Question DEGRADING Raptor lake CPUs

Page 15 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Kocicak

Golden Member
Jan 17, 2019
1,067
1,122
136
I noticed some reports about degrading i9 13900K and KF processors.

I experienced this problem myself, when I ran it at 6 GHz, light load (3 threads of Cinebench), at acceptable temperature and non extreme voltage. After only few minutes it crashed, and then it could not run even at stock setting without bumping the voltage a bit.

I was thinking about the cause for this and I believe the problem is, that people do not appreciate, how high these frequencies are and that the real comfortable frequency limit of these CPUs is probably at something like 5500 or 5600 MHz. These CPUs are made on a same process (possibly improved somehow) on which Alder lake CPUs were made. See the frequencies 12900KS runs at. The frequency improvement of the new process tweak may not be so high as some people presume.

Those 13900K CPUs are probably highly binned to be able to find those which contain some cores which can reliably run at 5800 MHz. Some of the 13900K probably have little/no OC reserve left and pushing them will cause them to degrade/break.

The conclusion for me is that the best you can do to your 13900K or 13900KF is to disable the 5800 MHz peak, which will allow you to offset the voltage lower, and then set all core maximal frequency to some comfortable level, I guess the maximum level could be 5600 MHz. With lowered voltage this frequency should be gentler to the processor than running it at original 5500 MHz at higher voltage. You can also run it at lower frequencies, allowing for even higher voltage drop, but then the CPU is slowly loosing its sense (unless you want some high efficiency CPU intended for heavy multithread loads).

Running it with some power consumption limit dependent on your cooling solution to keep the CPU at sensible temperature will help too for sure.
 
Last edited:

Hulk

Diamond Member
Oct 9, 1999
4,455
2,373
136
You're assuming that the performance impact of the fix will be noticeable. I wouldn't assume that. There are several possibilities I see for how Intel was delivering "too much voltage".

1- it was purely a bug, the tables delivered more voltage than was needed to achieve the desired frequency in certain performance modes, so the fix will not change the CPU's performance at all
2- they were overly aggressive with boosting voltage and frequency during turbo, the fix would have a minor impact in turbo (and might affect only ST, only MT, or both) but not make any difference in sustained performance
3- they were too aggressive with base frequency and pulling back the voltage there will affect both turbo and sustained performance

#3 is the only one that would make any real difference other than "oh no my GB6 score is a bit lower" (because GB6 pauses between subtests it allows CPUs to run at turbo for longer than a sustained benchmark like SPEC) but I view #3 as by far the least likely. I'd say 25% chance it is #1, 70% chance it is #2, 5% chance it is #3.

That all said, patches and other system changes increase or decrease performance ALL THE TIME, so a lawsuit on that is not going to be successful. If performance dropped by 20% or something then you might have had a case, but if performance drops by a single digit amount (and I'm betting on low to mid single digits, with low being the more likely and a 25% chance it is zero) that case isn't going to win in court. Too easy to point to examples of the exact same system getting different performance depending on where it is used (i.e. ambient temperatures in the house) or different BIOS options like power savings or DRAM timings.
Possibly but I've owned a couple of these and both require a lot of volts to reach specified frequency. Also check around at overlockers.net and you'll see lots of volts required to hit all-core and 1/2 boost core frequency.

Hit them with Stockfish or something compute rigorous like that and they fail without the voltage. Give them the volts and they fail a few weeks later.

We shall see how this goes for Intel.
 

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
26,061
15,199
136
Possibly but I've owned a couple of these and both require a lot of volts to reach specified frequency. Also check around at overlockers.net and you'll see lots of volts required to hit all-core and 1/2 boost core frequency.

Hit them with Stockfish or something compute rigorous like that and they fail without the voltage. Give them the volts and they fail a few weeks later.

We shall see how this goes for Intel.
What this means, is that you CAN'T hit stock frequency and have the system be stable. You have to run it underclocked and under-volted or its not stable ! Thats the whole point of this thread ! And its gets worse even under-volted, as its already damaged for life ! Its replacement must be started undervolted and underclockd.
 

Hulk

Diamond Member
Oct 9, 1999
4,455
2,373
136
What this means, is that you CAN'T hit stock frequency and have the system be stable. You have to run it underclocked and under-volted or its not stable ! Thats the whole point of this thread ! And its gets worse even under-volted, as its already damaged for life ! Its replacement must be started undervolted and underclockd.
Well, there are other options.
Run stock/auto and run the very possible risk of degradation.
Run lower volts with a custom loop or sub-ambient cooling and get performance and longevity but you need expensive and complicated cooling.
Get a golden sample that will work at sane volts and cooling and will last.

