Delidded Core i7 7700K runs 26 Degrees C cooler

Published by

Click here to post a comment for Delidded Core i7 7700K runs 26 Degrees C cooler on our message forum
data/avatar/default/avatar10.webp
This is your chance people. YOU can send a message to Intel and the motherboard mfg and not buy Kaby. Wait for Zen and see how it measures up. This isn't fanboy talk, it is consumer advocacy. If you think I sound like a AMD fanboy, read my comment history. The power is in your hands folks!
https://forums.guru3d.com/data/avatars/m/225/225084.jpg
That's pretty shocking to be honest. Intel must use earth/mud as it's tim.
data/avatar/default/avatar35.webp
SSD_PRO, I assume you overclocked ZEN to 5GHz, right? If no, ZEN please keep your offers private. Regarding Kaby Lake - I don't justify Intel's choice of TIM, but it shows what a huge potential lies in their CPUs, so if they really need, they will instantly release something which will trump ZEN.
https://forums.guru3d.com/data/avatars/m/248/248994.jpg
Either that individual CPU was flawed or the years of no competition have degraded the Intel folks into utter incompetents.
data/avatar/default/avatar38.webp
i do not care of this i7, i want ZEN. After ZEN we will see 300$ i7 8/16 core...till then i will not update to an intel quad core in 2017. Its octacore time.
https://forums.guru3d.com/data/avatars/m/248/248627.jpg
I bet after zen's release they start soldering and giving better performance across the board. The only reason they did this in he first place is it's the only way to make more other than increase prices they already have the entire market.
data/avatar/default/avatar39.webp
there is absolutely no way that a simple TIM change between the die and IHS reduces temps by 26c. Something else is going on here that has caused an exaggeration of the results. It would be ludicrous to think that Intel would change their TIM to a significantly lesser quality TIM in just 1 architectural update. LinusTechTips did a de-lid & test video on a 6700k and found it dropped their temps by less than 5c overall. those results are fabricated.
https://forums.guru3d.com/data/avatars/m/248/248627.jpg
Temps in my 3770k dropped by around 10 degrees after my de-lid.
https://forums.guru3d.com/data/avatars/m/242/242371.jpg
Wasn't there similar results with Ivy? When I had my 3570k I read delidding had a pretty big affect, but as I suffer from MS it wasn't something I could safely do. Edit:
Temps in my 3770k dropped by around 10 degrees after my de-lid.
Ah OK that answers my question, so nowhere near the same.
https://forums.guru3d.com/data/avatars/m/186/186763.jpg
there is absolutely no way that a simple TIM change between the die and IHS reduces temps by 26c. Something else is going on here that has caused an exaggeration of the results. It would be ludicrous to think that Intel would change their TIM to a significantly lesser quality TIM in just 1 architectural update. LinusTechTips did a de-lid & test video on a 6700k and found it dropped their temps by less than 5c overall. those results are fabricated.
When I delid my 3570K I saw a temp reduction of 23C These kinds of results are possible. He also used an old Heatsink on the 7700K then after the delid he used a very good aio cooler on, so his results are skewed a bit, however it is possible to get great temp reductions with a delid. Think I also saw a 12C temp reduction on my 4770K under single stage, not sure about air, never really used this CPU on air much before I delid it.
https://forums.guru3d.com/data/avatars/m/80/80129.jpg
Wasn't there similar results with Ivy? When I had my 3570k I read delidding had a pretty big affect, but as I suffer from MS it wasn't something I could safely do.
26 seems really high - I got 14c off my 3770K and I completely removed the IHS. Also want to point out that the quality of the TIM has little to do with the temps being lower. It's the gap caused by the glue that holds the IHS down.
https://forums.guru3d.com/data/avatars/m/235/235398.jpg
If that's true it is completely unacceptable on the part of Intel... just lazy engineering.
https://forums.guru3d.com/data/avatars/m/259/259654.jpg
SSD_PRO, I assume you overclocked ZEN to 5GHz, right? If no, ZEN please keep your offers private. Regarding Kaby Lake - I don't justify Intel's choice of TIM, but it shows what a huge potential lies in their CPUs, so if they really need, they will instantly release something which will trump ZEN.
See below: (Also fanboyism kills the joy of everything).
The genius who came up with this result used an old 6 pipe Thermalright air cooler before delidding. And then used Coollab Liquid Ultra with a Kraken X62 afterwards. 5Ghz with some random air cooler, great idea! Link to the Anandtech thread
https://forums.guru3d.com/data/avatars/m/239/239175.jpg
How do you prevent the Sandy Bridge "disaster" from happening again? You prevent people from overclocking by making your chip cook itself. People will then be forced to buy the same CPU with just a slight refresh every year. Thank you Intel for caring about your customers. We will be remembering this.
https://forums.guru3d.com/data/avatars/m/118/118821.jpg
The genius who came up with this result used an old 6 pipe Thermalright air cooler before delidding. And then used Coollab Liquid Ultra with a Kraken X62 afterwards. 5Ghz with some random air cooler, great idea!
any old thermalright is still quite competent, but it seems youre incorrect? the thread itself stated he used the thermalright for the chip under stock speeds, & then says
Before and after delid with Liquid Ultra - 5Ghz, 1.344v, Kraken X62 @ 50% fan and 65% pump (i.e. silent), Prime95 v27.9 - 26 degrees difference (~30 degrees max temp difference)!
anyway, theres no way load temp delta changes that much unless 1) improper initial mount or 2) literally defective/warped heatspreader ::edit:: yeah i just reread his soup-sandwich-tier posts - he used a corsair h110i set to low, & his temps were out of control (nearing 100) trying to get near 5ghz. after mounting the kraken he still reported insane temps...until delid, so it kind of sounds like a warped heatspreader that wasnt coming into proper contact with the core tbh
https://forums.guru3d.com/data/avatars/m/80/80129.jpg
The genius who came up with this result used an old 6 pipe Thermalright air cooler before delidding. And then used Coollab Liquid Ultra with a Kraken X62 afterwards. 5Ghz with some random air cooler, great idea! Link to the Anandtech thread
See below: (Also fanboyism kills the joy of everything).
He starts off with the Air Cooler, then buys the AIO water cooler still hits 96c - THEN he delids it, then it goes to ~66. All of the testing was at roughly ~5Ghz.
https://forums.guru3d.com/data/avatars/m/245/245459.jpg
If you look at the pictures you can see that the TIM layer was really thick in the stock configuration, I think that would have caused his high temperatures - I just think that unit was badly prepped in the factory. I'm sure delidding helped, but I think he got a duff TIM job from the factory in the first place.
https://forums.guru3d.com/data/avatars/m/90/90726.jpg
i do not care of this i7, i want ZEN. After ZEN we will see 300$ i7 8/16 core...till then i will not update to an intel quad core in 2017. Its octacore time.
Thats great and all, but honestly, most people will not need octocore at all. Its great that they are available, but its totally overkill, even for gaming. I'd much prefer to have a solid quad core with higher single core performance than 8 cores of weaker performance, right? I dunno, the whole "i have so many cores" argument isn't provocative to me.
https://forums.guru3d.com/data/avatars/m/266/266726.jpg
"Intel Introduces patented Mayonnaise based TIM" Q:So what prompted this change? A:We found that the usage of mayo greatly Improved the profitability of our cpus while maintaining adequate thermal performance.
data/avatar/default/avatar06.webp
How do you prevent the Sandy Bridge "disaster" from happening again? You prevent people from overclocking by making your chip cook itself. People will then be forced to buy the same CPU with just a slight refresh every year. Thank you Intel for caring about your customers. We will be remembering this.
This would be my guess also.