• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Intel Skylake / Kaby Lake

Page 581 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
I am not saying that the 7820x is a better gaming CPU than a 7700k. I am saying it is VERY FAR from a "poor gaming CPU".

The "poor gaming" CPU comes from the fact that it doesn't beat a 7700k clearly while costing double (actually more if you include mobo and especially quad channel RAM). Meaning it makes no sense to buy it just to get better gaming results. So poor = poor value for gaming.
 

We also get a confirmation of what's coming next...

Intel said:
The Ice Lake processor family is a successor to the 8th generation Intel® Core™ processor family. These processors utilize Intel’s industry-leading 10 nm+ process technology.

www.intel.com/content/www/us/en/design/products-and-solutions/processors-and-chipsets/ice-lake/overview.html

Edit: Already posted in the other thread.
 
What happens when you try to play a game and do something in the background with a 7700K vs a 7820X?

I have heard more than once that we need more cores so that we can do such.

Has anyone done a test like that?

Does the 7700K stumble?

Does Skylake-X do better than Broadwell-E?
 
The "poor gaming" CPU comes from the fact that it doesn't beat a 7700k clearly while costing double (actually more if you include mobo and especially quad channel RAM). Meaning it makes no sense to buy it just to get better gaming results. So poor = poor value for gaming.

Gaming results say more about the ability of these games to use more than 4 cores than the capabilities of the cpu that has more than 4 cores.

Blaming the cpu for not using it is a poor argument.
 
Gaming results say more about the ability of these games to use more than 4 cores than the capabilities of the cpu that has more than 4 cores.
Blaming the cpu for not using it is a poor argument.

Correct way is selecting cpu according its results in actual games , not mythical multicore games. Noone's blaming cpu for sw capabilities, but it doesnt change a fact that specific cpu is better in one type of sw, and worse in the other.
 
Last edited:
Correct way is selecting cpu according its results in actual games , not mythical multicore games. Noone's blaming cpu for sw capabilities, but it doesnt change a fact that specific cpu is better in one type of sw, and worse in the other.

Intel is marketing hedt for "mega-tasking" not for mythical multicore games and the tests posted earlier in this thread show it is not worse in games.
 
I agree completely. Total scores are pointless in this conversation. That is why I said "Looking at just graphics scores", and the percentages I posted were only on the graphics scores.

Good idea to drill down further and look at the individual tests though.

Can you post those @Darkhelmutt ?

Here are mine:
AFDZMxb.png


yUaDqUP.png

Here ya go. I have Doom but it has no benchmark tool. I rarely pay a premium for games so no GTA 5.

nOTBVhx.png


BaWLdyX.png
 
Intel is marketing hedt for "mega-tasking" not for mythical multicore games and the tests posted earlier in this thread show it is not worse in games.

no objections against that. I support the idea, that sk-x is not best solution for enthusiast gaming - cf-lake will be.
 
no objections against that. I support the idea, that sk-x is not best solution for enthusiast gaming - cf-lake will be.

I like how you agree that it is not worse and in the next sentence you say that it is not as good.

Until you show something with better performance your argument is void here.
 
I'd like to see more complex AI in games utilizing greater than 4 threads, seeing as how 4\6\8 threads seem to be inching toward mainstream. Maybe simple trainable neural nets each with a dedicated core\thread for supporting NPCs or bosses; homebrew projects like MarI/O really get me excited at such an idea.
 
no objections against that. I support the idea, that sk-x is not best solution for enthusiast gaming - cf-lake will be.

People like to say Coffee Lake will be better. Chances are it will be given previous history. It's not a sure thing though.
 
They are here to diminish the perception of sky-x ("poor", "worse", "not the best"), not to hype coffee lake, there's a difference.
I don't have a "SL-X = poor" perception. I think it's a great chip that's a little too power hungry and too expensive.
A common occurrence in the CPU and GPU world.

I also think all of these multi-core chips are out a little too early to really be able to show their stuff. Not enough software takes advantage of lots of cores yet.
 
I wonder if there's two separate events planned. They could unveil the Kaby Lake Refresh @ Facebook Live on August 21 and then Coffee Lake-S during Gamescom (August 23-26), in line with the DigiTimes leak from months ago.
 
Definitely get the feeling that Intel badly wants to get to Icelake. Can't blame 'em if it has EMIB.

I don't see EMIB coming on client for Icelake. EMIB is to reduce cost on solutions with high performance.

The eDRAM on Iris equipped chips uses neither interposers or EMIB. Interposers and EMIB are for high-bandwidth connections. Iris uses a proprietary connection on a standard MCM package. Same with the U chips. The thing is though their graphics do not need bandwidth much higher than what their proprietary connection provides.
 
Back
Top