Ryzen 7000 and 9000 see massive gains from Windows 11 24H2. Mostly.

[Naff8000]

One more thing, he gave his reasoning for ignoring the problem was that Intel was getting consistent results, so he just put it down to AMD, which speaks volumes in its self about his mind set, no wonder its so easy for him to publish results that are wrong and make a click bait video out of it raging at AMD.... but here's the point, when he published the 30% 9700X gains for 24H2 vs his original review he said Intel also gained 20%.

Roll the dice and you'll hear who Steve's shilling for now. I would not take Steve as an Intel shill,he generally favours AMD over the years I've been watching.

 

[humbug]

In his next video he will say sorry to AMD for calling the Zen 5 launch the worst CPU launch in the history of x86 and to be sure to call microsoft for answers as to why (since Intel have been struggling to keep the performance crown at any cost) recent Windows cripples AMD processors.

Seems to me a bit like Nvidia - TWIMTBP. Dodgy stuff

Never blame malice where stupidity can be attributed, he's a pretty stupid man, the problem with stupid people is they don't know they are stupid, i explained above what i think has happened.

But i don't blame anyone for thinking its malice because he keeps doing this.

 

[humbug]

This is what i think happened, he's been using that drive for some time switching it between AMD and Intel over and over again...

One more thing, he gave his reasoning for ignoring the problem was that Intel was getting consistent results, so he just put it down to AMD, which speaks volumes in its self about his mind set, no wonder its so easy for him to publish results that are wrong and make a click bait video out of it raging at AMD.... but here's the point, when he published the 30% 9700X gains for 24H2 vs his original review he said Intel also gained 20%.

You know what he was doing, he was just using that same drive with that same windows install for all his reviews, just plugging it in to Intel motherboards and then AMD for reviews..... Who does that????? No wonder it was broken.

 

[TNA]

Never blame malice where stupidity can be attributed, he's a pretty stupid man, the problem with stupid people is they don't know they are stupid, i explained above what i think has happened.

But i don't blame anyone for thinking its malice because he keeps doing this.

 

[Trick]

You know what he was doing, he was just using that same drive with that same windows install for all his reviews, just plugging it in to Intel motherboards and then AMD for reviews..... Who does that????? No wonder it was broken.

Doesnt he say about 5mins in that this isn't what they do? He says they do a fresh install each time and don't just swap cpus..

 

[pete910]

Roll the dice and you'll hear who Steve's shilling for now. I would not take Steve as an Intel shill,he generally favours AMD over the years I've been watching.

Always thought he was Italian to be honest, Switches sides to when suits....

 

[WARburton]

Link to the article you are referring to?

I was referring to the thread

 

[BS Dave]

I think it doesn't show up if virtualisation is disabled in the BIOS (SVM on AMD). The A and C are hexadecimal numbers by the way with A being 10 and C being 12.

That particular CVE is MSI specific and the likelihood of it being exploited is high according to this (it doesn't sound like the attacker requires physical access, but I'm not an expert):

CWE - CWE-123: Write-what-where Condition (4.15)

Common Weakness Enumeration (CWE) is a list of software weaknesses.

cwe.mitre.org

NVD - CVE-2024-36877

nvd.nist.gov

Hex! Amazing. So if I turn SME off in my bios it gets rid of the vulnerability? So need to upgrade the bios (this is a performance eater as well isn't it?)

OR if I upgrade the bios and turn off SME they'll be no performance loss??

It'd be great to know which

 

[ltron]

Hex! Amazing. So if I turn SME off in my bios it gets rid of the vulnerability? So need to upgrade the bios (this is a performance eater as well isn't it?)

OR if I upgrade the bios and turn off SME they'll be no performance loss??

It'd be great to know which

These vulnerabilities have nothing to do with virtualisation, virtualisation based security is something that can protect your computer against certain vulnerabilities that haven't been discovered yet, but it comes at a small performance cost. I was talking about two different things in my post.

If I were you I would flash the latest BIOS to close these security holes.

 

[iamdjdz]

Sounds positive, anyone seen more performance on 7700Xs?

 

[BS Dave]

These vulnerabilities have nothing to do with virtualisation, virtualisation based security is something that can protect your computer against certain vulnerabilities that haven't been discovered yet, but it comes at a small performance cost. I was talking about two different things in my post.

If I were you I would flash the latest BIOS to close these security holes.

Hex! Amazing. So if I turn SME off in my bios it gets rid of the vulnerability? So need to upgrade the bios (this is a performance eater as well isn't it?)

OR if I upgrade the bios and turn off SME they'll be no performance loss??

It'd be great to know which

OK. Thanks for the reply.

 

[WARburton]

Sounds positive, anyone seen more performance on 7700Xs?

The update is not out to general public yet l, but the beta tester are reporting all 7xxx cpus are doing better even if o it by a few percentage.

 

[Dirk Diggler]

I have a 7800x3d - do I need to tweak anything to get the performance boost or do I simply update Windows?

 

[WARburton]

I have a 7800x3d - do I need to tweak anything to get the performance boost or do I simply update Windows?

Other that’s a -25 offset there is nothing to do with a x3d chap

 

[Glanza]

 

[humbug]

Ok so after this 11% 23H2 vs 24H2 What AMD actually gained when the review matters was 2%, according to this when you add that difference to the slide below and be left with a 3% win to the 14700K, the absolute minimum you could do and argue its not margin of error.

I find it really interesting how you can get +11 % in 40 games 23H2 vs 24H2 but arguably nothing in 13 games when you're actually rereviewing the chip as it matters.

Ahhh...... wait a minute, all but 3 of those which gained more than 10% have not been included in this review, none of the top 5 have, but all of the bottom 20 have.

Put a dampener on it i guess, nothing actually changed, some as original conclusion. That's how its done, that's how AMD still lied.

 

[Tetras]

Why would he waste all this time comparing with an Intel CPU that nobody recommends you actually buy

 

[alec]

add another 2% from PBO "which does nothing", tune the memory (on both CPUs, but its more of a bottleneck on Zen 5), do the basic curve optimizer tweak (go ahead and try undervolting Intel if you want)
and suddenly the turns have tabled

Steve has decided to die on the hill that Zen 5 is a flop.
AMD launch is a flop, Zen 5 is good.

 

[humbug]

Why would he waste all this time comparing with an Intel CPU that nobody recommends you actually buy

Well he does now, even complained that AMD used the 125 watt Profile for the Intel CPU citing that Intel specifically said you shouldn't use that profile, so he of course uses the 253 watt profile, suddenly the last 6 months have been forgotten.
Steve Burke always uses the 125 watt profile for gaming reviews, always has, you know why he does that? Because Intel cite "125 watt TDP" all over their marketing material....

You couldn't make it up...

 

[Tetras]

Well he does now

He admitted himself at the end that it was a weird comparison, with two CPUs he doesn't even recommend you buy.

I would have been interested with the 7800X3D & 7700/7700X in the benchmark, but the 14700K is pointless, they're dead to me as something I'd consider recommending.

citing that Intel specifically said you shouldn't use that profile, so he of course uses the 253 watt profile, suddenly the last 6 months have been forgotten.

He is correct about that part. Intel never recommended anyone use the baseline profiles, they're under the stock spec.

The 253 watt limit is not why these CPUs have had issues within the last 6 months, they're still operating at that with the supposedly 'fixed' BIOS.

The average high-end board ignored the power limits entirely "out of the box".