Commented 11 months ago2023-12-17 11:22:01 UTC
in wiki page: Tutorial: Evolving GrassComment #105757
Maybe a bit too far-fetched, but I think it would be cool to have macro previews. It was something that I suffered from back when I was trying MESS out. Obviously, editor integration would be needed for this, and while it could be feasible with TrenchBroom (it being open-source and actively developed), it'd still take a frightening amount of effort, I figure. Add its lacking Half-Life mapping support on top of that. And of course J.A.C.K. is out of the question.
Where did hull 0 go? Aren't there four hulls? 0 is point hull, 1 is standing hull, 2 is big monsters, 3 is crouching. I don't get why the amount of clipnodes is 18/36 instead of 24/48.
Although I'm myself practically retired from mapping by now, and wasn't a big practitioner of mapping automation through tools like MESS, I found the tool remarkable and unique in what it does during my brief period of trying it out. In fact, I think that new mappers should absolutely try adding MESS to their compile process and utilizing the macros that it makes possible to engineer - it has great potential to reduce a lot of annoying and unnecessary work.
Brush entity merging is a really annoying technique to use manually: to reduce the amount of limits taken up by your map's entities this way, you have to sacrifice "readability" of your map source file. For example, it's rather time-consuming to take a single piece from merged entities and perform any individual operation on it. Your tool finally providing an abstraction over the technique gives me the hope that it can go as far as make this optimization zero-cost in terms of mapper's time spent, and this is extremely important - nothing kills the wish to map like the amount of brush paperwork that you have to do at times.
I'm excited to see where MESS will go in the future and would love to hear of any further improvements and new features!
EDIT:
Looking at the feature description, although this is a rather small detail that can be avoided if you use the feature carefully, I wonder how it determines the resulting "principal" or "primary" entity to merge all of the provided entities into. I think there should be a way to explicitly specify which one entity will serve as such.
It can make the map more refactoring-friendly: when you change things, some keyvalues may go out of sync with your changes and intent, and you may end up with bugs that you may not immediately notice.
Commented 11 months ago2023-11-30 12:47:44 UTC
in journal: So I drink coffee now...Comment #105677
I've noticed that i'm sleepy from 9 to 12, and after 12:30 i enter turbo mode, no matter how many energy drinks or coffees i'm on. Maybe that's a common thing
Commented 3 years ago2021-08-25 22:57:47 UTC
in vault item: Janitor SimulatorComment #103678
Lighting isn't anything special but I wouldn't call it poor. Some of the door lights are nice-looking. Too white indeed, but there's some color, which is good. What is good as well is that you're using texture lights.
The map's main problems crop up shortly after the start. A ghostly game_text tasks the player with getting some soda to drink. There's a tiny bit of ambience (should've added a tiny bit more in spots) and some scientists to entertain you.
But then comes the soda part. All the player has to do is walk up to the vending machine, only for another game_text to pop up and explain what's just happened. No feedback of any sorts - not a soda can popping out, nothing, sadly. Oh yeah, you forgot to add the word "mop" to "Grab the freaking and do the job". Nice testers you've got
The next task is to grab the mop and get rid of those stinky puddles all around. The mop has been in the start room all along, but you absolutely had to drink a soda first! No way around! It's sad that grabbing the mop doesn't produce any sound effects. You should've added at least one and it'd be a very nice improvement already!
This is where Janitor Simulator turns into Trigger Simulator. All the player has to do is run over the puddle. Yes, being a janitor is that simple! Jokes aside, I understand that there aren't any obvious improvements that may come to mind. As an idea, however, you could've made the player press the Use key on a puddle several times to gradually remove it, or something. And again, you at least should've added a splashy sound effect to make it more satisfying.
The Trigger Simulator continues, and now you have to get the garbage out of the trash bins. This is where feedback is completely thrown out the window. What you have to do is to walk up to each trash bin and that's it, with one caveat - all walking up does is increment an invisible counter. No sound, no animation or anything similar. No feedback.
Finally, to end the shift, our janitor needs to push a nearby cleaning cart into the now unlocked disposal unit. Another such cart, looking literally the same as the other, can be seen inside the start room. To my surprise, there's no way to push it out and trick the disposal unit. It is good you have considered it.
What you haven't considered, though, is that a dumbass like me might decide to jump into the disposal unit along with the cart! I got the "You're Winner" text as the unit was crushing me down along with the trash. Now, you could've clipped it off, however I don't remember if there's an easy way to do that to a pushable object. Alternatively, the disposal unit could've had a different shape. Or you could've taken the possibility into account. All up to you.
