• Crusader Kings III Available Now!

    The realm rejoices as Paradox Interactive announces the launch of Crusader Kings III, the latest entry in the publisher’s grand strategy role-playing game franchise. Advisors may now jockey for positions of influence and adversaries should save their schemes for another day, because on this day Crusader Kings III can be purchased on Steam, the Paradox Store, and other major online retailers.


    Real Strategy Requires Cunning

This thread is more than 5 months old.

It is very likely that it does not need any further discussion and thus bumping it serves no purpose. If you feel it is necessary to make a new reply, you can still do so though.

xenoargh

Recruit
Oct 19, 2014
9
2
Hi there. Hope the developers are still reading these pages; it appears to me that they're still working on patches for this game; if not, then perhaps they'll think about these things for a sequel. This is my stab at some constructive critique. I have done various flavors of game dev and modding for stuff, so this is probably not the usual kind of review.

I also don't know Estonian, but I've tried to make sure that this will translate well :)

Anyhow, here's the review:

Overall, I found Teleglitch, un-modded, to be a good game, but not wonderful, in terms of play and value.

We'll get into why it's just "OK" and not "wonderful" further down.

Modded, I have found it to be a lot more fun, because I think it plays a lot better if the emphasis is different.

The Good

1. I like the LOS system and the lighting system is wonderful.
2. I think the monster AI is reasonably good.
3. The combat, when the player is armed with anything but the knife, is pretty good.
4. The randomized levels are really great, in terms of replay value.
5. The plot and writing made me laugh a lot. There are typos and a few errors here and there, but it's a good story and it's funny.
6. The physics of things was occasionally pretty neat, although I felt like that could have been used more creatively.
7. The crafting system is a nice plus in a game of this type.
8. The boss fights were fun.
9. The final boss was fun.
10. The occasional scares from zombie swarms / robot swarms were fun.
11. I liked the Mag Shield, once I finally actually got one.

The Not-So-Good

1. I really don't like that the game won't render at larger than 640X480, other than the shadows. It looks really jagged and blocky and I just plain don't like the lack of anti-aliasing when I run it full-screen in 1080p. In fact, I didn't like it so much that I re-wrote the shaders to fix it where it's possible by aliasing things. That's the most I could do without having the engine source; I really think this should be fixed up by the developers or at least noted for a sequel.

2. I didn't like the auto-zoom feature at all. It just made the lack of zoom even worse. I'd much rather have a manual zoom level that could be adjusted via a slider, with a higher zoom level than the zoom level that we get when the auto-zoom is turned off.

3. I didn't like the way that randomization sometimes meant that a run just didn't work out, regardless of how skilled the player was. For example, the secret path on Level 3 that allows for a free trip to Level 5 doesn't always spawn, so far as I could determine. If you can find it, it saves a lot of ammo, because unfortunately, Level 4 isn't worth doing for the ammo. So there is no reason to continue a run where that secret doesn't spawn; it's sub-optimal play. There are a lot of little examples of that kind of problem; basically, I think it boils down the core issue I had with Teleglitch's current approach to balance, which is, basically, "use your knife as much as possible until Level 5".

That style of play I found counter-intuitive and ultimately pretty boring. I'd rather have more ammo and be able to shoot more enemies, but face more enemies. The game is strongest towards the end, if the player has saved up ammo; then it's intense and pretty satisfying. But it really should be intense and satisfying throughout, but get harder and more hectic later. In short, I feel that making it a game about ammo conservation during the first half is a mistake; newbies want to shoot stuff, and the knife is, in part because of its bad hitbox, such a lousy weapon that it is probably more frustrating than helpful to them. I would imagine that most new players quit before they've even gotten through Level 3, frankly.

For example... on my first, un-modded run where I got past Level 3 (and things got a lot easier, imo)... I got through the entire run but did not get a single Magnet. Not one. Yes, I probably missed a Secret somewhere, but you can imagine how annoyed I was when I realized that I couldn't do much with the final boss.

Same with the percentage drops in general. Sometimes you get a Motor, but otherwise, sorry? I don't like that; it means that I have a lot of excess choice anxiety when thinking about whether to kill Red Commanders. If they always dropped a Magnet, then I would take them down; but with a low drop rate, it's questionable whether any of them are ever worth killing after the first two.

