Intel Skylake / Kaby Lake

[scannall]

A 1080P TN panel @ 144Hz or even 240Hz is a fine choice if you are a competitive gamer.

That's a pretty small subset of users though. Not saying it isn't important to the few who are actually comptetive gamers, but really it's a small crowd.

 

[AdamK47]

Yours looks similar to mine - stuck around 2/3 speed. User benchmarks should show maximum performance or very close to it, lots of people have posted samples. Not sure whats going on with these boards, I can't figure it out.

My 512GB 960 Pro AS-SSD results look to be in line with or better than this Ars review from last year: https://arstechnica.com/gadgets/2016/10/samsung-960-pro-review-the-fastest-consumer-ssd-you-can-buy/

 

[Sweepr]

Well, it's an improved KL chip. Why shouldn't it single core turbo at least the same...?

How does better than Core i7-7700K's Turbo sound?

 

[ManyThreads]

My 512GB 960 Pro AS-SSD results look to be in line with or better than this Ars review from last year: https://arstechnica.com/gadgets/2016/10/samsung-960-pro-review-the-fastest-consumer-ssd-you-can-buy/

Their sequential read is 800MB/s higher and write speeds are 200 MB/s higher. Your benchmark (and mine) is a lot slower than what they got in that review. I just want the performance I paid for - right now I am not getting that. If I am not going to get it, I might as well return it for an EVO or something else.

 

[Ajay]

That's a pretty small subset of users though. Not saying it isn't important to the few who are actually comptetive gamers, but really it's a small crowd.

Nah, the paid competitive market is small but makes a big impact. So there are a fairly large number of people who play unpaid comp (ladders, etc) and wannabes - hence the proliferation of these types of monitors in recent years.

 

[AdamK47]

Their sequential read is 800MB/s higher and write speeds are 200 MB/s higher. Your benchmark (and mine) is a lot slower than what they got in that review. I just want the performance I paid for - right now I am not getting that. If I am not going to get it, I might as well return it for an EVO or something else.

Let's see if I can hotlink their graph.

It looks near the same to mine.

 

[secretanchitman]

Coffee Lake-S

Core i7-8700K 6C/12T 3.7 GHz base, 12MB L3, 95W
Core i7-8700 6C/12T 3.2 GHz base, 12MB L3, 65W
Core i5-8600K 6C/6T 3.6 GHz base, 9MB L3, 95W
Core i5-8400 6C/6T 2.8 GHz base, 9MB L3, 65W
Core i3 gets 4C/4T

https://www.cpchardware.com/coffee-lake-approche

Finally someone posted. Should I leak the Turbo clocks? I didn't believe at first.

Nice to see the clock speeds confirmed - I'm happy with this and I'll probably go with an i7 8700K for my upcoming build. I'm hoping they won't use crappy TIM but this is Intel...

 

[Arachnotronic]

That's a pretty small subset of users though. Not saying it isn't important to the few who are actually comptetive gamers, but really it's a small crowd.

The enthusiast/gaming PC market in general is a pretty small crowd, though

 

[ManyThreads]

Let's see if I can hotlink their graph.

It looks near the same to mine.

I was looking at this one - is that not the correct one?

https://cdn.arstechnica.net/wp-cont...w-chart-template-final-full-width-3.001-1.png

I guess regardless, tons of people are getting the full advertised maximum speeds, so we should be too. On that test I get 1000MB/s slower read and write than they get.

What do you get in the Samsung Magician benchmark?

 

[AdamK47]

I'll try Crystal Disk Mark and Samsung Magician tonight. I'm running Kaby Lake X and not Skylake X, so my results may not be totally comparative. My drives are all running off of the X299 PCH though.

 

[LTC8K6]

How does better than Core i7-7700K's Turbo sound?

It sounds unlikely, actually.

Unless there is something fundamental that we don't know about the new chips.

Or you mean Intel gave it a token 4.6 single core turbo so it would eke out ST wins.

 

[Sweepr]

Or you mean Intel gave it a token 4.6 single core turbo so it would eke out ST wins.

Personally I don't think Core i7-7700K owners will have any 'regrets' if they upgrade.

 

[tamz_msc]

Can you provide numbers?

rarely seen it in any review that ryzen shows better min fps (and if you look at the run chart it is an outlier), in fact oced 7700k shows performance advantage in min fps that shows THE reason to upgrade from sandy

except this forum talk of BF1 MP

I see the 6/6 CFL as THE cpu of choice. enough threads for most of desktop uses, low power and TEH enough coarz gaming performance

Two very recent examples: Prey and Dirt 4.

 

[Edrick]

Yours looks similar to mine - stuck around 2/3 speed. User benchmarks should show maximum performance or very close to it, lots of people have posted samples. Not sure whats going on with these boards, I can't figure it out.

How is that 2/3 speed? AdamK's posted speeds are expected full speed.

I am now getting the same speeds as he posted, which is the same speed I was getting on the Z170 platform. So I am now happy.

I WAS getting under 2k random read (not the 2.6k I am getting now). That issue was resolved.

 

[TheGiant]

Two very recent examples: Prey and Dirt 4.

And do you have and intel 6-8 core test to tell if this result is due to the core number or the ryzen architecture?

however, we all know that atm the 7700K oced is the king of gaming and what the mass and even lots of enthusiast crowd need (even if they don't say it)

 

[Edrick]

I was looking at this one - is that not the correct one?

https://cdn.arstechnica.net/wp-cont...w-chart-template-final-full-width-3.001-1.png

You are looking at the wrong chart.

