Intel Core i9-14900K

[ratirt]

I really find it hard to justify the editor's choice for this CPU. In general it is not bad (literally 13900k) but this is new gen so in that regard it sucks. The other thing is the power. It consumes so much more that I just can't overlook this power requirements. MSRP is higher than 13900K when released. The 14900k is a very weak release.

 

[Gica]

• Matches 13900KS performance, beats it in some cases

At a much lower price.
A bit more expensive than 13900K (new release, speculative price), but it will drop to its price very soon. I expect the 12th and 13th to drop in price in at most a month.
Let's see AMD's reaction.

Prices in Romania

 

[JustBenching]

1. You probably have done this long enough to know that best sellers are not always best choices. We can agree that OEM machines are trash and people buying them are idiots, right? Looking at retail sales, AMD dominates. That means most of these CPUs go for OEM machines.

2. and 3 You forgot something: that is AMD.

From performance summary, 13900K review: https://www.techpowerup.com/review/intel-core-i9-13900k/25.html

Applications:
i9-13900K Stock: 100%
Ryzen 9 7950X: 95.7%

Games 720p:
i9-13900K Stock: 100%
Ryzen 9 7950X: 89.2%

Based on that data only, you can claim Intel is 4.3% faster on applications and 10.8% faster on games. Therefore, you can claim Intel is faster on both games and applications. Now let's see if something has changed:

From performance summary, 14900K review: https://www.techpowerup.com/review/intel-core-i9-14900k/27.html

Applications:
i9-14900K Stock: 100%
Ryzen 9 7950X: 97.1%
Ryzen 9 7950X3D: 95.1%

Games 720p:
Ryzen 7 7800X3D: 105.7%
Ryzen 9 7950X3D: 100.4%
i9-14900K Stock: 100%
Ryzen 9 7950X: 88.2%

Based on that data only, you can no longer claim Intel is winner on gaming performance.

Overall for 2. and 3., while 13900K was performance king on both applications and games based on data you provided, that is no longer case on 13900K.

You call people idiots but you can't even do basic math right.

Going from 89.2 to 100 is not a 10.8% increase darling. My 9 year old nephew knows this. But he bought a prebuilt so that makes him an idiot I guess.

 

[lexluthermiester]

Going from 89.2 to 100 is not a 10.8% increase darling.

WHAT?!? Where did YOU learn mathematics? 100 - 89.2 is EXACTLY 10.8, "darling".

 

[JustBenching]

WHAT?!? Where did YOU learn mathematics? 100 - 89.2 is EXACTLY 10.8, "darling".

Okay man.. Not sure if you are trolling.

 

[mahirzukic2]

WHAT?!? Where did YOU learn mathematics? 100 - 89.2 is EXACTLY 10.8, "darling".

100 - 89.2 is EXACTLY 10.8, but you don't calculate it that way. You divide it to get a percentual increase, so it becomes: 100 / 89.2= 1.121 or about 12% increase.
And this works whether you compare against the same product or another product, you divide the performance metrics to get the percentual increase or the RELATIVE increase.
By subtracting the numbers you get the absolute increase, but the thing is NOBODY cares about the absolute increase, rightfully so, it is irrelevant. Only the relative increases (or reductions) matter.

 

[JustBenching]

100 - 89.2 is EXACTLY 10.8, but you don't calculate it that way. You divide it to get a percentual increase, so it becomes: 100 / 89.2= 1.121 or about 12% increase.

I think he was being sarcastic but not sure...

 

[lexluthermiester]

Okay man.. Not sure if you are trolling.

You said something skeewumpus. I was pointing that out. Such is not trolling.

100 - 89.2 is EXACTLY 10.8, but you don't calculate it that way. You divide it to get a percentual increase, so it becomes: 100 / 89.2= 1.121 or about 12% increase.

You're taking about statistical percentages. Where tech performance numbers are concerned, percentages are taken as direct face-value representations.

So the difference in performance of 10.8% is exactly 10.8%, whether an increase or decrease and is not adjusted to relative values.

And this works whether you compare against the same product or another product, you divide the performance metrics to get the percentual increase or the RELATIVE increase.
By subtracting the numbers you get the absolute increase, but the thing is NOBODY cares about the absolute increase, rightfully so, it is irrelevant. Only the relative increases (or reductions) matter.

Exactly.

I think he was being sarcastic but not sure...

Nope, 100% on the level...

 

[JustBenching]

You said something skeewumpus. I was pointing that out. Such is not trolling.


You're taking about statistical percentages. Where tech performance numbers are concerned, percentages are taken as direct face-value representations.

So the difference in performance of 10.8% is exactly 10.8%, whether an increase or decrease and is not adjusted to relative values.


Exactly.


Nope, 100% on the level...

Well then you were 100% wrong. Going from 50 to 100 is not a 50% increase.

 

[lexluthermiester]

Well then you were 100% wrong.

No I'm not and...

Going from 50 to 100 is not a 50% increase.

...yes, it is. If YOU fail to understand that, then YOU have the problem.

In measuring performance of a PC part or aspect of operation, if you START at 50% and go to 100%, you might be doubling the performance, but you're still only increasing by 50%. If you go from 100% to 150% you are not doubling the performance, but you are still only increasing by 50%.

That is how performance measurements in the PC industry are done. It is always measured by absolute numbers, not by relative values.

 

[JustBenching]

No I'm not and...

...yes, it is. If YOU fail to understand that, then YOU have the problem.

In measuring performance of a PC part or aspect of operation, if you START at 50% and go to 100%, you might be doubling the performance, but you're still only increasing by 50%. If you go from 100% to 150% you are not doubling the performance, but you are still only increasing by 50%.

That is how performance measurements in the PC industry are done. It is always measured by absolute numbers, not by relative values.

No. Just no. Absolutely not.

When you are talking about an increase, your base value is the lower performing product. Therefore the 50, which needs a 100% increase to get to a 100.

It's basic math, you are just completely wrong, please let's not spend 20 pages on this. Just admit it and move on. Please, for the love of God, we can't be arguing cause you don't understand basic math.

 

[theouto]

Are we really arguing over semantics here? If we take the big boy value as the value of reference, then it'd be a 10.8% reduction, if we take the small boy value, it's a 12.1% increase, if we specify neither and just say difference, it could go either way, but there is one thing we can all agree on: this conversation is absolutely pointless.

 

[lexluthermiester]

this conversation is absolutely pointless.

Accurately representing information is never pointless.

 

[Xaled]

Neither of the results change the fact that 14900K doesn't deserve the Editor's Choice award.
Any other try to justify this award is absolutely pointless.


Edit:
Although I though it is the other way around, It looks like lexluthermiester's arguenment is correct,

Source

 

[BoggledBeagle]

I have no idea why are so many people hating on a 14900K. It is a fine highly efficient and highly performant CPU. Only by a terrible mistake of the manucturer it is sold set with nonsensical parameters and needs to be toned down.

It is like an audio power amp, with a dial turned to 11 out of the factory. Like they forgot to turn it down after testing it. You will not get deaf by listening to it and hate it because of that. You will understand the problem, turn the dial down and enjoy your excellent music experience.

 

[Hxx]

Edit:
Although I though it is the other way around, It looks like lexluthermiester's arguenment is correct,

View attachment 318171

Source

first column is current period second column is previous period. Percentage increase decrease is always computed based on the previous numbers/periods. so income went down by 25% x 280 from 280 to 211. also looks like its missing the 2q22 column but probably the way it was cropped out from their statements otherwise that very last column doesnt make sense to me.

 

[btk2k2]

No I'm not and...

...yes, it is. If YOU fail to understand that, then YOU have the problem.

In measuring performance of a PC part or aspect of operation, if you START at 50% and go to 100%, you might be doubling the performance, but you're still only increasing by 50%. If you go from 100% to 150% you are not doubling the performance, but you are still only increasing by 50%.

