• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

14th Gen "Raptor Lake Refresh"

Soldato
Joined
22 May 2010
Posts
12,223
Location
Minibotpc
Got my replacement chip, the old one is indeed a dud. The new one will happily clock to 5.6ghz p core and 4.4 e core with ring clocked to 5ghz no problem. R23 bench sits around 80c at this clock speed with voltage at 1.33v under full load, voltage at idle is around 1.36v so it's right on the mark. Can do 5.7ghz but need 1.36v or higher under full load which means upping the llc which leads to much much higher temps.

Could ideally push more voltage using the same llc level so vdrop under full load sits on the mark but not comfortable with pushing 1.4v or higher requested voltage into the chip...
 
Last edited:
Soldato
Joined
15 Oct 2005
Posts
5,963
Location
Earth, for now
Got my replacement chip, the old one is indeed a dud. The new one will happily clock to 5.6ghz p core and 4.4 e core with ring clocked to 5ghz no problem. R23 bench sits around 80c at this clock speed with voltage at 1.33v under full load, voltage at idle is around 1.36v so it's right on the mark. Can do 5.7ghz but need 1.36v or higher under full load which means upping the llc which leads to much much higher temps.

Could ideally push more voltage using the same llc level so vdrop under full load sits on the mark but not comfortable with pushing 1.4v or higher requested voltage into the chip...

Great result for you. At least you can get refunded on the other CPU, hopefully..?

Cooling when stability testing or doing benchmarks can be somewhat different to games. So it looks like you have a lot of headroom there, for decent cooling without being overly noisy.
 
Soldato
Joined
22 May 2010
Posts
12,223
Location
Minibotpc
Great result for you. At least you can get refunded on the other CPU, hopefully..?

Cooling when stability testing or doing benchmarks can be somewhat different to games. So it looks like you have a lot of headroom there, for decent cooling without being overly noisy.

Honestly not sure, im tempted to reach out to Intel and see if maybe they could do something with this chip since its heavily degraded.

Yeah that's true, doing stability is necessary though in this case to make sure the chip actually works under those conditions especially since i do a lot of editing too which does max out the cpu.
 
Soldato
Joined
6 Feb 2019
Posts
17,857
Confirmation that the unreal engine game crashing on 14th Gen i9's is not due to random instability, those CPUs are actually degraded and damaged, but Intel wants to keep it very hush-hush, they even tried to not tell board partners that customer CPUs were damaged

 
Last edited:
Man of Honour
Joined
22 Jun 2006
Posts
12,207
Confirmation that the unreal engine game crashing on 14th Gen i9's is not due to random instability, those CPUs are actually degraded and damaged

I wouldn't call what they said confirmation that the crashing is due to degradation. It is confirmation that some CPUs have been degraded by the motherboard settings, but some CPUs don't work properly on "out of the box" settings even when they're brand new.
 
Soldato
Joined
28 May 2007
Posts
18,581
I wouldn't call what they said confirmation that the crashing is due to degradation. It is confirmation that some CPUs have been degraded by the motherboard settings, but some CPUs don't work properly on "out of the box" settings even when they're brand new.

It confirms what many expected would happen because of how hard Intel are forced to push its hardware.
 
Associate
Joined
21 Jun 2011
Posts
1,084
Location
London
The datasheets are meaningless.

It’s all the alleged secret dealings and back channel talk between Intel, motherboard manufacturers and reviewers that’s the only acceptable source of information.

It’s contagious. Most reviewers these days are nothing more than part time salesman.
 
Man of Honour
Joined
13 Oct 2006
Posts
91,678
I wouldn't call what they said confirmation that the crashing is due to degradation. It is confirmation that some CPUs have been degraded by the motherboard settings, but some CPUs don't work properly on "out of the box" settings even when they're brand new.

Yup some people can't even run some games, or versions of some benchmarks out the box even. But some boards especially Asus have been seen putting upwards of 1.6v through 14900s under "normal" circumstances which is almost certainly causing premature degradation as well.

Most of this problem blew up due to people being unable to run certain games out the box, hence the nVidia driver note about it.
 
Last edited:
Soldato
Joined
22 May 2010
Posts
12,223
Location
Minibotpc
Asus. The undisputed masters of throwing voltage at stuff.

It's no lie, having used Asus boards for the past few generations the AI overclocking feature especially is pretty nuts. The board will literally throw 1.5v at a cpu just so it can run a stupid unrealistic overclock and for an inexperienced user, they wouldn't even realise since it's supposed to be a one click overclock feature. The Asus multi core enhancement feature isn't much better either, voltages are all over the place mainly juicing cpu's far more than they actually need for a specific clock with LLC usually set incorrectly leading to insane temperatures which further adds to the degradation.