Good job either way. You should improve your lighting by having it less bright and less white, as pointed out above (just don't make it too dark), and also have more feedback and consideration put into your gameplay pieces.
Commented 3 years ago2021-08-09 22:50:29 UTC
in vault item: Scooby_RatsComment #103660
This was lovely. Played it with 2 other people only, but pretty sure it would've been brilliant with more. Lots, lots of gimmicks. The usage of trigger_gravity is fun as heck.
Commented 3 years ago2021-07-22 14:02:16 UTC
in vault item: meatworldComment #103620
Good, but I didn't like the following things:
Most enemies are easy to run past. You don't even have to shoot that much.
Could do more interesting open areas than those that are there. The portal frames are cool, etc., but the open areas are still rather underdetailed. Conveyors and those trench thingies don't help much.
Texture tiling is a bit too noticeable, mostly because of the previous point.
You get the shotgun but don't even get to use it properly.
The ending is abrupt. It's bad, because it made me and will likely make many other people try noclip'ing around the map to make sure it was truly the end.
Could've put something in the piss pool or whatever you call it, the yellow water thing. Either a goodie or a hazard. Could've also put more piss pools around the map.
The technical flaws that I've noticed:
You can skip the HEV suit, and will have to play the rest of the map without it. Nothing prevents you from doing that.
The third teleport's trigger_teleport seems to be active from the start, with no need to press the button.
The scientist before and after seems to change heads while you're not looking.
Don't be discouraged by all that, though. If I don't mention something, it's nicely done And no stars cuz I suck at rating.
Commented 4 years ago2020-03-10 20:48:30 UTC
in vault item: RazorbackComment #102577
@Dimbeak, @Admer456, @Bruce: Thank you for the words and ratings. I may be dependent on people's attention a bit too much but receiving it always is very motivating!
Bruce said:I still wish the lower interiors had a bit more contrasty lighting.
Yes, the lighting stays one of the major problems of this map. However, I think that the brightness levels are fine enough, and the real problem is the bland color scheme (unless you meant color contrast there ). But either way, I've failed to come up with a more original color combination than pink + gray.
Dimbeak said:Weapon placement could use some tweaking, though, I feel like it's hard to get my hands on a decent arsenal.
I came to agree with that while we were playing it on Quakeroach. Putting an MP5 instead of the revolver at one of the obelisks, as well as replacing the MP5 at the lower level with a shotgun, would certainly help. There have been trolls claiming that the map is terrible due to an excessively high amount of weapons but I don't think I can do anything about it: if it had little weaponry scattered around instead then someone else would surely complain about that, too. Besides, there are 16 spawnpoints, come on! There must be enough guns for everyone.
Bruce said:even though it lags
Man I know, honestly I'm even contradicting myself with this map. I have used some optimisation techniques, though, like using VHLT's CONTENTWATER tool texture to make the lamp holders one-plane instead of four. The same trick has been used on small chains. I have also tried to control the splits with setting proper detail levels on func_details but I didn't inspect the results a lot, so there may still be mistakes and unnecessary splits.
There are two main problems, though. The former is that the BEVEL texture is applied in-between many of the triangular prisms which compose the island's terrain. Simply put, BEVEL helps to improve collisions in certain cases like this one. But at the same time the splits on the terrain must be terrible, partly due to the compilers' fault, but also partly because I didn't use SOLIDHINT. This tool texture lacks the advantages provided by BEVEL and mostly works like NULL but helps the compilers split the geometry better in some cases. To be honest, I'm the one to blame here - some terrain pieces still have NULL in-between the triangles.
And the latter problem is the openness of the map. It's a "void map" (or so they refer to that kind of maps in one of Quake 1 mapping communities), which means that you can barely block visibility for players, therefore a lot of geometry is considered potentially visible and gets rendered. I wish I had the willpower and the skill to optimise it further, but for now 3500 wpolies on average it is.
I plan on maybe rolling out a patch (which will be pretty useless because it's too late, the map has already been downloaded many times), and even then I doubt that I'll alter anything crucial about the map. As someone said in some thread on this website: "Fix your mistakes in your future works". You may take it as laziness but I certainly want to finally be done with this map and move on.
