Intel has a Pretty Big Problem..

[Rroff]

Meanwhile I've been running a 14700K from launch - 2 years and 2 weeks without a hitch so far, paired up with a 4080 Super it is pretty much indistinguishable to my friends similar 7800X3D setup for 1440p/4K gaming with good quality settings, only slightly slower than the 9800X3D within that context while curb stomping both in most tasks outside of gaming and any difference in heat and power is pretty hard to discern in person (you'd need to be in a small room and running heavy multi-threaded tasks for hours) - in fact when measured over a whole day of average usage the Intel chip wins unless you spend the whole day gaming.

The 9800X3Ds are having their own problems at the moment and actually over the last few months there have been more recorded failure of those chips than the 13th and 14th gen, though how things pan out in the longer run is another story as the 9000 series failures seem largely correlated to specific early batches.

 

[Rroff]

And it may seem like a small thing, but AMD CPUs seem to boot a lot slower, and are slow and janky waking them up from hybernation and sleep states etc.

it's not much in terms of seconds, but it's very noticable to me and it really ruins my 'quality of life' experience on a constant dailly basis.

It can be a significant difference if talking DDR5 and memory context restore, etc. isn't usable and some USB implementations on AMD will also cause longer boot issues (or alternatively a delay in devices being recognised).

 

[mattyfez]

Nothing at all wrong with it, good CPU.

I know, If I had to choose between the 13600k and the 5700x3d, I's choose the intel, I wouldn't even think twice.

 

[humbug]

I know, If I had to choose between the 13600k and the 5700x3d, I's choose the intel, I wouldn't even think twice.

The 5700X3D is a fools CPU, its clocked so low the X3D bit doesn't help it at all, my CPU run's 20% higher clocks and guess what, its at least as good, the 5800X3D sure ok.... but the 5700X3D is just a reject binned 5800X3D with the X3D name to make people think its a special gaming CPU.

 

[mattyfez]

Meanwhile I've been running a 14700K from launch - 2 years and 2 weeks without a hitch so far, paired up with a 4080 Super it is pretty much indistinguishable to my friends similar 7800X3D setup for 1440p/4K gaming with good quality settings, only slightly slower than the 9800X3D within that context while curb stomping both in most tasks outside of gaming and any difference in heat and power is pretty hard to discern in person (you'd need to be in a small room and running heavy multi-threaded tasks for hours) - in fact when measured over a whole day of average usage the Intel chip wins unless you spend the whole day gaming.

The 9800X3Ds are having their own problems at the moment and actually over the last few months there have been more recorded failure of those chips than the 13th and 14th gen, though how things pan out in the longer run is another story as the 9000 series failures seem largely correlated to specific early batches.

JAy2c, linus and tech Jesus arn't whining about it very loudly though, so that means its not a problem? lol

 

[humbug]

JAy2c, linus and tech Jesus arn't whining about it very loudly though, so that means its not a problem? lol

They are, they just aren't blaming the CPU, they correctly identified ASRock pumping the SoC voltage way over recommended specs. Out of all 9800X3D failures 86% of them are on ASRock boards, it was JayZ2Cents who identified ASRock massively overvolting the SoC, later Steve Burke identified the same thing.

Doesn't happen with MSI or Gigabyte who share less than 5% of all failures between them and none with SoC pin burns.

 

[Grim5]

The 9800X3D as far as I know is an AsRock mobo issue, not an issue with the cpu

 

[Rroff]

The 9800X3D as far as I know is an AsRock mobo issue, not an issue with the cpu

The same issue appears to be happening on other boards as well just far far lower frequency, the ASRock boards are just particularly bad for exacerbating the problem and while the rate has slowed down there are still people experiencing those failures who have only ever used the latest BIOS. There are a whole bunch of graphs on Reddit of documented cases and there seems to be a high correlation with batch number - far too much to just be a specific motherboard issue.

