Guild Wars 2

The current build of the client is really taxing user’s CPUs. While tracing the stuttering/fps issue on my system, we discovered that launching the game would push my cpu temps up by 40 degrees celcius in a couple of seconds. (It was pushing my cpu to a temp of 102, which is “kill your chip” temps, and my fps problems were caused by my system throttling itself to stop from burning the chip out.)

Granted, I live on the prairie, and should clean out my system more often, but really…how many regular users even know how to disassemble their case, pull the heatsink, check the paste, clean it out and put it all back together? After doing all that, my system default cpu temp is 30-35c with 3 browsers, a couple of utilities and adobe cloud running.

If I turn everything off, and launch GW2, as soon as the coherent processes start spinning up, my CPU jumps to 75c. That’s pushing right up against the heat limit of non-overclocked cpus.

I’ve tested this now on 3 different computers, each with their own account, and on each system, there is a massive surge in cpu temp, directly before the FPS drop.

There are no heat issues running any other game or app that I’ve tried. Support has a log with all of the games installed on my system. I’ve launched them all, and none of them gloms on to the cpu and doesn’t let go like the latest gw2 patch.

Because development boxes live in controlled environments, and are generally beefy and well cooled, this sort of thing could slip past easily. I recommend that someone on the dev team follow tech support’s directions; download speccy and watch what happens to your cpu load when you launch the game.

If tech support hadn’t recommended downloading speccy, and I hadn’t seen that my chip was spiking into temps 20c degrees hotter than the recommended range, I would have ended up burning out components, and never been able to figure out why.

duckideva.6358:

There are no heat issues running any other game or app that I’ve tried.

Try prime95 (CPU stress test).

GW2 is CPU heavy, like most MMOs, whereas normal games tend to be GPU heavy. Look at your CPU usage when comparing temps.

Personally, I’m not seeing any abnormal temperature rises when playing GW2. After cleaning, I’m assuming the 75C was your new 102C? That sounds about right for a max load CPU with stock cooling.

Oh my gosh, no. Nobody who isn’t overclocking, and familiar with what specs they’re looking for, should run Prime95. Prime is contraindicated with many cpus, and in all cpus that are not running aftermarket coolers.

Prime95 stresses a computer far more than the majority of software based stress-suites, because the operating system usually shuts down the floating-point unit when unused by other programs, whereas Prime95 is well-optimized to continuously and effectively thread the FPU, causing it to be deeply pipelined, thereby generating significantly more heat because of elevated power consumption under the massive workload conditions.

Nobody, who doesn’t know exactly what they’re doing, should run prime against their system as a stress test. That’s a fast way to melt down a consumer grade box.

(That said; of course I ran it when I was building the box…but it is not a tool for the regular user.) Edit to add: I’m not using stock heat sinks or thermal paste.

duckideva.6358:

Oh my gosh, no. Nobody who isn’t overclocking, and familiar with what specs they’re looking for, should run Prime95. Prime is contraindicated with many cpus, and in all cpus that are not running aftermarket coolers.

Prime95 stresses a computer far more than the majority of software based stress-suites, because the operating system usually shuts down the floating-point unit when unused by other programs, whereas Prime95 is well-optimized to continuously and effectively thread the FPU, causing it to be deeply pipelined, thereby generating significantly more heat because of elevated power consumption under the massive workload conditions.

Nobody, who doesn’t know exactly what they’re doing, should run prime against their system as a stress test. That’s a fast way to melt down a consumer grade box.

(That said; of course I ran it when I was building the box…but it is not a tool for the regular user.) Edit to add: I’m not using stock heat sinks or thermal paste.

This is just false. Much like every other software, Prim95 is NOT going to break your system unless you have a pre-existing HARDWARE issue to begin with.

Running Prim95 is the easiest way to see if your thermal equipment (IE. Thermal Paste, Fans, Heat sinks, Water AIO kits) are functioning properly.

The fact that your system is jumping up to 75c while the launcher is running tells me you have thermal issues you will need to diagnose and fix. And it could be as simple as dried up TIM under the Heatsink on your CPU.

The thermal paste is less than 48 hours old. The system does not show any sort of heat spike with any other application, including number intensive crunching. It happens on multiple computers.

It happens when the new Coherent-UI starts spinning up processes. On my system, each individual case of of CUI causes a temp rise. CUI shouldn’t be spinning up multiple processes, first off. Secondly, child processes should terminate if the trading post isn’t being used. What appears to be happening is that CUI is loading up an entire browser’s worth of cpu process with each thread, and is not releasing it.