Commented 4 years ago2020-02-23 00:24:13 UTC
in vault item: dm_perthownedComment #102545
Despite the high detail level, the map could've been better. Sure, a lot of hard work has been put into fleshing out every little piece of geometry, but the layout is actually quite simple, if not boring. Maybe I have missed some places where gaussjumping could be applied, but I felt like there's not enough vents to crawl and ledges, platforms or pipes to jump on, climb or walk. The lack of interaction is a flaw too, the map could have had a lot more moving parts. It wouldn't have been a big issue if many static pieces of detail also didn't look distracting. They tend to attract attention as if they were useable, especially that one yellow valve which I thought does at least something. I often found myself walking up to things and pressing Use in hopes that something will happen, but nothing did. The optimisation... Well, not like anything could be improved as the map is a bunch of very open areas with barely any hallways to serve as VIS-blockers. In some spots the wpoly count would reach the ridiculous amount of +10,000, which didn't impact my framerate much, but Bruce had 10 fps on the map at average. I do understand that this map's excessive detail is part of the idea, but that doesn't make it a better HLDM map. I'd like to mention the good usage of angled geomtry, though. It gives some areas a very interesting look. The curved parts of the map are also well done!
Clearly, a lot of effort has been put into the map. I assume the main goals were to make a very, very detailed map and to only use the stock textures. But as a HLDM map, this one isn't very interactive nor fun, leaving you to admire only its rich detail.
And as for consistency, I do acknowledge the issue however I am afraid that I am not entirely sure how to resolve it.
Well, say we've got one crate that's pushable and breakable and another one which looks identical but is static. How will the player figure out which one to push or break? They'll try doing that to each one until they figure that out. But what if you have 10 crates in the room that all look the same? Let's imagine there's a vent in the room but you have to push a crate to it to get in, but the player will have to try all the 10 crates before they know which one to push. The solution is to make pushable/breakable crates distinguishable. You don't have to make all static/pushable crates in your mappack/mod look the same so that the player can make no mistake. You may have two pushable crates with different meshes, but they both should have at least some parts, for example, be of a color that will signify that these are interactive. So the player will establish that all crates that have this cue somewhere on them are pushable/breakable/whatever. BTW you're not limited to just color coding, use whatever you can come up with if you're sure it will work nicely.
Always name your maps so that it's clear what map it is and so that it won't possibly conflict with other maps in the player's maps folder;
Never surround your whole map with a box, even during construction. Sometimes you, of course, may be working on a map that's a bunch of floating islands or a small floating station, but in most cases you should have only the geometry that you really need. Boxes around maps increase compile time and cause overdraw. Oh yeah, they also make HLVIS cry.
You have very rich detail, but seem to have focused almost only on that. You should complete your map in the first place, detail comes later, so spend your energy on actually making the map and not sugaring it over and over. Also, maps with low detail often are better than many overdetailed ones. It doesn't mean that you always have to keep your maps' geo simple: add as much detail as your skill allows and as long as it doesn't take you away from finishing the map itself. As you get better, you'll be able to produce maps with even larger amounts of detail.
Try to restrict yourself from overusing game_text to tell the player what to do - show what to do instead. It's much easier to direct the player's attention towards something if there's no unnecessary detail clutter around. For example let's take that one blue monitor in one of the rooms: the whole room is dark, and the only thing that stands out in all this darkness is the blue computer screen. Of course the player will think that it's something important and, most likely, usable.
Try to establish consitensy. The first vent cap that the player has to break is broken by hitting the Use key. However, as you're crawling the vent you see another such vent cap on the left (if I recall correctly), but you can do nothing with it. And the third one breaks when you step on it, but not from hitting the Use key either. So it's not entirely clear how to actually break vent caps in your map, and which of them break. Mark breakable vent caps with a different texture or generally make them look different, and decide whether your vent caps break from touching them, Using them or smashing them with the crowbar or any other weapon. Same applies to doors (not in your map though, just in general). Sometimes some maps have a few doors that open by Using them and the rest are opened somehow else. Unless you have a very strong reason to break the rule, don't do that.
If I'm wrong at any of these points, correct me. It's 0:38 here and my brain is half-RIP. And happy mapping!
