Nightly Playtests - Tuesday - Friday (21st - 24th Aug) 20:00 BST / 3pm EDT / 12pm PDT

We'll be starting on Tuesday after this 2-day break. I only just got confirmation I'll be here this weekend. Same deal as usual, turn up, 30 minutes testing new additions, then we play.

connect epic-empires.co.uk:27051

See you all there.

 

Content server has been updated with the following changes:

• Fix: Vehicles not being repaired on a repair pad when being damaged by friendlies
  
• Show a message to join a squad instead of auto-joining when spawning
  
• Added more info to refinery debugging command

 

Another smaller test tonight, 10 people will be enough, though more are always welcome, but we won't be regularly playing. Hopefully, if we track down the resource bug issue, then we can go back to full playtesting.

 

After debugging the resource bug more, I believe that there's actually no resource bug.

If you calculate the amount of resources gained over a certain time period, e.g. 10 seconds, the amount of resources gained will be correct.

However, because of how the resource rate is updated on the client, the resources will sometimes appear to jump higher than they should be in a single update.

What happens is that each refinery is on an independent timer that gives resources every second. The server updates the client view of the resources at a variable rate. Because of this mismatch, sometimes the resource output on the client will seem to jump higher than it should be in a single update. Over a period of time though, the resource rate is accurate.

 

I think we timed it over 1 minute and received 244 res with 2 refs (1+2 = 3 res/sec) on mvalley yesterday.

 

I tested this just now creating my own server which will not have much lag. On mvalley with 2 refs (base + A4) so +3 res a second. Over a minute using an external stopwatch I got 3.05 res a second. Using the in game timer I got 2.95 res a second. I think the +-0.05 is probably cause it is kind of impossible to start it without some millisecond difference.

Also I play windowed and I noticed if you click out and than click in, it will display the + res you got while you clicked out. For instance I clicked out for 4 seconds and said i got +12 res when I clicked back in the game. That means nothing I know.

 

yes i knew it walls are smaller

when unbuilt, nf wall is smaller than the be wall by about 1 unit, however when fully built the nf wall is larger than the be wall by a more legible amount
proof: http://i.imgur.com/IFAvD.jpg
if you prone from one wall to another you notice a larger change in elevation.

 

I still believe there's something wrong. We'll see what happens with more players.

 

Two problems.
1: crashed. I believe several players crashed at the same time too.
https://dl.dropbox.com/u/22159878/hl2_4104_crash_2012_8_23T19_42_47C0.mdmp -- solved (jephir) 0d9d54180a705dc49bd995aa3c3e48cc5cf0ea69

2:This weird thing shows up when I turn left or right rapidly, not sure what that is.
https://dl.dropbox.com/u/22159878/emp_slaughtered0070.jpg

 

Looks like glove or something.
I thought that's someone nocliping around me lol.

 

It's the 3rd hand, it's for reloading.

 

btw, Gren now has a reminder for enemy mines?
I didn't get defusal yet I saw the reminders, is that normal?

 

I've counted res again on a local server and I'm now convinced that Jephir is correct: There is no bug.

But:

• The player_refinery_multiplier is now scaled directly with playercount, and does not 'jump' to the next x2/x3 res output when reaching 12/24/36/.. players.
  This means that we actually get more res on average than with the current steam version, so that wasn't just an illusion.
• The +/-res display in the top left does not show res/sec but about 10% less on average...

 

Looking at the file history, in empires steam release, the player multiplier is rounded before being applied to the resource output. This means the resource output will jump at 12/24/36 players regardless of refinery rate.

In the current version, there is no rounding until the very final calculation (the change happened when diminishing returns was implemented). This means that the resource output jumps at 12 / output_rate. The effect is that the multiplier is applied at smaller intervals instead of every 12 players.

 

Content server has been updated with the following changes:

• Default skills selection now chooses more sensible skills depending on the player's class
• Fix crash

 

Felcis was an Enginner(not 100% sure, because I saw him running beside his LT's side armor, looks like he was fixing it)
But after he get in the LT, the attack order says that he's a gren.

Felcis was being a bitch and refused to tell me what was his class so I am not 100% sure with this.

You might wanna look into it if you have extra time.

EDIT: And this weird stuff in my console, from yesterday's test.
https://dl.dropbox.com/u/22159878/emp_slaughtered0071.jpg

 

oh fuck me the one day i have time and i forget

 

Lamer, Eastborough needs the District sudden death. Just saying.

 

first, thats a coder thing.
second, i dont understand why. it ends/keeps going totally depending on the skill of the players, no endless grenade spam or stalemate