So are all Intel i5/i7/i9 13th/14th gen chips effectively defective?

[Iggy Wolf]

Based on some of the articles I'm recently reading, there's now talk of a class action lawsuit due to the instability of the Raptor Lake chips apparently being a manufacturing defect that's gonna result in most of them failing within the next 3-4 years. I mean, we're not talking about just a few gamer users here.

Entire companies and data centers probably have those chips. And I doubt they're just gonna take it lying down or let Intel off the hook that easily. It's crazy to think they could detect this issue earlier. People's PC's weren't BSODing or having the CPUs crash just because.

[r.linder]

Intel probably isn't telling the whole truth because they've already been caught lying and hiding the truth several times in this one isolated incident.

[Illusion of Progress]

Bad 💀 Motha původně napsal:

This is just nuts. I'm going to have tons of really PO'ed customers. Some of them don't want to ever hear the name "AMD"

Then that's on them. If anyone is sitting there with a product and encountering instability issues and refusing to consider alternatives, possibly because "it's more problematic/doesn't work as good", the irony of things would probably be lost on them and they shouldn't be taken seriously, especially if they are just parroting that FUD and never tried it for themselves, which is probably a fair share of them.

If you configured a system for someone who wanted Intel, and it has problems because of Intel, that's not on you. That's on Intel, and if the user continues to want Intel after having that experience, it's on both Intel and the user then. It's not on you. You probably already know this, but... don't let people walk all over you with blame when they chose that.

Bad 💀 Motha původně napsal:

So what is so different with 13th and 14th Gen compared to 12th Gen? I know the specs differences and such but maybe it comes down to a design flaw with 13/14th Gen I guess is what this is, huh?

That's been the million dollar question.

Some of the changes between Alder Lake and Raptor Lake are the following...

Core clocks are some hundreds of MHz higher.
Cache capacity is a bit higher.
There's more e-cores (12700 has 4, 13700 has 8, 14700 has 12, but the Core i7 was the only place the 14th generation saw further increases here)
The ring bus is clocked higher (apparently ~1 GHz higher from 3.6 GHz to 4.6 GHz, but I'm finding some conflicting information on exactly how much higher it is so maybe it depends on model, but it's way higher on average regardless).
There's a different manufacturing process (?)

And that's probably just the surface level differences.

[Crashed]

Bad 💀 Motha původně napsal:

This is just nuts. I'm going to have tons of really PO'ed customers. Some of them don't want to ever hear the name "AMD"

So what is so different with 13th and 14th Gen compared to 12th Gen? I know the specs differences and such but maybe it comes down to a design flaw with 13/14th Gen I guess is what this is, huh?

Sorry I don't already know more or have been caught up. This Summer in US has been hectic and I'm trying to get some real-life stuff outside of Tech all in order.

Anyways I appreciate the help and those who fact-check instead of just listening to these YouTube folk who think they know everything. What's his name from Gamer-Nexus didn't even know how a CPU was made until not too long ago. How can he and young folks like him expect to spend a couple days on this kind of issue and have that be the whole truth once they make a couple videos. It's fine that they can be informative and wave the red flag when needed to the general public but I seriously doubt any of these Tech folks on YouTube who have their hands full with dozens of other things, have enough people or time to dedicate to such an issue and it's need for finding truth. Too many of them are spreading fear and doubt, all via rumors.

Intel pushes a new more aggressive generation on the same process node. Traditionally "refresh" generations were on next generation process nodes (die shrink) which provided a boost in efficiency and performance headroom, but 13th and 14th Gen decided to just be hopeful in their capabilities despite it causing the processors to slowly burn themselves out.

Some are saying AMD's Zen 5 is underwhelming but they decided to prioritize efficiency over performance gains in their new generation.

[Bad 💀 Motha]

Illusion of Progress původně napsal:

Bad 💀 Motha původně napsal:

This is just nuts. I'm going to have tons of really PO'ed customers. Some of them don't want to ever hear the name "AMD"

Then that's on them. If anyone is sitting there with a product and encountering instability issues and refusing to consider alternatives, possibly because "it's more problematic/doesn't work as good", the irony of things would probably be lost on them and they shouldn't be taken seriously, especially if they are just parroting that FUD and never tried it for themselves, which is probably a fair share of them.

If you configured a system for someone who wanted Intel, and it has problems because of Intel, that's not on you. That's on Intel, and if the user continues to want Intel after having that experience, it's on both Intel and the user then. It's not on you. You probably already know this, but... don't let people walk all over you with blame when they chose that.

Bad 💀 Motha původně napsal:

So what is so different with 13th and 14th Gen compared to 12th Gen? I know the specs differences and such but maybe it comes down to a design flaw with 13/14th Gen I guess is what this is, huh?

That's been the million dollar question.

Some of the changes between Alder Lake and Raptor Lake are the following...

Core clocks are some hundreds of MHz higher.
Cache capacity is a bit higher.
There's more e-cores (12700 has 4, 13700 has 8, 14700 has 12, but the Core i7 was the only place the 14th generation saw further increases here)
The ring bus is clocked higher (apparently ~1 GHz higher from 3.6 GHz to 4.6 GHz, but I'm finding some conflicting information on exactly how much higher it is so maybe it depends on model, but it's way higher on average regardless).
There's a different manufacturing process (?)

And that's probably just the surface level differences.

Probably because that Bus might scale based on installed CPU. That's how it worked on AMD AM2 and AM3

[kingjames488]

well I'm still pretty sceptical about the whole thing... and I'm sure if I do something to my PC before intel tells me to they'll try to blame me if in fact I have a faulty CPU...

I'm really hoping my issue is just a software glitch, but it's kinda the sort of thing that lines up.

