Asus issues BIOS update to solve Intel CPU game crashing.

[humbug]

As a lot of you may already know there are reports of games crashing with the 13900K and 14900K, it seems to be specifically with Nvidia GPU's, Nvidia have publically thrown Intel under the bus by naming them as the culprit for an "out of video memory" GPU crash error on their GPU's. lol... "Out of video memory" i can see why they were so quick to throw Intel under the bus.
It makes sense as Nvidia drivers rely on the CPU to a much greater extent than for example AMD.

With the background on this given. Yes Asus strikes again, this is not an Intel problem, as such, Intel CPU's come with a TDP "Thermal Design Power" for example the 14900K has a TDP of 125 Watts, lol... yeah, this looks on a box better than a Ryzen 7950X TDP of 170 watts but in reality its more like 250 watts, it looks like Intel are in turn throwing motherboard vendors under the bus by pointing the finger at them for allowing the CPU to run far outside those TDP ratings, i can see Steve Burke already agreeing with Intel, but the truth IMO is Intel knowingly turn a blind eye to all of it, or even encourage it, because in reality at 125 watts the CPU is a lot slower than it is as they run in all reviews at about 250 watts, IE much slower than AMD's CPU's. so yes on the one hand they want to print a lower than AMD TDP on the box while at the same time want to appear like you're also getting the same or better than AMD performance for that.

So, it seems that Asus at least, and possibly others have taken this a little too far and the CPU's are not actually stable, they have issued a BIOS which basically backs the CPU off a little, like you would, well..... an unstable overclock.

Also, Intel, stop pretending your CPU's are something they are not. they either score 39K in Cinebench at 250 watts or they are a 125 watt CPU and score a lot less.

https://twitter.com/x/status/1781571284402356719

If you have an Nvidia GPU and get the "out of video memory" crash check your board vendor for a BIOS update.

 

[humbug]

PL2 is not what they put on the box, and without MCE it lasts just long enough to complete a Cinebench run. but not long enough to matter a _____ for anything useful.

The reason i go on about all this is because i find it all quite manipulative and cynical, and Steve Burke complaining over and over and every time he finds an opportunity to accuse AMD of "lying" about power consumption for going 5 or 10 watts over while explaining Intel do nothing wrong is systematic of a bias that exists simply because its good click bait and Intel at least understand the value of sucking up to people like him, to stroke their ego.

 

[humbug]

I think given that they are advertised as 125 watt CPU's they should be reviewed under 125 watt coolers.

The hilarious thing is Steve Walton (Hardware Unboxed) actually did something along those lines for AMD, back when AMD shipped all of their CPU's with Box Coolers Steve decided they should be reviewed under those coolers, it had no real performance penalty for them but for whatever reason AMD stopped that.... its quite sad as those coolers were nice little coolers, people actually look for them on the flea market to put in small builds.

Honestly IMO if you're going to advertise something as such then it should be tested as such, they soon stop that nonsense.

 

[humbug]

You make a good point on the memory but XMP / EXPO is such a common thing now IMO running it is not a problem, so long as its reasonable, for example with early AM5 while they could do 6000 it was not reasonable to expect that, they wouldn't always do it and even if it did it needed more work that just setting that in EXPO, 5600 was reasonable and that is what people like Hardware Unboxed used, some accused them of being bias because they would run Intel at 6400, but with Intel 6400 was a reasonable expectation, so its reasonable to do that.

These days AM5 can do 7000 even 8000 with some getting "its over 9000", AMD have put a lot of work in to that and its paid off, so i think its reasonable to review at 6000, 6400, maybe even 7000, i think HUB now do it at 6400.

 

[humbug]

In buildzoid's video he said the problem is that Intel only recommends their factory settings, but not requires and I note that in their statement they're still using the word recommends, albeit with strongly attached to it.

It seems like a no-brainer to me though, at the very least the BIOS should have a pop-up, e.g. "Asus overclocked settings enabled, press Y to keep, or N to use Intel/factory defaults".

The problem is they are like that out of the box, even if there was such a thing you're not going to get that warning if all you're doing is going in to the BIOS to set XMP.

They are pre-overclocking the CPU while the motherboard is in the factory. And by a lot.

 

[humbug]

By Intel's validation they are supposed to run at 125 watts with a power level state boost of 250 watts for a few seconds, they are run at that 250 watt PL state 24/7 out of the box, a 125 watt overclock.

With Intel's recommended BIOS.

 

[humbug]

Every reviewer Including Steve Burke who accuses AMD of lying about TDP for going 5 watts over has these things scoring around 40K points.

 

[humbug]

Not only are they becoming unstable as games are increasingly hard on the CPU but they are probably also slowly cocking themselves and Intel offer a 3 year no quibble return and replace warranty service.

