222 replies to this topic

[xengk]

System Specs
CPU: AMD Phenom II x6 1090T 3.2Ghz
RAM: 16Gb 1333 DDR3
GPU: Sapphire Radeon HD6850
M/B: Gigabyte GA-MA770T-UD3P
OS: Windows 7 64bit Ultimate
DX: DirectX 11
Country: Malaysia (South East Asia)

Settings
Resolution: 1440 x 900
Quality: Medium
Motion Blur: Off


Pro

• Frame rate remains the same for PTS and Live, getting 50+FPS average with occasional minor drop during brawls.
  
• HSR seem to be working most of the time for all weapon type.

Con

• Only suffer crash to mechlab once, that is before Matthew discover there are server port not utilized. After that, there have been no crash. [Fixed]
  
• Ping is slightly increased, from 280~350ping to 300~400ping.
  
• Missile firing seems to be delayed. Even with bay doors opened they suffer a 1 to 2 second delay in leaving the tubes.
  
• Charlie lance needs a different colour on the mini-map.
  
• Phantom damage still occurs, taking damage with no damage indicator or screen shake. Possible due to HSR and high ping.

Gameplay

• AMS and ECM are now very vital equipment for a team, as the number of missile tubes has increase generally.
  
• Having token back amour are now very dangerous, as flanking are more common and frequent.
  
• Fast light with JJ seems to be thriving in the chaos.
  
• Traffic jam in canyon map at middle and side valley lane, the slop area in the middle is now a death trap.
  
• Medium mechs are dying even faster due to high number of focus fire. I find myself trimming down ammo and DHS to install bigger engine.

Summery
Overall, 12v12 is working but with minor hiccups.

[Araara]

soarra, on 11 July 2013 - 11:29 AM, said:

good thing you showed your builds on both server..
i can put shs in a jenner and show similar results

Soarra, i also did the testing on the test server + normal servers for 6 ML and 15 DHS on a jenner F.

Alpha striking with 6 lasers do incur a massive heat penalty (38-39%) compared to shooting 3 med lasers.

There's also a slight penalty while shooting 4 med lasers (+2% more) and a bigger penalty at 5 med lasers (+5%) relative to normal servers.



edit : if paul said they didnt change anything heat balance wise, then i guess they ****-ed up the test server like the training grounds or something (aka no pilot skill or heat sinks value or something). In any case, an official post on what the problem is / official fix would be nice.

Edited by Araara, 11 July 2013 - 11:36 AM.

[Monky]

Chiming in on ammo concerns - I did run dry about halfway through a match with my 2x UAC5 configuration on my highlander - most shots hit and wasn't spamming it, normally my ammo will last until the exact end of a match on 8v8 - possible ammo capacity issue in 12 v 12s.

[C E Dwyer]

can't find the post now, but someone called for a 25% increase in ammo ..

I've not idea what the persons build was, but I ran a cat C-1 with 2x LRM 15 ran out twice in the 6 games I managed to start, but was there at the end for 5 of them..

I'd have to disagree, leaving things as they are, will give people a stark choice, stop using lrm70/ ac 40 boats or run the risk of being empty half way through the match

[Vraiment]

I got a message saying "This account is not authorized" anyone else?

[Zee SugarCube]

If you don't mind, I'll allow myself to express my thoughts in my native language:
1. FPS действительно скачет в зависимости от ситуации, как некоторые уже отметили, в среднем он немного ниже, чем в "обычной" версии.
2. Мой пинг(ping) был на 10 миллисекунд ниже, чем обычно.
3. Отличная идея, добавить вокруг иконок на мини карте чёрный контур
4. Действительно, нужно немного увеличить количество патронов на тонну, потому что в "12 на 12" они стали кончаться быстрее(хотя, наверное нужно просто метко стрелять)
5. На маленьких картах играть не здорово, весь бой превращается в обычную мясорубку без определённой цели. Если бы я мог - я бы убрал эти карты из списка.
6. И вообще, было бы неплохо в будущем делать ограничения на количество мехов на карте в зависимости от размера карты(5-6 человек для River City, например). Разумеется, для этого нужно ввести договорные матчи.

[PEEFsmash]

soarra, on 11 July 2013 - 11:29 AM, said:

good thing you showed your builds on both server..
i can put shs in a jenner and show similar results

I livestreamed the whole thing @ www.twitch.tv/peefsmash. Go there and watch "Previous Broadcasts." You can find the builds for yourself.

[soarra]

Araara, on 11 July 2013 - 11:33 AM, said:

Soarra, i also did the testing on the test server + normal servers for 6 ML and 15 DHS on a jenner F.

Alpha striking with 6 lasers do incur a massive heat penalty compared to shooting 3 med lasers.

There's also a slight penalty while shooting 4 med lasers (+2% more) and a bigger penalty at 5 med lasers (+5%) relative to normal servers.

must have been a bug, the guys i ran with and myself saw no difference in heat

[Sprouticus]

Oh, one additional note I just remembered:


The bug where the missile lock...pauses.....right before it turns red seemed worse on my Treb (2xALRM15, TAG).


If you are not familiar with the bug, sometimes when you try to lock on, it will freeze on you just a second before lock, sometimes for 2-4 seconds. It is really a bother because I often waste shots since I am assuming that the lock time will be linear and I launch too early trying to maximize the time locked.

[Roburn Bliss]

Vraiment, on 11 July 2013 - 11:35 AM, said:

I got a message saying "This account is not authorized" anyone else?

Test has ended Another one is on later today

[Koniving]

Tank Boy Ken, on 11 July 2013 - 11:18 AM, said:

Mech need more Hitpoints!

