How Is The Game Performance At Event Maps?

I'm just curious, to see if this is due to client limits, bad hardware, or both.

If you want, you could reply with your vote number and CPU/GPU specs so that we can see what kind of difference specs really make, and people with lower specs can see what they'd have to buy to have smoother game play.

**Hey Guys, when posting specs, please try to post the model of the cpu. Example:
"Intel Core i5-4570" Instead of "Intel Core i5 @ 3.20GHz". There are many different cpu releases with the same frequency and same common name. (About five i5 cpus rated at 3/20GHZ)
If you, don't I'll have to guess and pick the average one closest to the weakest.**

I'll post specs with vote number here for quick reference, and convert all specs to benchmark* scores. (Higher, better)

Freeze Ups When Near

PosterName	CPU Score	GPU Score

FPS Lowers Drasticly But Still Playable

PosterName	CPU Score	GPU Score
Oldie	1023	177
yiks	2172 - 3020	602
Doritos	6479	4969

Little to No SlowDowns

PosterName	CPU Score	GPU Score
jmmainvi	3229	456
Christopher Chance	12729	827
Matt	11249	4925
Goofy	7163	9036 + 9036

*http://www.cpubenchmark.net

Spoiler

 

(1)
CPU SCORE: 1023
GPU SCORE: 177

Okay so apprantly my gfx card is like one of the shittiest ones. Serves it right for being 8 years old!

 

I don't know how to get any figures but I'm playing on OSX Yosemite on my Macbook Air and I've faced slight lag; still playable though as long as

1. I turn my graphics to the lowest settings
2. I GFX when black boxes start appearing
3. I don't have more than one client open (with my characters being close enough to Slime to induce lag)

I can hit King Slime consistently and only back off from it when lag starts to worsen and I can't see myself hitting Slime repeatedly (earliest this has kicked in is 50% of Slime HP). I also back off when its HP reaches its last 10% or so to avoid any potential DC when Slime dies.

 

I don't have unbearable lag but I do get crashes that say C++ error or memory limit reached at times. I don't have exact error messages because I'm scrambling to relog my characters.

 

Barely any lag on parallels on my macbook air.

Mac client crashes instantly with an "unknown error" message.

1.8GHz Intel core i5
4 GB 1600 MHz DDR3
Intel HD Graphics 4000

Running Win XP sp3 on Parallels 10, from OS X 10.9.4 (haven't upgraded to yosemite. Probably should. Don't care.)

 

It's possible to do the event boss on mac OSX, but you'll have to hit it a couple of times and then walk away from the boss (preferably as far out of sight as possible). Then wait until it's dead. Sadly there's still a chance you'll dc, but it's much lower then when you stay right next to it.

 

I can't even get into the amoria map when the slime isn't summoned and it's just everyone AFK. It's an insta-crash since the slime was implemented. (I could go there to exchange wish tickets previously.) It's like I have the wrong .wz files, but I'm using the same files just fine on parallels so that can't be the problem.

 

Windows 7 Ultimate x64
3.20GHz Quad-core Intel Core i5
8GB 1600MHz DDR3 SDRAM
AMD Radeon R7

Seemed fine for the most part

 

3570k @ 4.8
2x 780ti @ 1300/7500
16gb

No lag at slime. Game just closes without warning.

The game was often unplayable at snowman (1 frame per 30 seconds--not exaggerating). Sometimes 1 client would be unplayable while the sandboxed client would be perfectly fine.

 

I'm pretty sure it remembers you hit it after you'ved relogged, as I dced multiple times after hitting it and still got a box,

You're right! I've always found that the sandboxed client lasts longer. Fun things I'll look into that.

Sorry I forgot to post where I get it from.

I'll convert to benchmark scores because you can't tell the performance of a processor from the GHZ rating.
For example, would you have thought that these two processors have similar performance?
One runs at 0.8GHZ, 2GHZ max, while the other runs at 2.8GHZ all the time.

http://www.cpubenchmark.net/cpu.php?cpu=Intel+Pentium+G2010+@+2.80GHz&id=1854

http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+M-5Y10+@+0.80GHz&id=2316

There is a time difference of when they were released, but again that wouldnt have been possible to check without a research while a benchmark score tells you off the bat.

 

Exactly, it's quite annoying when people just give a marketing designation and speed as opposed to an actual model number =/.

Spoiler: Relevant

Source

edit:
and my maple story clients do slow down a lot when events spawn, and i'm on a 2500k @4.5GHz and a 760 at whatever the hell the oem clocked it at. The game doesn't even max a thread though in fullscreen mode, it's just not designed to have so much going on.

 

Rough guesses on your benchmark scores since there are many i5's quad core 3.2GHZ and many R7's hehe.
edit: I'm not even going to guess... much too large of a range

Exactly! If only they marketed using # of instructions per cycle.
(How many instructions it can process per cycle, where x GHZ represents x billion cycles per second.)

@Doritos you must be talking about this 760!
http://www.videocardbenchmark.net/gpu.php?gpu=SiS+760&id=1900

 

Spoiler: offtopic

It's not as simple as x instructions in a cycle. Different instructions traverse the pipelines differently, some can be ran in parallel, weird caching sometimes even happens at the instruction level these days rather than higher up in the regular cpu caches. It's a very complex topic, which is why improvements are still being made. To be honest, you'd need to spend a lot more time than I have (only designed a RISC processor before) to even come close to understanding how you'd analyse the relative efficiency of various architectures without resulting to gross benchmark scores (which sane people do instead =D).

Famous example: bulldozer architecture consisting of a shared FP processing unit per two cores. Some multithreaded scenarios (integer heavy) it performs relatively well, others extremely poorly.

yeah, that's the one!

 

alright thanks. yeah I know it'd be more complicated than just ipc. just as how server cpus are sometimes designed different than desktop cpus.

 

**EDIT** I guess it was mostly client limits as some server edits solved most of the lag.

 

Too lazy to check on everything, but I'm using a completely brand new laptop with nothing but Royals downloaded on it and I still lag/crash every single time I get near the slime.

 

I have to do it the same way, though I still lagg out every now and than (on average 1,5 times per slime)