Or do what I did and give up 1/2 core boost, 100MHz all-core frequency (5.5 vs. 5.6) and HT. Runs on air fine and dandy and only draws 1.15V under load, generally about 175W max. Honestly really not giving up anything in day-to-day performance worth worrying about.
 

bigboxes

Lifer
Apr 6, 2002
40,253
12,219
146
Ahead of the launch, AMD found that “the initial production units that were shipped to our channel partners did not meet our full quality expectations."

Obviously they are not revising the architecture in two weeks so it makes sense that this is a binning/quality issue.

I think AMD did some additional quality testing in light of the ongoing Intel fiasco and reflected, "hey there are some niche situations where we might have similar problems." They are recalling the parts and re-binning to assure quality. Good on AMD.
I read where AMD had the wrong lithograph on the heatspreaders. Ryzen 9 instead of Ryzen 7 for the 9700X and Ryzen 5 for 9600X.
 

winr

Diamond Member
Feb 17, 2001
6,081
56
91
This has no contribution to to this thread but peaked my curiosity

Voltage, current, heat

I have a Phenom X II dual core 555 3.2 running at 4000 with stock voltage ( 1.4 ), never shut it off ( max watts 80 )

It is still running at the moment and I use it browse when I am not using my Core i7 2600 which is the newest PC I have built .. lol .....

Built it around 12 or 13 years ago, big cooler, used to play games that maxed the cpu out

I read the posts but do not understand CPU technology and such, could someone dumb it down for me perhaps ??


Ricky.
 
Reactions: lightmanek
Jul 27, 2020
19,613
13,477
146
I read the posts but do not understand CPU technology and such, could someone dumb it down for me perhaps ??
It's a lot more complicated to discuss it here but here's a quick (sort of) review:

CPUs are made up of units called transistors (well, generally. There are other things too but let's ignore them for now).

Transistors allow current flow in one direction but try to pass current in the other direction, and the transistor will block it. The state of current flowing is denoted as a 1 or true condition while blocking is termed as false or has value 0. Using some fun math called Boolean Algebra that only cares about ones and zeroes, scientists made up logic circuits with transistors. At first, these were primitive and could do limited stuff. Then someone discovered that transistors could be shrunk. So suddenly, a huge set of transistors doing something cool could be shrunk half in size, for example, or you could fit twice the transistor count in the same space.

This increased complexity allowed geniuses to design better and faster computers. As the transistor shrinks, it needs less voltage which allows it to run cooler. So smaller transistors are great for creating cooler running chips. However, the smaller the transistor, the barrier that stops the current from flowing in the false or zero state also shrinks. So they had to come up with all sorts of tricks to ensure that the barrier functioned adequately. But suppose a transistor that can tolerate 1.1V is fed 1.15V, it might hold up for a while. But sooner or later, the barrier will break down due to the tolerance level of the transistor being exceeded.

This is what seems to have happened with Intel 7 Ultra process technology (gross oversimplification though). They weren't able to figure out properly what voltage the transistor could tolerate happily. They made some miscalculations from limited data or they decided to tempt fate knowing full well that things would go bad for them. And so here we are. Intel transistors prematurely breaking down due to more volts, more heat, more current than they were designed to tolerate.
 

DrMrLordX

Lifer
Apr 27, 2000
22,000
11,560
136
This has no contribution to to this thread but peaked my curiosity

Voltage, current, heat

I have a Phenom X II dual core 555 3.2 running at 4000 with stock voltage ( 1.4 ), never shut it off ( max watts 80 )

It is still running at the moment and I use it browse when I am not using my Core i7 2600 which is the newest PC I have built .. lol .....

Built it around 12 or 13 years ago, big cooler, used to play games that maxed the cpu out

I read the posts but do not understand CPU technology and such, could someone dumb it down for me perhaps ??


Ricky.

Honestly you should make your own thread for that.
 

lakedude

Platinum Member
Mar 14, 2009
2,679
479
126
I read the posts but do not understand CPU technology and such, could someone dumb it down for me perhaps ??
Your Phenom has 758 million transistors which is a lot for sure but a fancy Raptor Lake (the CPUs Intel is having trouble with) has something like 26 billion transistors. The transistors are just tiny and it does not take much to kill something that small. They are just not as durable as they were.
 

DZero

Member
Jun 20, 2024
89
47
51
While I accept that the K versions are having the issue, makes me think... what the hell the non K got affected to badly?
And makes me think... maybe the lower tiers would be on eventual risk too?
 
Reactions: winr

Kocicak

Golden Member
Jan 17, 2019
1,067
1,122
136
While I accept that the K versions are having the issue, makes me think... what the hell the non K got affected to badly?
Because the frequency should be lower by 600-1000 MHz to reach long term stability, and not by 100 or 200 MHz, as the non-K models have?
 