Commented 5 years ago2019-09-01 11:07:21 UTC
in vault item: dm_eightballComment #102240
@ToTac glad to see that it wasn't just my brain being retarded and you've actually tested it you've got a sign in the credits room btw, as some other people
Commented 5 years ago2019-08-31 21:18:15 UTC
in vault item: dm_eightballComment #102237
well, thanks @Urby
the balance may still be off tho but I was way too lazy to do any changes to it in the end. Also I really wish I had added any verticality because the layout is roughly an eight, or visblockers because one area can be seen clearly from another, resulting in overdraw.
and i wonder if anyone will ever find the secret credits room (it's actually accessible, you just need to do a good gaussjump)
Commented 5 years ago2019-05-15 17:28:12 UTC
in vault item: Dynamic LightComment #101987
@stertyr
You probably can. It's a hacky setup and the keyvalue is used for some other internal purpose by the game most likely. You can experiment and try changing the value of the key to some other numbers. I recall that you can also get the yellow particles appear around the entity (usually you see them if a monster is stuck) or get a huge version of the dlight instead. Also you can get combinations of these effects, but I don't remember the exact values. I remember though that they loop at some point, so entering some huge number could give you the same result as entering something very small.
Commented 5 years ago2019-05-13 23:47:03 UTC
in vault item: Dynamic LightComment #101982
@stertyr
In Hammer-like editors (Hammer, Sledge, Jackhammer aka J.A.C.K., etc.), SmartEdit is a feature that replaces internal keyvalue names with more readable and comprehensible ones if there are any listed in the loaded fgd file. It also enables dropdown lists or color pickers on some keyvalues so you don't have to set the values by hand. But it's fgd-dependent too.
The point in turning SmartEdit off is that with it turned on you can't add custom keyvalues and can only edit the ones predefined in the fgd file. The "effects" key isn't listed in most fgds (or likely none at all), so you either have to add it by hand with SmartEdit off or add the key to your fgd file, making it appear in the entity's properties as you want it to with SmartEdit enabled.
This comment was made on an article that has been deleted.
Commented 5 years ago2018-11-27 01:44:18 UTC
in vault item: Tunnel VisionComment #101637
Looks like you were so busy replicating the original Valve style (and did a brilliant job on that!) that you almost forgot about ammo and health. Great work!
Commented 6 years ago2018-04-27 13:20:55 UTC
in vault item: tesis5xComment #21334
As far as I know func_* entities are always rendered no matter where you currently are and they also add up to the entity limit. Unlike func_detail. Don't forget that hints can help too.
Bad map design can also be a problem. If you have any badly-made parts of your map that are relatively expensive for the engine to render then try redesigning them, partially or completely.
But even if your map reaches 2000-2500 wpolys in certain spots then it doesn't mean it's trash, believe me. I mean, it's 2018 and nowadays people are definitely capable of running 2500+ wpoly maps at a framerate higher than 30 FPS.
lights.rad,info_texlights, orlight_surface?you're a badass :d
Although I'm myself practically retired from mapping by now, and wasn't a big practitioner of mapping automation through tools like MESS, I found the tool remarkable and unique in what it does during my brief period of trying it out. In fact, I think that new mappers should absolutely try adding MESS to their compile process and utilizing the macros that it makes possible to engineer - it has great potential to reduce a lot of annoying and unnecessary work.
Brush entity merging is a really annoying technique to use manually: to reduce the amount of limits taken up by your map's entities this way, you have to sacrifice "readability" of your map source file. For example, it's rather time-consuming to take a single piece from merged entities and perform any individual operation on it. Your tool finally providing an abstraction over the technique gives me the hope that it can go as far as make this optimization zero-cost in terms of mapper's time spent, and this is extremely important - nothing kills the wish to map like the amount of brush paperwork that you have to do at times.
I'm excited to see where MESS will go in the future and would love to hear of any further improvements and new features!
EDIT:
Looking at the feature description, although this is a rather small detail that can be avoided if you use the feature carefully, I wonder how it determines the resulting "principal" or "primary" entity to merge all of the provided entities into. I think there should be a way to explicitly specify which one entity will serve as such.
It can make the map more refactoring-friendly: when you change things, some keyvalues may go out of sync with your changes and intent, and you may end up with bugs that you may not immediately notice.
All I end up doing is just falling asleep anyway
It's pretty cool how you can find the strength to talk about something so personal in a relatively public place
Waited for this day
The map's main problems crop up shortly after the start. A ghostly game_text tasks the player with getting some soda to drink. There's a tiny bit of ambience (should've added a tiny bit more in spots) and some scientists to entertain you.
But then comes the soda part. All the player has to do is walk up to the vending machine, only for another game_text to pop up and explain what's just happened. No feedback of any sorts - not a soda can popping out, nothing, sadly. Oh yeah, you forgot to add the word "mop" to "Grab the freaking and do the job". Nice testers you've got
The next task is to grab the mop and get rid of those stinky puddles all around. The mop has been in the start room all along, but you absolutely had to drink a soda first! No way around! It's sad that grabbing the mop doesn't produce any sound effects. You should've added at least one and it'd be a very nice improvement already!
