took to make 

Most of the level was built over a long weekend (4 day) holiday.

The first part was to create the central tower and dish as this was the primary landmark which would determine the scale of the structure. I always build the landmark first because it sets the tone and style of the rest of the map.

The second day was to create a second tower with platforms that spirals upwards towards the dish tower. This was rough blocks, ramps and simple monster placement. I generally place AI as I go because I want them to suit the architecture for paths, line of sight etc.

The third day was creating the power cube idea of what makes the structures float. This was about the detail that went downwards and across to other platforms. I originally had solid walkways connecting the lower stuff but it looked too chunky so I went back to my test map and played with some entities and textures until I had something that was fun to play with and looked good.

The fourth day was to create the rest of the central platforms and AI placement using new Quoth units as accents to the encounters. I spent a while creating more textures, expanding the theme of runic metal and removing the brown metal as much as possible.

The next week was about getting the skybox sorted, it took me several attempts (over night renders) to get the right light source orientation and lower cloud setup. The latest version of Terragen has a lot of options which affect how fast or slow the render can be and the resolution was really high, so it took a lot longer than normal.

The second week was creating the final arena battle, playing with lightning and expanding the texture set with more trims and details which worked their way into the structure.

The third week was testing, testing and more testing. Trying to find bugs, balance the AI and build the skill selection area. I always build the first part of the map last so it looks the best with the textures. After a while of using a texture set you get into a certain rhyme of how they lock together and this usually leads to the best architecture / texture combinations.

Total time to build the map was about 3 weeks, but the majority of the time was spent in the first couple of days and there was a lot of overnight skybox rendering. 

I'm no mapper but that sounds like very very efficient mapping timewise. 

I must be crazy... I quite often will make my maps room by room (pretty much fleshing each one with bsp detail, then cutting up bsp to make trims and texturing, then lighting it... once the whole map is made only then do I generally add monsters). 

Perhaps that's a distinction between you and sock: architecture first vs gameplay first. 

That's how I made maps back in the day ... just string a bunch of interesting looks areas together and add monsters at the end. Sock definitely takes a more purposeful approach. 

Oh definitely. I will design areas with gameplay ideas in mind but I don't add the monsters until the end (in most cases)

Romero once mentioned that he made maps in this way... 

like 15 times trying to make a freakin' demo, GRRRRRRRRRR :(

I keep dying in a DIFFERENT way! I know that's a "good" thing but trying to supply a demo is pizzin' me off.

QS only let's me start recording before I enter the map? So loading a save and then recording from there won't seem to work...

meh... anyways,

Awesome map SOCK, jus' can't get it right to show me playing it. 

It's in two demo's though.

http://www.inchoate.com/qdemo/metmon_damage_demo.7z

THANKS for the map sock, totally ROCKS. 

I must be crazy... I quite often will make my maps room by room
That's how I made maps back in the day
Romero once mentioned that he made maps in this way

This is how most level designers start, the joy of building, exploring and playing with virtual space. Eventually you have to move beyond that and start thinking of architecture and spaces as gameplay.

The initial building part is perfect for working out scale and distance so that the AI have the best possible position. It is also good to try sketches or block-outs, but ultimately you should know where the AI is going to be in each area before the detail phase.

This is not to say I don't make architecture for the fun of it, I sometimes specially design some architectural to look a certain way, but that is usually landmark scale stuff. It really depends on how far the player is into the map.

@damage_inc, yeah this map can be unforgiving if you are not careful with ledges and where you jump. There is always two ways to get somewhere, early with tricky jumps or later with easier drop downs. Thanks for the demo, it was cool to watch. 

but if you change your mind, but if you make another small map, I'd love to see a kind of step by step development log from you! 

This just looks incredible. 

my favorite part of this map is the entire first half. The main player path through the level seemed linear but there are so many side areas, secret ledges, and so on that I spent most of my time jumping around, looking at different platforms and figuring out how to get there. In your readme you say the secrets aren't satisfying (due to the items they contain) but I think the satisfaction of finding them is good enough.

Your maps have all had good secrets, but I think what i especially like about these ledge-hunting secrets is that the player understands the rules of the game physics, so when looking around, you are using that understanding to judge the geometry you're looking at for it's climbability/jumpability. Other classes of quake secrets are commonly "shoot the unmarked button" type secrets where it is less about players applying their knowledge and more about random experimentation and luck.

One small negative I think is that for a first-time player, the different ledges and vertical "trunks" of the level all looked pretty similar. This alone would have been okay but then whenever i used a teleporter to get back up from a secret I completely lost my sense of orientation. There was a particular secret (blue armor i think) where every time i saw it from below, i could see how to get there, but after teleporting up i had no idea which direction that ledge was.

The final arena battle is very well done, good combat pacing and visual spectacle. Giving the players buttons to start each wave was a nice touch, it allows the player to control the pacing but not in a way that makes it too easy (since each wave was a pretty strong combat.) I am not a huge fan of these arena battles since so many maps seem to have them and the gameplay is always pretty similar, but like i said, it was still done.

Visually the level was very distinctive, the brushwork plus the overall structural shape of the level. It seemed that it might be inspured by Lun3DM5, but also didn't seem in any way to be a clone of that, just the general feeling of clusters of cubes with "ambient occlusion"-type lighting.

The translucent energy bridges were a nice addition as well. Good technique of training the player in the start map (though the first bridge in the main map is right in front of the player so that could have served the same purpose.) Also in the start map it took me about a minute to see the button. All the red runes looked much more noticable! Anyway, I managed to kill an NPC by turning off the bridge he was on, that was another nice emergent moment that didn't feel contrived.

Bobs were used well, the wide open sky was a good place to put them on big patrol paths and randomize how the player encountered them. I think what I have disliked about bobs in the past is how they were placed -- in super tight hallways where I can't dodge, in groups when I am underpowered, and sometimes at a distance where their small bounding box makes my SSG useless! In this case when i have a nailgun and they are usually solo, they felt pretty good.

Someone else said, sometimes they were in a tight place and wanted to be out in the open, other times they were in the open and they rushed for cover. That was a good feature of this map's combat. The small ledge with a Pyro on it was an example of this. 

And recorded a first run demo, hard skill. I technically died once from falling, but I guess a magic fairy grabbed me and floated me back onto the tower.

Lots of fast action, no time to let your guard down. No complaints from me. :)

By the way, this map for whatever reason doesn't work in enhanced glquake. It exits with an error:
PR_LoadProgs9: statement 50737: bad opcode 80

http://qrf.servequake.com/~orl/misc/orl_metmon.rar

Recorded in Fitzquake. 

Fun map. I probably wasn't really giving it the concentration it deserved (was more in the mood for a mindless blast) and died a few times for a variety of silly reasons. None of them felt unfair though, and I loved the fact that exploration was generally rewarded. I had the problem that metlslime mentioned about being very disoriented when teleporting back from secrets. Oh and I forgot to record a demo. Need to go back and play on hard and find all the secrets... 

@Drew, now if only all mappers would write down their thoughts and post them with their maps, it would be a wonderful world! :P

I have thought about creating a screen capture of the editor and compiling it into a video. That might be interesting to see, but it is a crazy amount of extra work just for a couple of people to watch. Plus I use GTKradiant and the Quake community is mostly WC/TB nowadays. 

I am surprised to see you comment, I had given up hoping that any of the original community mappers would play Quake anymore or even want to post feedback! Thanks for taking the time.

In your readme you say the secrets aren't satisfying
I think the problem with the secrets in this map is that they are all roughly the same type, either you are finding button or the route to a certain ledge. Often Quake maps have a much richer variety of types and the structure in this map is so open it was difficult to find good locations to hide stuff.

whenever i used a teleporter to get back up from a secret I completely lost my sense of orientation
Teleporters are like resetting the player and are always difficult to setup correctly. I tried to help with this by including arrows on corners and architecture for the player to see after teleporting but the arrows are too subtle. My other plan was to have the outer platforms explode after teleporting but I ran out time and will power.

I am not a huge fan of these arena battles since so many maps seem to have them
This is a problem with Quake, there is nothing to do for final battles besides wave spawning. The only two bosses in the game have been done to death and not really rewarding to fight. I am probably partly to blame because the wave spawning is an easy system to setup instead of trying to find more creative ways to finish.

So far I have tried nail shooters traps, giant crushers, classic DM style wave spawning and even staggered button pressing with lightning attacks for end battles. Each new map is certainly getting harder to find more tricks to include in the final arena without resorting to creating new boss models.

though the first bridge in the main map is right in front of the player so that could have served the same purpose
I really wanted the player to hit the bridge button first and drop the AI to their death and due to QPB (Quake Player Baggage) everyone just shoots first and then maybe explores later. So I thought it would be better to show the mechanic first without any distractions (enemies) so that the player might be tempted to push the button a second time. Funnily enough there was plenty of players who still ignored the buttons!
Oh well I tried ...

Your maps have all had good secrets
Now I am curious, have you played my other maps? Do you have any comments/feedback on those? :) 

@Orl, thanks the demo, you really dealt with the bob's well and you still had plenty of ammo left for the gug at the end! I suspect that GLQuake error you are getting is something related to Quoth, which version of Quoth are you using?

@mwh, I really try to create maps for players to explore and less about room clearing, but this map is tricky because of the limited floor space which does force greater awareness of surroundings. Hope you have fun with the higher skill level. 

Interesting. I did feel the Gug didn't work so well in this one, and that the lightning mechanic was underused.

It is a pain in the arse setting up puzzles in Quake but it could have been cool to have the balance of the final arena 70% puzzle 30% fight rather than the other way around.

So the player has to activate electrodes func_doors to block the lightning shafts in a timed secret or something. Then each time they do it a mini-horde spawns in, until the final one where the portal opens.

Setting up the logic for this kind of thing is extremely difficult though. I've lost count of the number of times I've tried and tried only to finally give in and drop in the big monster instead.

The lightning theme could also have been continued throughout the map as a signposting method, activating them in places where the player needs to go after teleporting in roder to orient them...

Don't take this too seriously, just thinking aloud. 

timed secret = timed sequence 

@ijed, sounds like an awesome idea! If I had the extra time for this project I would have done more interaction and consistent detail but this map was really just a quickie.

Creating wave spawning final encounters is far easier than trying to create entity logic and there is also the expectation of QPB (Quake Player Baggage) that players expect the game to work a certain way and anything different will need to be introduced gradually or beforehand. There is also the problem that even with introduction / training you might still get some players ignoring all the examples and still wander around aimless wondering what to do.

The best way to address final arena setups is to create something new and different with the aid of Quake C. One of my favourite Quake final battles is the boss at the end of The Altar of Storms. A classic example of good QC, models, sound and interesting boss arena design. 

The various new entities in Quoth do give you some additional control for game logic though.

But I admit for the electrodes idea I was thinking of using a NOT logic gate set up.

You fire a lightning bolt at a button that receives damage.

If it's receiving damage during a certain duration then you don't pass to the next phase.

To block it you activate an 'electrode' which is just a func_door that gets in the way of the bolt.

I should probably implement this in my map now :D 

I did play some of your previous levels, but not all. It's been a busy year for me, and I have found myself playing fewer new quake releases than in the old days. I played all of Zendar, and at least loaded up some of the other ones.

Regarding the teleporter problem, i don't have a great solution for them (other than putting in lifts or ladders instead.) but your subtle arrows were actually well done for helping players find the alpha path, it's just that they do no good helping them find that secret ledge they previously spotted and are trying to get back to. I guess the other solution, that i alluded to earlier, is to make each "trunk" distinctive so you could find it again after teleporting. Like if one was red brick, one blue brick, etc (i realize this doesn't match your asethetic, it's just an example.) So you see the secret armor, you say "okay, it's on the side of the red brick pillar" and then after teleporting, you look around and find that same pillar.

Regarding the boss battle, one idea that would have fit with this map is to build the battle around the yellow bridges. Imagine a multi-tiered area with the exit at the top, and the player is successively activating yellow forcefields to ascend. This collection of forcefields and other platforms creates an open combat area that is a stage for waves of monsters, but becuase the stage is interesting (instead of a big plain arena) it would probably be more memorable.

Or, maybe you could invert the mechanic by having the playfield be divided up by a bunch of vertical forcefields, so that as the player disabled each wall it would release more enemies -- players could see the enemies in advance and maybe plan the order of forcefields to open, but they would have to open them all eventually (to get to the exit.)

Or imagine a vertical room with player at the top, looking down through 3-4 transparent floors, each one a forcefield. You would have to disable each floor 1 by 1 to drop down to the next room, and each room would contain guys. So if you rush from button to button you'll have all the guys still alive on the bottom floor with you. Or you can clear each floor before dropping down.

In all cases, the idea is to use your level's unique gimmick which players already understand, to create a playfield that is changing over time and those changes are relevant to combat, and maybe introduce player choice into how they change the playfield. 

...got around to playing it. The usual, on easy with a couple of vodka martinis under the belt.

http://www.quaketastic.com/files/demos/dis_metmon.dz

I will replay this level again, and on different skill levels. The language of the level was crystal clear this time, which meant I felt comfortable exploring and never felt lost. It feels like there could be heaps of different routes to complete this one, and I want to test that feeling.

Thanks Sock! 

Also had some peace at home to play the map :)

http://www.quaketastic.com/files/demos/tr_metmon.rar

Another pearl from Sock, thank you very much for this fantastic map.

the routes possibilities were fantastic. loved the end fight :) was especting hard core final but was easyer that I tought ;)

Sock you ROCK!