Support NeoGAF

[Panajev2001a]

Uh, Watts and Amps aren't the same thing

Also waiting anxiously for @Leonidas in this thread

It is normal, they need to let the discussion and the fervour die out as they would get dogpiled at the moment and any hand waving / spin would get lost in that. So, they wait for most of the fervour to dissipate, they collect what good argument they think they can present and hopefully Intel also provides so good PR take on it, then they present it when they would not get dog piled and they think they can push the narrative somewhere else.

 

[Unknown Soldier]

watts = amps x voltage...so if you have 1.1V or if that's the voltage, you need to have current at around 280A for 300W draw.

Not sure what exact voltage on these are, tho.

Bruh, this is physically impossible. How do I know? Because your house literally cannot supply that much amperage to a single device.

Don't believe me? Your house has a circuit breaker box. It's that thing that you go to reset the breakers every time you pull too much power from one of your wall sockets and cause the power to that part of your house to cut off to protect your circuits and prevent your house from catching on fire. That breaker box supplies all the power to your whole house, and most houses have boxes which can supply between 100-200A. Total. To the whole house. Each circuit to a standard North American 120V socket typically supplies between 12-15A.

Your problem is that you are doing this calculation incorrectly. Those measurements of 300W draw are taken from the wall to the computer in the test lab. The socket in the wall is supplying 120V, not 1.1V. So the correct calculation where W = A x V is actually 300W = A x 120V, so A is actually 2.5A. Which is a sane number LMAO

 

[winjer]

Bruh, this is physically impossible. How do I know? Because your house literally cannot supply that much amperage to a single device.

Don't believe me? Your house has a circuit breaker box. It's that thing that you go to reset the breakers every time you pull too much power from one of your wall sockets and cause the power to that part of your house to cut off to protect your circuits and prevent your house from catching on fire. That breaker box supplies all the power to your whole house, and most houses have boxes which can supply between 100-200A. Total. To the whole house. Each circuit to a standard North American 120V socket typically supplies between 12-15A.

Your problem is that you are doing this calculation incorrectly. Those measurements of 300W draw are taken from the wall to the computer in the test lab. The socket in the wall is supplying 120V, not 1.1V. So the correct calculation where W = A x V is actually 300W = A x 120V, so A is actually 2.5A. Which is a sane number LMAO

The components in our PCs don't use the voltages/amperages that are supplied to our houses.
The current firsts passes through a transformer inside our PSUs.
Then voltages are stepdown into several rails, like the 3.3v, 5v, 12v, -12v.
And of course, amperages are stepped up, or used several lines to increase it.

 

[Unknown Soldier]

The components in our PCs don't use the voltages/amperages that are supplied to our houses.
The current firsts passes through a transformer inside our PSUs.
Then voltages are stepdown into several rails, like the 3.3v, 5v, 12v, -12v.
And of course, amperages are stepped up, or used several lines to increase it.

I understand all this but what I'm saying is the power draw measurements can only measure power the computer pulls from the wall, and those measurements will invariably be at 120V which is what the main is supplying, which is why calculating the computer is pulling 280A when the power draw measurement is 300W is hilariously wrong. It's pulling 2.5A.

 

[JohnnyFootball]

Uh, Watts and Amps aren't the same thing

Also waiting anxiously for @Leonidas in this thread

watts = amps x voltage...so if you have 1.1V or if that's the voltage, you need to have current at around 280A for 300W draw.

Not sure what exact voltage on these are, tho.

I’m a high school Physics teacher…and we ended the year discussing ohms law. It warms my heart seeing this discussion.

Back to regular Intel bashing….

 

[Celcius]

Is there any chance that Intel learned from this and 15th gen will be better?

Will they no longer use the dual lever sockets that warp the cpu over time?

Will they make it to where you build the system you simply load up a default profile that doesn’t fry the cpu and it’s super easy for the user to know they’re running the stock settings?

I assume the CPU’s won’t be furnaces anymore due to moving to a smaller node?

 

[StereoVsn]

More like 300W.
400W if the motherboard is not enforcing power limits.

And this is why I am running 7800x3D. That power consumption is nuts even if there was no degradation or other issues.

 

[winjer]

I understand all this but what I'm saying is the power draw measurements can only measure power the computer pulls from the wall, and those measurements will invariably be at 120V which is what the main is supplying, which is why calculating the computer is pulling 280A when the power draw measurement is 300W is hilariously wrong. It's pulling 2.5A.

The power characteristics we have outside our houses is different from the one inside our houses, and the power inside our PCs is different from the power at the outlets.

So if a CPU is using 1.3V and 200W, then the current is 153A. (Ignoring capacitance)
Though this is delivered through several different phases, to the CPU.

A simpler example is the 6+2 connector we use on our GPUs.
This uses 8A, with 3 lines, at 12V. Resulting in 288W of power.
Though the spec limits the delivery to 150W, as a safety measure.

 

[SHA]

This channel capitalize on shitty practices in the Tech industry, I literally can't unthink of hu without this stuff, it's like saying incomplete sentences without finishing them, I don't get the point of what they keep doing without showing alternative solutions, what they're doing is not heroic or revolutionary, mostly unprofessional.

 

[Chiggs]

Is there any chance that Intel learned from this and 15th gen will be better?

Yes, given their new node and architecture...but would I buy a first-gen design from them? Fuck. No. I didn't even do that for Ryzen.

This is really why some of you x86'ers need to be jumping for joy that a new player (Qualcomm) and a new architecture (ARM) is shaking things up. It's time to light a fire under the collective ass of Windows PCs.

 

[Frwrd]

@Leonidas whatcha think?

 

[Von Hugh]

@Leonidas whatcha think?