[Crashed]

kingjames488 původně napsal:

well I'm still pretty sceptical about the whole thing... and I'm sure if I do something to my PC before intel tells me to they'll try to blame me if in fact I have a faulty CPU...

I'm really hoping my issue is just a software glitch, but it's kinda the sort of thing that lines up.

Run HWiNFO64 and let us know what number you get for MCU. This is the currently loaded Microcode revision. MCU 125 fixes eTVB instability, and MCU 129 fixes overvoltages.

[Fandabidozi]

So ASUS have released BIOS updates for non K chips with the new Intel micro-code with recommended settings. I haven't had any issues with my chip but thought I'd try it out of interest. Upon running Intel default setting my chip actually operates at a higher voltage than with the ASUS OC profile. Intel voltage is 1.32v and ASUS voltage is 1.28v. The funny thing is that with Intel settings The TDP P1 is capped at 65w where as the ASUS P1 is capped at 254w. I think I know what profile I'll be using.

[C1REX]

I’ve checked my 7800x3D and I don’t see anything above 1.1v with default settings and no EXPO.
Expo puts 1.24v on SoC that includes memory controller. That 1.24v is the highest. Everything else remain below 1.1v. Closer to 1.01v actually.

I’m worried that current voltages on Intel CPUs are still dangerously high.

[kingjames488]

well I'm going to pull down the power limit on my CPU further til this is figured out...

these single core temps are all over the place and I expect the issue has to do with the auto boost boosting the single core performance of the chip and then over-loading it with another task without reducing the clock...

cuz really all of these boost things are reactive, and the only way they know to "boost down" is if they get over-loaded... kinda why I've always hated these auto-boost things.

[r.linder]

Boost has been all over the place since Coffee Lake, this issue isn't exactly new, it just didn't take such a bad turn until now

[kingjames488]

r.linder původně napsal:

Boost has been all over the place since Coffee Lake, this issue isn't exactly new, it just didn't take such a bad turn until now

even before that in GPUs... and they just do w.e they want since then... can't even overclock anything anymore cuz it just clocks its self down when it hits it's temp limits.

if one thing is clear to me it's that intel messed up... trying to push their chips too hard.

[Floid]

spaghettirella man původně napsal:

if one thing is clear to me it's that intel messed up... trying to push their chips too hard.

13th & 14th gen have proven to be problematic. But it's nothing compared to what's potentially in store for us with 15th gen Arrow Lake or beyond - with multi-layering.

If Intel pushes on with that, it effectively amounts to what is a research based approach to production.

And going back to the drawing broad on cup's doesn't sound fun at all :(

[Bad 💀 Motha]

r.linder původně napsal:

Boost has been all over the place since Coffee Lake, this issue isn't exactly new, it just didn't take such a bad turn until now

I can't remember the last timeI just left Intel Turbo Boost enabled. I always used to just disable that and then OC the Base to what the Turbo Boost would have been, sometimes higher if it can handle it.

I never liked or have a use for Turbo or Core Parking. I could give a crap about overall power usage, again as long as it's not prematurely killing the Motherboard or CPU and the thermals are all good.

Turning all that turbo, core parking and power saving crap off was the only way I could OC CPUs like 8350 to 4.4Ghz or higher. I've done that on quite a few and they ran that way for years no problems.

[r.linder]

Bad 💀 Motha původně napsal:

r.linder původně napsal:

Boost has been all over the place since Coffee Lake, this issue isn't exactly new, it just didn't take such a bad turn until now

I can't remember the last timeI just left Intel Turbo Boost enabled. I always used to just disable that and then OC the Base to what the Turbo Boost would have been, sometimes higher if it can handle it.

I never liked or have a use for Turbo or Core Parking. I could give a crap about overall power usage, again as long as it's not prematurely killing the Motherboard or CPU and the thermals are all good.

Turning all that turbo, core parking and power saving crap off was the only way I could OC CPUs like 8350 to 4.4Ghz or higher. I've done that on quite a few and they ran that way for years no problems.

I just manual OC at whatever runs stable and cool with more reasonable voltage, default settings for motherboards have been causing CPUs to go above Intel guidelines for years, it's only an issue now because it became a perfect storm of bad luck.

[🦊Λℚ𝓤ΛƑΛᗯҜᔕ🦊]

Bad 💀 Motha původně napsal:

r.linder původně napsal:

Boost has been all over the place since Coffee Lake, this issue isn't exactly new, it just didn't take such a bad turn until now

I can't remember the last timeI just left Intel Turbo Boost enabled. I always used to just disable that and then OC the Base to what the Turbo Boost would have been, sometimes higher if it can handle it.

I never liked or have a use for Turbo or Core Parking. I could give a crap about overall power usage, again as long as it's not prematurely killing the Motherboard or CPU and the thermals are all good.

Turning all that turbo, core parking and power saving crap off was the only way I could OC CPUs like 8350 to 4.4Ghz or higher. I've done that on quite a few and they ran that way for years no problems.

We're not supposed to do this with modern processors from either Intel or AMD. The reason why is because not all cores on the processor can all hit the same maximum speed. You may have 2 out of 8 cores for example that could do 5.8 Ghz while the other 6 cores could only do 5.4 Ghz (again this is just a theoretical example to get my point across). In that scenario if you were matching all cores to the same manual overclock then you would have all cores limited to 5.4 ghz and those 2 faster cores would never be able to hit their maximum peak that they were capable of.

If you use either AMD or Intel in automatic mode (Turbo Boost on for Intel) then when you play lightly threaded games like single-threaded games the fastest cores on the processor will boost up to their maximum speed, often 5.8 - 6.0 Ghz on 13900K and 14900K.

Manual all-core overclocking pretty much ended years ago. That is the completely wrong way to overclock any modern processor.