Intel Core i9 9900K processor review
Intel Core i7 9700K processor review
Intel Core i5 9600K processor review
MSI MEG Z390 ACE review
ASUS ROG Maximus XI Hero review
Crucial P1 M.2 1000GB SSD Review
Corsair Force Series MP510 M.2 SSD Review
MSI GeForce RTX 2070 Armor 8G review
ASUS Turbo GeForce RTX 2070 8GB review
Guru3D PC Buyers Guide Autumn 2018
Intel Shares Internal Security Fix Benchmarks Results for Desktop & Laptop
Intel has released a set of desktop and laptop benchmark results with all Windows security patches in place and the firmware up-to-snuff. They compare mobile (Kaby Lake) and Desktop CPUs (Coffee lake, Kaby and Sky lake).
The metrics are a bit all over the place, however as it seems their findings match up to ours, overall in Sky Diver (which is more CPU dependent), there's minimal performance loss that easily could be seen as an anomaly (~1 or 2%) otherwise.
When the workload changes to really heavy with say PCMark 10 / Sysymark, here on the desktop platforms you are looking at a 4% differential overall with an exception here and there. And in the massive stressed synthetic benchmark, Sysmark, the worst it got was 10% perf loss. The WebXPRT browsing benchmark does take a hit, here again, that is a heavy synthetic test with multiple workloads including DNA sequencing, not your normal browsing experience. Have a peek for yourself and load up the screengrab below.
The data sets do not include any server-side or storage benchmarks, where the biggest hit is to be found. Neither did Intel share numbers for the Haswell and older generations CPUs/Platforms. Intel did test the results with an SSD which would have an effect om Sysmark and PCMark alright, specs of the test setups are listed below the test-results.
Intel Showcases 10nm Updates and 64-Layer NAND For Enterprise - 09/20/2017 09:51 AM
Intel delivered updates at its Technology and Manufacturing Day held in Beijing, China, on Sept. 19. Disclosures included power and performance updates for Intel’s 10 nm process, high-level ...
Intel Shows 1.59x Performance Improvement in Upcoming Intel Xeon Scalable Family - 05/18/2017 07:57 AM
Intel unveiled performance advances in its upcoming Intel Xeon Processor Scalable family. At the SAP Sapphire NOW conference, Intel showed up to 1.59x higher Intel Xeon processor performance running i...
Intel Shows 28-core processor die-shot - 03/29/2017 10:43 AM
During its Technology and Manufacturing Day Intel has shared a slide of a server-processor with 28 cores. The die-shot as such must be the Skylake-SP series of products in the Xeon Platinum range....
Intel Shows Optane SSD writing at 2GB per second - 04/16/2016 10:33 AM
Over at IDF in China a prototype of an Optane SSD from Intel has been showing some staggering numbers. The SSD is based on 3D XPoint storage technology and while copying a 25GB file it did so at 2 GB/...
Intel shows Core i7-6950X specs at 3,5GHz and 25MB cache - 04/04/2016 10:34 AM
Intel posted info on their Core i7-6950X over the weekend on their website. The 10-core processor will get a 3,5GHz base clock and 25MB cache. Other then some snippets of info, nothing specific is m...
tsunami231
Senior Member
Posts: 8699
Joined: 2003-05-24
Senior Member
Posts: 8699
Joined: 2003-05-24
#5509766 Posted on: 01/11/2018 07:38 PM
how you see that, even HH said it matches what he has seen
So basically their test is just smoke and mirrors
how you see that, even HH said it matches what he has seen
Ricepudding
Senior Member
Posts: 397
Joined: 2017-02-17
Senior Member
Posts: 397
Joined: 2017-02-17
#5509772 Posted on: 01/11/2018 07:46 PM
Basically 1-2% difference in most cases, with some very heavy cases hitting around 4% difference in performance. Though general users would be hard pressed to see these differences, these would effect massive content creators/ servers and maybe even software programmers the most. Though as the table says these results are in a +/- of 3% meaning take them with a grain of salt as tests are never 100% accurate and go up and down all the time.