 

[Insane Metal]

Weird, still no sign of him! Wonder if he's ok.

 

[Unknown Soldier]

The power characteristics we have outside our houses is different from the one inside our houses, and the power inside our PCs is different from the power at the outlets.

So if a CPU is using 1.3V and 200W, then the current is 153A. (Ignoring capacitance)
Though this is delivered through several different phases, to the CPU.

A simpler example is the 6+2 connector we use on our GPUs.
This uses 8A, with 3 lines, at 12V. Resulting in 288W of power.
Though the spec limits the delivery to 150W, as a safety measure.

Yeah, alright. I get what you mean here and the calculation doesn't lie. I guess I never thought of it this way before huh

 

[M1chl]

Bruh, this is physically impossible. How do I know? Because your house literally cannot supply that much amperage to a single device.

Don't believe me? Your house has a circuit breaker box. It's that thing that you go to reset the breakers every time you pull too much power from one of your wall sockets and cause the power to that part of your house to cut off to protect your circuits and prevent your house from catching on fire. That breaker box supplies all the power to your whole house, and most houses have boxes which can supply between 100-200A. Total. To the whole house. Each circuit to a standard North American 120V socket typically supplies between 12-15A.

Your problem is that you are doing this calculation incorrectly. Those measurements of 300W draw are taken from the wall to the computer in the test lab. The socket in the wall is supplying 120V, not 1.1V. So the correct calculation where W = A x V is actually 300W = A x 120V, so A is actually 2.5A. Which is a sane number LMAO

Incredible

 

[winjer]

Yeah, alright. I get what you mean here and the calculation doesn't lie. I guess I never thought of it this way before huh

Not trying to be too annoying with this subject.
But look at the power delivery on motherboards.
You will find motherboards have several phases, each rated 20A, or 40A, or 60A.
Of course, they won't be used all the time. And not at full power. And some motherboards even have these over engineered.
But they are rated in relatively high current values.

 

[winjer]

Considering that the Performance profile can save 100-150W, while losing around 5% on average, this makes it the best choice for most people.

 

[Celcius]

Considering that the Performance profile can save 100-150W, while losing around 5% on average, this makes it the best choice for most people.

tl;dr: They’re tied at 4K and Intel gets smoked at 1440p and 1080p.

 

[winjer]

tl;dr: They’re tied at 4K and Intel gets smoked at 1440p and 1080p.

At native 4K, almost all CPUs are tied, because they are GPU limited.

 

[winjer]

Intel comments internally on the “13th and 14th Generation K SKU Processor Instability Issue” and finally brings a comprehensive update of its own investigation (Leak) | igor´sLAB

What do you think it means when Intel provides a document with the title “Title Enhanced Thermal Velocity Boost (eTVB) May Miscalculate Frequency Limits” and the subline “Issue Description An…

www.igorslab.de

What do you think it means when Intel provides a document with the title “Title Enhanced Thermal Velocity Boost (eTVB) May Miscalculate Frequency Limits” and the subline “Issue Description An incorrect frequency limit calculation may allow the processor to operate at a high frequency state at a high temperature”? That’s when it gets interesting, because the problem with potentially unstable 13th and 14th generation K-CPUs and possible consequential damage has been around for a long time.

Addendum from 15.06.2024 – 09.00 a.m.
Another piece of the puzzle, but not the so-called “root cause”. I’ll add this as a note, because once again I’ve been quoted incompletely and sometimes incorrectly. And since it doesn’t seem to be enough to recognize a comment from the context, I’ll put a heading above it so that you don’t have to spend a lot of time and effort using the wayback machine.

It is of course a pity that the following document has been provided with an NDA and not (yet) used as the basis for a public statement, but perhaps that will come. If not, I have something for you that should shed some light on the matter. The other comments are also interesting. Let’s start with the preceding summary and text description.

Root cause is an incorrect value in a microcode algorithm associated with the eTVB feature.
Implication Increased frequency and corresponding voltage at high temperature may reduce processor reliability.
Observed Found internally.
Impacted platforms Raptor Lake S, Raptor Lake Refresh S (CPUID 0xB0671)

What is eTVB (enhanced Thermal Velocity Boost)?​

In order to understand the following text correctly, I’ll start with a little reminder from my launch article on Raptor Lake S. Intel’s eTVB, or Enhanced Thermal Velocity Boost, is a technology developed specifically for Intel’s Raptor Lake S processors. This technology optimizes CPU performance by automatically overclocking the processor cores beyond the maximum turbo frequency, depending on the available thermal capacity and power consumption. This means that if the temperature of the processor is low enough and sufficient power is available, eTVB can increase the clock frequency to deliver improved performance for short-term, compute-intensive tasks.

This feature is particularly useful for games and applications that require high burst speeds. Intel integrates this technology into its latest desktop CPUs to ensure users get maximum performance, especially when combined with cooling solutions that support the low temperatures required for maximum boost performance. This also improves the overall efficiency of the processors as they can dynamically adapt to performance requirements and thermal conditions.

Failure Analysis (FA) of 13th and 14th Generation K SKU processors indicates a shift in minimum operating voltage on affected processors resulting from cumulative exposure to elevated core voltages. Intel® analysis has determined a confirmed contributing factor for this issue is elevated voltage input to the processor due to previous BIOS settings which allow the processor to operate at turbo frequencies and voltages even while the processor is at a high temperature. Previous generations of Intel® K SKU processors were less sensitive to these type of settings due to lower default operating voltage and frequency.

Intel® requests all customers to update BIOS to microcode 0x125 or later by 7/19/2024.
This microcode includes an eTVB fix for an issue which may allow the processor to enter a higher performance state even when the processor temperature has exceeded eTVB thresholds.