[SOLVED] i7-4790k 100°c after 5sec of benchmark

[Zzurhi]

Good day, i hope someone can help me with this problem..
I purchased a 2ndhand core i7-4790k online. I trusted the seller cause he got 5 positive feedback. I check first the processor using IPDT and its genuine intel core i7-4790k processor,
And when i play a games i notice its a bit laggy but the game i played is smooth on my core i5 4460.

When i touch the fan, on the side its not that hot. It feel it should be around between 50~60°c like when i compare it on i5, so i read a lot in forums and youtube and follow their guide.

When i check the temp using coretemp the idle temp is around 38~40°c and the temp when i used prime95 reached 100°c in about 5 sec i think, i immediately stop stress test..
My room temp is around 22°c
Heres the SS.



VID drop or core voltage drop to 0.7-0.8 it throttle down cause of heating right? Where it should be somewhere1.12-1.2v

Is it damage or do i need only to delid and replace TIM with liquid metal?

 

[TommyTwoTone66]

Stock i7 cooler.

Not all intel stock coolers are Equal. They come in various thicknesses and with a wide variety of fans. the 4th Gen i7 coolers were pretty tall if I remember correctly and had a big fat fan.

If you are using the original stock cooler that came with a 4790K from 8 years ago, and it's been in use all those years, then it is almost certainly time for a new one.

 

[Zerk2012]

Good day, i hope someone can help me with this problem..
I purchased a 2ndhand core i7-4790k online. I trusted the seller cause he got 5 positive feedback. I check first the processor using IPDT and its genuine intel core i7-4790k processor,
And when i play a games i notice its a bit laggy but the game i played is smooth on my core i5 4460.

When i touch the fan, on the side its not that hot. It feel it should be around between 50~60°c like when i compare it on i5, so i read a lot in forums and youtube and follow their guide.

When i check the temp using coretemp the idle temp is around 38~40°c and the temp when i used prime95 reached 100°c in about 5 sec i think, i immediately stop stress test..
My room temp is around 22°c
Heres the SS.



VID drop or core voltage drop to 0.7-0.8 it throttle down cause of heating right? Where it should be somewhere1.12-1.2v

Is it damage or do i need only to delid and replace TIM with liquid metal?

You just need to buy a better cooler the stock ones sux that is why Intel stopped including stock coolers on the K series processors.
EDIT in fact all stock coolers sux Intel and AMD!

For prime95 it's not something I use or recommend because their specific versions and settings that need to be used.
I recommend and use OCCT because it's a no brainer download and start the test.

 

[CompuTronix]

You just need to buy a better cooler the stock ones sux that is why Intel stopped including stock coolers on the K series processors.
EDIT in fact all stock coolers sux Intel and AMD!

For prime95 it's not something I use or recommend because their specific versions and settings that need to be used.
I recommend and use OCCT because it's a no brainer download and start the test.

Zerk2012,

Check out my previous post. While I agree that your suggestion to buy a better cooler is indeed very sound advice, your suggestion that OCCT is a "no brainer" is misleading, because it's missing some critical information. It's fine if users prefer to use OCCT over Prime95, but they still need to know which settings to use, so we need to be more specific.

If OCCT is allowed to use default settings, it runs "Instruction Set" in "Auto", which means that a processor with AVX Instruction Sets, such as the 4790K, will use it's AVX and AVX2 Instruction Sets. This will typically drive a 4790K running the stock cooler in a normal ambient environment to Throttle at 100°C, which is the same as Prime95 will do when allowed to use default settings.

As per Intel’s Datasheets, TDP and Thermal Specifications are validated “without AVX”. To eliminate AVX variables, both OCCT and Prime95 require proper configuration.

For everyone's benefit, we can not assume that users know which utilities and tests to run, and how to run them properly. It is necessary that we explain ourselves very clearly and completely, rather that just say "run OCCT" or "run Prime95" or "run AIDA64", which often leads users to inconsistent, conflicting and confusing test results. For example, AIDA64 has 4 CPU related stress test selections (CPU, FPU, Cache, Memory) which have 15 possible combinations that yield 15 different workloads and 15 different Core temperatures. That's a lot of variables.

These illustrations are from our Intel CPU Temperature Guide 2021:

Utilities that don't overload or underload your processor will give you a valid thermal baseline. Here’s a comparison of utilities grouped as thermal and stability tests according to % of TDP, averaged across six processor Generations at stock Intel settings rounded to the nearest 5%:

Although these tests range from 70% to 130% TDP workload, Windows Task Manager interprets every test as 100% CPU Utilization, which is processor resource activity, not actual %TDP workload. Core temperatures respond directly to Power consumption (Watts), which is driven by workload. As workload increases, CPU resource activity can only increase to a maximum of 100% Utilization, while CPU Power consumption can increase well above 100% TDP, especially if overclocked.

Prime95 Small FFTs (AVX disabled) provides the correct workload for testing thermal performance. However, if OCCT's first test, called "CPU", is configured for Small Data Set, Normal Mode, Steady Load and SSE Instruction Set (no AVX), then it's a steady-state workload at more than 97% that's nearly identical to the steady-state 100% workload of Prime95 Small FFTs with all AVX test selection disabled. A steady-state 100% TDP workload is key for testing thermal performance, so the CPU, cooler, socket, motherboard and VRMs can thermally stabilize.

Upon first run, either OCCT or Prime95 require about the same number of mouse clicks to get them properly configured for testing thermal performance.

As a footnote, when CineBench R23 (as well as R20) are run using the Multi Core test, although they use a realistic AVX workload, Power consumption and Core temperatures are nearly identical to both OCCT and Prime95 when configured as described above, where all three utilities are within a few watts and degrees of one another. Although the CineBench workload is relatively steady, since it pauses between renderings, which allows the processor to cool, it's not truly a steady-state workload.

Run these three utilities and compare Power consumption and Core temperatures for yourself. If you use an additional utility with graphs which overlay monitoring information, such as Intel XTU, AIDA64 or SpeedFan, then you can get a more complete overview and perspective of Power and thermal behaviors, rather than just observing numerical values.

CT

 

[Zzurhi]

Insufficient. And most likely improperly installed.
Get a better cpu cooler.

Gammaxx 400, CM Hyper 212 evo or similar.

insufficient - high probability
improper installed - negative.
well, i brought my cooler with me from home to test if it's really the cooler here is at fault.
MAELSTROM 240

Not all intel stock coolers are Equal. They come in various thicknesses and with a wide variety of fans. the 4th Gen i7 coolers were pretty tall if I remember correctly and had a big fat fan.

If you are using the original stock cooler that came with a 4790K from 8 years ago, and it's been in use all those years, then it is almost certainly time for a new one.

you have a point

Zerk2012,

Please read my previous post. While I agree that your suggestion to buy a better cooler is indeed very sound advice, your suggestion that OCCT is a "no brainer" is misleading, because it's missing some critical information. It's fine if users prefer to use OCCT over Prime95, but they still need to know which settings to use, so we need to be more specific.

If OCCT is allowed to use default settings, it runs "Instruction Set" in "Auto", which means that a processor with AVX Instruction Sets, such as the 4790K, will use it's AVX and AVX2 Instruction Sets. This will typically drive a 4790K running the stock cooler in a normal ambient environment to Throttle at 100°C, which is the same as Prime95 will do when allowed to use default settings.

As per Intel’s Datasheets, TDP and Thermal Specifications are validated “without AVX”. To eliminate AVX variables, both OCCT and Prime95 require proper configuration.

For everyone's benefit, we can not assume that users know which utilities and tests to run, and how to run them properly. It is necessary that we explain ourselves very clearly and completely, rather that just say "run OCCT" or "run Prime95" or "run AIDA64", which often leads users to inconsistent, conflicting and confusing test results. For example, AIDA64 has 4 CPU related stress test selections (CPU, FPU, Cache, Memory) which have 15 possible combinations that yield 15 different workloads and 15 different Core temperatures. That's a lot of variables.