For manual overclocks the boards are great, lots of fine tuning available but you should never touch any of the other "auto overclock" features.
 
Last edited:
Soldato
Joined
15 Oct 2005
Posts
5,963
Location
Earth, for now
Gigabyte, perhaps not on the level that some of the Asus boards seem to have been by default, can run with vcore voltages high and power limits set to maximum. My Z690 board has not had any updated BIOS to run with my 14700k at default settings and limits suggested by Intel. Even tho I have manually added those values myself, and ran with a -0.7v undervolt, you should be able to pop a CPU into a board and the default BIOS offer zero overclocks and the safe power draws etc recommended for the CPU. OOB experience should not be overclocked, perhaps Asus have taken that a little too far. You should not have to learn to apply settings that appear akin to alchemy just to not melt your CPU.
 
Soldato
Joined
22 May 2010
Posts
12,223
Location
Minibotpc
Gigabyte, perhaps not on the level that some of the Asus boards seem to have been by default, can run with vcore voltages high and power limits set to maximum. My Z690 board has not had any updated BIOS to run with my 14700k at default settings and limits suggested by Intel. Even tho I have manually added those values myself, and ran with a -0.7v undervolt, you should be able to pop a CPU into a board and the default BIOS offer zero overclocks and the safe power draws etc recommended for the CPU. OOB experience should not be overclocked, perhaps Asus have taken that a little too far. You should not have to learn to apply settings that appear akin to alchemy just to not melt your CPU.

The first thing i do usually when i enter the bios is enforce all limits. Asus as a way of setting that to auto which lets the board optimise the cpu's performance which cranks the voltages in my experience. Enforcing limits then working my way through the settings and reducing limits when i need to is the way in this case for Asus boards. That way the board won't do anything weird without me noticing and i will at least know it's set to intel stock settings.
 
Suspended
Joined
17 Mar 2012
Posts
48,325
Location
ARC-L1, Stanton System
Confirmation that the unreal engine game crashing on 14th Gen i9's is not due to random instability, those CPUs are actually degraded and damaged, but Intel wants to keep it very hush-hush, they even tried to not tell board partners that customer CPUs were damaged


Push enough power in to a CPU to power a GPU 4X the size caused degradation, who would have thought it?

Yes Intel did get to that stage.
 
Last edited:
Suspended
Joined
17 Mar 2012
Posts
48,325
Location
ARC-L1, Stanton System
I Also hate the way tech jurnoes use wording like "Compared to the performance you should get" when talking about the "performance profile" being lower performance. On the one hand they are saying Intel did this to look better on the bar charts while on the other hand saying its how they should perform.

Its those mental gymnastics that they have to pretzel their way through because they know running Intel CPU's in a state that's safe makes them less competitive with AMD.
 
Last edited:
Soldato
Joined
15 Oct 2019
Posts
11,900
Location
Uk
I Also hate the way tech jurnoes use wording like "Compared to the performance you should get" when talking about the "performance profile" being lower performance. On the one hand they are saying Intel did this to look better on the bar charts while on the other hand saying its how they should perform.

Its those mental gymnastics that they have to pretzel their way through because they know running Intel CPU's in a state that's safe makes them less competitive with AMD.
Probably just board vendors being a bit gung ho with voltage, if I remember something similar happened with Zen 4 CPUs getting fried on ASUS boards until AMD stepped in with an AGESA cap.
 
Suspended
Joined
17 Mar 2012
Posts
48,325
Location
ARC-L1, Stanton System
Probably just board vendors being a bit gung ho with voltage, if I remember something similar happened with Zen 4 CPUs getting fried on ASUS boards until AMD stepped in with an AGESA cap.

There is that, the difference being board vendors ignored AMD's very specific voltage limit guidance in the white paper on 'how to not fry our CPU's for idiots' because Asus and Gigabyte know better than the people who designed the bloody thing.

Oh but AMD didn't put physical limits in to the firmware, its called an unlocked CPU, i think AMD started that trend about 30 years ago, its nothing new, its for people who do competitive sub zero cooled overclocking, there is an assumption on AMD's part that people who build their own computers are not stupid, that they would read the 'don't be stupid' white paper, and for the most part that is true, people who build their own are a cut above normies, nerdy types.
So along come the people who build the motherboards and do exactly what all nerdy types know is a really really really bad idea, those are the people who are the reason we can't have nice things, the reason laws are made that are to most people hilariously daft.
AMD have now locked the SoC to 1.3v maximum, you cannot go above that.
 
Last edited:
Soldato
Joined
16 Jun 2005
Posts
14,646
Location
Doncaster
I currently have an i12700k, paired with a 4090 and i've really got the upgrade itch. Is it worth upgrading to a 14th gen intel/7950X3D or should i just wait until the next gen of cpus are released?
 
Back
Top Bottom