i7-8700k & Coffee Lake CPUs

[MageTank]

Just now, Taf the Ghost said:

It doesn't list the GPU on the post, nor are any of the links to the rest of the work present. However, I'm almost certain that game testing was done with either a 1080 or 1080 Ti. Those numbers are too high for a Fury card.

 

If it's an Nvidia GPU, there is a DX12 gaming problem with Nvidia + Ryzen. There's a larger performance gap between the same Nvidia GPU under DX12 than DX11. So the better trained RAM might simply be working around that issue that's likely on Nvidia's driver more than the performance of the System.

 

That being said, faster RAM in totality helps both Intel & AMD a lot in Gaming performance.

How can you make that assumption without knowing the game settings or resolution? That makes zero sense.

 

Nvida tends to scale negatively with DX12, so it would make little sense to intentionally choose DX12 as a benchmark, only to muddy the waters with a competitors GPU, when trying to show performance uplifts with your own platform. 

 

I just don't understand your reasoning at all, given the exact testing methodology was not released. I can only say it coincides with my findings because I posses both a Ryzen and Kaby system, and Ryzen saw very similar performance uplifts when comparing the multi-rank Samsung IC's at JEDEC, vs the single rank Micron IC's at JEDEC. I dialed in the exact same secondary and tertiary timings too, to make sure it was as close to one another as physically possible.

 

Intel sees less of a difference in regards to rank interleaving's impact on gaming results, but it's still there (about 3% in my findings, close enough to be margin, if not for the consistency).

[leadeater]

7 hours ago, MageTank said:

I wish I could afford the platform myself, so that I could test how well it scales with various memory configurations, but I'll see if I can't find someone that has it, and walk them through doing so for the sake of science. Either way, it's looking a little stronger in the IMC department all around compared to smaller Ryzen. This result also needs a little extra work as well. 90% read efficiency, 90% write efficiency, but only 81% copy efficiency, lol. Would prefer to see 95% on both read/write, and 90% on copy if possible. I suppose I am asking for too much out of quad channel though. 

That is a little odd considering it's the same dies as Ryzen, maybe it's all just binning or slight manufacturing maturity.

[EOZ]

i7-8700k will be better than Ryzen 7 1700 ?

[Drak3]

9 minutes ago, EOZ said:

i7-8700k will be better than Ryzen 7 1700 ?

Definitely in single core, most likely in programs that can use more theads than 4c/8t offers. Actual multitasking, no, not likely.

[Taf the Ghost]

11 minutes ago, EOZ said:

i7-8700k will be better than Ryzen 7 1700 ?

Gaming? Yes. 

Beyond that? Depends. We haven't had a Skylake design with 6 cores in a Ring Bus, so latency is going to be interesting.

[laminutederire]

20 hours ago, TechyInAZ said:

I'm really curious about the Zen+ core coming soon (i think very late 2018 or 2019), probably not a higher core count but if AMD can get in a better IMC and higher clocks, that'll be really good.

Saw a few weeks ago that they are already producing navi and zen 2 on 7nm as prototypes, or at least they'll have those before 2018 ends, so if all goes well, we could have both in 12 months. The scalability factor of the Ryzen line up could be beneficial for them to iterate faster, as they have less dies to engineer before having actual CPUs ready. (Could be the same for Navi, Where they could only tweak began on 7nm and scale those dies with infinity Fabric). At least it seems zen iteration time-frame is more dependent on 7nm processes being ready.

[Taf the Ghost]

21 minutes ago, laminutederire said:

Saw a few weeks ago that they are already producing navi and zen 2 on 7nm as prototypes, or at least they'll have those before 2018 ends, so if all goes well, we could have both in 12 months. The scalability factor of the Ryzen line up could be beneficial for them to iterate faster, as they have less dies to engineer before having actual CPUs ready. (Could be the same for Navi, Where they could only tweak began on 7nm and scale those dies with infinity Fabric). At least it seems zen iteration time-frame is more dependent on 7nm processes being ready.