Personally I've not seen or noticed anything of difference if there has been a 1/2 fps drops in games i would never personally notice something like that, Though CPU usage and temps seem about the same as before the patches and CPU-Z seems to give me similar results
Single thread - 547
Muti Thread - 4106.2
(not overclocked)
Using newest version of CPU-z
Basically 1-2% difference in most cases, with some very heavy cases hitting around 4% difference in performance. Though general users would be hard pressed to see these differences, these would effect massive content creators/ servers and maybe even software programmers the most. Though as the table says these results are in a +/- of 3% meaning take them with a grain of salt as tests are never 100% accurate and go up and down all the time.
Personally I've not seen or noticed anything of difference if there has been a 1/2 fps drops in games i would never personally notice something like that, Though CPU usage and temps seem about the same as before the patches and CPU-Z seems to give me similar results
Single thread - 547
Muti Thread - 4106.2
(not overclocked)
Using newest version of CPU-z
kruno
Senior Member
Posts: 246
Joined: 2017-09-25
Senior Member
Posts: 246
Joined: 2017-09-25
#5509779 Posted on: 01/11/2018 07:59 PM
They didn't test old machines, as i have noticed plenty of people here on forum have "old" 2000 or 3000 series cpu's.
Edit: That is one of more interesting stuff to test and that is the thing Intel is in unique position to test, of course i am not forgetting that they didn't post results for even older systems, and god knows there are plenty of Core2 CPU's in use
how you see that, even HH said it matches what he has seen
They didn't test old machines, as i have noticed plenty of people here on forum have "old" 2000 or 3000 series cpu's.
Edit: That is one of more interesting stuff to test and that is the thing Intel is in unique position to test, of course i am not forgetting that they didn't post results for even older systems, and god knows there are plenty of Core2 CPU's in use
Neo Cyrus
Senior Member
Posts: 8936
Joined: 2006-02-14
Senior Member
Posts: 8936
Joined: 2006-02-14
#5509795 Posted on: 01/11/2018 08:33 PM
According to MS the Haswell and older chips got hit harder so this doesn't apply to a lot of us even if Intel didn't twist the numbers.
According to MS the Haswell and older chips got hit harder so this doesn't apply to a lot of us even if Intel didn't twist the numbers.
tsunami231
Senior Member
Posts: 8699
Joined: 2003-05-24
Senior Member
Posts: 8699
Joined: 2003-05-24
#5509829 Posted on: 01/11/2018 09:51 PM
it should plainly obivous if current stuff have hit older stuff will have much bigger hit. im sure my dads 8400e will take much bigger hit, but requardles, you have to pick your posion, do the updates and take the hit or dont do the updates you dont regeret it.
Intel could do test on ALL there cpu that effect, but that is ALOT CPU, which isnt gona happen and no one in there right mind would expect them too, would be nice sure is it gona happen probably not, cause alot those cpu are EOL already.
it should plainly obivous if current stuff have hit older stuff will have much bigger hit. im sure my dads 8400e will take much bigger hit, but requardles, you have to pick your posion, do the updates and take the hit or dont do the updates you dont regeret it.
Intel could do test on ALL there cpu that effect, but that is ALOT CPU, which isnt gona happen and no one in there right mind would expect them too, would be nice sure is it gona happen probably not, cause alot those cpu are EOL already.
BlueRay
Senior Member
Posts: 239
Joined: 2015-11-18
Senior Member
Posts: 239
Joined: 2015-11-18
#5509833 Posted on: 01/11/2018 10:07 PM
While on most cases the drop is small Witcher 3 tells a different story. That drop was big.
While on most cases the drop is small Witcher 3 tells a different story. That drop was big.
WareTernal
Senior Member
Posts: 207
Joined: 2013-09-27
Senior Member
Posts: 207
Joined: 2013-09-27
#5509844 Posted on: 01/11/2018 10:37 PM
I would like to see numbers from these older CPU's as well, since they are said to see a more noticeable decrease in performance. I haven't seen results to confirm the claim that older CPU's are hit harder. I'm getting slightly better CPU score in Cinebench after the patch with my 2600/Win10. Hardly worth mentioning, but I'm not seeing a decrease in Cinebench, even though I've seen newer CPU's taking a bigger hit. This is not scientific. My 'before' result is the highest score from several runs I did a few weeks ago. The 'after' results are from 3 runs I did today.
Before: 694
After: 699 / 695 / 699 = 697.67 average
This guys results show the same thing on his 2600k - <1% difference, and in some cases faster after patch.
It would be interesting to see SQL benchmarks, as well as seeing how updated OS/firmware affects NAS units with weaker CPU's.
They didn't test old machines, as i have noticed plenty of people here on forum have "old" 2000 or 3000 series cpu's.
According to MS the Haswell and older chips got hit harder
I would like to see numbers from these older CPU's as well, since they are said to see a more noticeable decrease in performance. I haven't seen results to confirm the claim that older CPU's are hit harder. I'm getting slightly better CPU score in Cinebench after the patch with my 2600/Win10. Hardly worth mentioning, but I'm not seeing a decrease in Cinebench, even though I've seen newer CPU's taking a bigger hit. This is not scientific. My 'before' result is the highest score from several runs I did a few weeks ago. The 'after' results are from 3 runs I did today.
Before: 694
After: 699 / 695 / 699 = 697.67 average
This guys results show the same thing on his 2600k - <1% difference, and in some cases faster after patch.
It would be interesting to see SQL benchmarks, as well as seeing how updated OS/firmware affects NAS units with weaker CPU's.
H83
Senior Member
Posts: 2392
Joined: 2009-09-08
Senior Member
Posts: 2392
Joined: 2009-09-08
#5509846 Posted on: 01/11/2018 10:38 PM
I´m going to wait for benchmarks of more independent sources...
I´m going to wait for benchmarks of more independent sources...
tsunami231
Senior Member
Posts: 8699
Joined: 2003-05-24
Senior Member
Posts: 8699
Joined: 2003-05-24
#5509849 Posted on: 01/11/2018 10:47 PM
my take on the fact some game show diffrence and other dont, then same game show no difference with different cpu, tells me some engine/games dont care or are just program in way it dont mater, which would mean said games can be patchs so there is no hit.
But it also tells me these test for performance hits highly subjective and anything but consistent. from system to system cpu to cpu. it enterly possible 2 people with same cpu/gpu/amount ram cause have completely diffrent performance hits too, cause there is so many other things that factor in to peformance.
my take on the fact some game show diffrence and other dont, then same game show no difference with different cpu, tells me some engine/games dont care or are just program in way it dont mater, which would mean said games can be patchs so there is no hit.
But it also tells me these test for performance hits highly subjective and anything but consistent. from system to system cpu to cpu. it enterly possible 2 people with same cpu/gpu/amount ram cause have completely diffrent performance hits too, cause there is so many other things that factor in to peformance.
asturur
Senior Member
Posts: 184
Joined: 2010-05-12
Senior Member
Posts: 184
Joined: 2010-05-12
#5509854 Posted on: 01/11/2018 11:07 PM
Well i guess 3-5% is what we gained in performance per clock cycle from intel generation 4 to intel generation 7?
Well i guess 3-5% is what we gained in performance per clock cycle from intel generation 4 to intel generation 7?
kruno
Senior Member
Posts: 246
Joined: 2017-09-25
Senior Member
Posts: 246
Joined: 2017-09-25
#5509861 Posted on: 01/11/2018 11:33 PM
I would like to see numbers from these older CPU's as well, since they are said to see a more noticeable decrease in performance. I haven't seen results to confirm the claim that older CPU's are hit harder. I'm getting slightly better CPU score in Cinebench after the patch with my 2600/Win10. Hardly worth mentioning, but I'm not seeing a decrease in Cinebench, even though I've seen newer CPU's taking a bigger hit. This is not scientific. My 'before' result is the highest score from several runs I did a few weeks ago. The 'after' results are from 3 runs I did today.
Before: 694
After: 699 / 695 / 699 = 697.67 average
This guys results show the same thing on his 2600k - <1% difference, and in some cases faster after patch.
It would be interesting to see SQL benchmarks, as well as seeing how updated OS/firmware affects NAS units with weaker CPU's.
Well that is thing that bugs me.There is 99.99% chance that you will not receive new bios,mobo manufacturer want trouble itself with that "old" stuff.As it seems combination of new bios + MS patch can cause severe penalty. You and i don't have the resources to make new bios for mobo, but Intel has, and it would be fair from them to make and test their "old" systems. In any case time will show how this mess is going to end, but in my opinion you are scr**d, there will be penalty sooner or later.
I would like to see numbers from these older CPU's as well, since they are said to see a more noticeable decrease in performance. I haven't seen results to confirm the claim that older CPU's are hit harder. I'm getting slightly better CPU score in Cinebench after the patch with my 2600/Win10. Hardly worth mentioning, but I'm not seeing a decrease in Cinebench, even though I've seen newer CPU's taking a bigger hit. This is not scientific. My 'before' result is the highest score from several runs I did a few weeks ago. The 'after' results are from 3 runs I did today.
Before: 694
After: 699 / 695 / 699 = 697.67 average
This guys results show the same thing on his 2600k - <1% difference, and in some cases faster after patch.
It would be interesting to see SQL benchmarks, as well as seeing how updated OS/firmware affects NAS units with weaker CPU's.
Well that is thing that bugs me.There is 99.99% chance that you will not receive new bios,mobo manufacturer want trouble itself with that "old" stuff.As it seems combination of new bios + MS patch can cause severe penalty. You and i don't have the resources to make new bios for mobo, but Intel has, and it would be fair from them to make and test their "old" systems. In any case time will show how this mess is going to end, but in my opinion you are scr**d, there will be penalty sooner or later.
Ricepudding
Senior Member
Posts: 397
Joined: 2017-02-17
Senior Member
Posts: 397
Joined: 2017-02-17
#5509882 Posted on: 01/12/2018 12:48 AM
While on most cases the drop is small Witcher 3 tells a different story. That drop was big.
Strange all the other tests show no difference typically within the margins of error. Witcher shows a big difference I wonder if a patch of the game would fix that... Also is the effect as big on say the 8600/8700k cpus... Interesting to say the least in all the games I've played I've not seen anything noticeable as dropping 10fps
While on most cases the drop is small Witcher 3 tells a different story. That drop was big.
Strange all the other tests show no difference typically within the margins of error. Witcher shows a big difference I wonder if a patch of the game would fix that... Also is the effect as big on say the 8600/8700k cpus... Interesting to say the least in all the games I've played I've not seen anything noticeable as dropping 10fps
WareTernal
Senior Member
Posts: 207
Joined: 2013-09-27
Senior Member
Posts: 207
Joined: 2013-09-27
#5509884 Posted on: 01/12/2018 12:52 AM
I definitely agree. As Kruno noted, Intel would be in position to give us some real numbers with older systems, but I don't expect it with these EOL porducts.
Yes I agree, but I'm not sure if the mitigation is in effect without the needed microcode.
it also tells me these test for performance hits highly subjective and anything but consistent
I definitely agree. As Kruno noted, Intel would be in position to give us some real numbers with older systems, but I don't expect it with these EOL porducts.
it would be fair from them to make and test their "old" systems
Yes I agree, but I'm not sure if the mitigation is in effect without the needed microcode.
Aitortxo
Member
Posts: 37
Joined: 2004-08-15
Member
Posts: 37
Joined: 2004-08-15
#5509916 Posted on: 01/12/2018 04:14 AM
What I want to see is results from independent sources like this site on older CPUs (Intel 2000-3000 series) running W7 which is what the majority of people are using.
What I want to see is results from independent sources like this site on older CPUs (Intel 2000-3000 series) running W7 which is what the majority of people are using.
Click here to post a comment for this news story on the message forum.
Senior Member
Posts: 246
Joined: 2017-09-25
So basically their test is just smoke and mirrors