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 .
Intel releases a PDF detailing what processors they plan to release micro code updates for

One thing that caught my eye and it's not really related to the update is the mention of Coffee H/E/U(GT3e) I imagine these products will be launching soon AND the mention of 2 core Cannonlake with and WITHOUT a GPU

cfDWzPm.png

BAkqgig.png
 
Last edited:
oleNBR said:
soldered junk with 0 path to upgrade and 0 chance for a cpu/gpu replacement when things go bad
Click to expand...

I mean if you want to work on your laptop more power to you, but the number of people who want to work on their own laptops is negligible and you can't blame manufacturers for not caring about this irrelevant market. Clevos have very shitty touch pads, are bulky, look cheap, and have mediocre cooling solutions, which are the trade offs you make for that.

Everyone else wants small form factors, light weight, and good battery life. They don't care that their GPU and CPU aren't replaceable. Those things should never need replacement anyway: I have never had a gpu or cpu in a laptop die in over 15 years of laptop ownership.
 
Yup- the closest thing I've seen to a 'gaming laptop' that I'd actually want to use (read: not own) would be Razer's 14" Blade. Current life situation influencing that of course, if I wound up being exceedingly mobile I'd pick one up.

But a 13"-14" Ultrabook 2-in-1 just about hits the spot, and I don't really give a rats about the GPU. I do care about CPU performance for photography work on the go.

And upgrade-ability doesn't really bother me. I understand that the compromise for the exceedingly thin and light form-factor with pretty decent battery life is that nearly everything is going to be soldered on to the motherboard. Rule of thumb as always applies, same as in desktops: buy the best/most you can afford now, or pay later.
 
IdiotInCharge said:
Yup- the closest thing I've seen to a 'gaming laptop' that I'd actually want to use (read: not own) would be Razer's 14" Blade. Current life situation influencing that of course, if I wound up being exceedingly mobile I'd pick one up.
Click to expand...

LMAO a soldered low quality junk

rofl since people gets triggered and snowflake gets melted, report and delete my post i'll try again :D
 
Armenius said:
Get back to us after the liquid metal has begun to migrate to the copper and it looses some of its thermal conductivity. You're also going to have to regularly clean up the corrosion left behind.
Click to expand...

i thought liquid metal doesnt corrode copper
 
TaintedSquirrel said:
Stock 8700K beats the 1700X 4 GHz by 24%. Not perfectly fine.
Click to expand...

Well it is to be expected. 8700k has higher clocks, better IPC, so it should beat the 1700x. Specially considering the prices between the 2. 1700x is $299 8700k is $359. I would expect better performance from a chip that is $60 more.
 
Brackle said:
Well it is to be expected. 8700k has higher clocks, better IPC, so it should beat the 1700x. Specially considering the prices between the 2. 1700x is $299 8700k is $359. I would expect better performance from a chip that is $60 more.
Click to expand...
Question, how does the 50% performance disparity between Ryzen vs CFL compare to Sandy vs Vishera?
It's almost like GPU bottlenecks in modern games are hiding Ryzen's weaknesses... and as time goes on, Ryzen will only get comparatively weaker... If only someone had noticed that sooner.
 
Brackle said:
Huh Interesting. Other websites show totally different results. Here is just example.

http://www.guru3d.com/articles_page..._graphics_performance_benchmark_review,8.html

So tbh I think Ryzen is perfectly fine. But of course if results don't match someones agenda....Well you know the rest. Fake news my friend Fake news!
Click to expand...

Talking about someones agenda and then linking to a Guru3D review. :D

Minimizing the GPU botteleneck the stock 8700k is 24% faster than 8C Zen @ 4GHz on Guru3d and 28% on PurePC.
 
Brackle said:
Well it is to be expected. 8700k has higher clocks, better IPC, so it should beat the 1700x. Specially considering the prices between the 2. 1700x is $299 8700k is $359. I would expect better performance from a chip that is $60 more.
Click to expand...

You are comparing a stock Intel chip to an overclocked AMD chip. Compare stock vs stock or overclock vs overclock and then the performance gap increases up to 50%, whereas those $60 more (newegg?) only represent 20% higher pricing. So by 20% more dollars one obtains up to 50% more performance. Where is now the performance/price advantage of Ryzen?
 
I like AMD, and it hurts to say this, but the IPC difference is still huge and it really does matter for some things. I am an engineer and interested in CAD type benchmarks. All the ones I see make me cringe.
Puget Systems sells computers for professional tasks and they do lots of CAD and other professional benchmarks as well. I just recently discovered the site. https://www.pugetsystems.com/all_news.php

SolidWorks and Revit are multithreaded, but for many tasks the cpu has to calculate one step at a time, these can't be done in parallel. Extra cores and threads don't matter. IPC differences will stick out like a sore thumb:

https://www.pugetsystems.com/labs/a...offee-Lake-vs-Skylake-X-vs-Threadripper-1105/

https://www.pugetsystems.com/labs/articles/Revit-2018-CPU-Comparison-1002/

For those who don't click links, a Ryzen or Threadripper for creating models and model rotation fps is only about 50 to 70% as good as a Kaby Lake or Coffee Lake i7. It's a large and painful shortcoming. It's such a huge difference that no amount of overclocking the AMD will get it to where it's competitive.
 
Dayman said:
Mainboard Asrock H370 Performance
Mainboard Asrock B360 Gaming K4
Mainboard Asrock B360M Pro4
Mainboard Asrock H310-HDV/M.2
Mainboard Asrock H310-HDV

Gigabyte H370 Aorus Gaming 3 with 8700k appears on Sisoft

I5 8269u appears on Sisoft
Click to expand...

still nothing on z390 eh. im guessing at this point 10nm yields are not so good if z390 is planned for end of 2018 release then it'll most likely come with 8 core CFL 14nm++ so intel can buy themselves 1 more year for 10nm+. i remember seeing some 8+2 SKU leaks CFL donno when thats gonna make its way out to consumer.
 
oleNBR said:
still nothing on z390 eh. im guessing at this point 10nm yields are not so good if z390 is planned for end of 2018 release then it'll most likely come with 8 core CFL 14nm++ so intel can buy themselves 1 more year for 10nm+. i remember seeing some 8+2 SKU leaks CFL donno when thats gonna make its way out to consumer.
Click to expand...

More likely ICL.
 
oleNBR said:
still nothing on z390 eh. im guessing at this point 10nm yields are not so good if z390 is planned for end of 2018 release then it'll most likely come with 8 core CFL 14nm++ so intel can buy themselves 1 more year for 10nm+. i remember seeing some 8+2 SKU leaks CFL donno when thats gonna make its way out to consumer.
Click to expand...

So far as I know there is no 8-core CFL. 8-core ICL (10nm+) is scheduled for late 2018, early 2019.
 
8600ks are a very handily priced chip, out performing previous x700 chips.

Sadly I dont think I have a lottery winner although bear in mind this data is from my previous (now proven defective) board.

AVX 4.5ghz stable requiring 1.37v
AVX 4.8ghz stable requiring 1.42v
This scared me so didnt try to get 4.9 and 5.0 AVX stable.
Non AVX 5GHZ seemed stable in windows (intel extreme tuning bench and stress test, plus prime 95 and other tests) at 1.31-1.32v, but then failed to boot linux just kernel panics during boot.
So to get linux to boot I either had to drop the chip to 4.8ghz or increase vcore to a whopping 1.38v, 1.38v worked for both 4.9 and 5ghz.

I will try a higher clock again on the new board to see if its better.

Also interesting on the voltage I have this data.

With LLC level5 (max), under AVX load the vcore rises to 1.38v and occasionally over 1.4v, even at 4.3 clock. LLC level 4 or lower wasnt stable.
However my new board came with bios 1.00 which apparently has broken LLC, the vcore doesnt go above 1.34V for AVX loads, and is stable as well. The non AVX voltage loads are the same as LLC level5 on the fixed bios on the previous board, so this "broken" LLC behaviour I find myself liking.
 
You must log in or register to reply here.
Back
Top