News Rocket Lake Core i9-10900K and i7-11700K Blast Through CPU-Z Benchmarks

VforV

Respectable
BANNED
Oct 9, 2019
578
287
2,270
so be careful in trusting these results
I am!

As in careful, so I don't trust these results... not as much as the score per se, as to the fact that they mean nothing when it comes to gaming benchmarks.

It's still intel Skippy Lake. At best will have the same performance as Zen3 in gaming, but will run much more inefficient power wise and have less cores. And the prices will be a good laugh too.
 

Makaveli

Splendid
Those 11700k numbers look fishy for ST. It has lower boost clocks than the 11900K yet scores more.

Previous leaks




I know the last bios update helped with performance but I can't see single core going from 673 to 716 after a bios update.

Here is my own on 5800X my system.

 
Last edited:
Those 11700k numbers look fishy for ST. It has lower boost clocks than the 11900K yet scores more.

Previous leaks
The CPU-z scores are not from some review so standard clocks don't mean anything, you are showing the 11900k@5,2Ghz getting the same ballpark figure as these results and the 10700k can be overclocked to the same clocks, it would be a 200Mhz overclock on single core.
 

Makaveli

Splendid
The CPU-z scores are not from some review so standard clocks don't mean anything, you are showing the 11900k@5,2Ghz getting the same ballpark figure as these results and the 10700k can be overclocked to the same clocks, it would be a 200Mhz overclock on single core.

Yes the Chart above minus the 11900k at 5.2 look to all be stock numbers.

i7-11700K @ 5.30GHz

While this one posted today are all overclocked numbers. However there is no mention in the article that the numbers are from an overclocked system. If you just read the link here and don't look at whats on twitter it just looks like stock numbers. The comparison chart is basically overclocked RL numbers vs others cpu's at stock.
 

everettfsargent

Honorable
Oct 13, 2017
130
35
10,610
Yes the Chart above minus the 11900k at 5.2 look to all be stock numbers.

i7-11700K @ 5.30GHz

While this one posted today are all overclocked numbers. However there is no mention in the article that the numbers are from an overclocked system. If you just read the link here and don't look at whats on twitter it just looks like stock numbers. The comparison chart is basically overclocked RL numbers vs others cpu's at stock.
Several things are odd at my end. Like no AMD 5000 series CPU's in the overall ranking chart, the highest I see is a 3950X.

Also it looks like single core maximum OC as there is no multicore score for the i7-11700K (at 5.3GHz) and no single or multicore scores for the i9-11900K at 5.9GHz. Like saying I can boot into W10 at 5.9GHz and start CPU-Z and I can boot into W10 and run a single thread bench at 5.3GHz in CPU-Z (getting a 714 in multicore strongly suggests a single thread is all that CPU-Z could see). The i9-11900K score looks legit though (4.7GHz all core boost).

I am thinking the 5800X is tuned for all cores (or stock at say 4.45GHz) and not a single maximum thread clock (which might go to ~4.85Ghz according to THW 5800X review).
 
Last edited:
  • Like
Reactions: Makaveli

Makaveli

Splendid
I am thinking the 5800X is tuned for all cores (or stock at say 4.45GHz) and not a single maximum thread clock (which might go to ~4.85Ghz according to THW 5800X review).

Agreed pretty much what I've seen also for most samples at stock.

With PBO2 + CO I can get my chip to hit 5.05Ghz single core boost and maximum all core clocks depend on temps. I usually see 4.55-4.60Ghz and i'm using an AIO. Custom loops from what i've seen will hit the 4.65-4.70Ghz range in all core clocks.
 

danlw

Distinguished
Feb 22, 2009
137
22
18,695
We don't know why the 5800X makes up all its performance losses from the single-threaded test in the multi-threaded test, but it could be due to reduced turbo frequencies on the Core i9 part, as well as architectural differences between the two chips.

IIRC from Anandtech's review, they were speculating it was because the larger physical die size of Intel's 14nm part caused intra-core communication to be slower compared to AMD's 7nm parts. They reported core-to-core latency on the order of 28-30ns for the 11700 vs 19-24ns for their 10700k part. I couldn't find core-to-core latency for the 5800x, but the 5950 showed latencies on the order of 18ns for certain blocks of cores. And so it looks like the probable reason for the loss of performance when going multi-core on the intel part is the higher latency between cores due to the process node being bigger.