Especially irksome was the drops from Guards, Soldiers, etc. They drop guns that don't have ammo. That irritates me on a lot of levels; did they shoot their last bullets just before I killed them? Since there aren't any good uses for empty guns, especially ones I no longer want, it's just useless "loot". I also think that Zombies should drop more Canned Meat; they're about the only reliable source, but as things stand, I'd rather stay away from their triggers rather than taking my chances with a swarm to get some Meat. It probably shouldn't work that way.

In short, the drops are too random, and that makes runs too random; there is no guarantee that a skilled player will get all the items he / she needs for the Blue Commander or the final boss. Granted, really skilled players can probably live without, but I think this is unnecessarily frustrating; if hardcore players want a run where they will not get enough stuff to make a Magshield and Power Legs, fine; offer up a difficulty level setting that denies those things. The way things work right now is frustrating, rather than fun.

4. I didn't like the craft-able items that were pretty much useless. Can-guns, pzfausts, etc., are all theoretically cool... but the reality is that they're actually a waste of precious resources in the longer term. Most especially irritating was the lack of any way to craft the core components that you'll desperately need at various points, like M-Chips; there are motors with their own reactors... but they lack a computer chip, and you can't recycle the stators to harvest some magnets?

Stuff I thought was pretty useless, especially to craft:

Cangun
Pzfaust (and the AP version is even worse)- seriously, they were only useful as Secret finds
Meat Trap
Using up small explosives to make big ones (if the timer was longer and the explosion radius quite a bit bigger, then it might be OK)
The drum magazine for the Assault Rifle vs. saving up for the Minigun.
The Taser and the Tesla Coil (they weren't entirely useless, but mainly, because they compete too much with the lasergun2 for batteries and were pretty weak anti-swarm weapons vs. the shotgunX2).
The SMG (seriously, the Auto-Pistol is as far as I ever thought it was ever worth upgrading, especially since pistol rounds are artificially scarce later).
Stimulants (good as a secret item, lousy otherwise vs. health packs).
The Shock Blade (seriously useless and irritating, considering its cost- should do a sweep attack or have much longer reach or one-hit even bigger things- it appeared to do maybe 50-60 damage at most, and isn't 100% armor-piercing).
The AGL-3 is a really bad joke, because it misses even more often than the AGL-1, which makes a weak weapon even weaker. If the AOE / damage of the sticky explosive was better, the AGL-1 would still be better. I think that AGL-3 should be rebalanced with a better rate of fire, one shot per attack, but higher accuracy and damage, so that it's useful throughout the game and the player wants to keep collecting grenade ammo.


5. This version is basically the DLC version, except that all of the DLC weapons aren't actually available, except kind of, in the arenas. So I ended up modding recipes to enable all of that.

6. Modding support is pretty primitive and could really improve the game's replay value if improved in fairly easy ways. Was pretty surprised, honestly, considering the game is built with Lua on top of the core engine.

The level editor is OK, other than the total lack of documentation and a UI that is very, very hard to work with, due to the total lack of modern features like drop-down menus. That, and having to Select All before saving, which is a very yucky issue; by default, when saving, it should Select All before proceeding, IMHO.

There is no documentation of the Lua features.

There is absolutely no way to add new projectile types. Let alone support for per-frame events on projectiles (think bouncing stuff, flamethrowers and a bunch of other cool things) or callbacks, so that weapons can do something cool when they collide with things. This means modding is a bit bland.

There is no way to make a itemtype.weapon that has Lua trigger events. This is really disappointing; there are methods to add new bullets in the sim, but the only way to make that happen when a weapon fires. Doing it with a pzfaust-style device, which doesn't support having an ammo type, is very limiting.

There is no documentation of how to add new monster types, and it took quite a bit of study to determine how they're done. There is no way to add new Squid types, Blue Commander or Machinegun Snake types; they're all hard-coded, rather than being soft-coded via Lua.

The graphics don't allow for any scaling at all. The first thing I wanted to try and mod was to make a higher-rez version of the game with better art, but you cannot do that, because graphics are kept at 1:1 scale. That alone almost dissuaded me from bothering to mod it.

The engine is using PNG, but it's using the OpenGL render setting that forces pixels to be either transparent or 100% opaque, rather than allowing for the use of translucency. As a developer, I cannot see any logical reason for this, and it hugely limits what can be done with the game graphically.