These illustrations are from our Intel CPU Temperature Guide 2021:

Utilities that don't overload or underload your processor will give you a valid thermal baseline. Here’s a comparison of utilities grouped as thermal and stability tests according to % of TDP, averaged across six processor Generations at stock Intel settings rounded to the nearest 5%:

Although these tests range from 70% to 130% TDP workload, Windows Task Manager interprets every test as 100% CPU Utilization, which is processor resource activity, not actual %TDP workload. Core temperatures respond directly to Power consumption (Watts), which is driven by workload. As workload increases, CPU resource activity can only increase to a maximum of 100% Utilization, while CPU Power consumption can increase well above 100% TDP, especially if overclocked.

Prime95 Small FFTs (AVX disabled) provides the correct workload for testing thermal performance. However, if OCCT's first test, called "CPU", is configured for Small Data Set, Normal Mode, Steady Load and SSE Instruction Set (no AVX), then it's a steady-state workload at more than 97% that's nearly identical to the steady-state 100% workload of Prime95 Small FFTs with all AVX test selection disabled. A steady-state 100% TDP workload is key for testing thermal performance, so the CPU, cooler, socket, motherboard and VRMs can thermally stabilize.

Upon first run, either OCCT or Prime95 require about the same number of mouse clicks to get them properly configured for testing thermal performance.

As a footnote, when CineBench R23 (as well as R20) are run using the Multi Core test, although they use a realistic AVX workload, Power consumption and Core temperatures are nearly identical to both OCCT and Prime95 when configured as described above, where all three utilities are within a few watts and degrees of one another. Although the CineBench workload is relatively steady, since it pauses between renderings, which allows the processor to cool, it's not truly a steady-state workload.

Run these three utilities and compare Power consumption and Core temperatures for yourself. If you use an additional utility with graphs which overlay monitoring information, such as Intel XTU, AIDA64 or SpeedFan, then you can get a more complete overview and perspective of Power and thermal behaviors, rather than just observing numerical values.

CT

the moment of truth.

 

[TommyTwoTone66]

the moment of truth.

Welp, looks like you solved your issue. Looking at the layers and layers of dust and what looks to be.. sand? covering a lot of your components, I would say you probably live in a very dusty, hot place. Please clean everything thoroughly with isopropyl alcohol, get it all inside a case with very good dust filters, seal up any little holes or ways for dust to get into the case with electrical tape or stickers maybe... And change your dust filters often!

 

[TommyTwoTone66]

I've just opened this pic full size and seen the cooler you took off in the bottom right! I could not believe that would be the size of cooler that Intel would ship with an 88w TDP chip, it looks more like a Celeron cooler! But I looked up the original spec and that was the correct retail cooler for the i7 from 2013 when the chip came out. It's so tiny!

In addition to it being way too small for the heat that the chip pumps out, it has a copper core surrounded by aluminium fins. The core is bonded to the outer fins using some kind of adhesive, which doesn't get great thermal transfer to begin with, and I bet you that layer has worn away or evaporated over time leaving you with bad thermal transfer from the copper core. With nowhere for the heat to go it just builds up to 100C and the cpu shuts down or thermal throttles to protect itself.

 

[Zzurhi]

Welp, looks like you solved your issue. Looking at the layers and layers of dust and what looks to be.. sand? covering a lot of your components, I would say you probably live in a very dusty, hot place. Please clean everything thoroughly with isopropyl alcohol, get it all inside a case with very good dust filters, seal up any little holes or ways for dust to get into the case with electrical tape or stickers maybe... And change your dust filters often!

Dust, The image i posted was in the my workplace office.

I've just opened this pic full size and seen the cooler you took off in the bottom right! I could not believe that would be the size of cooler that Intel would ship with an 88w TDP chip, it looks more like a Celeron cooler! But I looked up the original spec and that was the correct retail cooler for the i7 from 2013 when the chip came out. It's so tiny!

In addition to it being way too small for the heat that the chip pumps out, it has a copper core surrounded by aluminium fins. The core is bonded to the outer fins using some kind of adhesive, which doesn't get great thermal transfer to begin with, and I bet you that layer has worn away or evaporated over time leaving you with bad thermal transfer from the copper core. With nowhere for the heat to go it just builds up to 100C and the cpu shuts down or thermal throttles to protect itself.

That fan was the stock of processor when it came to me, purchased it 2nd hand in online store.