That is how performance measurements in the PC industry are done. It is always measured by absolute numbers, not by relative values.

The correct notation for that scaling is %pt (percentage point, p.p and pp are also other shorthands). So in your example it would be a 50%pt increase for each step.

When doing a ranking with the baseline as the test product and indexing other products off of that baseline then yes, absolute measures are what you would use and we clearly see that with the performance charts.

When comparing 2 products then you would use a relative measure to describe the change in performance from a to b which is what you see in those a vs b charts that get made.

If we take your 50%, 100% and 150% and substitute in some measured values we can do 60, 120 and 180 fps. Each 50%pt increase is clearly 60fps but the relative % changes depending on what two values you compare. 60 to 180 is a 200% increase. 60 to 120 is a 100% increase and 120 to 180 is a 50% increase.

 

[Imouto]

Accurately representing information is never pointless.

You lost to menoscats. Shame on you, dude.

 

[JustBenching]

I have no idea why are so many people hating on a 14900K. It is a fine highly efficient and highly performant CPU. Only by a terrible mistake of the manucturer it is sold set with nonsensical parameters and needs to be toned down.

It like an audio power amp, with a dial turned to 11 out of the factory. Like they forgot to turn it down after testing it. You will not get deaf by listening to it and hate it because of that. You will understand the problem, turn the dial down and enjoy your excellent music experience.

That's why I hate my TV. By default it was set to maximum brightness and volume and I can't bother to get into the settings and tone it down a bit. /s

 

[Hxx]

I have no idea why are so many people hating on a 14900K. It is a fine highly efficient and highly performant CPU. Only by a terrible mistake of the manucturer it is sold set with nonsensical parameters and needs to be toned down.

It like an audio power amp, with a dial turned to 11 out of the factory. Like they forgot to turn it down after testing it. You will not get deaf by listening to it and hate it because of that. You will understand the problem, turn the dial down and enjoy your excellent music experience.

Yeah I dunno your analogy doesn’t make much sense to me lol.
you don’t buy this chip to tone it down. Why ? Cause you could achieve the same result paying less with a lesser chip. I hope folks don’t buy this chip to dumb it down . And this doesn’t even answer your initial point of confusion which is the hate . It stems from intel charging a crapton of money for little innovation at the expense of insane power requirements

 

[JustBenching]

Yeah I dunno your analogy doesn’t make much sense to me lol.
you don’t buy this chip to tone it down. Why ? Cause you could achieve the same result paying less with a lesser chip. I hope folks don’t buy this chip to dumb it down . And this doesn’t even answer your initial point of confusion which is the hate . It stems from intel charging a crapton of money for little innovation at the expense of insane power requirements

Which cheaper cpu can achieve the same result? You realize that even limited to 150w there is only one other cpu that is as fast as the 14900k? Yes?

 

[Hxx]

Which cheaper cpu can achieve the same result? You realize that even limited to 150w there is only one other cpu that is as fast as the 14900k? Yes?

Before anyone can answer that question for you , you need to define what “same result” means to you. Is it overclocking , benchmarking , productivity , gaming , etc because not all chips are created equal

 

[rv8000]

I have no idea why are so many people hating on a 14900K. It is a fine highly inefficient and highly performant CPU. Only by a terrible mistake of the manucturer it is sold set with nonsensical parameters and needs to be toned down.

It is like an audio power amp, with a dial turned to 11 out of the factory. Like they forgot to turn it down after testing it. You will not get deaf by listening to it and hate it because of that. You will understand the problem, turn the dial down and enjoy your excellent music experience.

FTFY

Fast, but horribly power hungry/inefficient even when putting restrictions vs the competition.

 

[JustBenching]

Before anyone can answer that question for you , you need to define what “same result” means to you. Is it overclocking , benchmarking , productivity , gaming , etc because not all chips are created equal

I assume you meant MT performance, what did you mean since you brought it up?

 

[trparky]

I have no idea why are so many people hating on a 14900K. It is a fine highly efficient and highly performant CPU.