Intel i7 7700k overclocking temps too high?

staticmindstudios

Prominent
Oct 8, 2017
4
0
520
Previously, I posted a question about my overclocking temps being to high at 5.15Ghz 1.35 volts and I had temperatures of 87-88°c.

"I tried to overclock my processor (Intel i7 7700k @4.2Ghz) to 5.0Ghz. I was able to achieve a successful overclock of 5.15Ghz at 1.35v. The issue is, the Cinebench r15 benchmark gives me temperatures of 88°c. I decided I'd tone it down to a more reasonable 5.0Ghz since that was the goal in the first place; yet I still received temps of 87°c. I have a Kraken x62 AIO water cooler so I don't understand whats going on. Am I putting in too many volts perhaps? My idle temps are anywhere from 35-40°c and my ambient temp is usually around 20-24°c. I was anticipating lower temps when I first built this PC initially so I'm not sure if I got a faulty cooler or a slightly less performing cooler."

Since then, I have been able to reach a stable overclock of 5.05Ghz at 1.28v but I still hit temperatures of 88°c. The only benchmark I've been using is Cinebench r15 because I don't want to perform stress test like ida64 for too long and risk destroying my processor. I've bought some Artic Silver 5 so I can reseat the waterblock with TIM other than the pre-applied that came with the kraken. That's the only solution I can think of other than underclocking. I do however think reseating the waterblock may provide a significant difference, but until then, any suggestions?https://imgur.com/BKcjzSC
https://imgur.com/7pSE0VB
I've also decreased the temperature of my air conditioner by 2°f and now I'm freezing.
 
Solution
I don't think you need to "underclock", you just need to reduce your overclock. You need to have a very special win in the silicon lottery to get a chip that can continuously, as a daily driver, handle a 4.8ghz or higher overclock on any of the last three or four Intel architectures, especially without a very high end custom loop. Your temps are definitely too high. No question. At least, if you want the chip to last very long at all.

Everything you need to know about Intel CPU temperatures is right here:

http://www.tomshardware.com/forum/id-1800828/intel-temperature-guide.html
I don't think you need to "underclock", you just need to reduce your overclock. You need to have a very special win in the silicon lottery to get a chip that can continuously, as a daily driver, handle a 4.8ghz or higher overclock on any of the last three or four Intel architectures, especially without a very high end custom loop. Your temps are definitely too high. No question. At least, if you want the chip to last very long at all.

Everything you need to know about Intel CPU temperatures is right here:

http://www.tomshardware.com/forum/id-1800828/intel-temperature-guide.html
 
Solution
No, it doesn't look fine. Anything over 85°F is bad. Anything over 75°F is undesirable. I don't know where you are coming from, but you're mistaken. Especially if that's not during an pretty intense thermal test using Prime or another utility.

This is directly from the Anandtech testing on the 7700k:

At 4.8 GHz, the Core i7-7700K passed POV-Ray with ease, however the 1.400 volts needed at that point were pushing the processor up to 95C during OCCT and its mixed AVX workload. At that point I decided to call an end to it, where the CPU was now drawing 122W from idle to load. The fact that it is only 122W is surprisingly low – I would have thought we would be nearing 160W at this point, other i7 overclockable processors were at this level in the past.

The second set of results is with the AVX offset. This afforded stability at 4.8 GHz and 4.9 GHz, however at 5.0 GHz and 1.425 volts the CPU was clearly going into thermal recovery modes, as given by the lower scores in POV-Ray.

Based on what we’ve heard out in the ether, our CPU sample is somewhat average to poor in terms of overclocking performance. Some colleagues at the motherboard manufacturers are seeing 5.0 GHz at 1.3 volts (with AVX offset) although I’m sure they’re not talking in terms of a serious reasonable long term stability.
 
That CPU is inherently hot, because Intel no longer uses solder between the IHS and CPU die. You are doing very well if you have not delidded and applied liquid metal.

The following is an image of the hottest I can get my CPU, just a FEW seconds after the benchmark ran. I am using a Noctua U-14S, an air cooler, that is why the CPU temperatures dropped so quickly.

My CPU at 5.2GHz ~1.47 volts ran from 74 to 77 degrees C on a very CPU demanding benchmark. That is much higher voltage than you used, therefore your CPU will run cooler than mine if you delid the CPU and apply Thermal Grizzly Conductonaut http://www.thermal-grizzly.com/en/products/26-conductonaut-en

There are YouTube videos on how to apply this, although some are better than others. Thermal Grizzly provides instruction for the tool.

NOTE: Cut a tiny portion at the tip of the applicator to let liquid flow slower. For much more flow control, hold the plunger in your right hand between thumb and for finger [at the "wings" on the tube]; that will permit very positive flow control - otherwise the liquid metal can squirt out, a lot of it, and exactly what you do not want. Use very little. Also, cover the resistors on the PCB [that will be covered by the IHS] with liquid tape, or common black electrical tape will work as well [use scissors to cut a small amount, that is all you need].

I use delid, relid tools from Rocket Cool https://rockitcool.myshopify.com/products/rockit-88

I use a very thin line of head gasket maker or any good high temperature silicone with an extended tip applicator in the package. I only put the silicone on two sides of the bottom of the die. I do not cover the entire bottom edge for two reasons. 1.) Two sides is more than enough to secure the CPU to the die. 2.) Any gas buildup has a chance to escape under the sides not covered with silicone [that is the reason Intel leaves a small area with no sealant].

Instructions are at Rocket Cool and can be found on YouTube.com

Without applying liquid metal there is little chance of lowering your CPU temperature further. Take your time and follow instructions and you will accomplish the process successfully.

I hope this helps.

EDIT: I forgot to add an image.

YRW0kYp.jpg


 
Regardless that delidding works, it's a gamble, and is not recommended for the average user to attempt. If you know somebody that does this professionally or has several of them successfully under their belt, great, have them do it. For everybody else, it's a recipe for disaster.
 
darkbreeze,

A person that has worked with computer hardware enough to be comfortable with doing it has the technical skills to delid and relid modern Intel CPUs - if they are careful and methodical in follow instructions from Rockit Cool will have no problems as a result of the delid/relid process. The process is simple. Following instructions is vital.

You are correct that some people should not delid. Indeed, there are you tube videos that show a delid/relid that are completely wrong and could lead to disaster.

There are some correct youtube videos that show proceedures that do work correctly. The problem is the uneducated do not have a clue as to what is correct or incorrect and dangerous.

To the latter class of individuals I would suggest this video, it is clear [easily understood] and correct:

https://www.youtube.com/watch?v=uGSF2mIzSxA
 
You and I understand what qualifies as "a person that has worked with computer hardware enough to be comfortable with doing it". The average person does not. Every 13 year old out there that has replaced a graphics card or installed a set of fans successfully is likely to think that puts them in that category, which of course, is fanciful at best.

We've had discussions regarding this in other, more private settings, and even a lot of folks who've been working with hardware for the last 20-25 years, or more, say they would not be comfortable taking the risk associated with doing it themselves.

Personally, I'm sure I could do it as well, but I value my hardware and so unless it was a chip that I could stand to lose if something went wrong, I'd prefer to send it out to a service that could be held responsible in the even there was a problem.