 

[LTC8K6]

Personally I don't think Core i7-7700K owners will have any 'regrets' if they upgrade.

Will 7800X owners have regrets?

 

[dullard]

What expected changes ?
So you think 8700K will match 7700K 4.5Ghz turbo ?

I expected more cores, slightly higher TDP, lower base clock, and about the same turbo. I think the turbo will be a bit lower than the 7700k turbo, but they might surprise us with the same turbo. Crude math says that a 6 core 4.0 GHz turbo is doable with the numbers that Intel released in the spring, but maybe they are better than that now. Even if not, Intel could set the turbo to 4.5 GHz but the desktop chips probably will down-throttle a lot more. You can't put in 50% more cores in a process that is about 23% more efficient with about the same TDP and expect to be able to stay at full turbo as long.

 

[Sweepr]

Will 7800X owners have regrets?

If they don't care about the extra PCIe lanes / memory channels, AVX-512 and X299's upgrade path, then yes.

 

[LTC8K6]

I expected more cores, slightly higher TDP, lower base clock, and about the same turbo. I think the turbo will be a bit lower than the 7700k turbo, but they might surprise us with the same turbo. Crude math says that a 6 core 4.0 GHz turbo is doable with the numbers that Intel released in the spring, but maybe they are better than that now. Even if not, Intel could set the turbo to 4.5 GHz but the desktop chips probably will down-throttle a lot more. You can't put in 50% more cores in a process that is about 23% more efficient with about the same TDP and expect to be able to stay at full turbo as long.

Maybe you can get really close to doing that.

 

[TheGiant]

This myth needs to die, sure not relevant for "Your uses", but creates a false sense of "enough performance for 4K@60hz".

There is in fact a huge difference. Of course impact depends on what games You are playing, but a faster CPU is faster. Frames per second is not "end all" stat for CPU's, even if we ignore the elephant in the room called "minimum frames".

Where a faster, "built for gaming" CPU matters is various games, think about turn times in strategy games like Civilization series / Endless series. Also simulation times in games like EU4 / Stellaris / HOI4. There is a world of a difference to play those on fast CPU. Just ask around Arma, Nintento emulation, flight sim crowds, fastest CPU there will be 7700K @ 5Ghz and no one will have "too much" of performance
And fool yourself no more, once you realize that faster CPU is helping in playing those games both directly and indirectly ( for example game/save load times ), you will come to realize that those 1080P tests are in fact a good proxy of well built CPU with good cache subsystem that facilitates inter thread communication, and those CPUs that don't do well there, tend to suffer in general gaming. All those guys with 4Ghz Ryzens are giving up 30-40% of that sweet ST performance vs current best in market.

I've been hearing that GPU is the king since GeForce 1. As former pro gamer (Q3 in the 2000s) I am super sensitive to smoothness. I can only say this is very true. For the competitive gaming or high difficulties the CPU is the bottleneck (unless you clearly play at resolution out of your league which no sensible gamer will do-like 4k witcher 3 with gtx780)

The CPU plays a major role. And currently still the 7700K oced is the king of the hill and we all expect to be beaten only by CFL i7 6C in newer games (we will see how the ring performs in CLF 6C and if the latency actually doesn't hurt the gaming performance). Atm no ryzen or TR of SKL-X or bdw-e without the exception defeats it.

Unless AMD throws out 1 CCX ryzen 6C at 4,5 GHz which can handle 3600MHz DDR4...

 

[tamz_msc]

And do you have and intel 6-8 core test to tell if this result is due to the core number or the ryzen architecture?

however, we all know that atm the 7700K oced is the king of gaming and what the mass and even lots of enthusiast crowd need (even if they don't say it)

Outside of high FPS competitive gaming, an OC Ryzen 8-core, an OC Sandy Bridge i7 2600K, and an OC i7 7700K all perform within 5% of each other, using setups which reflect the real world, like GTX 1080Ti for 4K, GTX 1080 for 1440p, and GTX 1060/RX 580 for 1080p, in relatively modern AAA games, barring a few exceptions.

 

[ManyThreads]

How is that 2/3 speed? AdamK's posted speeds are expected full speed.

I am now getting the same speeds as he posted, which is the same speed I was getting on the Z170 platform. So I am now happy.

I WAS getting under 2k random read (not the 2.6k I am getting now). That issue was resolved.

These are NOT my results - My results are all around 2/3 of these numbers and in some cases 1/2. The below is what I should be getting but I'm way less on both identical tests. The biggest differences are in Seq Q32T1 and Seq where I am a full 1000MB/s slower.

 

[ManyThreads]

I'll try Crystal Disk Mark and Samsung Magician tonight. I'm running Kaby Lake X and not Skylake X, so my results may not be totally comparative. My drives are all running off of the X299 PCH though.

Ah ok maybe that makes a difference - both are X299 though so shouldn't be too much different. Is there another way to run them other than through the PCH on X299? Regardless I would think on a higher end MOBO, the speeds of modern M2 drives shouldn't be bottlenecked.

 

[wildhorse2k]

Kaby-Lake-X delidded:

http://www.overclex.net/news/mise-au-point-delid-skylake-x/

Notice how the 7740X doesn't have that small interposer than 7900X does. As if Intel initially tested 7900X on LGA 1151 and due to release being rushed didn't have time to switch to LGA 2066 package properly. Could be also the manufacturing reason why 7900X can't be soldered. If you look at the small 7900X interposer and imagine 7980XE its clear it won't fit in there. 7980XE will probably be just like the 7740X.