This is where Janitor Simulator turns into Trigger Simulator. All the player has to do is run over the puddle. Yes, being a janitor is that simple! Jokes aside, I understand that there aren't any obvious improvements that may come to mind. As an idea, however, you could've made the player press the Use key on a puddle several times to gradually remove it, or something. And again, you at least should've added a splashy sound effect to make it more satisfying.
The Trigger Simulator continues, and now you have to get the garbage out of the trash bins. This is where feedback is completely thrown out the window. What you have to do is to walk up to each trash bin and that's it, with one caveat - all walking up does is increment an invisible counter. No sound, no animation or anything similar. No feedback.
Finally, to end the shift, our janitor needs to push a nearby cleaning cart into the now unlocked disposal unit. Another such cart, looking literally the same as the other, can be seen inside the start room. To my surprise, there's no way to push it out and trick the disposal unit. It is good you have considered it.
What you haven't considered, though, is that a dumbass like me might decide to jump into the disposal unit along with the cart! I got the "You're Winner" text as the unit was crushing me down along with the trash. Now, you could've clipped it off, however I don't remember if there's an easy way to do that to a pushable object. Alternatively, the disposal unit could've had a different shape. Or you could've taken the possibility into account. All up to you.
Good job either way. You should improve your lighting by having it less bright and less white, as pointed out above (just don't make it too dark), and also have more feedback and consideration put into your gameplay pieces.
And no stars cuz I suck at rating.
- Most enemies are easy to run past. You don't even have to shoot that much.
- Could do more interesting open areas than those that are there. The portal frames are cool, etc., but the open areas are still rather underdetailed. Conveyors and those trench thingies don't help much.
- Texture tiling is a bit too noticeable, mostly because of the previous point.
- You get the shotgun but don't even get to use it properly.
- The ending is abrupt. It's bad, because it made me and will likely make many other people try
- Could've put something in the piss pool or whatever you call it, the yellow water thing. Either a goodie or a hazard. Could've also put more piss pools around the map.
The technical flaws that I've noticed:noclip'ing around the map to make sure it was truly the end.- You can skip the HEV suit, and will have to play the rest of the map without it. Nothing prevents you from doing that.
- The third teleport's
- The scientist before and after seems to change heads while you're not looking.
Don't be discouraged by all that, though. If I don't mention something, it's nicely donetrigger_teleportseems to be active from the start, with no need to press the button.And no stars cuz I suck at rating.
CONTENTWATERtool texture to make the lamp holders one-plane instead of four. The same trick has been used on small chains. I have also tried to control the splits with setting proper detail levels on func_details but I didn't inspect the results a lot, so there may still be mistakes and unnecessary splits.There are two main problems, though. The former is that the
BEVELtexture is applied in-between many of the triangular prisms which compose the island's terrain. Simply put,BEVELhelps to improve collisions in certain cases like this one. But at the same time the splits on the terrain must be terrible, partly due to the compilers' fault, but also partly because I didn't useSOLIDHINT. This tool texture lacks the advantages provided byBEVELand mostly works likeNULLbut helps the compilers split the geometry better in some cases. To be honest, I'm the one to blame here - some terrain pieces still haveNULLin-between the triangles.And the latter problem is the openness of the map. It's a "void map" (or so they refer to that kind of maps in one of Quake 1 mapping communities), which means that you can barely block visibility for players, therefore a lot of geometry is considered potentially visible and gets rendered.
I wish I had the willpower and the skill to optimise it further, but for now 3500 wpolies on average it is.
I plan on maybe rolling out a patch (which will be pretty useless because it's too late, the map has already been downloaded many times), and even then I doubt that I'll alter anything crucial about the map. As someone said in some thread on this website: "Fix your mistakes in your future works". You may take it as laziness but I certainly want to finally be done with this map and move on.
The lack of interaction is a flaw too, the map could have had a lot more moving parts. It wouldn't have been a big issue if many static pieces of detail also didn't look distracting. They tend to attract attention as if they were useable, especially that one yellow valve which I thought does at least something. I often found myself walking up to things and pressing Use in hopes that something will happen, but nothing did.
The optimisation... Well, not like anything could be improved as the map is a bunch of very open areas with barely any hallways to serve as VIS-blockers. In some spots the wpoly count would reach the ridiculous amount of +10,000, which didn't impact my framerate much, but Bruce had 10 fps on the map at average. I do understand that this map's excessive detail is part of the idea, but that doesn't make it a better HLDM map.
I'd like to mention the good usage of angled geomtry, though. It gives some areas a very interesting look. The curved parts of the map are also well done!
Clearly, a lot of effort has been put into the map. I assume the main goals were to make a very, very detailed map and to only use the stock textures. But as a HLDM map, this one isn't very interactive nor fun, leaving you to admire only its rich detail.
studying sucks, but probably still better than ending up as a cargo hauling drunkard
this has been the biggest revelation for me for the last few months
but O it is
- Always name your maps so that it's clear what map it is and so that it won't possibly conflict with other maps in the player's
- Never surround your whole map with a box, even during construction. Sometimes you, of course, may be working on a map that's a bunch of floating islands or a small floating station, but in most cases you should have only the geometry that you really need. Boxes around maps increase compile time and cause overdraw. Oh yeah, they also make HLVIS cry.
- You have very rich detail, but seem to have focused almost only on that. You should complete your map in the first place, detail comes later, so spend your energy on actually making the map and not sugaring it over and over. Also, maps with low detail often are better than many overdetailed ones. It doesn't mean that you always have to keep your maps' geo simple: add as much detail as your skill allows and as long as it doesn't take you away from finishing the map itself. As you get better, you'll be able to produce maps with even larger amounts of detail.
- Try to restrict yourself from overusing
- Try to establish consitensy. The first vent cap that the player has to break is broken by hitting the Use key. However, as you're crawling the vent you see another such vent cap on the left (if I recall correctly), but you can do nothing with it. And the third one breaks when you step on it, but not from hitting the Use key either. So it's not entirely clear how to actually break vent caps in your map, and which of them break. Mark breakable vent caps with a different texture or generally make them look different, and decide whether your vent caps break from touching them, Using them or smashing them with the crowbar or any other weapon. Same applies to doors (not in your map though, just in general). Sometimes some maps have a few doors that open by Using them and the rest are opened somehow else. Unless you have a very strong reason to break the rule, don't do that.
If I'm wrong at any of these points, correct me. It's 0:38 here and my brain is half-RIP.mapsfolder;game_textto tell the player what to do - show what to do instead. It's much easier to direct the player's attention towards something if there's no unnecessary detail clutter around. For example let's take that one blue monitor in one of the rooms: the whole room is dark, and the only thing that stands out in all this darkness is the blue computer screen. Of course the player will think that it's something important and, most likely, usable.And happy mapping!
glad to see that it wasn't just my brain being retarded and you've actually tested it
you've got a sign in the credits room btw, as some other people
the balance may still be off tho but I was way too lazy to do any changes to it in the end. Also I really wish I had added any verticality because the layout is roughly an eight, or visblockers because one area can be seen clearly from another, resulting in overdraw.
and i wonder if anyone will ever find the secret credits room (it's actually accessible, you just need to do a good gaussjump)
You probably can. It's a hacky setup and the keyvalue is used for some other internal purpose by the game most likely. You can experiment and try changing the value of the key to some other numbers. I recall that you can also get the yellow particles appear around the entity (usually you see them if a monster is stuck) or get a huge version of the dlight instead. Also you can get combinations of these effects, but I don't remember the exact values. I remember though that they loop at some point, so entering some huge number could give you the same result as entering something very small.
In Hammer-like editors (Hammer, Sledge, Jackhammer aka J.A.C.K., etc.), SmartEdit is a feature that replaces internal keyvalue names with more readable and comprehensible ones if there are any listed in the loaded fgd file. It also enables dropdown lists or color pickers on some keyvalues so you don't have to set the values by hand. But it's fgd-dependent too.
The point in turning SmartEdit off is that with it turned on you can't add custom keyvalues and can only edit the ones predefined in the fgd file. The "effects" key isn't listed in most fgds (or likely none at all), so you either have to add it by hand with SmartEdit off or add the key to your fgd file, making it appear in the entity's properties as you want it to with SmartEdit enabled.
HB anyway
Don't forget that hints can help too.
Bad map design can also be a problem. If you have any badly-made parts of your map that are relatively expensive for the engine to render then try redesigning them, partially or completely.
But even if your map reaches 2000-2500 wpolys in certain spots then it doesn't mean it's trash, believe me. I mean, it's 2018 and nowadays people are definitely capable of running 2500+ wpoly maps at a framerate higher than 30 FPS.