Intel's 8th Generation Core Family - Coffee Lake (LGA 1151, 6C/12T)

Where do you expect Core i7-8700K's Turbo to land?

  • 3.8/3.9 GHz

    Votes: 0 0.0%

  • 4.0/4.1 GHz

    Votes: 3 23.1%

  • 4.2/4.3 GHz

    Votes: 6 46.2%

  • 4.4/4.5 GHz

    Votes: 3 23.1%

  • 4.6/4.7 GHz

    Votes: 1 7.7%
  • Total voters
    13
  • Poll closed .
I still can't decide between i7 8700k and the i7 7820X, coffee lake will be faster, but skylake x 8 core might last longer.
 
Skyblue said:
I
m not sure I can tell if its good or bad. 7700K results are all over the place. Some even scoring higher than the 8700k in multithreaded, which I would think was somewhat unlikely.
Click to expand...

I would imagine those 7700k are under LN2
 
I continue to see prices a little bit higher than even gouging levels on reddit. Hopefully this thing has the same RRP as the 7700k, but it feels like it's somewhere between $30 to $50 US more.
 
AbRASiON said:
I continue to see prices a little bit higher than even gouging levels on reddit. Hopefully this thing has the same RRP as the 7700k, but it feels like it's somewhere between $30 to $50 US more.
Click to expand...
It's going to be more expensive.
If it's not retailers raising the price (99% chance they will) than Intel themselves. It's their first consumer 6C chip, a 7700K with 2 more cores, aka god tier. Retailers know a golden goose when they see one.
 
TaintedSquirrel said:
It's going to be more expensive.
If it's not retailers raising the price (99% chance they will) than Intel themselves. It's their first consumer 6C chip, a 7700K with 2 more cores, aka god tier. Retailers know a golden goose when they see one.
Click to expand...

They are also fiercely competitive though and they often run a pretty slim margin on some premium parts. (After the initial 4 - 10 week gouging run)
That being said, the prices are beating the nromal launch prices of 7700k / 6700k etc. Suspect Intel did at least $30 US
 
Since there is apparently no Coffee lake 2 core dies, what happens to Pentiums? Rebranded Kaby i3s still on 14NM+? Cut down Coffee I3s?
 
Dayman said:
Since there is apparently no Coffee lake 2 core dies, what happens to Pentiums? Rebranded Kaby i3s still on 14NM+? Cut down Coffee I3s?
Click to expand...

They already went 2C/4T with KBL. The only 2C/4T die on a newer node is Cannon Lake on 10nm.
 
realworld said:
What are the chances we'll see another gorgeous board like this one in the age of RGB eye cancer?
P9X79-WS.jpg
Click to expand...

I wouldn't call it pretty, I don't like Asus mainstream boards colours, its to outlandish with 50 shades of blue, gray, white and black. I think Gigabyte did well with the carbon black mainstream boards, the colour works for any setup
 
AbRASiON said:
What numbers to compare to?
Click to expand...

Hard to compare, as every comparable part runs different all core turbos that affects the results here and there. Of course you can easily do a clock for clock comp by taking the database scores and work them all to something attainable like 4ghz and compare the 6700K, 7700K, 6800K, 6850K and get a delta that way.
 
7800x with overall lower clocks can get higher score than that. Looks like Turboboost isn't working on that video.
 
Dayman said:
I don't believe that multicore score at all if im honest, something else is at play here.
Click to expand...

As in no turbo? Another leak gave 1400 something for MT. Also fits if that was with 4.3Ghz or so and the 1230 is with 3.7Ghz.

CB scores is pretty much just clock.
 
I'd expect around 1400 at stock considering its clocked higher than the 7800x. People with the chips I know collectively said "wat" to the scores.
 
Shintai said:
As in no turbo? Another leak gave 1400 something for MT. Also fits if that was with 4.3Ghz or so and the 1230 is with 3.7Ghz.

CB scores is pretty much just clock.
Click to expand...

The test was apparently in a HP Omen, it could be thermals, early bios' with messed up turbos (though single seems to be fine so I doubt it) it could also be power limiting if they used a god awful stock board for the chip, who knows.
 
Each day more leaks, sooner or later someone is going to provide some pretty credible data with pics to back it up.
 
Cinebench and Geekbench really need to die in a fire.

The companies promoting those tests are simply massively disingenuous (AMD and Apple resp.).

If you are going to use a totally pointless synthetic benchmark you should at least use something that is actually optimized, actually represents current workloads for the respective applications they are simulating, and actually uses all available instructions.

Cinebench is an ancient test running on an essentially GPU workload that has zero modern instructions used for no other reason than that it is ancient.
For an application that is literally doing 100% GPU work to not have any modern instruction support to be something that people use as a CPU benchmark absolutely boggles the mind.
 
Communism said:
Cinebench and Geekbench really need to die in a fire.

The companies promoting those tests are simply massively disingenuous (AMD and Apple resp.).

If you are going to use a totally pointless synthetic benchmark you should at least use something that is actually optimized, actually represents current workloads for the respective applications they are simulating, and actually uses all available instructions.

Cinebench is an ancient test running on an essentially GPU workload that has zero modern instructions used for no other reason than that it is ancient.
For an application that is literally doing 100% GPU work to not have any modern instruction support to be something that people use as a CPU benchmark absolutely boggles the mind.
Click to expand...

It works perfect for seeing the IPC increase from one generation to another. Just like any benchmark it just gives you a idea of performance.
 
Gideon said:
It works perfect for seeing the IPC increase from one generation to another. Just like any benchmark it just gives you a idea of performance.
Click to expand...
^^^^^^^^^^

Any technical person worth their salt will simply benchmark across a HUGE suite of applications, games, in single and multi-core scenarios. In high memory usage or high compute scenarios.

What is annoying and I'll agree with 10,000 fold is idiotic leaks where they make a post on a forum and include just 1 or 2 numbers from 1 benchmark (such as, cinebench) - if you're gonna leak some goddamn numbers, leak at LEAST 3 or 4 for goodness sakes.
 
AbRASiON said:
^^^^^^^^^^

Any technical person worth their salt will simply benchmark across a HUGE suite of applications, games, in single and multi-core scenarios. In high memory usage or high compute scenarios.

What is annoying and I'll agree with 10,000 fold is idiotic leaks where they make a post on a forum and include just 1 or 2 numbers from 1 benchmark (such as, cinebench) - if you're gonna leak some goddamn numbers, leak at LEAST 3 or 4 for goodness sakes.
Click to expand...

The bigger concern is that there are leaks, it shows a complete lack of integrity by those entrusted to portray high levels of integrity.
 
Gideon said:
It works perfect for seeing the IPC increase from one generation to another. Just like any benchmark it just gives you a idea of performance.
Click to expand...

Not really. As soon as you enter the single cycle SSE loop capability its pretty much just plain clocks. And benches testing only a handful of instructions never give a clear picture.
 
Based on rumors, I'm thinking that the i3-8350k may be the sweet spot for gamers. Should be a decent price, with 4 cores that have a base clock that should be higher than the 8600k or 8700k. Hopefully it will have some decent headroom to overclock since the boost clock in the 8700k is 4.7Ghz.

Hoping for the next 2500k or 2600k? (Still rocking the 2500k while I wait. :) )
 
You must log in or register to reply here.
Back
Top