Why is the inter core latency slower with the 11700 vs the 10700? Probably something to do with the fact that the 11700k was designed and optimized for a 7nm node but had to be back-ported to 14nm.
 
  • Like
Reactions: Makaveli and VforV

watzupken

Reputable
Mar 16, 2020
1,022
516
6,070
I am not sure what it the value add of this CPUZ benchmark result when there is a proper review done by Anandtech that already painted a clear picture of Rocket Lake's performance and efficiency. And as a few already pointed that the numbers don't seem correct/ consistent.
 
  • Like
Reactions: Makaveli and VforV
As others have pointed out, something in these results just seem to be "off". The biggest problem with CPU-Z is the validation / benchmark test doesn't really push a system very hard making unrealistic benchmark scores possible. I've been overclocking every system I've ever owned (going way back to Core 2 Duo) and use to get into the "how high can I validate" a long time ago, but years ago I stopped because really (unless your going to a LN2 world record) it doesn't mean anything. Now when I overclock a system I first check for total system stability, run a suite of benchmarks and stress tests to ensure stability then and only then validate the results with a full suite of benchmarks to show a much fuller overclock result. At the end of the day a high overclock (AMD or Intel) means nothing if all you can do is load into windows long enough to validate it.

I have my system currently rendering a project in the background (running a CPU-Z benchmark now would be fruitless) but I have this validation from when I first obtained full stability on my Ryzen 5900X. This benchmark was run on a Asus Dark Hero motherboard so I am utilizing the dynamic OC switcher with PBO curve optimization for single core. My single core boost is 5Ghz and for this run my all core overclock was 4.65Ghz (my 24/7 overclock, my max overclock is 4.7Ghz).

https://valid.x86.fr/5d32t2

For reference with this overclock I get ~650 in CB R20 single core and ~9240 in multicore. I am at thermal limitation with my setup as at my max all core overclock (4.7Ghz) I am hitting temps of 86C, at 4.65Ghz all core my temps spike around 80C which I believe is much safer long term. I am on full air cooling with a Noctua NH-U14S, so someone with a better AIO or custom loop could get better overclocking results.

I believe that the earlier CPU-Z benchmarks we have seen from Rocket Lake are much more realistic. The 11900K scoring ~700, 706 in single core @ 5.2 - 5.3Ghz would be in line with the IPC advantage Zen 3 still has which is exactly what AnandTech found in their early review of a retail 11700K. In single core a Zen 3 processor @ 5Ghz will be more or less equal to Rocket Lake @ 5.2 - 5.3Ghz in some benchmarks. I also expect to see that benchmarks that push the processors harder will show Zen 3 retain their max single core boosts longer than Intel's thermal velocity boost. In these reported benchmarks we don't know what the system configuration was, what cooling was used, and what overclocks were applied. I would make an educated guess that the Rocket Lake processors were hitting single core clocks of 5.3 - 5.4Ghz and were more than likely unstable past running a quickie validation. Saying I am already getting ~700 CPU-Z single core scores with a fully stable overclock I'm sure if I were to push my system past stability I could hit single core boosts near 5.05, maybe 5.1Ghz and match or beat these posted CPU-Z results for Rocket Lake. I just don't see any reason to as if the system is not stable and will crash under any real workload, then what is the point?

We will have to wait for the real unbiased reviews, but I fully expect at the end of the month we will see Rocket Lake's flagship CPUs obtaining relative parity with Zen 3 in single core applications and gaming while being outperformed in multi core workloads (AMD's SMT is simply superior to Intel's hyperthreading). The 11700K and Ryzen 5800X will be the closest match up for single and multi core performance, but if Intel prices their 11900K in the same price bracket as the 5900X its going to be bloodbath. The two processors will more than likely have near parity in gaming and single core execution (depending on Rocket Lake's latency and achievable sustained thermal velocity boost) but the 5900X will decimate the 11900K in all multi core workloads. While I believe Intel has made the best out of the bad situation they find themselves in I think there is a reason why we have such relatively easy availability of retail processors a month before unbiased reviews can be published.
 
Last edited:
  • Like
Reactions: Makaveli
IIRC from Anandtech's review, they were speculating it was because the larger physical die size of Intel's 14nm part caused intra-core communication to be slower compared to AMD's 7nm parts. They reported core-to-core latency on the order of 28-30ns for the 11700 vs 19-24ns for their 10700k part. I couldn't find core-to-core latency for the 5800x, but the 5950 showed latencies on the order of 18ns for certain blocks of cores. And so it looks like the probable reason for the loss of performance when going multi-core on the intel part is the higher latency between cores due to the process node being bigger.

Why is the inter core latency slower with the 11700 vs the 10700? Probably something to do with the fact that the 11700k was designed and optimized for a 7nm node but had to be back-ported to 14nm.

There is no doubt that while Intel was able to increase IPC they suffered a regression in latency and core count. It will be interesting to see what effects this has on gaming. The biggest issues facing Ryzen through the 3000 series was the latency penalty. The 3000 series processors performance in gaming was handicapped with the latency penalty and although they outperformed their Intel counterparts in almost every other dynamic, they still came in second best in gaming. Intel is going to try to sell Rocket Lake as the best gaming processor because they know they have no chance in matching Zen 3 in productivity. Rocket Lake won't even be able to match Comet Lake in productivity due to core regression. With the latency penalty that Rocket Lake has vs Comet Lake and Zen 3 how will its gaming performance actually be...
 
  • Like
Reactions: Makaveli
I am not sure what it the value add of this CPUZ benchmark result when there is a proper review done by Anandtech that already painted a clear picture of Rocket Lake's performance and efficiency. And as a few already pointed that the numbers don't seem correct/ consistent.

Everything about Rocket Lake's release up till this point has been extremely "off", and Intel wants "leaks" like this to build hype. All the early leaks were showing CPU-Z and Geekbench dominance of Rocket Lake in single core. Then we have a German and a French retailer sell their 11700K retail processors a month before Intel's review embargo will lift, I suspected then it was with Intel's blessing and they wanted to sell as many processors as they could out of the mainstream's sight before unbiased reviews went live. Now we are seeing what I believe Intel wanted, people overclocking their processors to unstable levels so they can validate high scores to post, build the hype... Of course it also backfired on Intel when AnandTech bought one of the early released processors and did a full early review on it. I think that review is very telling in how Intel has approached the release of Rocket Lake, they are concerned about unbiased full reviews and want as many CPU-Z and Geekbench quick run benchmarks as possible to "leak" out and build hype for a product they know isn't going to outperform AMD (at best it might match it, and then only in single core execution).
 
  • Like
Reactions: Makaveli
build hype for a product they know isn't going to outperform AMD (at best it might match it, and then only in single core execution).
Prices are confirmed, the equivalent of the 5800x is going to be the 11700kf at an msrp of $374 making it a good choice against the $450 5800x ,weaker but also cheaper.
If you go with the "locked" 11700f at $298 and enable MCE it will run as fast if not faster than the stock 11700k at even cheaper with only the power being a concern.
View: https://www.youtube.com/watch?v=-pZQz2sp5yY
 
Prices are confirmed, the equivalent of the 5800x is going to be the 11700kf at an msrp of $374 making it a good choice against the $450 5800x ,weaker but also cheaper.
If you go with the "locked" 11700f at $298 and enable MCE it will run as fast if not faster than the stock 11700k at even cheaper with only the power being a concern.
View: https://www.youtube.com/watch?v=-pZQz2sp5yY

Newegg is listing the 11700K for $419.
https://www.newegg.com/intel-core-i..._11700k-_-19-118-233-_-Product&quicklink=true
They are listing the 11900K for a whopping $614:oops:
https://www.newegg.com/intel-core-i..._11900K-_-19-118-231-_-Product&quicklink=true
I haven't seen prices for the the 11700KF yet, but it should be a reasonable value, however when looking at a full system build you can build a Ryzen 5800X system with older and cheaper X470, B450 and B550 boards whereas you will have to buy a more expensive motherboard for Rocket Lake... Full system builds will probably be similar in price between a 5800X and 11700KF depending on the motherboards used.

The elephant in the room is of course the 11900K. $614 for a 8 core processor!! The 5900X and 11900K will have near parity in single core performance and gaming. The 5900X will absolutely obliterate the 11900K in multi core workloads (12 core 5900X vs 8 core 11900K), yet The 11900K is retailing for $614 vs the $550 5900X. It defies all logic. No wonder Intel is "launching" these processors today, two weeks before they will lift their own review embargo. They want people to buy based on their cherry picked benchmarks and CPU-Z / Geekbench scores where we don't know what settings were used. Sell all you can before unbiased reviews smack you down... Biggest rope-a-dope Intel has tried in awhile.
 