 

[humbug]

The same issue appears to be happening on other boards as well just far far lower frequency, the ASRock boards are just particularly bad for exacerbating the problem and while the rate has slowed down there are still people experiencing those failures who have only ever used the latest BIOS. There are a whole bunch of graphs on Reddit of documented cases and there seems to be a high correlation with batch number - far too much to just be a specific motherboard issue.

I've seen those charts, the effected unit counts are recorded from 1 unit to 25 units with no discernible pattern of older vs newer batches being a higher or lower failure rate, when the CPU was made or which batch it came from has no bearing on it. The only commonality is the ASRock boards, 86% of all failures. the remaining 14% is shared between MSI, Asus, BioStar and Gigabyte. Primarily Asus who are in a big lead for 2'nd vs the other 3, probably because of all the stupid OC / overvolting limit unlocks they allow that are supposed to be for LN2 overclocking but we all know a lot of people don't unlock those for that, they think their Noctua D14 it that good with ambient case temp air flow. people who overpay for stuff don't have great critical thinking skills....

There is no data yet about BIOS versions as ASRock themselves have only recently passively admitted the issue by releasing BIOS updates for SoC voltage corrections.

 

[mattyfez]

If I recall correctly Asus got the blame initially for the so called issues with Intel.

Sounds kinda familiar.

 

[mattyfez]

repeat after me:
"it's technically in spec, we just misinterpreted the parameters"

How do I reeach deez kiidz?

 

[jigger]

Honestly i think the 13600K is fine, nothing wrong with it, doesn't seem to have the burnout issues, has good performance, its just a bit pricey still but i think that's because its still Intel's best seller.

Honestly I think you’re wrong and here’s why. The performance is meh, power consumption is poor, motherboard platform is on a super crapola level and the CPU failure rate is ridiculous. The 13600 didn’t even make a value argument.

The 13/14th are hot garbage.

 

[humbug]

If I recall correctly Asus got the blame initially for the so called issues with Intel.

Sounds kinda familiar.

mmm... no Nvidia, Nvidia got the blame initially for the very real issues with Intel, early signs of a problems with these CPU's was an "Out of Video Memory" error on Nvidia GPU's, Nvidia soon piped up and said the problem was with Intel CPU's, they were right.

 

[Rroff]

Honestly I think you’re wrong and here’s why. The performance is meh, power consumption is poor, motherboard platform is on a super crapola level and the CPU failure rate is ridiculous. The 13600 didn’t even make a value argument.

The 13/14th are hot garbage.

You do talk some rubbish, the performance is fine, yes power consumption isn't the best but it isn't as terrible as the reviews make out because people aren't sitting there doing benchmarks 24x7 and whole system power consumption at the wall is far less dramatically different than the difference at the CPU socket, the failure rate is not ridiculous* and there are hardly any 13600s failing due to degradation, I don't actually know of a single documented case, and aside from poor PCI-e 5.0 provisioning there is nothing wrong with the motherboard platform (though it would be nice if the USB4 support was less clunky). You are insulting people's intelligence trying to push this as reality.


* At least as of yet, the jury is still out on the longer term which does put a cloud over the platform.

 

[jigger]

Meanwhile I've been running a 14700K from launch - 2 years and 2 weeks without a hitch so far, paired up with a 4080 Super it is pretty much indistinguishable to my friends similar 7800X3D setup for 1440p/4K gaming with good quality settings, only slightly slower than the 9800X3D within that context while curb stomping both in most tasks outside of gaming and any difference in heat and power is pretty hard to discern in person (you'd need to be in a small room and running heavy multi-threaded tasks for hours) - in fact when measured over a whole day of average usage the Intel chip wins unless you spend the whole day gaming.

The 9800X3Ds are having their own problems at the moment and actually over the last few months there have been more recorded failure of those chips than the 13th and 14th gen, though how things pan out in the longer run is another story as the 9000 series failures seem largely correlated to specific early batches.

