• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

Question Intel 14nm and 5 Ghz

TheHolyLancer

Member
Aug 10, 2005
49
0
66
So I got a thought, I kind of wanted to know just how much of a difference if you OCed your Intel chips across the 14 nm family would an upgrade to a newer 14 nm family would do.

I know that back in 8700K / 8600K you can start to get more "widespread" 5 Ghz all core OCed ships (still called a silicon lottery).

And once that happened you can find I5/I7 K series processors hitting that 5Ghz all core fairly easily in all subsequent releases.

So if anyone has any data across these generations, with a set 5 Ghz OC with I guess 6 cores (seeing as how 11gen dont have 4 core K processors anymore) with or without hyper threading comparing against each other.

Basically, I wanted to know how much is the raw IPC improvement for gaming / productivity / etc. And perhaps more importantly (to me) how much is there actually to gain.

Maybe throw in a nice Ryzen 5600x but it is more of an apples to orange comparison then.
 

IntelUser2000

Elite Member
Oct 14, 2003
7,503
2,283
136
There's ZERO difference between 6th Gen and 10th Gen desktop chips. At the same clocks they perform identically, because they are all Skylake. Only 11th Gen is truly different since that uses the new Cypress Cove core.
 

TheHolyLancer

Member
Aug 10, 2005
49
0
66
So if it wasn't for the increase in frequency (with OCing mostly) then there is no difference? I thought that each gen had IPC increase (by a bit) due to their microcode being updated, and they keep patching the older processors with new microcode that reduce their performance due to security bugs?

IE the newer stuff are same but with security fixes baked in to "increase" IPC.
 

JoeRambo

Golden Member
Jun 13, 2013
1,119
979
136
The cores themselves are pretty much the same, newer generations starting from Coffee Lake have hardware fixes to some security problems, so they dont need as nasty OS software workarounds as early chips that have uCode updates. For example 9700K:

* CPU explicitly indicates not being vulnerable to Meltdown/L1TF (RDCL_NO): YES
* CPU explicitly indicates not being vulnerable to Variant 4 (SSB_NO): NO
* CPU/Hypervisor indicates L1D flushing is not necessary on this system: YES
* Hypervisor indicates host CPU might be vulnerable to RSB underflow (RSBA): NO
* CPU explicitly indicates not being vulnerable to Microarchitectural Data Sampling (MDS_NO): NO
* CPU explicitly indicates not being vulnerable to TSX Asynchronous Abort (TAA_NO): NO
* CPU explicitly indicates not being vulnerable to iTLB Multihit (PSCHANGE_MSC_NO): NO
* CPU explicitly indicates having MSR for TSX control (TSX_CTRL_MSR): NO
* CPU supports Transactional Synchronization Extensions (TSX): YES (RTM feature bit)
* CPU supports Software Guard Extensions (SGX): YES
* CPU supports Special Register Buffer Data Sampling (SRBDS): NO
* CPU microcode is known to cause stability problems: NO (family 0x6 model 0x9e stepping 0xc ucode 0xa0 cpuid 0x906ec)
* CPU microcode is the latest known available version: NO (latest version is 0xde dated 2020/06/03 according to builtin firmwares DB v165.20201021+i20200616)

Few were fixed in hw.

Other than that two things changed:

1) Uncore -> there is HUGE difference between 8MB on 6700K and 20MB of L3 on 10900K, more cache results is smoother multitasking performance
2) Platform, newer Skylake derivatives can clock memory better. It is all "OC", things like higher memclk, tighter latencies and more stability came with platform experience ~Z390 - Z490 timeframe. Not relevant for stock, but very relevant for enthusiast.
 
Last edited:
  • Like
Reactions: Tlh97

Mopetar

Diamond Member
Jan 31, 2011
5,797
2,566
136
You're basically asking about IPC gains across the generations and those are pretty well established if you bother to pick through the reviews.

Maybe there are some differences now that won't show up in the original reviews due to Spectre/Meltdown mitigation effects, but even that probably comes out in the wash.

Even before 14nm you could get top chips to 5GHz so you could go back even further if you really wanted. I think Sandy Bridge could hit 5GHz, or it wasn't too unusual at least.
 

ASK THE COMMUNITY