70 replies to this topic

[LordNothing]

RickySpanish, on 09 May 2018 - 10:10 AM, said:

A few things in this thread triggered me, but I'm just going to bite down on this particular bait for now.

Memory leaks are not good, they are atrocious.

Furthermore, HBS used Unity of all things to develop this game which may have led to a faster development time, but is also the root cause of every major reported bug.

Long load times? You have Unity's craptacular asset management to thank for that - just check out the "Unity standard" file structure in your Bt install folder, and wheep.

Multitudes of loading screens, plus having to load the "mission start" screen before loading your actual saved battle, plus the main thread blocking when loading? That's Unity's amazing scene based game design system right there.

Bugs with not being able to hunt down all reinforcements / targets? I wonder if those units didn't fall out of the world?

Don't get me wrong, I love BT, but as a game developer who has had (and continues to have) the misfortune of using Unity, it's pretty clear that it was in many ways, a poor choice for engine.

FYi in case it hasn't been mentioned, dropping texture detail and turning of cloud saves speeds loading up a tad, as does removing old save games - I'd love to blame Unity for taking its time to populate the saved game list, but that's probably all on whatever muppet decided to parse every save file to validate it instead of just the header...

again, sarcasm.

also, not that im a fan of rent an engine development. its always used as an excuse not to have a few good programmers on staff, and the inability to fix problems is the result. this is a problem with the dev team, not the engine.

unity doesn't have these problems in other games. you have to throw a pile of mods at kerbal space program to get it to crash. granted ksp is a mature title and had a great dev team (hbs are noobs in comparison). unity games have been pretty stable for me (a lot more stable than cryengine for sure).

Edited by LordNothing, 09 May 2018 - 04:59 PM.

[RickySpanish]

LordNothing, on 09 May 2018 - 04:06 PM, said:

again, sarcasm.

also, not that im a fan of rent an engine development. its always used as an excuse not to have a few good programmers on staff, and the inability to fix problems is the result. this is a problem with the dev team, not the engine.

unity doesn't have these problems in other games. you have to throw a pile of mods at kerbal space program to get it to crash. granted ksp is a mature title and had a great dev team (hbs are noobs in comparison). unity games have been pretty stable for me (a lot more stable than cryengine for sure).

HBS certainly aren't noobs, most Unity games are a hot mess. The fact that they managed to get a game out of it is an astonishing achievement. There's a great deal of complexity to BT, lots of stats and events to keep track of, plus the combat mechanics themselves, not to mention multiplayer mode. Unfortunately, Unity just does not lend itself to professional game development, particularly if you are a developer who likes to do things in the generally industry accepted way. Another example: Unity was built around the idea of separating your game out into discrete scenes, with all objects in the world having a physical presence, and everything you see being composited together from behaviours. A light might have an illumination and a light switch behaviour. Sounds good, right? Well, what about behaviours that have no physical presence? Where do you put your game data loader? Where does your event manager live? Do you just create an invisible, dummy object with all of your "meta" behaviours on it? What happens when you change scenes? Presumably, you also need this God object to never have any duplicates, so you put it into a splash scene that runs once. Ok, great... But now you want to playtest one scene with just a subset of the features from that God object... And things just spiral out of control. There are workarounds of course, but that's the issue - they are workarounds, to problems that should never have existed in the first place. When you use Unity you fight an uphill battle, and you fight more or less every good programming practice you learnt. Personally, I started creating loads of Singletons. Disgusting, but it gets the job done. And now, factor game state loading and memory management into all of this (C# uses a garbage collector so, oh boy, we all know how fun GCs are).

