• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Intel Core i9-13900K Reportedly 60% Faster Than i9-12900K in 7-Zip Decompression Test

Joined
Mar 31, 2020
Messages
1,519 (0.82/day)
The upcoming flagship Intel Core i9-13900K processor has recently appeared in a 7-Zip benchmark screenshot where the chip beat its predecessor by 20% and 60% in compression and decompression tests respectively. The i9-13900K looks set to feature an additional 8 High-Efficiency cores for a total of 24 cores along with a higher boost clock of 5.8 GHz. This increased core count and clock speed account for the majority of the performance improvements with the i9-13900K reaching a max single thread clock of 5716 MHz and 4611 MHz on 16 threads compared to 5021 MHz and 4060 MHz with the i9-12900K. The processors were both paired with a 32 GB set of DDR5-6400CL32 memory on an unspecified motherboard. These results have not been confirmed with Intel expected to officially unveil the new lineup on September 27th.



View at TechPowerUp Main Site | Source
 
Not bad. This is where Intel is lacking in ZIP.
 
Who is actually leaking these screenshots? Intel high command? Can't see supply chain organisations would have any business doing Zip screenshots. Be nice if this was representative of the general performance uplift on a regular setup and not liquid nitrogen.
 
While consuming between 20-60% more power...
 
I've got a feeling the 13900K isn't going to impress me over the 12900K in gaming, only threaded apps.

How high it will clock is going to impress me, hopefully.
 
Extra 8 E cores is showing it's muscle , can't wait to see if the E cores on 13th gen will clock to 4,5Ghz or higher
 
While consuming between 20-60% more power...

13900K has the same TDP as 12900K.

Edit 1: AMD has been lying about its CPUs power consumption for years now. Nowhere on their website there's any mention that their CPUs consume roughly 35% more watts than their TDP specifies, e.g. 5800X - 105W TDP, real power consumption, i.e. PPT, 142W.
Edit 2: Intel is not lying. From the specs page: Maximum Turbo Power 241 W.

TDP =/= powerconsumption

This used to be the case until AMD decided to use their peculiar "logic" which its fans find impeccable.
 
Last edited:
Who is actually leaking these screenshots? Intel high command? Can't see supply chain organisations would have any business doing Zip screenshots. Be nice if this was representative of the general performance uplift on a regular setup and not liquid nitrogen.
Well we had times where 'Another day another Intel leak' actually means it.

We all know why.
 
I've got a feeling the 13900K isn't going to impress me over the 12900K in gaming, only threaded apps.

How high it will clock is going to impress me, hopefully.
if we believe the leak results 13900k can clock 5.75Ghz to 6 ( even more ) with ease :) and 13900ks will be 6Ghz out of the box compare to 12900ks ~5.5Ghz out of the box

which is 241Watts while boosting? Nice :roll:
is it? that would be pretty nice then
 
is it? that would be pretty nice then
It is a sarcasm and a reply to a previous post.
How a 241w be a nice thing for a CPU huh?
 
Might be some nice last gen deals this upcoming Black Friday
 
Is it the extreme performance mode (350W) in 700 series boards? (Like an evolution from the i9-12900KS case)
 
I feel like there is a fake carrot here somewhere. When has intel ever been accurate in its "faster" claims in the last 5 years?
 
I've got a feeling the 13900K isn't going to impress me over the 12900K in gaming, only threaded apps.

How high it will clock is going to impress me, hopefully.
Who needs more than a 12900K (or a 10900K to be honest) for gaming ?

I feel like there is a fake carrot here somewhere. When has intel ever been accurate in its "faster" claims in the last 5 years?
Yes. Alder Lake was noticeably faster than Rocket Lake.
 
more like 10yrs.
I feel like there is a fake carrot here somewhere. When has intel ever been accurate in its "faster" claims in the last 5 years?

How will this impact the laptop market, my laptop can hit the processor limit of 117w but throttles down to about 110w, 12900h.
Doubling the e-cores will add more wattage usage.
 
13900K has the same TDP as 12900K.

Edit 1: AMD has been lying about its CPUs power consumption for years now. Nowhere on their website there's any mention that their CPUs consume roughly 35% more watts than their TDP specifies, e.g. 5800X - 105W TDP, real power consumption, i.e. PPT, 142W.
Edit 2: Intel is not lying. From the specs page: Maximum Turbo Power 241 W.



This used to be the cased until AMD decided to use their peculiar "logic" which its fans find impeccable.
I don’t like the “Intel vs AMD” logic but AMD definition of TDP is hilarious.
We should realize that AMD is no different from Intel, lately.
And Zen 4 prices will be another punch in the stomach of customers.
 