We can't add new Swarm trigger objects. That is a big missed opportunity; it really should be done via Lua triggers rather than engine-side code. I can't see how to make monsters spawn at a given location on the map via Lua code at this time, though, so I can't play around with that idea.

Lastly, the game uses shaders and the source is exposed and modders can change stuff, which is great; now I have a aliased modded version rather than a very blocky one. But there's a pretty cool thing that's missing; characters and background objects, if they were allowed to have a secondary texture, could then use that with a shader to do various things that would add a huge amount to the look and feel. For example, one character might use second texture for glowing eyes.

7. The in-game UI is... ugh.

I can't click on the item menu? I can't click on items to combine them? I have to use keys to move up and down a list? There are keys that are bound to various things, but it's not documented and they cannot be changed?

It felt like the late 1980s, UI-wise. I realize that's too big of a thing to fix at this point- but if the sequel has a UI this bad, I won't buy it.

8. I honestly feel that the EM Detector ruins the game. I know it's optional and it means one less Teleporter and all that; I still feel it ruins the core of the game.

9. I was pretty disappointed to find out that you can't even mod in new hand-to-hand weapons.

10. Weapons lack many special qualities that make them moire interesting.

For example, I think that giving weapons a "weight" (some float number in the engine), where heavier weapons slow aiming movement and maybe even turning speed down more than others, would have a huge impact, by making the different weapons have a lot more differentiation. The Minigun would feel heavy and bulky; the SMG would feel light and fast, the Assault Rifle could be in-between (and then it could do decent damage, rather than less than the Pistol, without being massively OP, too).

I also agree with the person who said that the Shotgun was the, "worst videogame shotgun ever". Boost the damage by 5-10 points at least per bullet; right now it's hard to justify using it past Level 4...

11. There isn't a flashlight. Ever? Really? I'd rather have a flashlight than the knife, honestly. Then it could give you a light source and a weapon you could bash zombies with, which would be cool.

12. Getting stuck on physics objects that block doors or drag on a character is funny only once. It would be very nice to be able to destroy that "bench" that is currently slowing down my character, rather than having to figure out which series of counter-intuitive movements will slide it out from under me.


Summary:

I know that this review probably sounds like I just wanted to say mean things to you, dear developer, but it's meant to be a good critique. I really like a lot of things about this game- the random levels, the scary feeling of running into rooms where you might just suddenly die, the frenetic firefights where you try to keep the zombie hordes away from you. I took the time to write this because of the things I liked; this is an idea with a lot of promise and it's clearly a much better game than when it was initially released.

However... I feel that there are a still a ton of rough edges on this title right now.

The ability to mod it beyond the basics is still very much a missed opportunity; with better support for Lua code to interact with objects and weapons (which is not a huge job; a couple of callbacks and a few more function calls to utility code that's almost certainly already in your engine-side code) would make things so much better for modders; having more serious API-like access to game objects (like, say, getting lists of all the monsters and their vital stats, so that Lua code can do distance sorts, etc. before doing stuff) would enable the game to have a lot more long-term appeal, despite its other blemishes, simply because the mods would help iron out gameplay ideas for different play-styles and difficulty levels.

The core game's main problem, in my opinion, is that it wants to be a survival-horror, low-resource game, especially at first. This doesn't work really well, and I think it probably drives a lot of new players away after trying for a few minutes. Further into the game, it's a game about luck more than skill, in some ways; if you get a bad run, all the skill in the world won't get you good gear. I think that's a problem; if luck is why I lose, in a single-player title that involves twitch skill, I tend to walk away, fast.

Sure, you have those players' money, but they won't buy your sequel or recommend the game to friends, and I think this is an issue for the game. I only found this title through Steam because of the review of Runers on RockPaperShotgun, for example, and I bought it despite them saying they never made it past Level 2.

Anyhow, I've modded it quite a bit, reversed the difficulty curve to make it more newbie-friendly, and have made it more of a zombie shooter with survival-horror elements and RPG development via crafting, and the graphics look much better. I'll probably get around to making it public at some point. If the game's ever really moddable, or even better yet, supports sprite scaling and translucency too, I'll perhaps revisit it. Right now, other than adding in some monsters, I think I've done about as much as feel like doing with it, though.
 
Last edited:
  • 1
Reactions: