A set of benchmarks (opens in new tab) have appeared on-line for what seems to be Intel’s Thirteenth-generation Core structure, generally known as Raptor Lake. The outcomes, which seem like from a submission to the SiSoftware database, seem to point out the chip’s hybrid structure offers a Ryzen 9 5900X a run for its cash, although there isn’t any direct apples-to-apples comparability to be made between the chips chosen on account of differing core counts. Take the information extra as a sign of what to anticipate, relatively than strong truth till official benchmarks might be verified.
|Specs||Intel Core i9-13900 8C+16c/32T (RPL)||Intel Core i9-12900 8C+8c/24T (ADL)||Intel Core i9-11900K 8C/16T (RKL)||AMD Ryzen 9 5900X 12C/24T (Zen3)|
|Structure||Raptor Cove + Gracemont / RaptorLake||Golden Cove + Gracemont / AlderLake||Cypress Cove / RocketLake||Zen3 / Vermeer|
|Cores / Threads||8C+16c / 32T||8C+8c / 24T||8C / 16T||2M / 12C / 24T|
|Rated Velocity (GHz)||?||2.4 massive / 1.8 LITTLE||3.5||3.7|
|All/Single Turbo Velocity (GHz)||3.7 massive / 2.76 LITTLE||5.0 massive / 3.8 LITTLE||4.8 – 5.3||4.5 – 4.8|
|Rated/Turbo Energy (W)||65 – 200 [=]||65 – 200||125 – 228||105 – 135|
|L1D / L1I Caches||8x 48kB/32kB + 16x 64kB/32kB||8x 48kB/32kB + 8x 64kB/32kB||8x 48kB 12-way / 8x 32kB 8-way||12x 32kB 8-way / 12x 32kB 8-way|
|L2 Caches||8x 2MB + 4x 4MB (32MB) [+2.3x]||8x 1.25MB + 2x 2MB (14MB)||8x 512kB 16-way (4MB)||12x 512kB 16-way (6MB)|
|L3 Cache(s)||36MB 16-way [+20%]||30MB 16-way||16MB 16-way||2x 32MB 16-way (64MB)|
|Particular Instruction Units||VNNI/256, SHA, VAES/256||VNNI/256, SHA, VAES/256||AVX512, VNNI/512, SHA, VAES/512||AVX2/FMA, SHA|
|SIMD Width / Items||2 x 256-bit||2x 256-bit||512-bit (1x FMA)||2x 256-bit|
|Worth / RRP (USD)||$549?||$519||$539||$549|
The benchmarks use the 5900X – a 12-core, 24-thread chip – alongside Intel’s eleventh and Twelfth-gen i9s, one an 8C/16T and the opposite a 16C(8+8)/24T hybrid design. The Raptor Lake contestant has extra of the effectivity cores than the Twelfth-gen CPU, boasting eight energy and 16 effectivity cores, for 32 threads general and certain the identical chip that leaked (opens in new tab) earlier this yr..
Picture 1 of 2
Little surprise, then, that it comes out in entrance. The interpretation of the numbers can also be type to the Raptor Lake chip: “With non-SIMD code, we see big efficiency uplift in each integer (previous’ Dhrystone) and floating-point (previous’ Whetstone) of 33% to 100% over ADL [Alder Lake] that assist push (method) previous even AMD’s Zen3 with 12 massive cores! The additional eight Little Atom cores vastly assist right here.
“Thus for regular, non-SIMD code – RPL [Raptor Lake] will carry out vastly and supply a fantastic improve over ADL.”
This seems to be good for gaming, however will not be the total story, because the second set of outcomes, utilizing vectorized SIMD code, present. Raptor Lake is available in second to Zen3 in three out of 4 exams: “With closely vectorized SIMD workloads RPL can not beat ADL by greater than 4-6% and thus usually even lose to the AVX512-enabled RKL [Rocket Lake] with simply 16 threads (vs. 32T on RPL). This exhibits simply how a lot AVX512 may also help regardless of the additional energy required – and right here it’s vastly missed. Nevertheless, AMD’s Zen3 (5900X) with [fewer] threads (24T) however 12 massive cores (vs. 8C + 16c) nonetheless beats RPL by an honest margin.”
The ‘respectable margin’ right here being nearly 900 factors within the Integer SIMD check, which sees the Thirteenth-gen Intel chip come final within the comparability.
The preview of the brand new chips and its conclusion are price studying, bringing collectively threads comparable to software program help for hybrid architectures, the impact of cache ranges on efficiency, and whether or not Socket 1700 PCs are price upgrading to, on condition that Meteor Lake will use a unique interface. In any other case, Intel’s Thirteenth-generation Raptor Lake structure looks as if the multi-threading monster we have been promised, however solely when fed the suitable sort of code.