We should realize that AMD is no different from Intel, lately.
Where have you been? When was AMD that much different than Intel? PRs say it all.
And Zen 4 prices will be another punch in the stomach of customers.
Prices are out. See front page. 10-15% higher than Z3.
 
While consuming between 20-60% more power...
Lol I hope not. Probably not. I think it'll either be the same or low. 241W maximum turbo boost power isn't good, but it's where the maximum bar is right now. Since they want to (have to) increase performance with the 13900K (because AMD became competitive), I'd say it's not going down
 
Lol I hope not. Probably not. I think it'll either be the same or low. 241W maximum turbo boost power isn't good, but it's where the maximum bar is right now. Since they want to (have to) increase performance with the 13900K (because AMD became competitive), I'd say it's not going down
Raptor Lake uses more power at stock speeds which are higher than Alder Lake and OC'd RL is like 350W TDP. The best test will be 13700K vs 12900K as they have same core configuration, as long as we clock 13700K same as 12900K. Also we can disable E-cores and vice versa and see if any efficiency improvements have been made at same clocks.

Anyway this benchmark is nothing special. 50% more cores and higher clocks, so 60% improvement is about what you would expect for decompression test. It actually does worse per core for compression than 12900K.
 
Raptor Lake uses more power at stock speeds which are higher than Alder Lake and OC'd RL is like 350W TDP. The best test will be 13700K vs 12900K as they have same core configuration, as long as we clock 13700K same as 12900K. Also we can disable E-cores and vice versa and see if any efficiency improvements have been made at same clocks.

Anyway this benchmark is nothing special. 50% more cores and higher clocks, so 60% improvement is about what you would expect for decompression test. It actually does worse per core for compression than 12900K.

Do you mean worse clock for clock?

edit: it seems you do, I'm just clarifying. Comparing the 13700K to the 12900K will be interesting. I wonder when reviews will be out
 
13900K has the same TDP as 12900K.

Edit 1: AMD has been lying about its CPUs power consumption for years now. Nowhere on their website there's any mention that their CPUs consume roughly 35% more watts than their TDP specifies, e.g. 5800X - 105W TDP, real power consumption, i.e. PPT, 142W.
Edit 2: Intel is not lying. From the specs page: Maximum Turbo Power 241 W.



This used to be the cased until AMD decided to use their peculiar "logic" which its fans find impeccable.
...
You do know that TDP and wattage are not the same thing, and that neither AMD not intel list them plainly like that?
They're both an average, not a hard cap.

Intel: 65W TDP = 250W for 60 seconds, then 65W after that.
AMD: 105W TDP = 140W (similar time limit, not sure if also 60 seconds)

Depending on motherboard, that 60 second rule can be completely ignored. That happens on both sides.

Neither lists TDP as the maximum value, they're the average.
Why the one sided view there?
You do understand that you're looking at different numbers and values and mixing them up to create something that doesn't exist?


Let's use tomshardware today since they measure CPU power only and not total system like TPU.

1660793163326.png


AMD state the 5600x is 65W TDP
5600x uses 75W (115%), or 102W (156%) with PBO enabled

5900x 105W TDP
136W (129%), or 165W (157%) with PBO


11600k
125W TDP
95W TDP down
203W (162%)

Hmm. Maybe that's an oddity, lets go the 11700k
Nope, 125W with a 95W TDP down.
Oh... 239W. (191%)

Okay lets ignore the 11th gen and move to 12th gen.
Theres a whole new bag of kittens to go through with the boost values as those are supposed to be temporary and quite often they're not - so I'm comparing both values.

12700K:
125W, with 190W boost.
But it uses 224W. (179%/117%)

12900K then!
125W, with 241W boost:
and it hits the same 224W (179%/92%) limit as the 11900K, because the boards just have power limits there. In this case it doesn't even reach the advertised level.


And for bonus points, the still-for-sale 10700
Intel® Core™ i7-10700 Processor
65W TDP, no other wattages listed

Also, zero shits given for accuracy, with the 65W CPU using more power than the 125W CPU
Power Consumption - Intel Core i7-10700 vs Core i7-10700K Review: Is 65W Comet Lake an Option? (anandtech.com)
1660793808359.png

329% advertised TDP. Nice. Needs an award for that.

So @birdie tell me again why intel listing one single CPU advertised with a maximum TDP value so high that most boards wont even reach it, while all the others are outright false makes AMD the one who's lying here?
I have no problem with you pointing out TDP values are inaccurate but for gods sake don't just spew out fanboy drivel you've read online without the most basic of research
 
Last edited:
Back
Top