night scenario to see the stars, how about waiting until night, going outside and actually LOOKING up at the REAL night sky.

[Battlefront.com]

As for Star Charts in CMx2: instead of playing a night scenario to see the stars, how about waiting until night, going outside and actually LOOKING up at the REAL night sky.

I do that all the time I'm blessed with extremely low levels of ambient light where I live. A friend's place I used to take care of didn't have a man made source of light for 20 miles and on the other side of a mountain. Talk about star power!

Steve

[Dillweed]

Well, I imagine that if the setting is appropriate we will see some recreations of some of the more famous maps from the CMx1 series (Last Defense, Chance Encounter, etc) courtasy of the mod community.

[Scipio]

Originally posted by Claw:

... instead of playing a night scenario to see the stars, how about waiting until night, going outside and actually LOOKING up at the REAL night sky.

Now that's really a weird idea! With what will you come up next? Real daylight??? :eek:

[flamingknives]

Look up at the night sky?

Err, dull and slightly orange?