Thursday, July 21st 2022

Intel Core i5-13600K Geekbench Results Pop Up

It's a busy day when it comes to Intel Raptor Lake benchmark leaks and this time around, we're looking at a pair of Core i5-13600K CPUs, but in two different motherboards. The first CPU sits in an ASUS ROG Maximus Z690 Extreme board and is paired with 32 GB of DDR5 memory, whereas the second CPU is in an ASRock Z690 Steel Legend WiFi 6E board, this one also paired with 32 GB of RAM, but DDR4 rather than DDR5. Both systems are running Windows 10 and unlike the Core i9-13900K results that we saw a couple of weeks ago, these two results are both valid according to the Geekbench results browser.

Courtesy of @harukaze5719 we have a nice graph with the results below, even comparing them to the average Geekbench results for a Core i5-12600K, as well as a Ryzen 9 5950X and a Ryzen 5 5600X. The Core i5-13600K beats everything in the single threaded results and the ASUS system is not far behind the Ryzen 9 5950X in the multithreaded tests. The ASUS system is significantly faster than the ASRock one though, so it seems like Raptor Lake is really making the most out of DDR5 memory. As we don't know the memory clocks, it's hard to say if that is also a contributing factor here, but it's likely that the ASUS system uses much faster memory, as in some of the sub-tests like Rigid Body Physics, the two systems are on par in terms of performance. It'll be interesting to see how Intel prices these upcoming CPUs, especially considering that the company has already warned it'll raise its prices in time for the retail availability of Raptor Lake.
Sources: Geekbench (ASUS), Geekbench (ASRock), via @harukaze5719
Add your own comment

44 Comments on Intel Core i5-13600K Geekbench Results Pop Up

#26
Minus Infinity
People are gloating about a cpu that is not even 7% faster in ST compared to its predecessor and has 40% more cores and a lot more cache, and is about 40% faster in one MT benchmark.
Posted on Reply
#27
hs4
The number of cores (6P+8E) is consistent with the high end of Alder lake-H, so those are the reference. For example, the 12900HK in the MSI Raider GE76 12U runs at 77 W and scores over 17000 in Cinebench R23 (5900X @ 120 W PPT is 21000 and 5800X @ 120 W is 15000). Running it at 105-120 W, it would be on par with the 5900X and probably faster than the 7800X. I would guess this benchmark to be in the 150-180W range.
Posted on Reply
#28
Richards
hs4The number of cores (6P+8E) is consistent with the high end of Alder lake-H, so those are the reference. For example, the 12900HK in the MSI Raider GE76 12U runs at 77 W and scores over 17000 in Cinebench R23 (5900X @ 120 W PPT is 21000 and 5800X @ 120 W is 15000). Running it at 105-120 W, it would be on par with the 5900X and probably faster than the 7800X. I would guess this benchmark to be in the 150-180W range.
The e cores are powerful ibtel should rename them to medium or middle cores.. because they are 80% performance of the big cores
Posted on Reply
#29
hs4
ppn4E core are taking +33% surface area than a single 1C2T ~ 3E, 1C2T ~ 1000 CPUz. 3E ~ 1200, so 20-30% better performance at what power, that would be interesting to know. are they that much more efficient. But just give me the 10C20T part instead.
I've measured mine in the past in a variety of ways:
At the highest clocks: E-core has 1.5x power efficiency and 1.2x area efficiency relative to P-core
At the same clock: E-core has 1.0x power efficiency and 1.6x area efficiency

