Sunday, March 22nd 2020

Intel Rocket Lake-S Platform Detailed, Features PCIe 4.0 and Xe Graphics

Intel's upcoming Rocket Lake-S desktop platform is expected to arrive sometime later this year, however, we didn't have any concrete details on what will it bring. Thanks to the exclusive information obtained by VideoCardz'es sources at Intel, there are some more details regarding the RKL-S platform. To start, the RKL-S platform is based on a 500-series chipset. This is an iteration of the upcoming 400-series chipset, and it features many platform improvements. The 500-series chipset based motherboards will supposedly have an LGA 1200 socket, which is an improvement in pin count compared to LGA 1151 socket found on 300 series chipset.

The main improvement is the CPU core itself, which is supposedly a 14 nm adaptation of Tiger Lake-U based on Willow Cove core. This design is representing a backport of IP to an older manufacturing node, which results in bigger die space due to larger node used. When it comes to the platform improvements, it will support the long-awaited PCIe 4.0 connection already present on competing platforms from AMD. It will enable much faster SSD speeds as there are already PCIe 4.0 NVMe devices that run at 7 GB/s speeds. With RKL-S, there will be 20 PCIe 4.0 lanes present, where four would go to the NVMe SSD and 16 would go to the PCIe slots from GPUs. Another interesting feature of the RKL-S is the addition of Xe graphics found on the CPU die, meant as iGPU. Supposedly based on Gen12 graphics, it will bring support for HDMI 2.0b and DisplayPort 1.4a connectors.
Intel Rocket Lake-S Platform
Some things like Direct Media Interface (DMI) will double the bandwidth and now there will be eight links present, compared to four of the previous platforms. Announced at CES 2020, ThunderBolt 4 will also be present along with USB 3.2 20G. Additionally, Intel Software Guard Extensions (SGX) have been removed to improve the security of the platform, as the SGX has proved to be quite vulnerable to many kinds of attacks and exploits. There are some updated media encoding standards as well, like 12-bit AV1/HEVC and E2E compression. Source: VideoCardz
Add your own comment

113 Comments on Intel Rocket Lake-S Platform Detailed, Features PCIe 4.0 and Xe Graphics

#51
notb
@ARF
OMG, you're on fire man. And I'm called a fanboy around here...

Can you point me to the source of these graphs?
I would like to meet the person who got more fps from GT2 than from GT3e.

"Creators unleashed" slide figures don't make much sense as well - even just comparing Intel CPUs.
Posted on Reply
#54
bug
Valantar
Let's wait for some independent/third party benchmarks, yeah? I'm not doubting the new APUs will kick currently available Intel chips' butt, but vendor-provided benchmarks are always fishy.

Also, how is this relevant to the topic, isn't the topic about upcoming desktop chips?
Dude, after 2,000 posts you're still stuck on "relevant"? :D
Posted on Reply
#55
Cheeseball
Not a Potato
ARF
Exactly double the performance. You need to have the software right optimised.


https://www.notebookcheck.net/AMD-Ryzen-7-4800U-Laptop-Processor-Benchmarks-and-Specs.449937.0.html


https://www.notebookcheck.net/Intel-Core-i7-1065G7-Laptop-Processor-Ice-Lake.423851.0.html
I would hope the 4800U is near-double the performance of the i7-1065G7, considering that the AMD is a 8C/16T and the Intel is a 4C/8T.

Also, I don't see why Ice Lake is "terrible". It is a really good mobile x86 (e.g. low-power) chip. Intel just failed at bringing it to desktop, that's all.
Posted on Reply
#56
ARF
In the same power envelope! This means Renoir has got up to 100% higher performance per watt!

If Rocket Lake gets released in maximum 8-core flavour, imagine how competitive it will be against Vermeer with at least 16 cores, potentially even more.
Posted on Reply
#57
bug
ARF
In the same power envelope! This means Renoir has got up to 100% higher performance per watt!

If Rocket Lake gets released in maximum 8-core flavour, imagine how competitive it will be against Vermeer with at least 16 cores, potentially even more.
Wth are you smoking? Zen2 doesn't have 100% better perf/W than current Intel chips, how would making it into a mobile part do what you seem to think it does?