Speaking of rubbish, can we see your data for the failures?

I bought a 7800X3D and Asus ROG murder board after everyone claimed there was a fault and the chip would fail. That system is still running just fine on the OG murder BIOS. The 13900 system I bought however, has had two RMA replacements and the last refused. I would request a new RMA now Intel has admitted the faults, but I CBA dealing with a hot running system that will almost certainly become flakey over time and fail again. I now have a 12 core Zen in place and will likely upgrade to a dual X3D part once released.

There is zero question to which system is better between Intel and AMD right now. AMD are in another league in every metric, pretty much across the board. Intel needs to massively improve. The thread title remains accurate

 

[Rroff]

Speaking of rubbish, can we see your data for the failures?

I've linked to loads of data in past posts, not my problem if you refuse to take it onboard.

Meanwhile where is the evidence of these ridiculous failure rates? these forums would be full of people posting about their problems if the failure level was even 20% never mind the claims of 50-100% that mysteriously GN, etc. were unable to verify.

 

[jigger]

I've linked to loads of data in past posts, not my problem if you refuse to take it onboard.

Meanwhile where is the evidence of these ridiculous failure rates? these forums would be full of people posting about their problems if the failure level was even 20% never mind the claims of 50-100% that mysteriously GN, etc. were unable to verify.

You claimed you have some level of access to Intel returns data (that disproves the issues Intel have already admitted to) but have never produced any.

The fact is, failure rates are ridiculous and Intels desktop platform is crap.

 

[Rroff]

You claimed you have some level of access to Intel returns data (that disproves the issues Intel have already admitted to) but have never produced any.

The fact is, failure rates are ridiculous and Intels desktop platform is crap.

No I didn't and you know it - I mentioned I have access to returns data for some UK parts suppliers through work.

There are no facts supporting failure rates are ridiculous, just some unsubstantiated claims by unverified sources, while the most credible source for the scale of it is likely the statements put out by RAD Games Tools who by the nature of what they do are pretty tapped into it.

 

[jigger]

No I didn't and you know it - I mentioned I have access to returns data for some UK parts suppliers through work.

There are no facts supporting failure rates are ridiculous, just some unsubstantiated claims by unverified sources, while the most credible source for the scale of it is likely the statements put out by RAD Games Tools who by the nature of what they do are pretty tapped into it.

You know you’ve just contradicted yourself in the same paragraph, right? Your argument was (in the face of all the evidence to the contrary) the claims made in the media were false and you, yourself had access to data that would go some way to disproved the many articles. You even doubled down and claimed to data that shown AMD had an issue.

Are you going to provide the evidence or just continue to argue in face of the facts?

 

[Rroff]

You know you’ve just contradicted yourself in the same paragraph, right? Your argument was (in face all the evidence to contrary) the claims made in the media was false and you, yourself had access to data that would go some way to disproved the many articles. You even double down and claimed to data that shown AMD had an issue.

Are you going to provide the evidence or just continue to argue in face of the facts?

Where is the contradiction? as I said I don't have access to Intel's return data but I have enough insight into some retailer's returns data to have an idea of the scale of the problem - it would certainly show up there if it was remotely close to the 20, 50, 100% failure rates some claim.

I also pointed out that the claims in the media made by certain parties of ridiculously high failure rates weren't able to be verified by the likes of GN while we have claims by a substantial source who have a verifiable industry presence which are likely more accurate.

There are no facts I'm arguing against - I've even repeatedly linked to facts which don't support the claims you keep referring to.

No idea what you are referring to with the "You even double down and claimed to data that shown AMD had an issue." comment.

EDIT: Unless you are referring to:

Return rates at retailers who publish details vary between 0.56% and 1.1% for the 9800X3D, I know one retailer has nearly 300 pieces (out of well into the 10s of thousands sold) returned with the majority having symptoms consistent with this failure.