[LordNothing]

RickySpanish, on 09 May 2018 - 05:30 PM, said:

HBS certainly aren't noobs, most Unity games are a hot mess. The fact that they managed to get a game out of it is an astonishing achievement. There's a great deal of complexity to BT, lots of stats and events to keep track of, plus the combat mechanics themselves, not to mention multiplayer mode. Unfortunately, Unity just does not lend itself to professional game development, particularly if you are a developer who likes to do things in the generally industry accepted way. Another example: Unity was built around the idea of separating your game out into discrete scenes, with all objects in the world having a physical presence, and everything you see being composited together from behaviours. A light might have an illumination and a light switch behaviour. Sounds good, right? Well, what about behaviours that have no physical presence? Where do you put your game data loader? Where does your event manager live? Do you just create an invisible, dummy object with all of your "meta" behaviours on it? What happens when you change scenes? Presumably, you also need this God object to never have any duplicates, so you put it into a splash scene that runs once. Ok, great... But now you want to playtest one scene with just a subset of the features from that God object... And things just spiral out of control. There are workarounds of course, but that's the issue - they are workarounds, to problems that should never have existed in the first place. When you use Unity you fight an uphill battle, and you fight more or less every good programming practice you learnt. Personally, I started creating loads of Singletons. Disgusting, but it gets the job done. And now, factor game state loading and memory management into all of this (C# uses a garbage collector so, oh boy, we all know how fun GCs are).

problem is it the good days of olde, game developers painstakingly designed an engine in house that met the precise requirements that their game needed. developers today are like kindergardeners gluing mass produced bobbles together to make something resembling a game. they go in thinking they can bend the code blob to their liking to meet their goals. when they fail to do this they make excuses, blame engine limitations and no quantity of kindergarten paste can fix it for them. you end up with a game made out of dirty hacks and messy workarounds.

ive used unity before, and while i have yet to find fault with the engine itself. the practice of gluing massive blobs of libraries together to make a game tends to lead developers down a path of broken dreams and shattered promises. maybe hbs arent noobs but they certainly aint masters either. i would prefer developers write their own damn code, but this is the world we live in. unity is still superior than the engine mwo uses. it is kept up to date and is at least somewhat modern. its not ue, but who cares. it does what it says on the tin. hbs has bugs to fix, nothing more.

everyone knows oop slows things down, thats why i like c where its all moving memory around, or c++ which while providing objects, doesnt force you to use them for everything. thats incidentally the thing i like about lua too. you arent forced into paradigms you don't want to deal with. also every game has to deal with dynamic memory, its just a thing you need to do.

Edited by LordNothing, 09 May 2018 - 10:53 PM.

[meteorol]

It shows were PGI stands as a developer of a battlemech themed game: a lot better than people give them credit for.

People constantly bashed PGI for producing a buggy, badly balanced, badly optimized game with stale gameplay and horrible loading times etc. When BT was announced, people praised HBS as saviour of BT, drooling over the "real BT experience, not the crap that MWO is" they are about to deliver...

Now we got it, and HBS delivered... a buggy as hell, terribly optimized game with loading times worse than MWO could ever dream of, with stale, repetitive, one dimensional gameplay and terrible balance.

Still, the mechdads are praising it like the second coming of christ, because it's the single player experience they hoped for so badly. Yet, if someone is able to see past the hype and his rose tinted BT glasses, he will see it for what it is: a mediocre turn based game. Seeing how the "saviour" fell flat on his face aswell, it makes PGI look a bit better. Or better: it gives a more realistic view on what PGI has actually achieved with MWO.

Edited by meteorol, 10 May 2018 - 12:26 AM.

[STEF_]

meteorol, on 10 May 2018 - 12:25 AM, said:

It shows were PGI stands as a developer of a battlemech themed game: a lot better than people give them credit for.

People constantly bashed PGI for producing a buggy, badly balanced, badly optimized game with stale gameplay and horrible loading times etc. When BT was announced, people praised HBS as saviour of BT, drooling over the "real BT experience, not the crap that MWO is" they are about to deliver...

Now we got it, and HBS delivered... a buggy as hell, terribly optimized game with loading times worse than MWO could ever dream of, with stale, repetitive, one dimensional gameplay and terrible balance.

Still, the mechdads are praising it like the second coming of christ, because it's the single player experience they hoped for so badly. Yet, if someone is able to see past the hype and his rose tinted BT glasses, he will see it for what it is: a mediocre turn based game. Seeing how the "saviour" fell flat on his face aswell, it makes PGI look a bit better. Or better: it gives a more realistic view on what PGI has actually achieved with MWO.

BT is just running very fine in my pc.

Ripetitive and one dimensional?
The campaign is not at all..... but maybe you are here only to praise pgi....
so, goodbye

[B L O O D W I T C H]

DovisKhan, on 07 May 2018 - 11:52 PM, said:

even after having sunk well above 1000+ hours into it.

>1000+ hours

[meteorol]

Stefka Kerensky, on 10 May 2018 - 12:40 AM, said:

BT is just running very fine in my pc.

Ripetitive and one dimensional?
The campaign is not at all..... but maybe you are here only to praise pgi....
so, goodbye

Top
f*cking
kek.

Feel free to check my post history if you think i'm here to praise PGI. I'm seeing things objectively, which most people not seem able to when it comes to mechs in general.

Not repetitive and one dimensional?

What mission type in BT doesn't come down to "knockdown spam all enemy mechs and be done with it"? That's literally it. Load into the map, knockdown everything, do whatever irrelevant objective there is after nuking all enemies with knockdown and precision strikes. That is how deep BTs gameplay actually goes. Load into the tiny map, move 1 inch until you detect enemy forces, knockdown spam, rise repeat.

Edited by meteorol, 10 May 2018 - 01:24 AM.

[Nema Nabojiv]

meteorol, on 10 May 2018 - 12:25 AM, said:

Yet, if someone is able to see past the hype and his rose tinted BT glasses, he will see it for what it is: a mediocre turn based game.

I'd say its more of a "good but non great" turn based game at its core. There are tons of highly hyped games worse than BT, and as for plainly better ones that I played, I can think only of X-Coms

Edited by Nema Nabojiv, 10 May 2018 - 01:30 AM.

[STEF_]

meteorol, on 10 May 2018 - 01:23 AM, said:

Top
f*cking
kek.

Feel free to check my post history if you think i'm here to praise PGI. I'm seeing things objectively, which most people not seem able to when it comes to mechs in general.

Not repetitive and one dimensional?

What mission type in BT doesn't come down to "knockdown spam all enemy mechs and be done with it"? That's literally it. Load into the map, knockdown everything, do whatever irrelevant objective there is after nuking all enemies with knockdown and precision strikes. That is how deep BTs gameplay actually goes. Load into the tiny map, move 1 inch until you detect enemy forces, knockdown spam, rise repeat.

then top KEK here too, if you see BT like that!

LOL

[B L O O D W I T C H]

meteorol, on 10 May 2018 - 01:23 AM, said:

What mission type in BT doesn't come down to "knockdown spam all enemy mechs and be done with it"? That's literally it. Load into the map, knockdown everything, do whatever irrelevant objective there is after nuking all enemies with knockdown and precision strikes. That is how deep BTs gameplay actually goes. Load into the tiny map, move 1 inch until you detect enemy forces, knockdown spam, rise repeat.

Kinda true, it feels a lot more immersive, tho. All the little details add up. Salvage mechanic, trashtalking with pirates. The litte choices for 30 day bonuses with the crew. If only MWO had a bit of the BT immersion Battletech got on the table.

[STEF_]

Toha Heavy Industries, on 10 May 2018 - 01:29 AM, said:

Kinda true, it feels a lot more immersive, tho. All the little details add up. Salvage mechanic, trashtalking with pirates. The litte choices for 30 day bonuses with the crew. If only MWO had a bit of the BT immersion Battletech got on the table.

But meteo cannot see that. He's not able.

Also something like this, he is not able:
https://imgur.com/Spb4IQY

Because he goes with assaults, and no more than that....just like in mwo

[jss78]

HBS BT is like two weeks in. They have some major issues ongoing, but they're technical and solvable. Otherwise it's already a hell of a Battletech experience.

I anticipate to see 5+ years of increasingly epic stuff from them.

[meteorol]

Stefka Kerensky, on 10 May 2018 - 01:44 AM, said:

But meteo cannot see that. He's not able.

Also something like this, he is not able:
https://imgur.com/Spb4IQY

Because he goes with assaults, and no more than that....just like in mwo

That's the difference between objectively judging the game and looking through rose tinted glasses.

Adding the illusion of depth by using less than optimal mechs and builds doesn't make the combat system objectively better.

[RickySpanish]

LordNothing, on 09 May 2018 - 10:47 PM, said:

problem is it the good days of olde, game developers painstakingly designed an engine in house that met the precise requirements that their game needed. developers today are like kindergardeners gluing mass produced bobbles together to make something resembling a game. they go in thinking they can bend the code blob to their liking to meet their goals. when they fail to do this they make excuses, blame engine limitations and no quantity of kindergarten paste can fix it for them. you end up with a game made out of dirty hacks and messy workarounds.

ive used unity before, and while i have yet to find fault with the engine itself. the practice of gluing massive blobs of libraries together to make a game tends to lead developers down a path of broken dreams and shattered promises. maybe hbs arent noobs but they certainly aint masters either. i would prefer developers write their own damn code, but this is the world we live in. unity is still superior than the engine mwo uses. it is kept up to date and is at least somewhat modern. its not ue, but who cares. it does what it says on the tin. hbs has bugs to fix, nothing more.

everyone knows oop slows things down, thats why i like c where its all moving memory around, or c++ which while providing objects, doesnt force you to use them for everything. thats incidentally the thing i like about lua too. you arent forced into paradigms you don't want to deal with. also every game has to deal with dynamic memory, its just a thing you need to do.

I can definitely get behind the sentiment of "kids nowadays!" but I think the ideal solution lies somewhere in the middle. For most games, building your own engine is impractical. Look at the the evolution of textures in your modern graphics engine - in ~1998 they were 128x128 images packed into .wad files in Half-Life. Now, they are materials governed by real world properties, often made from several different textures each describing the exact value of a property at a per-pixel level of accuracy. You've also got the recent explosion of games onto anything with a screen and a processor, which all have their own APIs and hardware to develop for. Thus, it's not unreasonable for developers to glue blobs together to make their game. BUT! As you point out, you run the risk of not fully understanding the thing you have built or rather, its underlying technology. Unity definitely is an expedient development tool, but it has some serious downsides which can give your game that "Unity" feel. I do get that vibe from BT, and yeah I sorta wish UE had been used instead, but not everyone understands its beauty or cpp for that matter.

With all that said, have you *seen* the amount of json data in the game's streamingAssets folder? I do wonder if maybe someone is reloading portions of that more often than is needed, and perhaps that's contributing to the slow loading times...

[Dragonporn]

On HBS topic in general, can't say these guys are "noobs", but they have quite niche approach and set of skills that wouldn't work for every single game out there. I have played Shadowrun series of theirs, and have to say that particularly Dragonfall and HK are superb in terms of visuals (environment design) writing, music and especially story. I have played many old and new school RPGs of different caliber, but I can swear that these games are one of the best ones (for me THE best) in genre, period. Those games are quite minimalistic both in terms of gameplay and technologically, but it works nevertheless, and pretty spectacularly. SRs are built on Unity by the way, they run and perform very well and are stable with very few bugs, mostly small ones.

I have also played Necropolis a little, it completely built on gameplay formula, but I couldn't enjoy one bit of it, although I'm a big fan of Dark Souls series. They just couldn't make decent game with gameplay on it's front page.

I read and watched many articles and footage on BT so far, and it doesn't appeal to me. I don't know why, I like TBS in general very much, but game itself doesn't seem to have anything (aside from mechs maybe) to draw me in. Plus technical issues stop me from picking it up too. Maybe one day though, but still pretty hyped about MW5.

[Xeno Phalcon]

Once they patch out the major issues (bugs mostly) I don't see any real issue TBH. At the moment its sort of like MWO fanboys are throwing dirt at the BT fanboys cause a dog took a dump on the BT lawn. Once there has been some stoop and scooping the other guys grass is suddenly gonna seem a lot greener.

[STEF_]

meteorol, on 10 May 2018 - 03:11 AM, said:

That's the difference between objectively judging the game and looking through rose tinted glasses.

Adding the illusion of depth by using less than optimal mechs and builds doesn't make the combat system objectively better.

And that's not understanding the point.

You should enjoy a game not bringing meta only, but because you enjoy the game itself and its challenges.

Also, if you say the campaign's scenarios are always the same, you are simply lying to yourself, wearing black glasses

[meteorol]

Stefka Kerensky, on 10 May 2018 - 05:13 AM, said:

And that's not understanding the point.

You should enjoy a game not bringing meta only, but because you enjoy the game itself and its challenges.

Also, if you say the campaign's scenarios are always the same, you are simply lying to yourself, wearing black glasses

Pretty much every game has a "meta". Even single player games. Difference between good and mediocre games? The meta in good games isn't as boring, dull and repetitive.

BT has a very obvious flaws in its combat system (evasion pretty much being a nonfactor in late game, knockdowns and pricisions strikes being the more than obvious way to go) that are objectively bad. "Well don't use it if you don't like it" is no justification for objective flaws that make a game objectively worse.

Could i use all kinds of (subpar) builds, play specific mechs or subjugate myself under some made up rules to actually make combat more engaging? Sure. I could do that. The whole point is, i shouldn't have to do it to make the game more engaging.

Immersion? Well i'm not exactly drowning in immersion fighting against a ragtag group of bandits at the end of space that... well, field like 8 assault mechs that should be an incredibly rare sight and not the base standard for everyone in the whole goddamn universe.

Logistics? Resources? Don't even know what that is anymore because i'm drowning in both assault mechs (the rare sight) and money. It becomes irrelevant so quickly it might aswell not be in the game, wouldn't make much of a difference.

I'll give you the campaign missions have some degree of variety (it's there to some degree, but it's nothing to write home about). Aside of that? Dull grind with zero variation in combat. The AI being dumb as a rock doesn't really help to provide diverse gameplay.

I'm calling out BT on it's flaws just like a have always called out MWO on them. Or every other game for that matter. If you can't see how BT came up way short in many areas, well to each his own. I'm seeing it, and i'm seeing it pretty clearly.

Not an issue for me though. Got the game for 30, played it for like 50 hours, got good value out of it. Doesn't mean i have, or will, praise it for more than it actually is though.

Edited by meteorol, 10 May 2018 - 06:13 AM.

[Mystere]

LordNothing, on 08 May 2018 - 08:37 PM, said:

i was kind of being sarcastic. im actually a big fan of lua. id rather use that than dot net or java. python might be worth picking up though. still i started programming with c and id rather have a good c programmer over a room full of java devs any day.

Well, if you want to get into AI (deep/machine learning), the most popular tools use Python, unless you have your own toolset, of course.

[VaudeVillain]

I moved all but 2 of my most recent save files and it sped up loading times.  
I've played my Stalker, Atlas, and King Crab builds from Battletech in MWO and the Atlas is a little weak but the other 2 do good damage. Since I only use Level 1 tech in MWO, I tried my 2 PPC, 5 Md Laser MWO Awesome build in Battletech but it ran way too hot.

Edited by VaudeVillain, 10 May 2018 - 07:14 AM.