Tapeouts for 7nm are "soon", which means the design is finalized but tweaks before full production are still to be done. Navi & Zen2 should both be in 2019, though Zen2 might show up in Q4'18. 

 

Zen+ is a refinement phase for the current Ryzen/TR/Epyc CPUs. Hopefully tweaked IMC and maybe 200-300 Mhz more in clocks. We'll hear more in a few months.

[Bananasplit_00]

21 minutes ago, Taf the Ghost said:

Tapeouts for 7nm are "soon", which means the design is finalized but tweaks before full production are still to be done. Navi & Zen2 should both be in 2019, though Zen2 might show up in Q4'18. 

 

Zen+ is a refinement phase for the current Ryzen/TR/Epyc CPUs. Hopefully tweaked IMC and maybe 200-300 Mhz more in clocks. We'll hear more in a few months.

I thought Navi was going to be 2018? 

[leadeater]

8 minutes ago, Bananasplit_00 said:

I thought Navi was going to be 2018? 

Not likely, the return on investment for Vega would be shocking if they released it that soon. Even if it was a down shift and all Polaris based GPUs are now Vega and used to make up the mid range then Navi is the new top end that's a lot of money to have wasted on Vega for a 6-8 month sale window.

[laminutederire]

3 hours ago, leadeater said:

Not likely, the return on investment for Vega would be shocking if they released it that soon. Even if it was a down shift and all Polaris based GPUs are now Vega and used to make up the mid range then Navi is the new top end that's a lot of money to have wasted on Vega for a 6-8 month sale window.

Unless if vega is their base design for Navi sub dies Ala Zen arch

[Taf the Ghost]

8 hours ago, Bananasplit_00 said:

I thought Navi was going to be 2018? 

Vega 11 and Vega 12, which will be the Polaris replacement design. (So, likely 32 CU + cut downs.) Oddly enough, the RX 550 probably just gets a rebrand as it's actually a different "design", but one that's going to be around for a while. (It's like the GTX 1030 in & of the fact that it does everything it needs for at least 3-4 years, so there's little reason to redo the design until you want to drop doing driver updates.)

[TheCherryKing]

14 hours ago, Taf the Ghost said:

https://en.wikichip.org/wiki/intel/xeon_bronze

 

There doesn't seem to be any announcement of a real Xeon D replacement, from what I can find.  There's been Skylake E3s since 2015, and it looks like that's staying for now as the level under "Bronze". At least, that's the information I can find.

 

@MageTank

 

It should be noted that most of those results are DX12 with Nvidia high-end GPUs. Most of the benchmarkers had stopped doing that with Ryzen because there's such a huge performance drop off between DX11 and DX12 in comparative situations. (Nvidia's driver team has come up with a brilliant way of exploiting the Sandy Bridge uArch, which doesn't seem to work as well in DX12. And it works a lot worse with Ryzen.) So part of that uplift might be related to that specifically. (Vega 64 vs 1080 Ti on DX12 on Ryzen are going to be very interesting results.) 

The non-scalable Coffee Lake Xeons for the LGA 1151v2 socket will probably be called "Xeon Cooper".

[Taf the Ghost]

1 minute ago, TheCherryKing said:

The non-scalable Coffee Lake Xeons for the LGA 1151v2 socket will probably be called "Xeon Cooper".

I didn't know they opened a UK fab!

[Centurius]

I mean honestly, at this point Intel could get a chip that outperforms Ryzen in both core count and performance and I'd still go AMD just to keep competition going.

[Averletum]

Tested: identical machine, identical settings, difference in cores only, 5960X on 4.4 GHz.

 

As follow: 4c, 4c/8t, 6c, 6c/12t, 8c

 

https://m.imgur.com/a/FSqhW

 

Taken from another forum, credits goes to Tonči

 

http://m.bug.hr/forum/topic/ostale-hardverske-teme/cpu-chat-tema/224306?page=319

Edited August 13, 2017 by Averletum

[done12many2]

Just now, Averletum said:

Tested: identical machine, identical settings, difference in cores only, 5960X on 4.4 GHz.

 

As follows: 4c, 4c/8t, 6c, 6c/6c, 8c

 

Are you just letting us know you tested or did you forget something? 

[Averletum]

22 minutes ago, done12many2 said:

 

Are you just letting us know you tested or did you forget something? 

Sorry, link was missing, I edited now..

[Cookybiscuit]

53 minutes ago, Averletum said:

Tested: identical machine, identical settings, difference in cores only, 5960X on 4.4 GHz.

 

As follow: 4c, 4c/8t, 6c, 6c/12t, 8c

 

https://m.imgur.com/a/FSqhW

 

Taken from another forum, credits goes to Tonči

 

http://m.bug.hr/forum/topic/ostale-hardverske-teme/cpu-chat-tema/224306?page=319

The difference between 6 vs 8 cores is kind of surprising, I question his testing methodology. Numerous review sites have the 5960X (8 cores) only 5% ahead of the 6700K (4 cores) in Rise of The Tomb Raider, even considering the 6700K is clocked higher than the 5960X.

 

Maybe he is testing minimums?

[Brooksie359]

On 8/12/2017 at 2:57 PM, done12many2 said:

 

Not really.  Real cores are stronger than threads so a 6c/6t would still generally outperform a 4c/8t, but there are times when it will be close. 

Yeah but why buy anything  but an i3 for gaming  if it is basically  a 7700k. They will have the same issue as they had when they created the Pentium g4560

[Drak3]

18 minutes ago, Brooksie359 said:

Yeah but why buy anything  but an i3 for gaming  if it is basically  a 7700k. They will have the same issue as they had when they created the Pentium g4560

We're moving towards higher core counts in games. It might not affect average FPS that much, but it does bolster minimums and frame times, which translate to a better gaming experience for most people more so than having 150FPS averages instead of 140FPS.

[Brooksie359]

34 minutes ago, Drak3 said:

We're moving towards higher core counts in games. It might not affect average FPS that much, but it does bolster minimums and frame times, which translate to a better gaming experience for most people more so than having 150FPS averages instead of 140FPS.

Yeah but the 7700k still has really good minimums  and the price increace from an i3 to an i7 is so big that  I highly  doubt  many would buy anything  other than a i3 for a pure gaming rig or even gaming rig with some light work as well. This is all dependant  if they release an i3 with 4c8t which I highly doubt  because they wouldn't want to mess with the sales if i5s and i3s which it absolutely  would.

[Drak3]

Just now, Brooksie359 said:

Yeah but the 7700k still has really good minimums  and the price I create from an i3 to an i7 is so big that  I highly  doubt  many would buy anything  other than a i3 for a pure gaming rig or even gaming rig with some light work as well. This is all dependant  if they release an i3 with 4c8t which I highly doubt  because they wouldn't want to mess with the sales if i5s and i3s which it absolutely  would.

I'd imagine any 4c/8t i3 will either have low base and boost clocks, and be locked, or be in a similar price range as the lower end i5's.

[Brooksie359]

Just now, Drak3 said:

I'd imagine any 4c/8t i3 will either have low base and boost clocks, and be locked, or be in a similar price range as the lower end i5's.

I believe there was a rumor for a k variant for the 8th gen i3s so I'm  not exactly sure. Bottom line is having an i3 with 4c8t is just plain dumb on Intel's part.

[Drak3]

Just now, Brooksie359 said:

I believe there was a rumor for a k variant for the 8th gen i3s so I'm  not exactly sure. Bottom line is having an i3 with 4c8t is just plain dumb on Intel's part.

Not really. Logical processors don't scale quite as well as physical cores. Some members have speculated that the best a 4C/8T i3 could do to a 6C i5, at the same clocks, is match it at some workloads. That's ignoring any additional cache the i5 would have, and any disabled features. And that is best case scenario to them, and that it won't manage that under most scenarios.

[adams]

better start saving...