Microsoft Kernel Patch CPU Before and After Benchmarks Thread

[Mockingbird]

Oh oh!

 

[Yogurth]

The performance drop does not look as bad as I have feared but it seems that the Windows patch is only a part of the equation as Intel started releasing Firmware upgrades. It would be interesting what is the penalty of the patch if any, in conjunction with the new Firmware.

[AlexTan]

HP has just pushed BIOS update.

 

TITLE: HP NoteBook System BIOS Update
VERSION: 01.12 REV: A PASS: 1

 

Spoiler

TITLE: HP NoteBook System BIOS Update

VERSION: 01.12 REV: A PASS: 1
DESCRIPTION:
This package creates files that contain an image of the System BIOS (ROM) for
the supported notebook models with a P80 family ROM. This package is used to
flash the System ROM on a supported notebook.

This package includes several methods for updating the BIOS version as follows:

- Use the HPBIOSUPDREC Utility to update the BIOS directly in a Microsoft
Windows Operating System environment.
- Use System Software Manager (SSM) to update the system BIOS on PCs in a
network.

NOTE: To determine the ROM family and ROM date, press F10 on the notebook/laptop
during startup to run the Setup Utility, and then view File/System Information.

PURPOSE: Critical
SOFTPAQ FILE NAME: SP84329.exe
SOFTPAQ MD5: ca35a50549e6fc4205d61bd52fc5ad67
SUPERSEDES: SP82334
EFFECTIVE DATE: January 04, 2018
CATEGORY: BIOS
SSM SUPPORTED: Yes

PRODUCT TYPE(S):
Notebooks

HARDWARE PRODUCT MODEL(S):
HP EliteBook x360 1030 Notebook PC

OPERATING SYSTEM(S):
Microsoft Windows 10 All Editions (64-bit)

LANGUAGE(S): Global

ENHANCEMENTS:
- Provides the updated Realtek UEFI drivers.
- Provides improved security on systems using the Kaby Lake processor.
- Provides improved security on systems using the Skylake processor.
- Adds a feature to disable all integrated audio devices on the system.
- Adds a feature to disable all integrated microphones devices on the system.
- Provides support for HP Thunderbolt Dock 120W/230W G2.
- Provides the keyboard self-test feature.
- Provides a feature to allow the IT administrator to use the BIOS Configuration
Utility or the F10 BIOS menu to control HP Sure View on the system.
- Provides improved security of UEFI code and variables in an Intel system. HP
strongly recommends transitioning promptly to this updated BIOS version which
supersedes all previous releases.

NOTE: Due to the security changes in this release, attempts to install older
BIOS versions require the user to be physically present to accept the older
version.

FIXES:
- Fixes an issue on a system with an Intel processor where the CPU usage is
unexpectedly high when Wake On LAN is used.
- Fixes an issue where the response of the cursor is unexpectedly slow after the
lid is closed and then reopened.
- Fixes an issue where the Save/Restore GPT feature does not function properly.
- Fixes an issue where the Windows Management Instrumentation (WMI) Physical
Presence Interface setting does not function properly after the setup password
is configured.
- Fixes an issue which causes the system to stop functioning (hang) after Intel
Active Management Technology (AMT) is disabled and then the BIOS is updated in
Non-Delay Mode.
- Fixes an issue where the system exits the F10 BIOS menu unintentionally when
the Fn+F3 key combination is used in the setup menu.
- Fixes an issue where the system exits the F10 BIOS menu unintentionally when
the Fn+F4 key combination is used in the setup menu.
- Fixes an issue where the system unexpectedly wakes from Sleep (S3) or
Hibernate (S4) modes after the system is reconnected to a USB-C Dock.
- Fixes an issue where a system that is powered off while connected to a USB-C
Dock and then disconnected from the dock wakes unexpectedly after the system is
reconnected to the USB-C Dock.
- Fixes an issue where the Fast Charge option is no longer available in the F10
BIOS setup menu after a BIOS update is run.
- Fixes an issue where the response of a device connected to the system using
the Thunderbolt port is unexpectedly slow after a BIOS update is run.
- Fixes an issue where the battery in a USB device connected to the system
continues to recharge from the system when the device is connected to a system
that has the Charging Port disabled when the battery is below a level that is
set in the F10 BIOS menu.

 

[Mando]

9 hours ago, jamester64 said:

Strange that AsusFanControlService is still running in the background so at least I got that :-) I just cant get the GUI to open :-(

yep same here.  suspect its the bios side of things in AISuite. if you try to rerun it, UAC prompt pops up but it never opens its ui, but you can see it in task mgr.

[AlexTan]

Benchmark Comparison Between:
After OS Patched, Before BIOS Patched
&
After OS Patched, After BIOS Patched

 

 

Computer:  HP HP EliteBook x360 1030 G2
Processor: Intel Core i7-7600U (GenuineIntel Family 6 Model 142 Stepping 9)

 

PassMark PerformanceTest (CPU)
==============================
Before: 6054, 6106, 6033 (Average: 6064)
After : 6166, 5957, 5738 (Average: 5954)
Result: -1.81%

 

Geekbench 4 (CPU)
=================
Before (Single-Core): 4715, 4701, 4707 (Average: 4708)
After  (Single-Core): 4629, 4657, 4667 (Average: 4651)
Result: -1.21%

 

Before (Multi-Core): 9254, 9301, 9294 (Average: 9283)
After  (Multi-Core): 9111, 9142, 9165 (Average: 9139)
Result: -1.55%

 

Score link for before BIOS patched:
https://browser.geekbench.com/v4/cpu/6136429
https://browser.geekbench.com/v4/cpu/6136510
https://browser.geekbench.com/v4/cpu/6136562

 

Score link for after BIOS patched:
https://browser.geekbench.com/v4/cpu/6137171
https://browser.geekbench.com/v4/cpu/6137243
https://browser.geekbench.com/v4/cpu/6137313

 

CPU-Z
=====
Before (Single-Core): 407.4, 414.7, 414.3 (Average: 412.1)
After  (Single-Core): 269.2, 253.2, 248.2 (Average: 256.9)
Result: -37.66%

 

Before (Multi-Core): 1164.0, 1169.4, 1171.8 (Average: 1168.4)
After  (Multi-Core): 1140.7, 1144.2, 1138.2 (Average: 1141.0)
Result: -2.35%

 

[PsYcHoKiLLa]

17 hours ago, Elliot B. said:

So, no difference.

 

The whole thing was overblown.

Not really, on newer CPUs the difference is nominal, on older models, especially those with less cores (dual core) or no hyperthreading, it'll be a lot more damaging to performance.

[satukoro]

22 hours ago, Mando said:

Great idea Warwagon

 

here is the start of mine 

 

Benchtools used on sig rig @1080p

 

• Cinebench 4D
• 3D Mark (Timespy Extreme)
• Novabench
• Rog RealBench
• Passmark

Before Patch 

 

Passmark CPU test BEFORE

 

Passmark CPU AFTER

 

Cinebench 4D BEFORE

• CPU Benchmark = 966CB
• OpenGL Bench = 140 FPS

AFTER

• CPU Benchmark = 975CB ???
• OpenGl Bench = 149.95 FPS ???
   

3D Mark timespy extreme BEFORE

• 3998

AFTER

• 3997 ????

Novabench 

• 2713

After

• 2703
   

RealBench BEFORE

• Image Editing 212,486/ Time 25.0745
• Encoding 93,501/time 56.9829
• OpenCL 113,529/ Ksamples per sec 20877
• Heavy Multitasking 110,647
• Time 68.9761
• System Score 132,540

RealBench AFTER

• Image Editing 208.830/Time 25.5135
• Encoding 92,841/time 57.3881
• OpenCL 113,529/ Ksamples/sec 20922
• Heavy Multitasking 103,867/Time 73.4779
• System Score 129,766

 

Overall slightly slower with some benches, Cinema 4D benches.....anyones guess!

 

before and after had all the same processes running and apps, what i have noticed however, Asus AI Suite 3 wont run

 

btw Webroot Secure anywhere working no problem at all post patch,

Cinema 4D doesn't give a crap about kernel patches apparently. Thanks for the laugh lol

[Mando]

2 minutes ago, satukoro said:

Cinema 4D doesn't give a crap about kernel patches apparently. Thanks for the laugh lol

lol so it seems, but tbh outside of work, i dont give a crap about kernel patches either  lol 

[Mockingbird]

Realbench 2.56

 

Core i7-8700K

 

[Yogurth]

5 hours ago, Mockingbird said:

Realbench 2.56

 

Core i7-8700K

 

Looks really bad when both BIOS and Windows patch work together. Gonna hold off the updates for now and see if any consequent update will fix the performance.

