Alienware R4 2017 - Are these normal temps?

If you absolutely need to use prime95, just use the non AVX version of it.

 

How to get that version any links?
I use a custom script on AVX2 to generate max. heat on CPU its fun to see numbers hitting 90C+ on PCs with improper cooling.

 

Since you're not capable of reading and processing information I'll stop replying to you from this point on.

@Mobius 1
What does prime95 exacly do to harm the CPU? Can you give some details on it? I'm interested to know.

 

The problem mostly stems from the retarded design of Haswell(-E)'s FIVR going out of control when AVX is run by P95.

The CPU draws so much current even with a mild overclock that it has the potential to degrade the silicon. Unless you cap the maximum power consumption, but even then I would still be really careful about it.



Prime95 v26.6 and before is okay, anything after I wouldn't use due to safety.






Personally for my own CPU stability test
OCCT large data + linpack, AIDA64 FPU, and ROG Realbench (must download latest version) is a good tool.

Can also use the x264 tool found on overclock.net

 

@Danishblunt check my post I edited some things.

 

So isnt this a haswell / Broadwell only problem then? Didn't skylake and kaby lake get rid of the FIVR?

 

That's a shame, I was really looking forward to your reply...



Sent from my SM-G935F using Tapatalk

 

It's a reason Intel pushed for AVX Instruction Core Ratio Negative Offset: " This setting reduces CPU core frequencies by the applied value when an AVX workload is run. The thermal output of AVX workloads is an order of magnitude higher than for non-AVX workloads, which is why this setting has been introduced. By using this feature, we can utilize a higher operating frequency for light-load applications while heavy-load applications that contain AVX code will downclock the processor to help keep core temperatures below the throttling point." Maybe you haven't seen or heard about this feature?

Use Aida64 with FPU+GPU for stress testing with Unified Heatsink. Better option than a pure single Cpu stress test. You can also combine Aida64 FPU with Heaven. All you really need.

 

Even GPU Caps Viewer can be used if you want to see a compute heavy CPU+GPU rendering. Julia 4D rendering on OpenCL CPUs w/ MSAA 4x or 8x at custom resolution or default resolution. As for GPU, Rain Forest using OpenGL 3.2 hammers the GPU.

 

Hello guys, I am here to tell some good news.

Today, after 2 weeks waiting for the new heatsink kit that came abroad my wait was worth it.

The Dell technician came and the first thing that I noticed is the size difference of this new heat sink from CCI (Fans are still SUNON) compared with my previous one.

LEFT = OLD
RIGHT = NEW

One is 85W and the other one is 125W

During the replacement I removed all the crap paste of Dell and put my new Thermal Grizzly Hydronaut (only a thin layer on the die of the CPU and the GPU), all thermal pads came properly.
now my worst temp was 82 celsius , idle now reduced to 37-42 CPU, GPU is alwaus under 40, stressed 71 at most
I am very happy the results, now my CPU can sustain 3.9Ghz without any Undervolt.

Quick facts
CPU BEFORE(Celsius)
Idle: 57-60
Full 92-99

GPU BEFORE(Celsius)
Idle: 57-60
Full 92-99

CPU AFTER (Celsius)
Idle: 38-45
Full: 67-71

GPU BEFORE(Celsius)
Idle: 54-60
Full 78-84

My 3D Mark benchmarks have also increased.

I will run my tests and let you know.

 

What are you core temp differentials?

 

Very low now, 3-4
Celsius so far


Enviado do meu iPhone usando Tapatalk Pro