Reactions: winr

Kocicak

Golden Member
Jan 17, 2019
1,067
1,122
136
...
IMO the most practical scenario is to release "new" CPUs with lower frequencies that will last a reasonable amount of time at a reasonable usage intensity. For example lasting 5 years at 100% load 6 hours a day under a 6 heatpipe air cooler. I personally really do not expect more from a consumer CPU.

Then replace all original CPUs regardless if they are failing or not with these new CPUs.
...
I have been thinking about this and there is no point in replacing the CPUs because the collosal amount of work, just provide the new slower and stable CPUs to the customers for free that prove that they bought one of the affected CPUs, they can then decide if they run their by chance still functioning original CPUs and wait untill they start failing, or just replace their CPUs right away as a precautionary measure.
 

DrMrLordX

Lifer
Apr 27, 2000
22,000
11,560
136
While I accept that the K versions are having the issue, makes me think... what the hell the non K got affected to badly?
And makes me think... maybe the lower tiers would be on eventual risk too?

Potentially anything that is Raptor Lake (as opposed to an Alder Lake refresh) is vulnerable.
 

Hitman928

Diamond Member
Apr 15, 2012
6,049
10,379
136
While I accept that the K versions are having the issue, makes me think... what the hell the non K got affected to badly?
And makes me think... maybe the lower tiers would be on eventual risk too?

Despite the willingness to jump on all RPL products, Intel didn’t say the lower tier were affected, but that they might be. Now, that could be because they just don’t want to admit to it yet, or that they really aren’t sure because they aren’t confident that they’ve actually root caused the issue yet. I’m leaning towards the latter at the moment but time will tell. So far, the data we have suggests they aren’t affected, or at least to the point of failing within a short time frame.
 

DZero

Member
Jun 20, 2024
89
47
51
Despite the willingness to jump on all RPL products, Intel didn’t say the lower tier were affected, but that they might be. Now, that could be because they just don’t want to admit to it yet, or that they really aren’t sure because they aren’t confident that they’ve actually root caused the issue yet. I’m leaning towards the latter at the moment but time will tell. So far, the data we have suggests they aren’t affected, or at least to the point of failing within a short time frame.
Indeed, while before I tought that non K laptop products are safe, now due constrained voltages, made me think if those are screwed too.
 

moinmoin

Diamond Member
Jun 1, 2017
5,063
8,025
136
So far the only mobile Raptor Lakes strongly suspected of degradation are the ones using the 8+16 die (HX CPUs).
I think it would be easier to state that there is only one die specific to Raptor Lake, that is the 8+16 die. All other dies are Alder Lake Refresh dies and shouldn't be affected.

Probably not easy since on desktop Intel really obfuscates what chips use the Raptor Lake die, which a Alder Lake Refresh die, and which a random pick of those two. At least with mobile HX does cover all cases.
 
Reactions: Tlh97 and DZero

DZero

Member
Jun 20, 2024
89
47
51
I think it would be easier to state that there is only one die specific to Raptor Lake, that is the 8+16 die. All other dies are Alder Lake Refresh dies and shouldn't be affected.

Probably not easy since on desktop Intel really obfuscates what chips use the Raptor Lake die, which a Alder Lake Refresh die, and which a random pick of those two. At least with mobile HX does cover all cases.
Made me think if first was Alder Lake ones but then switches to Raptor without knowing with the microcode error that comes with it.
 

DrMrLordX

Lifer
Apr 27, 2000
22,000
11,560
136
I think it would be easier to state that there is only one die specific to Raptor Lake, that is the 8+16 die. All other dies are Alder Lake Refresh dies and shouldn't be affected.
Yeah but Intel's mobile lineup is so confusing, sometimes I forget if they ever bothered with a 6+8 Raptor Lake die.
 
Reactions: lightmanek

ikjadoon

Senior member
Sep 4, 2006
205
445
146
Cross-posting to the 2nd thread:

Intel is now adding +2 years of warranty coverage for all 13th and 14th gen CPUs.

KS goes from 1yr to 3yr.
All others go from 3yr to 5yr.
OEM owners need to contact their OEM and...see what happens?


More details coming later.

This scandal will be remembered for a decade.
 

bigboxes

Lifer
Apr 6, 2002
40,253
12,219
146
Cross-posting to the 2nd thread:

Intel is now adding +2 years of warranty coverage for all 13th and 14th gen CPUs.

KS goes from 1yr to 3yr.
All others go from 3yr to 5yr.
OEM owners need to contact their OEM and...see what happens?


More details coming later.

This scandal will be remembered for a decade.
Yes, that is better, but I'd expect that my CPUs last longer than 5 years of usage. So, whatever actual SKUs this ultimately affects it's not a matter of if, but when. Intel should really just replace them all. Like Toyota replacing all those engines. It was the right thing to do. With my Camry, Toyota put the onus on me to prove it was burning too much oil before the extended warranty deadline before rebuilding the engine. It started burning too much just as the extended warranty expired.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |