• 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"

The way intel's handled this has been shockingly poor. I'm hopeful the media doesn't lose sight of it and keep on them.

At the very least the process for RMA should be:
1. A central link that any consumer can go and download the proper BIOS for their setup
2. Clear instructions to set everything to intel specs
3. A stress test for consumers to run at anytime to check the status of their CPU
4. If the test above fails, send report via one click to a service for an advanced RMA

Anything less than above is inadequate.
 
Robert896r1 said:
The way intel's handled this has been shockingly poor. I'm hopeful the media doesn't lose sight of it and keep on them.

At the very least the process for RMA should be:
1. A central link that any consumer can go and download the proper BIOS for their setup
2. Clear instructions to set everything to intel specs
3. A stress test for consumers to run at anytime to check the status of their CPU
4. If the test above fails, send report via one click to a service for an advanced RMA

Anything less than above is inadequate.
Click to expand...


I like apart from the stress test, the real test is the CPU running at the advertised clocks while doing the tasks the user does, not some random intel benchmark
 
Robert896r1 said:
3. A stress test for consumers to run at anytime to check the status of their CPU
Click to expand...

Doesn't necessarily need a stress test, though may on some CPUs - seems the issue can be exposed by certain operations without putting much load on the CPU. Though not sure if that only applies to CPUs already significantly unstable or very poor out the box or all cases.
 
Robert896r1 said:
The whole point of a stress test is to tease out the issue on a consistent basis across the board….
Click to expand...

No such wrap around software exists. There is plenty of stress tests already and systems that can pass it but fail in something else. For example a server provider recently complained about their 14900k that would pass any stress test they could throw at it and any benchmarks, yet trying to host minecraft which only needs 1 core and high memory would cause crashes
 
Last edited:
Grim5 said:
No such wrap around software exists. There is plenty of stress tests already and systems that can pass it but fail in something else. For example a server provider recently complained about their 14900k that would pass any stress test they could throw at it and any benchmarks, yet trying to host minecraft which only needs 1 core and high memory would cause crashes
Click to expand...

A stress test isn't a product. You can easily create a pre packaged stress test that ranges from anywhere from load spikes on a single cores with sustained and variable workloads, impacting both cache and mem to doing that for all cores or any number of threads in between.
 
Robert896r1 said:
A stress test isn't a product. You can easily create a pre packaged stress test that ranges from anywhere from load spikes on a single cores with sustained and variable workloads, impacting both cache and mem to doing that for all cores or any number of threads in between.
Click to expand...
Of course my speculation on the other thread (and over on AT previously) is that since that "major [American] System Integrator which had the high failure rate at intake QA" are probably not running days of test on each CPU that they have such a tool. It might even be an Intel tool for their OEMs.

If such a tool exists (and it may be a simple cli thing) Intel should package it up and use it for RMA process.

That they haven't yet: could be indifference, or that it is just too widespread. Who knows? We are all left guessing.
 
Robert896r1 said:
A stress test isn't a product. You can easily create a pre packaged stress test that ranges from anywhere from load spikes on a single cores with sustained and variable workloads, impacting both cache and mem to doing that for all cores or any number of threads in between.
Click to expand...
In the reddit thread that the Intel rep posted on, I think they recommended to watch a video that would help you know if you were affected.

It might have been this one?
www.youtube.com

Intel Has Responded! How to STOP Instability and Protect your CPU!

Link to official Press Release - https://community.intel.com/t5/Processors/July-2024-Update-on-Instability-Reports-on-Intel-Core-13th-and/m-p/1617113#M74792C...
www.youtube.com www.youtube.com
 
Robert896r1 said:
The way intel's handled this has been shockingly poor. I'm hopeful the media doesn't lose sight of it and keep on them.

At the very least the process for RMA should be:
1. A central link that any consumer can go and download the proper BIOS for their setup
2. Clear instructions to set everything to intel specs
3. A stress test for consumers to run at anytime to check the status of their CPU
4. If the test above fails, send report via one click to a service for an advanced RMA

Anything less than above is inadequate.
Click to expand...

TBH I think it's been blown out of all proportion. Youtube influences and the press are pretty desperate for clicks, due to a lack of new releases, so need to generate hype when they can. Pooping on Intel is just a good way for them to do this, as it generates much traffic.