Its also Intel's fault because there is no way they didn't know what was going on, all they had to do was look at benchmark results, they turned a blind eye to it, if not encouraged it.

 

[humbug]

Take my GPU as an example.

Its an MCM GPU so i'll explain what on the GPU die and what isn't.
In MSI OSD AMD report the total board power, that's the core, the MCM chips, the power stages, the Memory IC's, the fans.... everything , it has a TDP rating of 260 Watts, i've never seen it pull more than 250, usually around >240, stock.

GPU Core Power: 127 watts
GPU SoC power: 43 watts
GPU USR Power: i don't know what that is so i'll add it here: 14 watts

That's it, 184 watts total. that die is 200mm

The rest.
GPU Memory Power VDDIO: those are the IMC's: 30 watts
GPU Memory Power VDDCI_Mem: memory IC's: 12 watts

The rest to make up the 268 watt board power draw (its overclocked) will be board components not listed, which includes the fans.

So 184 watts for a 200mm die (This is overclocked) here the GPU is running at 63c, its a good cooler.
--------

The 14900K is 250mm, its pulling 287 Watts in Blender, 200 watts in Cyberpunk, holy ____!

Doesn't seem too bad for a 250mm die, however most of that power is concentrated in this area. An area less than half of the die. I don't know what that heat density is but its probably double what mine is, its why they are so difficult to cool.

The Intel Problem: CPU Efficiency & Power Consumption | GamersNexus

CPUs The Intel Problem: CPU Efficiency & Power Consumption January 20, 2024 Last Updated: 2024-01-20 We put a commenter's claim of Intel efficiency to the test -- and the claim didn't hold up. The Highlights We ran Intel and AMD CPUs through efficiency calculations for gaming, power-normalized...

gamersnexus.net

 

[humbug]

Funnily enough, in Buildzoid's video he mentioned that the throttling (power and/or heat) is maybe an important factor in the binning being valid, since he said that he knows people who buy a bunch of these CPUs for their PCs and on "out of box" settings a large number of the samples will fail to run Cinebench. When the recommended settings are applied it throttles them down to the point that they don't run in an unstable config anymore.

That's just crazy to me, if you're relying on the red lining to bin the CPU! It shows how far these things are being pushed beyond the voltage/efficiency curve they were probably intended to run at. AMD don't need to be involving themselves in these shenanigans.

Right, AMD just need to keep doing what they are doing....

 

[humbug]

My Asus z790 dark hero motherboard default boost power settings is for only 96 seconds

So which motherboards do you know of which runs the boost settings for 24/7

Its supposed to be 25 seconds.

 

[humbug]

Its still not 96 seconds.

We all know why these limited boost timers exist, its to make them look better than they are on bar charts, because reviewers don't benchmark anything for more than a minute. Intel shenanigans.

If what @chaparral says is true then if he lets R23 loop for 10 minutes instead of just one 30 second run he's no longer scoring 40K (as every single review shows) but around 30K, he bought it believing its a 40K CPU. I would call that at best misleading. At worst a cynical lie.

 

[humbug]

Thats far fetched. Even a gimped 250w pl2 13900k will still score high 38 to 39k..

Mine has been solid in gaming, rendering, ycruncher and cinebench at 6ghz 4 core and 5.7 all..

The reason these chips have have issues is when the bios on Auto as ASUS (and others) like to ram 1.4 to 1.45 volts through the CPU's neck which is WAAAY ott...

125 watts vs 287 watts (quote Gamers Nexus) its not far fetched at all, its reality, you don't even need anyone to put that to the test to realise it, its obvious.

You see this, you can loop that in R23 all day long and it will come out the other end scoring with in a couple of % what the review bar charts show, around 40K.

 

[humbug]

Most people buying a 13900 and 14900 really don't care what TDP is on the box.. If you buy one of them and gimp it to 125 watts, then you should just buy an i5 or AMD equivalent.

saying that the games I play I see mostly 80 to 110watts on the cpu, so 125watt limit wouldn't make much difference in that.

I'm not an intel fanboi either, I got both systems here - they have their pros and cons

If I was to recommend a CPU for gaming, I would say the 7800x3d is the choice..

If you do work and gaming etc, then Intel is the way to go..

I didn't say you were, i didn't even think that. I don't think @chaparral is either.

What annoys me, what gets me worked up are these cynical shenanigans, its cynical because they are taking advantage of people why unlike sad no life ______ like me aren't nerdy enough to be clued in to all this ____ I talked about all this on this forum long before the inevitable that is happening now, even before Steve Burke did in the above video.

Its narcissistic bordering on sociopathic.

I'll be honest with you, i don't like Intel, as a company, never have, but i have good reason to.

 

[humbug]

I'll say this, not just for balance but because it's true.