First days until everybody has enough supply is always much more expensive, prices will normalize.
I haven't seen prices for the the 11700KF yet, but it should be a reasonable value, however when looking at a full system build you can build a Ryzen 5800X system with older and cheaper X470, B450 and B550 boards whereas you will have to buy a more expensive motherboard for Rocket Lake... Full system builds will probably be similar in price between a 5800X and 11700KF depending on the motherboards used.
You only need a good mobo if you plan on overclocking, running stock and even MCE can be done on very cheap mobos without any problems, only top noch mobos can handle the crazy TDPs you see in reviews, cheaper mobos will run much more reasonable settings.
The elephant in the room is of course the 11900K. $614 for a 8 core processor!! The 5900X and 11900K will have near parity in single core performance and gaming. The 5900X will absolutely obliterate the 11900K in multi core workloads (12 core 5900X vs 8 core 11900K), yet The 11900K is retailing for $614 vs the $550 5900X. It defies all logic. No wonder Intel is "launching" these processors today, two weeks before they will lift their own review embargo. They want people to buy based on their cherry picked benchmarks and CPU-Z / Geekbench scores where we don't know what settings were used. Sell all you can before unbiased reviews smack you down... Biggest rope-a-dope Intel has tried in awhile.
The way things are right now intel would be stupid to not charge more for the same thing, it's not like people can buy anything else.
Performance doesn't even matter right now, they will sell as many as they can make.
Also all the locked CPUs and everything i5 and below has stayed at the same price and overclocking has been a waste of time and money anyway for years now so yeah, let anybody that still insist on O/C pay premium for it.
Both ryzen and intel for the last few gens now you can only overclock to below single core clocks, so what's even the point anymore?
 
The way things are right now intel would be stupid to not charge more for the same thing, it's not like people can buy anything else.
Performance doesn't even matter right now, they will sell as many as they can make.
Also all the locked CPUs and everything i5 and below has stayed at the same price and overclocking has been a waste of time and money anyway for years now so yeah, let anybody that still insist on O/C pay premium for it.
Both ryzen and intel for the last few gens now you can only overclock to below single core clocks, so what's even the point anymore?

Stock has been a challenging issue this year, no doubt there. Took me what seemed like forever to get my 5900X and RTX 3080, but I wasn't going to pay over MSRP. Its challenging, but not impossible. For Intel to again take full advantage of its customers by overcharging like this is disgraceful. AMD didn't raise their MSRP because there was high demand, they didn't set the price the scalpers are charging and they don't have much control over the issues TSMC is facing right now. Intel is setting its price high because they know there is a processor shortage and they are acting like a scalper themselves. $540 (that they know will be more like $600 retail) for an 8 core processor that at best matches a 12 core $550 processor in single core (while being demolished in multi core) is doing what they have done best for years... Price gouging their own faithful customers.

As far as overclocking... Its definitely a mixed bag now... Gone are the good old days of getting an extra 800Mhz - even a full 1Ghz extra "free" performance... That's not to say that today's processors aren't better, they are much better, but overclocking now generally has diminishing returns. I have personally overclocked my 5900X, I have one of the best motherboards, quality RAM and a powerful EVGA Supernova PSU, so why not get every last bit of performance possible out of the build... For me it comes down to with it running stock "project X" will take an hour to complete, but overclocked I can get it done in 45, 50 min... add up enough projects in a day and you can be saving yourself a couple hours a day... I will agree with you though, it feels more like fine tuning than overclocking with these new processors.
 
Stock has been a challenging issue this year, no doubt there. Took me what seemed like forever to get my 5900X and RTX 3080, but I wasn't going to pay over MSRP. Its challenging, but not impossible. For Intel to again take full advantage of its customers by overcharging like this is disgraceful. AMD didn't raise their MSRP because there was high demand, they didn't set the price the scalpers are charging and they don't have much control over the issues TSMC is facing right now. Intel is setting its price high because they know there is a processor shortage and they are acting like a scalper themselves. $540 (that they know will be more like $600 retail) for an 8 core processor that at best matches a 12 core $550 processor in single core (while being demolished in multi core) is doing what they have done best for years... Price gouging their own faithful customers.
With the extremely small benefits of the 11900k and the much lower prices of the other CPUs, especially the "locked" ones are the same price as last gen, intel is basically forcing you to buy the cheaper options, it's also better for them, cheaper to make and less risk.
It's not like they only released the 11900k forcing people to buy only that, you know like how AMD did it, releasing only the top margin chips and no low tiers at all, although intel didn't release any i3 or below either.
Both companies are the same crap.