News Dev reports Intel's laptop CPUs are also suffering from crashing issues — several laptops have suffered similar failures in testing

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.
My i5-3570k still works after all these years. The P8Z77-I deluxe mobo still mostly works, minus a few usb ports.
Usually the mobo dies before the CPU ever reaches its mtbf.
 
  • Like
Reactions: Amdlova
My i5-3570k still works after all these years. The P8Z77-I deluxe mobo still mostly works, minus a few usb ports.
Usually the mobo dies before the CPU ever reaches its mtbf.
Oh, if we're talking about old machines still working, I recently removed a Sandybridge i7-2600K from service. The only things replaced on it were the PSU, a case fan (both for getting noisy), the RAM & SSD (both for more capacity). The CPU and motherboard are still original. The motherboard is a Genuine Intel, BTW. I'm half-embarrassed to admit that it was running the stock CPU cooler, the entire time.
 
Intel best cpus range from sandy bridge to broadwell... before you can see the cpus dying from age... I have here my 2500k still working after all these years. I think I will sell my 13600t and get a 12900k :) (will be cheaper because all this Intel drama).
12900k at 50w? Anyone?
 
Intel best cpus range from sandy bridge to broadwell... before you can see the cpus dying from age... I have here my 2500k still working after all these years. I think I will sell my 13600t and get a 12900k :) (will be cheaper because all this Intel drama).
12900k at 50w? Anyone?
Im running my 12900k at 50w. Incredible CPU

image-2024-07-20-092002778.png
 
  • Like
Reactions: Amdlova
I think I will sell my 13600t and get a 12900k :) (will be cheaper because all this Intel drama).
12900k at 50w? Anyone?
My desktop at work has a regular i9-12900 (non-K). I actually boosted the PL1 to 80W (the default is 65W), because that improves my build times measurably (I'll have to go back and check how much) and it's still low enough that the temperature plateaus below the throttling point. It helps that it lives in an air conditioned server room.

If you want to tame that beast, I'd recommend tamping down the PL2. If you do that (or raise PL1), it'll actually boost for longer without having to touch Tau. It all has to do with how the EWMA algorithm works - that's what governs turbo, on these CPUs.
 
  • Like
Reactions: Amdlova
However, one has to wonder what went wrong for Intel to miss the mark so badly. Maybe merely having such a target isn't the level of assurance I thought it was, if they're not doing all the necessary diligence to make sure they're hitting it.
Intel chips infamously run hotter, are designed to use more power, so they are that much more justified in focusing on the issue.

I presume it is the copper conductors that paralyzed Intels 10nm projects for years and may still be coming back to bite them, but the physics gets more challenging to any and all processes as the geometry shrinks.
 
  • Like
Reactions: bit_user
I presume it is the copper conductors that paralyzed Intels 10nm projects for years and may still be coming back to bite them, but the physics gets more challenging to any and all processes as the geometry shrinks.
Based on what I've seen about Intel 4, I believe they retained the use of cobalt throughout the Intel 7 series of nodes. This page has a pretty good overview of the issue, as well as a discussion of what they changed in the Intel 4 node:


Perhaps the optimized version of Intel 7, used in Raptor Lake, tinkered with that aspect.
 
Whatever correlation people have come up with is definitely not the situation given official statements going around: https://www.tomshardware.com/pc-com...er-blames-common-software-and-hardware-issues
That's what Intel says, but we should point out that the founder of Alderon Games directly contradicted them, as noted in that article:

"The laptops crash in the exact same way as the desktop parts including workloads under Unreal Engine, decompression, ycruncher or similar. Laptop chips we have seen failing include but not limited to 13900HX etc.

Intel seems to be down playing the issues here most likely due to the expensive costs related to BGA rework and possible harm to OEMs and Partners," he continued. "We have seen these crashes on Razer, MSI, Asus Laptops and similar used by developers in our studio to work on the game. The crash reporting data for my game shows a huge amount of laptops that could be having issues."

Given that, I wouldn't be so quick to brush aside concerns about the HX-series Gen 13 & 14 CPUs.
 
That's what Intel says, but we should point out that the founder of Alderon Games directly contradicted them, as noted in that article:
"The laptops crash in the exact same way as the desktop parts including workloads under Unreal Engine, decompression, ycruncher or similar. Laptop chips we have seen failing include but not limited to 13900HX etc.​
Intel seems to be down playing the issues here most likely due to the expensive costs related to BGA rework and possible harm to OEMs and Partners," he continued. "We have seen these crashes on Razer, MSI, Asus Laptops and similar used by developers in our studio to work on the game. The crash reporting data for my game shows a huge amount of laptops that could be having issues."​

Given that, I wouldn't be so quick to brush aside concerns about the HX-series Gen 13 & 14 CPUs.
The implication there is that they're not just the HX parts having problems.
 
  • Like
Reactions: bit_user
Nope. Here's an entire article on the differences:

They changed:
  • the manufacturing process.
  • the cache sizes, parameters, policies, and algorithms.
  • optimized critical paths in the P-cores to make tighter timing constraints
  • the number of E-core clusters
  • the memory controller

Taken together, those add up to a proper generational increase, performance-wise.
"Intel says that compared to 12th Generation Core, formerly codename Alder Lake, you can expect to see up to 15% better single-thread performance. The majority (2/3 or 67%) of the single-thread performance comes from frequency improvement, 14% from memory improvements, and 20% from cache changes."​

So, even the single-threaded performance increase isn't purely from higher frequencies. According to that, IPC should be about 5% higher, for cache/memory-intensive workloads.
I don't think these chips saw anything more then a 2-3% IPC improvement. Intel marketing likes to count "performance gain" and call it IPC improvement. yes there was about a 15% improvement gain, almost all of it from increased clock speeds. If you normallize 12th and 13th gen to 5ghz the performance gains are negligable, around 2-3% and most of the time so minimal as to be within the margin of error.

I don't debate the "changes" made from 12th to 13th just like i don't argue about the changes from athlonx64 to phenom, but like with that old athlon to phenom change the only changes in performance came from clock speeds not from ipc.
 
Status
Not open for further replies.