As long as there is a 240 W power limit, 10P+0E can only deliver MT performance equivalent to 8P+4E or 6P+8E, and 8P+0E is better if you want to concentrate on ST performance.
RichardsThe e cores are powerful ibtel should rename them to medium or middle cores.. because they are 80% performance of the big cores
When Lakefield (intel's first hybrid architecture) was announced at Hot Chips 2019, they officially called it as "Big-Bigger", not big-little.
Recently, more and more ARM chips are composed of three types of cores (prime-big-little), and the relationship between P-core and E-core is similar to that between ARM's prime and big cores.
Posted on Reply
#30
Frank_100
thomaskoellnWish they had recent AMD processors to compare it to, instead of 2 year-old hardware.
The two year old hardware still beats in real world (multi-core) work.
izyIm mostly working on android apps but im keeping a virtual machine (ubuntu) open at the same time , photoshop time to time and stuff like this (and sometimes light gaming , like WOW, very CPU bound) . I am rendering videos for youtube but the project is paused at the moment and is not the reason for the upgrade. The 3700x does a decent job , i can not complain , but i can get an 5800x for like 200E (and maybe the x3d one for 250-280) new and the 3700x will lose most of its value soon and i was thinking to upgrade.(the system i want to upgrade has a b450 tomahawk max , not sure if it can keep up with 5800x as it should but i think is still OK, maybe i can go for 5700x even)
Edit: I dont tend to upgrade very often and thats why i was thinking at the i5, it seems that its better than the 5800x ST and MT, and the extra ST performance can make things snappier. (i have to see at what price first anyway)
Upgrading for small gains is more work then it is worth. The time spent reinstalling software and configuring os's is lost forever. If you are going to upgrade, don't be cheap.
Make it worth your time. Build the most powerful system you can afford. Then use it until it dies. Trying to recover a few bucks selling old hardware is nice but unless you have a whole rack of servers, worrying about the depreciation in a three hundred dollar cpu may not be worth your time. Good luck.
Posted on Reply
#31
AlwaysHope
izyI am not even sure if regular people need so much CPU performance tbh. , better GPUs yes but CPU i dont think they need more than what is currently on the market, thow i might be wrong. (for gaming at least)
All one needs is enough performance to push their systems right up to the maximum refresh rate of their monitor(s) with gaming. That's all that matters to me anyway... :)
Posted on Reply
#32
Frank_100
Tiggeri could agree with that. i often wonder why some even buy 5950x CPU's, must be doing some heavy gaming :laugh:
DAW's usually like to have a core for each track of music. Drums alone can use up five or six tracks. Guitars can use up lots of tracks. Vocals - ditto.
16 cores can feel awful cramped.
Posted on Reply
#33
hs4
Frank_100DAW's usually like to have a core for each track of music. Drums alone can use up five or six tracks. Guitars can use up lots of tracks. Vocals - ditto.
16 cores can feel awful cramped.
The E core does not have HT/SMT, so there is no competition for resources between threads co-located on the same physical core, which also seems to make it suitable for DAW. A similar story is told by Youtube game streamers, who report fewer dropped frames compared to CPUs with uniform cores because the OS is oriented to distribute games and encoding to the P and E cores, respectively, reducing interference between the tasks.
Posted on Reply
#34
Gica
NDownyeah man, i mean what else they're supposed to be compared to lmaooo
If you want AMD now, you have no other offer. Probably for this reason they reduced the prices and launched processors that should have been launched at least a year ago.
Posted on Reply
#35
ratirt
not much difference with SM performance and the MT is expected to be higher due to more ecores but is it really that much higher? I would not say so.
I will wait for legit reviews and see how the new 13th gen Intel CPUs perform but I think there will not be much increase in the IPC rather the higher cache will boost some games apps using it.
I wonder what is the difference between the ASUS and ASrock? The cpu is the same but results vary noticeably. Power limits perhaps? More juice with the ASUS set-up. Wonder, how much more.
Posted on Reply
#36
Tsukiyomi91
now this is news. For the 13600K to keep up with a 5950X is pretty impressive, more so it's only an ES chip.
Posted on Reply
#37
ratirt
Tsukiyomi91now this is news. For the 13600K to keep up with a 5950X is pretty impressive, more so it's only an ES chip.
Yeah this is pretty good considering it is a 14c20t CPU but I really am gonna wait with celebration until I see what was the power limit are here since the Asrock and Asus Z690's boards bench differs a bit and these are probably the top models. I wonder how these results can be compared to a lower tiered boards.
Posted on Reply
#38
Tsukiyomi91
ratirtYeah this is pretty good considering it is a 14c20t CPU but I really am gonna wait with celebration until I see what was the power limit are here since the Asrock and Asus Z690's boards bench differs a bit and these are probably the top models. I wonder how these results can be compared to a lower tiered boards.
14c (10p, 4e)?!?! No wonder! =O
Posted on Reply
#39
persondb
Tsukiyomi9114c (10p, 4e)?!?! No wonder! =O
No it's 6p 8e
Posted on Reply
#40
Tsukiyomi91
persondbNo it's 6p 8e
oh. Still, that's pretty amazing. If the price is somewhere around $300 or less, then I think it probably be a better pick over the i7, assuming everything we know is true.
Posted on Reply
#41
zaku49
What I predict is going to happen, Intel will win in ST, AMD will win in MT. However, AMD will probably release 3D versions of their CPUs this year which will make them the best in gaming scenarios while consuming less watts.
Posted on Reply
#42
Unregistered
zaku49What I predict is going to happen, Intel will win in ST, AMD will win in MT. However, AMD will probably release 3D versions of their CPUs this year which will make them the best in gaming scenarios while consuming less watts.
Well i would say my 12700k only consumes 50 gaming but i'll get stomped on for having an old GPU so the power use gaming is meaningless.
Posted on Edit | Reply
#43
AleXXX666
izyIm mostly working on android apps but im keeping a virtual machine (ubuntu) open at the same time , photoshop time to time and stuff like this (and sometimes light gaming , like WOW, very CPU bound) . I am rendering videos for youtube but the project is paused at the moment and is not the reason for the upgrade. The 3700x does a decent job , i can not complain , but i can get an 5800x for like 200E (and maybe the x3d one for 250-280) new and the 3700x will lose most of its value soon and i was thinking to upgrade.(the system i want to upgrade has a b450 tomahawk max , not sure if it can keep up with 5800x as it should but i think is still OK, maybe i can go for 5700x even)
Edit: I dont tend to upgrade very often and thats why i was thinking at the i5, it seems that its better than the 5800x ST and MT, and the extra ST performance can make things snappier. (i have to see at what price first anyway)
wow, that's a decent and good mobo. sell the 3700 and get plain 5800x, 5700x is pointless in terms of performance and 3d is pointless in terms of price/performance ratio and performance diff from plain 5800x. also i'm concerned 3d will heat up like f*&k.
Posted on Reply
#44
springs113
TiggerSo lots of people not doing the same for Intel. At least there is a good idea what Intel has to offer.
Aww me2? I guess lol. In all honesty fanboys suck but to each their own. As of right now efficiency is king in my book and until I get solar that will always trump weak benchmark #s. Ppl in both camps need to just wait 'til zen 4 n raptor lake is released and then compare. Quite frankly amd is in the best position out of the two because of the DC market.
Posted on Reply
Add your own comment
Dec 18th, 2024 14:04 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts