Monday, June 13th 2022
Intel Core i9-13900 "Raptor Lake" Processor Gets a Preview
Intel is preparing to launch its 13th generation of desktop processors codenamed Raptor Lake. Succeeding Alder Lake, the 13th gen design will implement up to eight P-cores with 16 E-cores manufactured on Intel's improved 7+ technology node. Today, we got a performance preview from SiSoftware that has collected SiSoftware Sandra database scores of Intel Core i9-13900 Raptor Lake-S processor. They present an overview of a few benchmarks. Firstly, the SoC features 36 MB of unified L3 cache versus 30 MB in Alder Lake. With DDR5 memory running up to 5600 MT/s and PCIe 5.0, the SoC features the latest IO and memory standards. The big P-cores now lack AVX-512 and feature 2 MB of L2 cache per core. We see 4 MB of L2 cache for a cluster of small E-cores. An exciting addition to E-cores is the AVX/AVX2 support, which is a first for Atom cores.
Regarding testing, the author has collected a few tests that seemed appropriate to compare to the equivalent Alder Lake model. Starting with ALU/FPU tests that benchmark basic arithmetic tasks, Raptor Lake delivered 33% to 50% improvement over Alder Lake. The Raptor Lake design achieved this with 3.7 GHz P-Core and 2.76 GHz E-Core frequency. In vectorized and SIMD tests, the 13th gen design showed only 5% to 8% improvement over the previous generation. For more benchmarks and accurate results, we have to wait for TechPowerUp's test, which will be coming on the release day.
Source:
SiSoftware
Regarding testing, the author has collected a few tests that seemed appropriate to compare to the equivalent Alder Lake model. Starting with ALU/FPU tests that benchmark basic arithmetic tasks, Raptor Lake delivered 33% to 50% improvement over Alder Lake. The Raptor Lake design achieved this with 3.7 GHz P-Core and 2.76 GHz E-Core frequency. In vectorized and SIMD tests, the 13th gen design showed only 5% to 8% improvement over the previous generation. For more benchmarks and accurate results, we have to wait for TechPowerUp's test, which will be coming on the release day.
31 Comments on Intel Core i9-13900 "Raptor Lake" Processor Gets a Preview
Now I'm super excited to buy a 6800U laptop which was launched and reviewed five months ago. Currently the only one available is a dinky little 13" OLED one with piss-awful cooling and OLED burn-in. Where are my regular 14", 15.6" models without OLED that normally make up about 98% of the models people buy? Five f*cking months and nothing really even announced yet from Asus, Lenovo, Acer etc. They're still pushing 5000-series laptops with awful Vega graphics that we had 4.5 years ago.
If you went into a store after reading the launch reviews of the 3080 and the best they could offer you was and 4.5-year-old GTX 1080 from two generations ago, you'd be equally annoyed right? That's how it is with most AMD laptops and a solid majority of Intel laptops.
The advantages cannot be understated, because fewer steps towards a full chip is a boost to yields and to quality of the silicon. So an N6 process might very well clock a lot higher, for example. Its not just about density, its about consistency.
Take note of the power envelope on Intel's '7'. Its not pretty, it won't really clock that well above 4.8~5.0 Ghz. And it took them a LONG time to get there, even so, when they do, its grossly inefficient.
Similarly, compare TSMC's 7nm GPUs (AMD) versus Nvidia's Samsung 8nm GPUs. The former is built up out of (part) EUV. The latter is DUV. We're looking at gaps of 500mhz and more. The TL DR here is that even if you 'shrink' chips based on DUV patterning they won't quite work as you'd want. Its clearly the end of DUV to facilitate further 'true' shrinks that make economical sense.
That is in a nutshell, the whole reason Intel stagnated beyond 14nm. Going smaller on DUV is absolute hell. And it is why Intel is moving as it does today. They saw what TSMC realised first, and have admitted defeat by embracing EUV, and even pushing it harder than TSMC wrt testing new aperture sizes and stuff. They want that node advantage again.