On some systems, perhaps not yours, and certainly not in every case, but in some systems, the current patch is causing cpus to overwork.

duckideva.6358:

The thermal paste is less than 48 hours old. The system does not show any sort of heat spike with any other application, including number intensive crunching. It happens on multiple computers.

It happens when the new Coherent-UI starts spinning up processes. On my system, each individual case of of CUI causes a temp rise. CUI shouldn’t be spinning up multiple processes, first off. Secondly, child processes should terminate if the trading post isn’t being used. What appears to be happening is that CUI is loading up an entire browser’s worth of cpu process with each thread, and is not releasing it.

On some systems, perhaps not yours, and certainly not in every case, but in some systems, the current patch is causing cpus to overwork.

I have many systems which play GW2, and not a single one of them is having thermal issue’s at the loader or ingame.

if you are having issues its going to be your heatsink(or even the Heatpipe(s)), TIM, or fans.

But also, you dont tell us anything about your system. what CPU, is it over clocked, or your thermal cooling method. so really, who knows about your issue.

How many CoherentUI_Host.exe processes are there?

Having 2 of them is normal. One acts as the host while the other is the renderer, which is a child process of the host. When clicking play on the launcher, the host process along with its child should be killed and a new host process should start. After opening the trading post, a new renderer process should start, which should stay alive as long as the trading post is open. The host process should stay alive as long as the game is.

Watching the CPU usage of CoherentUI, loading the trading post uses roughly 1-4%.

Where is yours going wrong? If you have multiple copies, are they still using the CPU?

If your cooling system is functioning properly (and is adequate); you will be able to max all cores of your CPU to 100% usage for hours – without overheating.

Three days ago I started having seemingly random computer reboots while playing GW2. It happens pretty quick, anywhere from 2-10 mins. I’ve started monitoring my CPU/GPU temps and found that they shoot up while playing GW2. I’ve tried to “cook” my machine playing different games but they don’t don’t go high enough. Restarts seem to occur at 70 degrees c plus, which is about the cap for my machine. I’ve been running GW2 for 2 years on this machine every day and never had anything like this happen. I had my machine cleaned of dust with new silver paste put on the heat sink about 2 months ago.

The restarts didnt occur immediately after the last patch, though there have been many minor patches since the big one. I’m wondering if some overheating issue caused by the game itself has burnt through my paste so now nothing’s there to stop it from hitting my CPU’s cut out. But I’m not sure if that’s even possible.

I don’t know if it’s at all related but the issue did arise after I was swapping monitors around while the game was running.

Machine freezes, reboots and graphic driver resets are a symptom of hardware issues. Often, it is a heat problem. Sometimes compatibility issues between hardware components.
A properly built computer is able to endure hour-long prime95 sessions and furmark gpu test, even in parallel. Or GW2.

If it freezes/reboots after a few seconds or minutes, it’s often a heat issue. My machine is an example of this: when I built it 2 years ago, I ran prime95+furmark over night perfectly fine and the temperature rose high, but never over the documented cpu/gpu limit.

Just a few weeks ago, graphics driver resets started to happen when I was recording videos of my gaming sessions. I identified a heat problem indeed: I have a fan-throttling program running to keep the machine silent when there is no load, and I misconfigured it so the fan speed was too low on maximum load.
I fixed the configuration, so the fans ran at maximum speed when there was maximum temperature, and I cleaned out the dust in the machine.
After that, no graphics driver reset any more.

So if you detect a reboot due to heat problems, you cooling system is probably underperforming. There are big differences between cpu coolers. Often the stock coolers are the culprit: they are barely able to transport away the heat on maximum load, even if they are correctly mounted and not throttled. Most customers never run into this issue, because they never run CPU-intensive programs but only office applications. But people who do this, and GW2 is a CPU-intensive program, will definitely run into this. Here is a difference between a computer built for office use and a computer built for gaming use as well.

If you can melt a consumer box, it’s either a design flaw issue or a hardware issue with the cooling system. Period.

The idea that you can run a piece of software on any hardware and cause thermal issues and it’s the SOFTWARE’s fault is absurd…..it’s 100% a hardware being improperly cooled issue.

All of the above is assuming a reasonable ambient temp and adequate air flow to meet the manufacture specs.

I suggest before jumping to any conclusions about things that you take a look inside your PC case and carefully clean it of dust and any other debris which tends to accumulate on cooling fans and heat-sinks.

A soft dry paint brush, can of compressed air or a vacuum cleaner are great for cleaning the dust magnets. Obviously, be careful and don’t attempt to clean while the PC is powered on and running.