Monday, 2 May 2016

Mapping Progress

Yup, I'm still alive and I haven't quit UT4.

Most of the maps I'm working on are getting quite large though and I don't want to subject players to constantly having to download large new paks all the time. So while I've slowed down on uploading new paks I am still working on the maps.

Lately I've been doing a lot of work on my DM-Bishop remake. Mesh work, showdown support, cleaning up the navmesh, collision, lighting and so on.

A couple progress screenshots:



Wednesday, 20 April 2016

DM-Codex for Unreal Tournament 4


DM-Codex-A6 - Released 20 Apr 2016


Remake based upon Cliff (CliffyB) Blezinski's original map from UT99. Codex is a smaller, fast paced map with lots of vertical action.

Server Info:

[/Script/UnrealTournament.UTDMGameMode]
MapRotation=/Game/Maps/DM-Codex-A6

[/Script/UnrealTournament.UTBaseGameMode]
RedirectReferences=(PackageName="DM-Codex-A6-WindowsNoEditor",PackageURLProtocol="https",PackageURL="ut.rushbase.net/MoxNix/DM-Codex-A6-WindowsNoEditor.pak",PackageChecksum="7d200438c10e642fdd83525d51412306")

Changes:

  • Changed, replaced and/or rescaled many materials.
  • Added physical materials properties to blocking volumes so when you walk on a wooden surface it sounds like you're walking on wood.
  • Completely reworked the lighting (again).
  • Added a few new light sources and removed several unsourced fill lights.
  • Added several windows to bring in more ambient light.
  • Made the outdoor lighting a fair bit darker, changed sun and skylight colour to suit the sunset on an overcast day look of the sky.
  • Added atmospheric details like water, rocky cliffs and castle/fortress type walls to the outdoor (not playable) part of the map.
  • Raised ceilings and/or roof beams in several places to better facilitate movement.
  • Redid the dripping water sounds in the lowest parts of the map.
  • Added more trim and decorative bits.
  • Added several new meshes.
  • Replaced some old placeholder meshes with meshes that fit the theme better, removed a few meshes entirely and added many more mesh instances.
  • Added 3x UTGenericObjectivePoint for Showdown.

The assets (meshes, materials, emitters, etc.) in this map are mostly a mix of stock assets that come with the UT4 Editor and assets from the UE4 Infinity Blade Firelands pack (free on the UE4 marketplace), along with a few assets from other UE4 demos and a few self made assets.

Feedback is always welcome!

Thursday, 7 April 2016

DM-Eliminator for Unreal Tournament 4

DM-Eliminator V18 - 07 Apr 2016



A medium sized original map that plays well in FFA and Duel (assuming powerups off in Duel). Untested in TDM and Showdown (feedback would be greatly appreciated). Previous versions (1-17) were named DM-Octagon.

Many of the meshes and materials in this map are from the UE4 Infiltrator Demo.

Version changes:

- Removed Redeemer
- Removed Invisibility
- Added 3x UTGenericObjectivePoints for Showdown
- Added 1x sniper ammo and 1x rocket launcher ammo

- Many (mostly minor) tweaks to geometry and blocking volumes (collision).
- A great many cosmetic changes, more meshes, materials changes, lighting tweaks, etc.
- Lots of finicky little details, polishing, fit and finish...

Server Configuration Info (goes in Game.ini):

[/Script/UnrealTournament.UTDMGameMode]
MapRotation=/Game/Maps/DM-Eliminator-V18

[/Script/UnrealTournament.UTBaseGameMode]
RedirectReferences=(PackageName="DM-Eliminator-V18-WindowsNoEditor",PackageURLProtocol="https",PackageURL="ut.rushbase.net/MoxNix/DM-Eliminator-V18-WindowsNoEditor.pak",PackageChecksum="9d4cb94240cc6f23d49576e4744a4dc1")

Additional Notes:

A full list containing links to the most recently released versions of all maps I'm working on can be found here: http://merchantmonarchy.blogspot.ca/p/unreal-tournament-4.html

Feedback is welcome!

Wednesday, 23 March 2016

Connection Issues Fixed

I've been having some issues with my internet for quite a while. For browsing, email and other routine things everything seemed to work fine. There was the odd time where I noticed my internet connection dropped for a few minutes but I figured that was just minor ISP trouble.  It never lasted for very long and I didn't worry about it since it wasn't really affecting me.

But since I started playing Unreal Tournament 4 pre-alpha I noticed a couple problems.

1. I run my own UT4 LAN server on a 2nd machine at home. In the in-game server browser my server would show up listed in red for other machines on the LAN but worked fine for remote users connecting from the internet  (not on the same LAN). The only way to connect to the server from the LAN was connecting by IP.

I figured that was just a problem with the game. After all it is pre-alpha and quite buggy.

2. When uploading large map files to rushbase, the upload would timeout after a while and just stop. It was difficult uploading maps much over 100M, often having to restart the upload multiple times before it finally uploaded successfully. The larger the file the more difficult it was to upload, once a map got around 200M it was impossible.

That definitely should not be happening and I was certain it wasn't my computer. Not hardware, software, nor drivers. I'd checked and rechecked everything (configuration issues, driver versions, etc.) many times looking for something, anything wrong with my computer that could potentially be causing connection issues.

I also noticed when a file upload failed my internet connection would often go down for several minutes. Later I realized during that time I couldn't even login to the modem's built-in router which made me start to suspect it might be a problem with the modem... But I'd already replaced the modem twice.

After doing a lot of searches on Google I finally found some information about others fixing similar connection problems by replacing the built-in router in their ISP provided modem with their own router and a plain modem (one without a built-in router).

Getting my own router to replace the built-in router in my ISP's Hitron modem might help with the file upload problem.

I did some research on routers hoping to find a decent router in the $50-125 range and see if that helps. I read a whole bunch of "best" router reviews and articles, most of them on reputable technical sites. I noticed one particular router kept getting mentioned as a really good cheap router.

TP-Link TL-WR841N

Practically all of the reviews mentioned this router and the ones that did all had really good things to say about it. It's also the best selling router on Amazon and gets great reviews from buyers there too.

It's really cheap too, around $30.00 CDN regular price, $25.00 when on sale. I figured at that price I might as well just grab one and see if it fixes the upload problem... If nothing else, it'll make a good cheap backup.

I got one the next day, got a plain modem (with no built-in router) from the ISP to replace the Hitron modem and hooked everything up.

And it worked!

It solved my upload problem, now I can upload large files without trouble. Uploads work on the first try, every time now! :)

Even better and completely unexpected... When playing UT4 my LAN server is now listed white in the server browser list and I can join it just like any other server in the list without having to use "Join by IP".

All in all it's the best 33 bucks (the local price was a little high) I've spent in a while. I'm very pleased with this router and highly recommend it to anyone looking for a good but cheap router.

Saturday, 19 March 2016

Unreal Tournament 4 Dumb and Dumber

UT4's rAgetard is still at it, constantly posting garbage "statistics" in his never ending crusade to somehow "prove" the incredibly versatile shock rifle is gimped and most other weapons (but especially the flak cannon) are far more powerful than shock.

 As always he separates shock combos from shock primary and secondary fire while combining damage from all firing modes for other weapons. If we add up the damage for shock primary and combos too we get.

DM-Coma
450 + 431 =881 damage

DM-Erase
1205 + 850 = 2055 damage

2055 total damage from shock on Erase very nearly double that of the next closest weapon, the rocket launcher. Flak is in 4th place.

Damage from shock secondary (which wasn't included at all) would push total damage from shock up even further though probably not a whole lot.

What the idiot actually just proved is that the shock combo alone (completely disregarding all other damage from shock rifle) is super efficient. On Erase it was better than all firing modes combined for any other weapon. When taking all damage from shock rifle into account (including primary, secondary and combos rather than just combos alone) shock was twice as good as the next best weapon (Rockets).

Sure shock isn't tops on Coma but Coma is a tighter map and flak is strictly a short ranged weapon, while shock is more of a medium and long ranged weapon, so that's pretty much how it should be.

Here he takes damage values from single hits with combos, comparing individual combos to what he calls "highest damage dealt by" all the other weapons "in one fight" coming up with values like 144 for link and 224 for mini. Considering mini and link don't do anywhere near that kind of damage from a single hit (single flak and rocket hits don't do 258 and 220 damage either) he's obviously adding up damage from all hits during "one fight" for everything except shock.

As always he's not even counting shock primary or secondary fire at all. Only combos and just one combo at a time.

And what's "one fight" anyhow?

- A single engagement, counting all damage until one player manages to break off and escape?
- A longer period of time counting all damage taken between deaths, which in a duel between top players engaging, disengaging, topping up health and armor repeatedly, might be several minutes?
- Something else altogether?

The numbers in this second screenshot are completely meaningless. The only thing it proves is the instantaneous damage from a single hit with one weapon (even a powerful hit like a shock combo) doesn't match sustained damage over time from multiple hits with even the lightest hitting weapons.

Hardly an objective comparison at all, these so-called "statistics" are about as biased as it gets, totally misleading and painting a completely inaccurate picture of what actually happened.

But hey, this clown is allowed to keep spewing lying, misleading bullshit like this all over the UT forums, while disagreeing with and debunking his garbage will get you banned from the forums just like that (snap fingers)... Go figure!

Saturday, 12 March 2016

DM-Chill Visual Clarity

There's been a lot of gnashing of teeth and throwing of tantrums over visual clarity on the newly released fully meshed version of DM-Chill.

While it's not nearly as bad as some players claim, there are a few minor problems with seeing other players in some spots. Especially necris characters with their white skin and nano black camo clothing blending into the predominately dark grey stone and white snow background.

In addition the shadows are so dark in some spots it's nearly pitch black, making necris practically invisible and other character models very difficult to see, especially if they aren't moving.

But instead of complaining about it, why don't these players who think they know everything and keep trying to tell everyone else how to do everything "properly" (their way that is) just fix it themselves?

It's not like it's difficult to do or takes a whole lot of time...

Here's what I managed in about 20 minutes of editing time (not counting build time).

1. Trees. A lot of players don't like the moving, swaying trees in the background.

- There are only 2 different tree meshes in the map with multiple instances of each. I selected one tree, then used right mouse button menu to select all matching meshes and hit delete.
- Rinse and repeat for the other tree mesh.

No more trees... That took about 5 seconds.

With a few seconds more one could easily do the same for those dangling rope meshes or anything else considered unnecessary eyecandy that hurts visual clarity or performance... Or you could leave them in and set LoD on such items so they only show on the highest settings but not on lower settings.

2. There are many instances of a decal on the floor for shiny, highly reflective ice (actually I think it's supposed to be frozen mud).

- I removed all instances of that decal. Since we can't use select all on decals to find decals with matching materials I had to find and delete them 1 at a time and there were a lot of them.

That took a while, maybe 10 minutes in total. Longer than any other single thing.

3. Lighting, brightening up the dark spots and changing PPV settings to make eye adaptation transitions to/from inside to outside and dark to light parts of the map as unobtrusive as possible.

- The map has 5 active (enabled) Post Processing Volumes (PPVs) and a 6th redundant PPV that is disabled.
- There are a real lot of disabled ("affects world" unchecked) fill lights all over the place.

First I tweaked the PPV settings.

- I set Auto Exposure minimum and maximum brightness to 1 in all PPVs. Effectively (sort of anyhow) turning Auto Exposure off and minimizing eye adaptation when moving between bright and dark areas.
- Then I tweaked AE Exposure Bias in all PPVs to brighten up the inside areas and darken the outside area. In the the unbound PPV I turned it down from 0.5 to 0 making the outside part of the map a little darker. I turned it up in the other PPVs, setting a couple to 1, a couple to 2 and one to 0.7.
- One of the PPVs had AE Blendables causing some really bright odd looking shiny spots. I removed the blendables to fix that.

Then I turned on 2 or 3 of the disabled fill lights and tweaked the intensity setting on one of them to brighten up a few of the darkest spots left in corners, doorways and beneath ledges.

That all took another 10 minutes or so.

Roughly 20 minutes in total not counting build time. Building with lighting quality set to production takes about 40 minutes but that's afk time so I'm not counting it.

And here's the result:

DM-Chill VC

The lighting could still use more tweaking in some spots but this isn't intended as a be-all, end-all, visual clarity version. It's simply intended to show those who want a version of Chill with better visual clarity just how easy it is to do it themselves.

Edit: While FPS wasn't a factor in my reasoning for making this version, I have been informed by several players they see a 15-20 FPS improvement on this version over the unedited map.


Sunday, 6 March 2016

Unreal Tournament 4 and the Village Idiot

There are a lot of bickering idiots pushing hidden agendas on the UT4 forums. Posting ridiculous bullshit and in some cases even outright lies in an attempt to convince Epic to change the game to favour their preferred playstyle to the detriment of everyone else.

But UT4 has it's "special" problem child, constantly posting ridiculously contrived scenarios, laughable "proofs" and other nonsense while crying for weapon nerfs just about everytime someone kills him.

The village idiot likes posting what he calls statistics, taking small carefully selected samples and manipulating the numbers to make it seem like:

1. The Flak Cannon is the best weapon in the game, extremely overpowered and in dire need of nerfs.
2. The best, most versatile and one of the most used weapons in the game (Shock Rifle) is severely underpowered and in dire need of buffs.

For example his latest idiot "Statistics" post:


This raging lunatic puts Flak as the number 2 killing weapon on his list after adding up the deaths from both Flak firing modes. But for Shock he splits it up into primary fire and combos as if they were completely different weapons and just leaves Shock secondary firing mode out altogether to make Shock look much less powerful than it actually is.

234 deaths from Shock combos + another 180 from Shock primary fire = 414 already slightly more than from Flak and that's not counting deaths from Shock secondary fire which would bump it up even further ahead of Flak.

Busted again!

And of course like always he makes absolutely no mention of rockets which are pretty much always the number 1 killer. It's difficult to find a scenario where rockets aren't number 1 even when trying to manufacture one. You can look at virtually any player's stats and clearly see everyone both gets the most kills with rockets and dies the most to rockets.

The village idiot's solution to overpowered rockets?

Nerf flak! Then once Flak has been turned into a Marshmallow Cannon, nerf Enforcer, Link and Mini... Probably Bio too for good measure.

Basically he wants to nerf everything except Shock, Snipe and Rockets... Can you guess which weapons are his favourites?

Note to Epic:

Stop reacting to idiots, just remove the collision radius from projectiles (mainly rockets and flak) and see how that changes things.

Then take a long, hard look at the netcode (ping compensation in particular) and fix it so more players can actually see incoming fire, where the other guy is, what he's doing and react skillfully rather than just randomly hopping about and praying the ping comp gods favour them.

A server side option to turn ping compensation completely off would be a great start. Then we could find out how much of the so-called weapons balance problems are actually ping compensation problems. Beyond that an option to set how much ping is compensated for (from 0 to say 120) would be nice too.

Oh, and don't you think it's way past time to set default player count numbers to reasonable values so everything besides duels aren't just giant spamfests?

8 players for DM/TDM
10 players for CTF