[adrynalyne]

5 hours ago, Mockingbird said:

Realbench 2.56

 

Core i7-8700K

 

What motherboard?

[nekrosoft13]

this recent patch slowed down my 4k x265 10bit encoding from 24fps to 21fps...

when doing large batches that is somewhat significant, FU intel.

[Mockingbird]

1 hour ago, adrynalyne said:

What motherboard?

ASUS PRIME Z370-A

 

It doesn't have to do with that specific motherboard though.

 

It's the microcode update from Intel (in the BIOS) that's causing additional performance drop.

[adrynalyne]

9 minutes ago, Mockingbird said:

ASUS PRIME Z370-A

 

It doesn't have to do with that specific motherboard though.

 

It's the microcode update from Intel (in the BIOS) that's causing additional performance drop.

Duh. Thats not why I was asking. I was curious who had released an update so soon.

 

How do you know this isn't the Intel ME microcode update? It doesn't specify and getting the BIOS out the same day as Intel released the update to OEMs sounds kinda crazy. You'd think at least it would be a beta BIOS.

Edited January 6, 2018 by adrynalyne

[nekrosoft13]

Threadripper here i come, FU intel

[adrynalyne]

2 minutes ago, nekrosoft13 said:

Threadripper here i come, FU intel

And then they patch Spectre and you wish you hadn’t switched  There are also rumors of more CPU vulnerabilites out there too. Might want to wait and see how this all plays out. 

[LimeMaster]

10 minutes ago, adrynalyne said:

And then they patch Spectre and you wish you hadn’t switched  There are also rumors of more CPU vulnerabilites out there too. Might want to wait and see how this all plays out. 

1% or 2% hit per patch. Now that's a scary thought. 

[adrynalyne]

1 minute ago, LimeMaster said:

1% or 2% hit per patch. Now that's a scary thought. 

Maybe, or we could instead see optimization.

 

I am tempted to switch to AMD, but I am watching first. I don't want to walk into a new mess and be out of pocket for it.

[nekrosoft13]

22 minutes ago, adrynalyne said:

And then they patch Spectre and you wish you hadn’t switched  There are also rumors of more CPU vulnerabilites out there too. Might want to wait and see how this all plays out. 

Threadripper is already faster in type of task that I care about, and now intel is 12-20% slower than before, so benefit of running Intel is less and less.

[adrynalyne]

1 minute ago, nekrosoft13 said:

Threadripper is already faster in type of task that I care about, and now intel is 12-20% slower than before, so benefit of running Intel is less and less.

*shrug* Its your money. I am just saying that we don't know how much a hit Spectre will be.

[LimeMaster]

1 minute ago, adrynalyne said:

Maybe, or we could instead see optimization.

 

I am tempted to switch to AMD, but I am watching first. I don't want to walk into a new mess and be out of pocket for it.

Honestly, I would have preferred if they had went with optimization first. I get that security is important, but so is usability & performance.  Especially when you consider the fact that most people don't have top of the line CPUs.

[adrynalyne]

1 minute ago, LimeMaster said:

Honestly, I would have preferred if they had went with optimization first. I get that security is important, but so is usability & performance.  Especially when you consider the fact that most people don't have top of the line CPUs.

I've a feeling that maybe they would have, had a certain company not blabbed sooner than they expected.

[LimeMaster]

Just now, adrynalyne said:

I've a feeling that maybe they would have, had a certain company not blabbed sooner than they expected.

Yeah I suppose. They are kinda screwing themselves over a bit, since they have less time to create their own patch.

[adrynalyne]

42 minutes ago, adrynalyne said:

Duh. Thats not why I was asking. I was curious who had released an update so soon.

 

How do you know this isn't the Intel ME microcode update? It doesn't specify and getting the BIOS out the same day as Intel released the update to OEMs sounds kinda crazy. You'd think at least it would be a beta BIOS.

Disregard the question--I saw how they checked it.

Edit: Seems like there are conflicting reports.

 

10 minutes ago, LimeMaster said:

Yeah I suppose. They are kinda screwing themselves over a bit, since they have less time to create their own patch.

True that.

Edited January 6, 2018 by adrynalyne

[Mando]

6 hours ago, adrynalyne said:

Disregard the question--I saw how they checked it.

Edit: Seems like there are conflicting reports.

 

True that.

dang, jjust checked for my 270 mobo no bios update yet  (Strix H) 7700k