Would stopping the guessing and speculation be an option for you?
Posted on Reply
#58
ARF
bug
Wth are you smoking? Zen2 doesn't have 100% better perf/W than current Intel chips, how would making it into a mobile part do what you seem to think it does?

Would stopping the guessing and speculation be an option for you?
Are you hurt or what?
Ryzen 7 4800U at 15W achieves 3300 points in Cinebench 20.
Core i7-1065G7 at 15W achieves 1650 points in Cinebench 20.

What is the performance per watt difference according to you?
Posted on Reply
#59
bug
ARF
Are you hurt or what?
Ryzen 7 4800U at 15W achieves 3300 points in Cinebench 20.
Core i7-1065G7 at 15W achieves 1650 points in Cinebench 20.

What is the performance per watt difference according to you?
From your own link:
The TDP of the APU is specified at 15 Watt (default) and can be configured from 10 to 25 Watt by the laptop vendor.
Posted on Reply
#60
ARF
I've got another CB R20 result for the Ryzen 7 4800U and it is 3100 points https://www.cpu-monkey.com/en/cpu-amd_ryzen_7_4800u-1142

Also, let's compare Ryzen 7 3700X at 65W - 22792 points in PassMark https://www.cpubenchmark.net/cpu.php?cpu=AMD+Ryzen+7+3700X&id=3485
Core i7-9700K at 65W - 14554 points in PassMark https://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i7-9700+@+3.00GHz&id=3477

Difference in performance per watt 57%.

Ryzen 7 3700X at 65W - 4760 points in CB R20 https://www.guru3d.com/articles-pages/amd-ryzen-7-3700x-ryzen-9-3900x-review,10.html
Core i7-9700K at 65W - 3656 points in CB R20 https://www.cpu-monkey.com/en/cpu-intel_core_i7_9700k-888

Difference in performance per watt 30%.
Posted on Reply
#61
bug
Ok, you've got numbers to prove anything. We get it.
Posted on Reply
#62
ARF
bug
Ok, you've got numbers to prove anything. We get it.
It may be a hint that the N10 process is broken..
Posted on Reply
#63
notb
ARF
In the same power envelope! This means Renoir has got up to 100% higher performance per watt!
It's apparent that you know very little (which is absolutely fine), but I'm not sure if you're willing to learn something or not.
Because if you do, I'll try to be patient and helpful. If not, I'll keep having fun.

Clearly, you read a lot of benchmark results (more or less credible...). I hope you're using that PC for something else. :)
BTW, I'm waiting for source of those earlier graphs. Honestly.
ARF
It may be a hint that the N10 process is broken..
There is no N10 process.
Posted on Reply
#64
ARF
notb
It's apparent that you know very little
Better than you that still knows nothing :laugh:
Posted on Reply
#65
notb
ARF
Better than you that still knows nothing :laugh:
I'll know more when you finally share the source of the graphs you used in this thread.
So?
Posted on Reply
#66
bug
notb
I'll know more when you finally share the source of the graphs you used in this thread.
So?
Leave him be. You asked whether he was willing to learn, he answered. There's nothing more you, me or anyone else can do here.
Posted on Reply
#67
notb
bug
Leave him be. You asked whether he was willing to learn, he answered. There's nothing more you, me or anyone else can do here.
But he's extremely active - flooding more and more threads with random comments (graphs, data). Don't you care at all? :)
Posted on Reply
#68
bug
notb
But he's extremely active - flooding more and more threads with random comments (graphs, data). Don't you care at all? :)
I asked him a few questions, so that everybody can see he's clueless. Now I'm moving on. I'm not going to address each of his rants.
Posted on Reply
#69
Valantar
notb
I'll know more when you finally share the source of the graphs you used in this thread.
So?
If you are talking about the graphs in post #50 they are from AMD's Renoir launch presentation slides. As such not the most trustworthy source, hence my suggestion to wait for independent reviews above. We can't know the entire circumstances around the testing done for various leaked test results (such as TDP configuration) which adds too much uncertainty. Nonetheless we know that Ice lake has slightly higher IPC (~18% over CFL, which Zen 2 beats by ~7% (AnandTech), so ICL should beat it by ~10%) but Renoir clocks significantly higher as demonstrated by base clock numbers in the same TDP.

