• 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 596 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
If this wasn't an Intel thread, that would be trolling. As it is, it just tries to insult my intelligence. More cores and at a reasonable speed, and less expensive is a smart buy in many use cases.

Most consumer tasks are highly sensitive to per-core speed, from web-surfing to video games.

Very few consumer or even enthusiast tasks will take advantage of a lot of cores, so I don't think frozentundra123456's point is trolling at all.
 
14++ (CFL) has 10% higher transistor performance vs 14+ (SKL 7700k/7740x) and 4.4% higher single core turbo (4.7 Ghz vs 4.5 Ghz). 7740X is hitting 5 Ghz avg OC easily. I think 8700k has a good chance of hitting 5.2 Ghz avg OC for 24x7.

Agreed, I definitely think the clock ceiling for CFL will be higher than that of KBL. The higher perf transistors should help, for sure.
 
The only thing that worries me is 300 chipsets' future gen support. If they turn out to not support Ice lake (which apparently will be the direct successor of CFL instead of CNL), that would be extremely infuriating.

I don't think it will; but at the same time Icelake is likely to be at best no faster in games (and has a good chance at being slower) so it's not like people would upgrade anyway.
 
I don't think it will; but at the same time Icelake is likely to be at best no faster in games (and has a good chance at being slower) so it's not like people would upgrade anyway.

You keep spreading this FUD without any sources. It's really annoying and has no basis in fact/reality.
 
You keep spreading this FUD without any sources. It's really annoying and has no basis in fact/reality.

So you're pretty much ignoring Intel's chart saying the transistor quality is worse on 10+ versus 14++? Not by much but still. It'd be like Broadwell vs Haswell without the magic edram.

Then you have the possibility of the mesh, Intel's renewed focus on the server, if it has MorphCore that hurts Single Thread a touch, etc.
 
If this wasn't an Intel thread, that would be trolling. As it is, it just tries to insult my intelligence. More cores and at a reasonable speed, and less expensive is a smart buy in many use cases.
That is my honest criteria for the uses I have for a cpu (gaming, and light encoding, which if I wish, I can leave running overnite). I wasnt trying to "insult" anyone, but you are certainly free to view my comments through your personal filter.
 
So you're pretty much ignoring Intel's chart saying the transistor quality is worse on 10+ versus 14++? Not by much but still. It'd be like Broadwell vs Haswell without the magic edram.

Then you have the possibility of the mesh, Intel's renewed focus on the server, if it has MorphCore that hurts Single Thread a touch, etc.
That chart was from late winter and it was very slightly less performance. Things may have changed. But even if 10+ is a bit less performance (or a bit more energy) than 14++, I'd take a bit less performance per transistor if I get twice as many transistors.
 
If we compare this prices to Ryzen Unlocked CPU lineup, well what to say.....................................
4fvgdaq_th.gif


https://forums.anandtech.com/thread...s-out-page-554.2428363/page-594#post-39042961

2017_08_20_171129.jpg


http://www.microcenter.com/product/..._am4_boxed_processor_with_wraith_spire_cooler
 
So you're pretty much ignoring Intel's chart saying the transistor quality is worse on 10+ versus 14++? Not by much but still. It'd be like Broadwell vs Haswell without the magic edram.


10nm+ is better than today's 14nm+ according to this chart, it would would translate in only a 100-200 Mhz loss which can be easily counterbalanced with a slightly better IPC. But even this is way too early to predict, even Intel can't know how 10nm+ is going to work in 1.5 years with a final Icelake silicon.
 



i7-8700k should be easily faster than Ryzen 1800X in reviews with mixed workloads, even a price near 500 USD is worth it. Below 400 USD is really nice. 1700 is no match and the 1700x pricing looks too high, I wonder if AMD is forced to lower their pricing of some Ryzen.
 
So you're pretty much ignoring Intel's chart saying the transistor quality is worse on 10+ versus 14++? Not by much but still. It'd be like Broadwell vs Haswell without the magic edram.

Then you have the possibility of the mesh, Intel's renewed focus on the server, if it has MorphCore that hurts Single Thread a touch, etc.

MorphCore? Seriously?
 
MorphCore? Seriously?

The meme that never dies 🙂

Realistically, Ice Lake will be an evolution of the Sky Lake microarchitecture, so IPC will go up. And the FUD about the mesh needs to stop, the client parts will continue to use descendants of the ring bus and the inclusive LLC.

The change was made in server for ultra-high core count CPUs; it's not applicable to client stuff.
 
i7-8700k should be easily faster than Ryzen 1800X in reviews with mixed workloads, even a price near 500 USD is worth it. Below 400 USD is really nice. 1700 is no match and the 1700x pricing looks too high, I wonder if AMD is forced to lower their pricing of some Ryzen.

This is not a "Speed Comedy Club for beginers", at logical level you failed absurdy.😀
 
CFL 8700k will most probably be the fastest gaming CPU edging out even the 7740X and 7700k. My guess is 8700k will be atleast 10% faster against 7820x. Here is a comparison of 7700k at 4.9 Ghz vs 7800x at 4.7 Ghz.

In terms of gaming, TahoeDust's 7820 has gone up against a few 7700s here with the same clock and in general, stays with them. I think your guess of 10% lead for the 8700K in gaming is probably a pretty accurate estimate.

You'll have to forgive me - even though logic says the 8700K is the way to go, my heart still wants to go HEDT. 🙂
 
You know, not everyone overclocks. You really think that EVERYONE that buys an unlocked KBL or CFL runs at higher than 4Ghz? Sure, some of the gamer / enthusiasts on THIS board do, but the vast majority of Intel customers probably dont.

Wouldn't someone who doesn't overclock generally be best advised to go with the fastest processor?
 
Realistically, Ice Lake will be an evolution of the Sky Lake microarchitecture, so IPC will go up. And the FUD about the mesh needs to stop, the client parts will continue to use descendants of the ring bus and the inclusive LLC.

The change was made in server for ultra-high core count CPUs; it's not applicable to client stuff.

It'd save Intel a helluva lot of money... and that's what matters right now.
 
the FUD about the mesh needs to stop, the client parts will continue to use descendants of the ring bus and the inclusive LLC.

The change was made in server for ultra-high core count CPUs; it's not applicable to client stuff.
Wait, I thought all Skylake-X dies used mesh. Interesting and odd. Do you have a link? Can't find anything proper.
 
Is there still a glimmer of hope that Z270 will be compatible with coffee lake?
I don't think so man 🙁. It would have been awesome if z270 boards supported this lineup, if prices are as rumored it would be a great upgrade for i5 and even i7(8T) users.
 
Back
Top