AMD have done the same thing in the past, the FX-9590, that thing was not stable at 5Ghz all core nor was it safe in terms of the CPU's longevity to be run like that, it only pulled 220 watts and we though that was way too high, it was.

 

[humbug]

I'm all for piling onto Intel over this, it annoys me that they allow the board makers to just ignore their specs.

But you're wrong about how PL1/PL2 and TAU work, that's why I'm still disputing that part of your post.

It doesn't help that the article you referenced is poorly written and causing confusion.

TAU is allowed to be unlimited and so far as I'm aware, the fixed Asus profile does not change this behaviour.

It:
- Disables MCE (boost clock ratios are fixed, i.e. the CPU can't just use the highest multiplier for every core).
- Restores Intel's recommended power limit for PL2 (from 4096 watts to 253 watts).
- Enables Intel's voltage/frequency table.

You're 100% right that it will have an impact on the top-end performance in something like Cinebench, or long-run workloads like blender. However, the impact will be less than the numbers you quoted, because as stated in the article: their CPU was an engineering sample and artificially power-limited lower than PL2.

Reverting to PL1: 125 or 150 watt (depending on the CPU) is NOT considered stock behaviour and Intel do not require it.

The performance loss for i9-K CPUs in highly multithreaded benchmarks or long-run workloads would be noticeable (enough to change their relative benchmark position in those apps) just with MCE disabled and PL2 being enforced.

The reported stability issues which have prompted the BIOS updates (with the Intel-default profiles) are not due to unlimited running @ PL2. A case in point: the compiling shaders would often instant-crash the PC, not only after 30 seconds (or whatever).

To be explicit here: if a CPU can't run at PL2 forever (assuming the motherboard and cooling are sufficient) then it is just broken and you can RMA it. That's not just my opinion, buildzoid said the same thing (with running Cinebench, or whatever) and it is Intel's opinion too.

Reverting to PL1: 125 or 150 watt (depending on the CPU) is NOT considered stock behaviour and Intel do not require it.

So motherboard vendors did nothing wrong and there is no need to make a change?

Is this because its guidance, not required? Be that as i may then yes this is entirely Intel's doing and fault, nothing at all to do with the motherboard vendors, this is where i have a problem with it, on the one hand Intel are saying this is a fault with motherboard vendors, but actually they don't have any requirements to follow any of your guidance.

With that said all of this is performative, its public relations, "oh its because motherboard vendors are doing it wrong" while at the same time tell them they still don't actually have to change anything at all... You as the consumer think this problem has been fixed, but it hasn't.

 

[humbug]

Also, how can i as a consumer be expected to trust any Intel product?

If the specification guidance are just that and nothing more then how do i trust a third party product to treat that CPU out of the box any more competently than my mate trying his hand at overclock with my rig? Because we obviously can't.

 

[humbug]

Intel know what safe is for their CPU's, hence the guidance, but they have no requirements for third parties to run them safely through the products sold to you, what am i supposed to make of that?

 

[humbug]

@Tetras on your point about AMD being strict and Nvidia being very strict when it comes to enforcing their standards.

Nvidia can afford to be very strict, they don't need to be taking any chances with stability issues caused by third parties trying to one up eachother, and Nvidia aren't looking over at AMD thinking we need to pull out all the stops to keep up with those guys.

HUB just released a video testing this and it is actually worse than I thought.

Gigabyte's baseline profile underclocks the CPU by setting PL2 below spec. I guess we know what they meant by "Superior stability" now.

Intel CPUs Are Crashing & It's Intel's Fault: Intel Baseline Profile Benchmark

See Antec’s Range of C8 Cases below:C8 ARGB White: https://bit.ly/4a9i42tC8: https://bit.ly/4an0PuASupport us on Patreon: https://www.patreon.com/hardwareunb...

www.youtube.com

Its going to be interesting to see how this develops, MSI might look at that and be thinking they can push it 5% harder than that to get ahead on the bar charts, Asus will be looking at MSI and thinking the same thing, which is exactly how we got here in the first place, which is why i'm wondering if Intel are actually sick of their _____ and are scramming ENOUGH!!!

 

[humbug]

So is it an nvidia problem, who have thrown Intel under the bus, like the OP implies? It is literally the opening statement of this thread. Should we all therefore be buying AMD GPUs?

It's a rhetorical question btw, I already know what humbug wants people to think.

No you should be buying whatever GPU you feel is right for your needs, this is not an Nvidia problem, at all, Nvidia were right to throw Intel under the bus.

It was just a joke, having had criticism for not offering enough VRam on past GPU's, rightly or wrongly here not being the argument, i just thought the error message, that being "out of video memory" was amusing, imagine Nvidia's reaction... upon seeing it.

I have a childish situational sense of humour, no ill will toward Nvidia at all...