Among the more trustworthy leaks I've seen: NoteBookCheck apparently failed to exclude their test data for the Lenovo Yoga Slim 7 with the Ryzen 7 4800U from their searchable results database, meaning it could be added to comparsion tables in any laptop review. Sadly the person screenshoting the results didn't add an Ice Lake machine to all of the comparison graphs, so some of them are rather AMD-heavy. Interesting nonetheless. Some sources report that laptop as configured to 25W, while others say it's a 15W but with an extended 25W turbo window when thermals allow for it - I guess we'll see which it is when the NBC review is published.
Posted on Reply
#70
notb
Valantar
If you are talking about the graphs in post #50 they are from AMD's Renoir launch presentation slides.
From the CES keynote? Nope. They only shown 1065G7 vs 4800U. Values are the same for these 2 SoCs.

I've found this:
https://www.techspot.com/news/84400-amd-details-ryzen-mobile-4000-performance-architecture-features.html
So the slides are from something called "AMD Tech Day".

So I tried to get that presentation from AMD's IR site, but it wasn't added:
https://ir.amd.com/events-and-presentations/past-events
Nothing here either:
https://ir.amd.com/search?query=tech+day&op=Search

Basically, the figures are weird and unrealistic - for example with GT3e being beaten in games by GT2, which has roughly half of the cores and performance.
And the results aren't coherent between games, so where different laptops used or what? Is this cherry-picked from multiple runs?
Bottom of the slide:
"See endnotes RM3-227. Results my vary."
Yes. "MY":
https://static.techspot.com/images2/news/bigimage/2020/03/2020-03-16-image-27.jpg
Posted on Reply
#71
Valantar
notb
From the CES keynote? Nope. They only shown 1065G7 vs 4800U. Values are the same for these 2 SoCs.

I've found this:
https://www.techspot.com/news/84400-amd-details-ryzen-mobile-4000-performance-architecture-features.html
So the slides are from something called "AMD Tech Day".

So I tried to get that presentation from AMD's IR site, but it wasn't added:
https://ir.amd.com/events-and-presentations/past-events
Nothing here either:
https://ir.amd.com/search?query=tech+day&op=Search

Basically, the figures are weird and unrealistic - for example with GT3e being beaten in games by GT2, which has roughly half of the cores and performance.
And the results aren't coherent between games, so where different laptops used or what? Is this cherry-picked from multiple runs?
Bottom of the slide:
"See endnotes RM3-227. Results my vary."
Yes. "MY":
https://static.techspot.com/images2/news/bigimage/2020/03/2020-03-16-image-27.jpg
Sorry, guess I got those two releases mixed up. Beyond that, you did see that I said that I don't see these as particularly valuable and that we should wait for independent reviews? You're barking up the wrong tree, my friend. Chill.
Posted on Reply
#72
ARF
notb
Basically, the figures are weird and unrealistic - for example with GT3e being beaten in games by GT2, which has roughly half of the cores and performance.
That can be due to everything - different notebook cooling design, different configurations, different conditions which could have caused the i7-10710U to throttle its iGPU speed quite a fair bit.

In Audio LAME, the 8-thread Ryzen 7 4700U is faster than the 16-thread Ryzen 7 4800U.
How do you explain it ?
Posted on Reply
#73
notb
ARF
That can be due to everything - different notebook cooling design, different configurations, different conditions which could have caused the i7-10710U to throttle its iGPU speed quite a fair bit.
That's why I wanted to get to the full presentation and read the footnote.
In Audio LAME, the 8-thread Ryzen 7 4700U is faster than the 16-thread Ryzen 7 4800U.
How do you explain it ?
Why would I explain that? :o
I'm contesting these slides. You're using them to support your theories.
Posted on Reply
#74
Valantar
ARF
That can be due to everything - different notebook cooling design, different configurations, different conditions which could have caused the i7-10710U to throttle its iGPU speed quite a fair bit.
Which is exactly why we ought to wait for independent reviews so that we can a) be aware of any configuration differences, and b) try to account for them when comparing.
Posted on Reply
Add your own comment