I've used a 13900k from release, 0 issues. Currently using it for work, as my 7950X3D is superior in games and uses 150-200W less power (big deal in the Summer!). Both are amazing CPU's, 100% stable (at stock and tuned/OC'd), I've had no drama.
 
I'm certainly interested to see what things look like once the dust settles. IMO if the issue was as big as the limited sources some of these tech media are talking about we'd be seeing forums like this overwhelmed by complaints, even the manufacturer forums where they exist only have a small number of complaints and at least half of those aren't talking about the main symptoms of this issue - though the CPUs can become degraded enough to exhibit more general problems, but mostly it is quite specific out of memory errors (due to an instruction trying to reference 0 for the memory location for data) and specific memory related BSODs (due to trying to execute incorrect memory location for next instruction) which indicate this problem.
 
Last edited:
Dave2150 said:
TBH I think it's been blown out of all proportion. Youtube influences and the press are pretty desperate for clicks, due to a lack of new releases, so need to generate hype when they can. Pooping on Intel is just a good way for them to do this, as it generates much traffic.

I've used a 13900k from release, 0 issues. Currently using it for work, as my 7950X3D is superior in games and uses 150-200W less power (big deal in the Summer!). Both are amazing CPU's, 100% stable (at stock and tuned/OC'd), I've had no drama.
Click to expand...

out of interest have you dialed down the settings now / using intels default settings ? as you are all about benchmarks probably getting lower now
 
Dave2150 said:
TBH I think it's been blown out of all proportion. Youtube influences and the press are pretty desperate for clicks, due to a lack of new releases, so need to generate hype when they can. Pooping on Intel is just a good way for them to do this, as it generates much traffic.

I've used a 13900k from release, 0 issues. Currently using it for work, as my 7950X3D is superior in games and uses 150-200W less power (big deal in the Summer!). Both are amazing CPU's, 100% stable (at stock and tuned/OC'd), I've had no drama.
Click to expand...

Not sure how a sample size of 1 is a definitive statement. I don't have an issue with my 14700k or my 13600k but I know plenty of people who'd had them.

One of the guys in our private group builds custom PC's and he started noticing the deg issue with 13th gen about a year ago and then with 14th gen continued the trend and if anything, it sped up due to higher frequencies and voltage boosts.
Look at the date below. This is based off a sample size of about 400 systems he personally built btw:
image.png
 
uscool said:
out of interest have you dialed down the settings now / using intels default settings ? as you are all about benchmarks probably getting lower now
Click to expand...

If you were running MCE and XMP presets without any manual voltage tuning, you're at most risk for deg issues.

If you manually set your voltages and locked down your voltages, then the chances of deg are non existent.
 
Last edited:
Robert896r1 said:
If you were running MCE and XMP presets without any manual voltage tuning, you're at most risk for deg issues.

If you manually set your voltages and locked down your voltages, then the chances are deg non existent.
Click to expand...

This isn't really clear right now - there seems to be some unexpected voltage behaviour (which also isn't clear if it is the cause of the main issue or just something Intel also found while investigating) which seems to still happen even if you set things manually as the unexpected behaviour underlies the intended algorithm behaviour.
 
Rroff said:
This isn't really clear right now - there seems to be some unexpected voltage behaviour (which also isn't clear if it is the cause of the main issue or just something Intel also found while investigating) which seems to still happen even if you set things manually as the unexpected behaviour underlies the intended algorithm behaviour.
Click to expand...

If you run offset with TVB, then you can subject yourself to the spikes. If you run manual/fixed and set your frequencies without TVB, you don't.

You can control the spikes with even offset + TVB if you run CEP but disabling that in this configuration leaves you vulnerable. But then with the spikes happen and CEP kicks in, your performance tanks....
 
Last edited:
Robert896r1 said:
All of them. (GB/Asus/MSI). Asrock sales are too low for a solid sample size.
Click to expand...

I mean percentage wise - I've had a conversation with someone who does similar and despite 1/3rd more systems built with Asus than Gigabyte and MSI about half of those two there is a big difference in the ratio of number of returned systems exhibiting CPU failure for each of those brands.
 
You must log in or register to reply here.
Back
Top Bottom