Band-aid to underlying problems 1) Insane heat thresholds after 22 DHS + 20% threshold increase and 14% faster cooling with 2x basic efficiencies and 2) Many single-shot ACs in a lore where the ACs are supposed to be multi-shot bursts to get said damage, and concentrated pinpoint PPC weapons. In other words pinpoint point-and-click convergence akin to Call of Duty's sniper rifles.

Problem 1 allows insane amounts of alpha strikes that would otherwise destroy a mech instantly. i.e. 6 PPCs. A proper threshold is at 10 standard heatsinks, where if you double the cooling would still never allow even 4 PPCs to be fired at the same time without a shutdown. You'd recover faster, but a proper threshold would never let you alpha 4 PPCs. If you were to fire those same 4 PPCs a second time at 50% heat you'd self destruct. Having a bigger heatsink on a processor doesn't change the temperature at which the processor fries, just how fast it cools. The same should be true for our mechs. Faster cooling with DHS only, remove incrementally rising thresholds based on how many heatsinks we have. "1.4" heatsinks should not be 3 times superior in overall efficiency than standard heatsinks, which they are due to increased thresholds.

Problem 2 causes several issues. Aside from not being lore friendly (by allowing us the 203mm AC/20 which is the only single shot AC/20 in existence in BattleTech. It is a weapon listed as can only be mounted on certain tanks because it would cause an Atlas firing it while walking to fall over on his rear -- hence why it is not mounted on mechs in lore), it allows insanely fast and cheap kills that required our double armor, double internals and even then still causes balance issues. This just taps the surface, honestly.

Edited by Koniving, 11 July 2013 - 12:07 PM.

[Sprouticus]

Vraiment, on 11 July 2013 - 11:35 AM, said:

I got a message saying "This account is not authorized" anyone else?

They shut down access 38 minutes ago.

[Vraiment]

Roburn Bliss, on 11 July 2013 - 11:37 AM, said:

Test has ended Another one is on later today

Oh God, I got it all wrong, I tought it was from 12 to 2 PM, that's a shame I guess I'll see you guys this afternoon

[Hellen Wheels]

Araara, on 11 July 2013 - 10:27 AM, said:

For whoever testing live, check out your heat to see if there's any change. A.....firing 6 medium lasers on a jenner F with 15 double heat sinks without moving/JJ brings me up around 38-39% on forest colony, which would be around 10% more than usual. (and thats while moving + JJing)

it also seems 4 and 5 medium lasers is hotter than what I'm used to in the normal game server. Can anyone confirm?

I can confirm that you are the first of the "Excel-Warriors" to post their Excel-lent observations.

Please disregard your findings while the rest of us play MECHwarrior and not EXCELwarrior.

[Lazy Eye]

River City (Day) seemed significantly lower FPS than usual. In general, there was a LOT more tearing than I would normally get, but overall, it was the same as live.

Edited by Lazy Eye, 11 July 2013 - 11:52 AM.

[Sidekick]

First 3 matches were fine. After that, I had only dcs and crashes. Cancelled the experiment after 20 fruitless minutes.

[MrPenguin]

Karl Berg, on 11 July 2013 - 11:21 AM, said:

A lot of the 12v12 optimizations did indeed focus on HUD and render costs. Would you mind posting your CPU/GPU, as well as anyone else who noticed a significant drop in performance for this test? It would be very much appreciated.

>50% FPS loss on my end compared to 8 v 8.

Intel core i5-2450M CPU @ 2.50 ghz

6 GB of RAM

Geforce 525m running latest GPU drivers

OS: Windows 8

Edited by MrPenguin, 11 July 2013 - 11:54 AM.

[Wolf Ender]

Only had one match with a problem, got kicked back to the mechbay after trying to drop. only happened once.

overall FPS is only 5-10 FPS lower than normal, hardly noticeable on average.

basically it was freaking awesome...so can we get 16 v 16 ?????

[William Rahn]

PEEFsmash, on 11 July 2013 - 11:24 AM, said:

Firing 6 MLs DOES incurr extra heat penalty on Test Server. I might have been wrong about it happening at 4. Here is proof that shooting 6MLs, (not OVER 6, but exactly 6) incurs extra heat.

Test Server: http://www.twitch.tv...smash/c/2560918

Normal Server same build: http://www.twitch.tv...smash/c/2560957

No - this is proof that when you fired 6 lasers you saw the red bar reach a certain level and a certain number was displayed on the HUD. The actual numbers are (or were - if PGI didn't keep the data) on the server. One of the things working differently between the test server and the usual one might be the netcode and it may result in the client showing something different (like reacting faster/slower to "reality").

A useful test would be to find a hunchback build with lots of MLs and few heatsinks, that on the normal server goes to 95-99 heat. If there is extra heat for MLs it should shutdown on the test server (provided that it is the same map and that pilot inefficiencies work correctly on both servers).

[Nightcrept]

Cathy, on 11 July 2013 - 11:34 AM, said:

can't find the post now, but someone called for a 25% increase in ammo ..

I've not idea what the persons build was, but I ran a cat C-1 with 2x LRM 15 ran out twice in the 6 games I managed to start, but was there at the end for 5 of them..

I'd have to disagree, leaving things as they are, will give people a stark choice, stop using lrm70/ ac 40 boats or run the risk of being empty half way through the match

I didn't run short with lrms but I only used them a few times.

What I did was I tried to run at least all my build styles at least a few times.

So ballistic, energy, and missile etc.


With almost any ballisitc boat you're going to run low or out of ammo and be less combat effective then a energy weapon based mech.

I had that basic experience when I used my ballistic mechs and energy mechs.
I also ran into many enemy mechs during gameplay who had ran out of ammo. Not fun for me to win because they ran out of ammo and I had a mehc without ballistics.

So with the current 12 vs 12 set-up energy/gauze sniper builds are even stronger and will stay so unless the ammo per ton limit is increased.