Overclock.net - An Overclocking Community - Reply to Topic
Thread: [Phoronix] Intel's Mitigation For CVE-2019-14615 Graphics Vulnerability Obliterates Gen7 iGPU Performance Reply to Thread
Title:
Message:

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in


  Additional Options
Miscellaneous Options

  Topic Review (Newest First)
01-20-2020 06:56 PM
skupples
Quote: Originally Posted by rluker5 View Post
The death of core should be here soon enough.
Check out the max single core of a locked 3.9 single/3.5ghz multi throttle prone 15w sunny cove vs the max single locked 5ghz skylake. https://cpu.userbenchmark.com/Compar...16115vsm888368
Willow cove (tiger lake) will be faster and I'm really hoping golden cove will have a 8c16t ddr5 desktop sku. It will make ryzen look like fx and that may be being generous to AMD.

What I have will do for now, but I'm pretty hyped about golden cove. I wonder if the supposed redone cache will fix these security flaws that seem pretty inconsequential for the home user, but viable for shared public servers.
I'm sure they'll rebound just fine, the question is when.

i'm on a 9700k atm, and i've gotta say. I'm an idiot for putting money into something with only 16 lanes. I'm jumping ship over to whatever winner AMD drops at the end of the year.
01-20-2020 07:43 AM
EniGma1987
Quote: Originally Posted by SwitchFX View Post
Does this affect QuickSync?
I would think so. QS has its own hardware blocks in the iGPU, but does use the shader cores for compute stuff. So switching between QS compute and graphics rendering of applications will trigger the context switch and flush the EUs. You could probably mitigate this by turning off as much gpu rendering of windows as possible.
01-20-2020 06:38 AM
rluker5
Quote: Originally Posted by skupples View Post
only problem is anyone with a bit of knowledge in their tech division is telling them to renew on AMD products. MSPs will also pimp AMD solutions going forward, if they haven't already. That only leaves home users victim to these shady tactics, which rarely end well (see Apple's battery update that ended in class action and free warranty replacements)

example - we ordered 100 Vega 10 books last week, BEFORE Ryuk made the news, weeks after windows defender started finding it... Now i'm like a prophet, even though I've been saying the EXACT SAME THING thru all of 2019. We'll be buying nothing but AMD products for the next 3-4 years assuming things keep this way. Maybe Shart's hitting the fan hard enough @ Intel CPU division to speed up the death of core.
Quote: Originally Posted by Paradigm Shifter View Post
I was (mostly) joking. I'm working in a place where when I asked for a Threadripper system, I got funny looks... "Why don't you want Intel?"
The death of core should be here soon enough.
Check out the max single core of a locked 3.9 single/3.5ghz multi throttle prone 15w sunny cove vs the max single locked 5ghz skylake. https://cpu.userbenchmark.com/Compar...16115vsm888368
Willow cove (tiger lake) will be faster and I'm really hoping golden cove will have a 8c16t ddr5 desktop sku. It will make ryzen look like fx and that may be being generous to AMD.

What I have will do for now, but I'm pretty hyped about golden cove. I wonder if the supposed redone cache will fix these security flaws that seem pretty inconsequential for the home user, but viable for shared public servers.
01-19-2020 09:57 PM
Paradigm Shifter
Quote: Originally Posted by skupples View Post
only problem is anyone with a bit of knowledge in their tech division is telling them to renew on AMD products. MSPs will also pimp AMD solutions going forward, if they haven't already. That only leaves home users victim to these shady tactics, which rarely end well (see Apple's battery update that ended in class action and free warranty replacements)

example - we ordered 100 Vega 10 books last week, BEFORE Ryuk made the news, weeks after windows defender started finding it... Now i'm like a prophet, even though I've been saying the EXACT SAME THING thru all of 2019. We'll be buying nothing but AMD products for the next 3-4 years assuming things keep this way. Maybe Shart's hitting the fan hard enough @ Intel CPU division to speed up the death of core.
I was (mostly) joking. I'm working in a place where when I asked for a Threadripper system, I got funny looks... "Why don't you want Intel?"
01-19-2020 07:22 PM
skupples
Quote: Originally Posted by Paradigm Shifter View Post
I think the point was that what impact will a future mitigation have on Gen.9 performance? Obsolescence by security mitigation? Too many people still using old systems? Need a quick injection of cash from companies replacing their kit? Announce security vulnerability, "mitigate" it (and destroy performance in the process), tell everyone "new chips don't see a performance drop!".... profit?
only problem is anyone with a bit of knowledge in their tech division is telling them to renew on AMD products. MSPs will also pimp AMD solutions going forward, if they haven't already. That only leaves home users victim to these shady tactics, which rarely end well (see Apple's battery update that ended in class action and free warranty replacements)

example - we ordered 100 Vega 10 books last week, BEFORE Ryuk made the news, weeks after windows defender started finding it... Now i'm like a prophet, even though I've been saying the EXACT SAME THING thru all of 2019. We'll be buying nothing but AMD products for the next 3-4 years assuming things keep this way. Maybe Shart's hitting the fan hard enough @ Intel CPU division to speed up the death of core.
01-19-2020 07:17 PM
SwitchFX Does this affect QuickSync?
01-19-2020 06:00 PM
Paradigm Shifter
Quote: Originally Posted by rluker5 View Post
They tested the mitigation on gen 9 in the article. Minimal performance impact.
I think the point was that what impact will a future mitigation have on Gen.9 performance? Obsolescence by security mitigation? Too many people still using old systems? Need a quick injection of cash from companies replacing their kit? Announce security vulnerability, "mitigate" it (and destroy performance in the process), tell everyone "new chips don't see a performance drop!".... profit?
01-18-2020 12:54 PM
rluker5
Quote: Originally Posted by sumitlian View Post
The interesting this is; how do we know for sure that the Gen9 iGPUs don't suffer from the similar vulnerabilities and its performance won't be "obliterated" in future; say after 2 or 3 years?
They tested the mitigation on gen 9 in the article. Minimal performance impact.
01-18-2020 10:03 AM
EniGma1987
Quote: Originally Posted by ThrashZone View Post
Hi,
Who is using onboard graphic's now days ?
Besides maybe a cheap laptop.
I have some friends who play CS in the office on Fridays when work is slow and they all use integrated intel graphics from Ivy Bridge CPUs.
01-17-2020 10:05 PM
sumitlian
Quote: Originally Posted by Liranan View Post
Well done Intel, well done.
The interesting this is; how do we know for sure that the Gen9 iGPUs don't suffer from the similar vulnerabilities and its performance won't be "obliterated" in future; say after 2 or 3 years?
This thread has more than 10 replies. Click here to review the whole thread.

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off