Intel Core i7-8086K Listed, First 5.00 GHz Processor

[R-T-B]

I didn't think the 9590 was intel's doing

The title is slightly misleading though.

dedicated VRAM.

eDRAM, technically. On die ram. That didn't have to be used as VRAM, and could be repurposed as an L4 cache.

 

[Hood]

What a "useless" product. The last thing we need is a new i7 with even more aggressive boost.

Wouldn't it be a trip if they soldered the heatspreader? Or better yet, sold it naked? Anything to make it worth the extra money. But no, it will just be a binned 8700k with joke "improved thermal paste", like Devil's Canyon was supposed to have.

 

[Deleted member 67555]

If we're digging through memories, Cyrix actually had 5x86. I'm not sure if they also had a 6x86 after that.

But yes, the price of this one isn't randomly picked.

True...and yes there was a Cyrix 6x86 and AMD went K5....
But for Intel...
It was all about the pentiums baby.

 

[R-T-B]

Wouldn't it be a trip if they soldered the heatspreader? Or better yet, sold it naked? Anything to make it worth the extra money. But no, it will just be a binned 8700k with joke "improved thermal paste", like Devil's Canyon was supposed to have.

That's implying they'll actually try to improve the thermal paste again.

I think since Devil's Canyon, they gave up on that, since from what I can tell playing with both, the "NGIM" (Next Gen Interface Material) was essentially the same Dow Corning shit with a dab of something resembling oil added in.

My personal theory is they asked Dow Corning to improve the paste, and some guy in the factory said "hmm... must need more lube," spit in the thermal mixture, and that was about the end of the science behind it. Intel has been afraid to try since.

 

[Hood]

I think since Devil's Canyon, they gave up on that, since from what I can tell playing with both, the "NGIM" (Next Gen Interface Material) was essentially the same Dow Corning shit with a dab of something resembling oil added in.

Noun. spooge (usually uncountable, plural spooges) Any sealant or lubricant applied during the assembly of electronic equipment. (informal) Semi-liquid gunk. (vulgar, slang) semen.

 

[R-T-B]

Noun. spooge (usually uncountable, plural spooges) Any sealant or lubricant applied during the assembly of electronic equipment. (informal) Semi-liquid gunk. (vulgar, slang) semen.

I didn't want to go so far as to claim "semen."

 

[HimymCZe]

I was hoping it would be 8 core 16 thread so I could replace my 4790k.

I was hoping it would be 8 core 16 thread so I could replace my Q9550.

 

[Hood]

I didn't want to go so far as to claim "semen."

https://en.wiktionary.org/wiki/spooge - I was talking about "Any sealant or lubricant applied during the assembly of electronic equipment". Wiktionary supplied the 3rd definition. Those crazy Dow-Corning workers may have read the wiki and decided their additive was appropriate...

 

[CrAsHnBuRnXp]

Yet you bought i5 with 6c/6t?
8c/16t would mean lower clocks, which is worse for games and most applications.

Indeed I did. Why try and pick apart my purchase choices at the time that I bought it? Maybe I couldnt afford a then $350 CPU but maybe Im better off now where I could afford a $486 one. Try and give people the benefit of the doubt instead of trying to be a dick.

Whose to say the boost clock 5GHz wont be faster than an i7 8700k at 5GHz? We wont know for sure until the benchmarks are released.

 

[eidairaman1]

Wouldn't it be a trip if they soldered the heatspreader? Or better yet, sold it naked? Anything to make it worth the extra money. But no, it will just be a binned 8700k with joke "improved thermal paste", like Devil's Canyon was supposed to have.

Bare die should be just called modders edition, be the same price as the IHS version.

 

[Hood]

Bare die should be just called modders edition, be the same price as the IHS version.

Offer it with no warranty and it will still sell, and save Intel money. I only mention these possible options because Intel needs to get more creative/aggressive with their marketing strategy. Too much negative publicity lately, what with Spectre, Meltdown, people howling about "toothpaste", rumors that Intel engineered the quick exposure of AMD exploits (Ryzenfall, Masterkey, Fallout, and Chimera), loss of market share to Ryzen, etc. Intel needs to secure their core customer base, the enthusiasts, gamers, and power users who drive the perception of Intel as the source of top performing parts. Do something that nobody can possibly complain about, not even misguided AMD fans. Naked 8086k would be a good step in the right direction, and even a soldered heat spreader would shut up the toothpaste whiners.

 

[las]

I think they called that Planned obsolescence. How many de-lid say 4-5 years of hard use to see what is under the IHS.

I delidded a i7-3770K a few months ago and TIM was fine.

Indeed I did. Why try and pick apart my purchase choices at the time that I bought it? Maybe I couldnt afford a then $350 CPU but maybe Im better off now where I could afford a $486 one. Try and give people the benefit of the doubt instead of trying to be a dick.
Whose to say the boost clock 5GHz wont be faster than an i7 8700k at 5GHz? We wont know for sure until the benchmarks are released.

I'm not trying to be a dick. I just find it funny that you absolutely wants 8c/16t yet bought 6c/6t.

 

[BoinkyBoingBoing]

Ah AMD FX CPUs. Most of the time they went best from the tray to the trash.

It also doesn't help that a lot of AM3 boards were garbage.

 

[Octopuss]

I delidded a i7-3770K a few months ago and TIM was fine.

I did that back in 2012 or 13.
Looked like dried semen mixed with dust.

edit: The CPU was brand new then.

 

[efikkan]

Wouldn't it be a trip if they soldered the heatspreader? Or better yet, sold it naked? Anything to make it worth the extra money. But no, it will just be a binned 8700k with joke "improved thermal paste", like Devil's Canyon was supposed to have.

So the TIM complaint again? Intel chose TIM for longevity, and it works just fine for anyone except extreme overclockers.

Forgive me for being old-school, but back in the day overclocking used to be about higher performance for a lower price, trading off warranty and some longevity. There was a time when you could buy like a Celeron, a Sempron or a lower Athlon and get extreme gains out of them without pushing voltage too much. Nowadays, CPUs have already taken out most of the OC headroom at stock. So what would be the purpose of an "overclockers model" which is already pre-overclocked and have even less headroom left than i7-8700K?

An i7-8086K for overclocking would be just pointless. Just think about it for a moment, you really would have to push voltage to get this a little higher, and then you'll quickly kill it.
- If you want more gaming performance, then going beyond 5 GHz is pointless, since i7-8086K will "never" be bottlenecking games anyway.
- If you want more multithreading performance for "cheap", pushing voltage is going to require expensive cooling and will eventually kill the CPU. Just buy HEDT instead.
So what you're left with is people who do overclocking just for the purpose of overclocking, which is fine, but a pre-overclocked model serves no purpose here.
(Sorry for bursting some people's dreams…)

 

[mouacyk]

Intel chose TIM for longevity, and it works just fine for anyone except extreme overclockers.

Longevity of what exactly? The post above yours show the common case of the TIM drying out, while the CPU is still completely usable. In respect to solder, many still run 2500K up to 5GHz from 7 years ago.

 

[efikkan]

Longevity of what exactly? The post above yours show the common case of the TIM drying out, while the CPU is still completely usable. In respect to solder, many still run 2500K up to 5GHz from 7 years ago.

The TIM should last >6 years of normal usage, if it doesn't then it's defective.
It's not bad just because it looks dry, that changes as it heats up.

 

[Hood]

So what you're left with is people who do overclocking just for the purpose of overclocking, which is fine, but a pre-overclocked model serves no purpose here.

I agree with what you're saying, because the high stock clocks won't leave much room for overclocking. Similar to the 4790k (4.0 - 4.4 turbo). And yet the 4.4 figure is for one core turbo boost. Under heavy multi-core load, the stock clock of 4.0 prevails. My stable OC to 4.6 on all cores is a 15% increase over stock. The i7-8086K also has a 4.0 base clock, so 5.2 on all cores would be fairly impressive. Don't confuse one core turbo boost with an all core overclock. The performance gain is easily shown in comparative benchmarks, or by timing multi-threaded apps. So this OC won't improve single core very much, but will greatly improve overall performance in apps that can use all cores. The de-lidded or naked die model would allow lowest temps/highest OC

 

[efikkan]

I agree with what you're saying, because the high stock clocks won't leave much room for overclocking. Similar to the 4790k (4.0 - 4.4 turbo). And yet the 4.4 figure is for one core turbo boost. Under heavy multi-core load, the stock clock of 4.0 prevails. My stable OC to 4.6 on all cores is a 15% increase over stock. The i7-8086K also has a 4.0 base clock, so 5.2 on all cores would be fairly impressive. Don't confuse one core turbo boost with an all core overclock. The performance gain is easily shown in comparative benchmarks, or by timing multi-threaded apps. So this OC won't improve single core very much, but will greatly improve overall performance in apps that can use all cores. The de-lidded or naked die model would allow lowest temps/highest OC

Remember that modern CPUs operate at many clock speeds, example:
i7-8700K: 3.7 GHz base (AVX), 4.3 GHz 6 cores - 4.7 GHz 1 core.
i7-8086K: 4.0 GHz base (AVX), x GHz 6 cores - 5.0 GHz 1 core.
(I think we can estimate what x will be)
So you can see the gains will be minimal.

 

[mouacyk]

The TIM should last >6 years of normal usage, if it doesn't then it's defective.
It's not bad just because it looks dry, that changes as it heats up.

Still haven't come across a good reason for Intel to make this change, since AMD has been able to afford soldering its octa-core CPUs and sell it competitively as well.

 

[Hood]

Intel Core i7-8086K likely not to be soldered
The information posted at OCUK forums by Gibbo is a response to other retailers listing this 40th Anniversary Edition processor before the official introduction, which is expected on June 8th.

Gibbo confirmed that the price of the new SKU should not exceed 400 GBP. For comparison boxed version of retail i7-8700K is offered for 350 GBP, while the OEM variant is sold for 330 GBP at OCUK.

Same as 8700K, the new processor will be offered in pre-binned/de-lidded version. This would confirm that 8086K is not soldered as some rumors have suggested.

https://videocardz.com/76355/intel-core-i7-8086k-arrives-june-8th-for-under-400-gbp
Also mentioned in link; binned skus with a copper/silver IHS.

 

[xorbe]

Let us hope that no major games do ... if (intel 8 series) { if (<8300) { nope }}

 

[John Naylor]

Um what about the FX 9590?

... and others .... from what I read, doesn't fit in the sentence's criteria.

.... making it the first mainstream desktop processor from Intel to hit the 5.00 GHz mark, out of the box.

 

[cadaveca]

well, "the first mainstream desktop processor from Intel to hit the 5.00 GHz mark, out of the box. " i didn't think the 9590 was intel's doing

IT says from Intel... so many people's reading comprehension skills are lacking!

It's not that hard to bin for 5Ghz. I'm certain that's the route they've gone rather than solder. And they don't really seem to comprehend a better paste exists. They even ship the same brand dow corning shit with their heatsinks. I opened a rare tube of it once with one of their actual decent LGA1366 heatsinks and the stuff was dry as powdered Sahara sand... longevity my ass.

Um, wax is normally dry? If a wax-based TIM is wet, there's something actually wrong. So yeah, actual longevity is higher than "wet" pastes, because those use an oil base, and they tend to separate over time, while wax-based ones do not. Glad you understand TIM so well though, it'd be a shame if you'd make all these claims without knowing anything about how a TIM is actually made...

 

[R-T-B]

Um, wax is normally dry? If a wax-based TIM is wet, there's something actually wrong. So yeah, actual longevity is higher than "wet" pastes, because those use an oil base, and they tend to separate over time, while wax-based ones do not. Glad you understand TIM so well though, it'd be a shame if you'd make all these claims without knowing anything about how a TIM is actually made...

It wasn't waxy. It was an absolutely completely unusable powder that literally had trouble exiting the tube.

This isn't me making any overreaching claims either. It's one personal ancedote. Take it as nothing more. (The rest of the composition comments were me kidding around).