Thursday, August 4th 2022
Intel "Raptor Lake" Core i9-13900 De-lidded, Reveals a 23% Larger Die than Alder Lake
An Intel Core "Raptor Lake" engineering sample was de-lidded by Expreview giving us a first look at what will be Intel's last monolithic silicon client processor before the company switches over to chiplets, with its next-generation "Meteor Lake." The chip de-lidded here is the i9-13900, which maxes out the "Raptor Lake-S" die, in featuring all 8 "Raptor Cove" P-cores and 16 "Gracemont" E-cores physically present on the die, along with 36 MB of shared L3 cache, and an iGPU based on the Xe-LP graphics architecture.
The "Raptor Lake-S" silicon is built on the same Intel 7 (10 nm Enhanced SuperFin) silicon fabrication node as "Alder Lake-S." The "Raptor Lake-S" (8P+16E) die measures 23.8 mm x 10.8 mm, or 257 mm² in area, which is 49 mm² more than that of the "Alder Lake-S" (8P+8E) die (around 209 mm²). The larger die area comes from not just the two additional E-core clusters, but also larger L2 caches for the E-core clusters (4 MB vs. 2 MB), and larger L2 caches for the P-cores (2 MB vs. 1.25 MB); besides the larger shared L3 cache (36 MB vs. 30 MB). The "Raptor Cove" P-core itself could be slightly larger than its "Golden Cove" predecessor.
Even with the larger die, there's plenty of vacant fiberglass substrate inside the IHS. Future client sockets such as the LGA1800 have an identical package size to the LGA1700, with the additional pin-count coming from shrinking the "courtyard" in the land-grid (the central empty space). This indicates that future MCM chips such as the "Meteor Lake" have plenty of real-estate on the substrate, and Intel can maintain package-size and cooler-compatibility across several more generations. That said, "Raptor Lake-S" will be a Socket LGA1700 processor, will work with Intel 600-series and upcoming 700-series chipset motherboards; but will likely not be compatible with future LGA1800 platforms.
Sources:
Expreview (BiliBili), VideoCardz
The "Raptor Lake-S" silicon is built on the same Intel 7 (10 nm Enhanced SuperFin) silicon fabrication node as "Alder Lake-S." The "Raptor Lake-S" (8P+16E) die measures 23.8 mm x 10.8 mm, or 257 mm² in area, which is 49 mm² more than that of the "Alder Lake-S" (8P+8E) die (around 209 mm²). The larger die area comes from not just the two additional E-core clusters, but also larger L2 caches for the E-core clusters (4 MB vs. 2 MB), and larger L2 caches for the P-cores (2 MB vs. 1.25 MB); besides the larger shared L3 cache (36 MB vs. 30 MB). The "Raptor Cove" P-core itself could be slightly larger than its "Golden Cove" predecessor.
42 Comments on Intel "Raptor Lake" Core i9-13900 De-lidded, Reveals a 23% Larger Die than Alder Lake
Wonder how the 13900k would perform with the extra 8 e-cores
Maybe 350W ?
Would probably end up faster overall and I would feel better about my thermal load.
2.5kW spikes? Transients are handled by capacitors in VRMs and capacitors on the PSU output (most are capable of spikes way higher than 2.5kW)
Edit: If I had a 4090Ti which took 400W running FurMark, and I had a next gen CPU that was overclocked in a way that made it take 400W running Prime95 small FFT, the PSU I'd buy would be an 850W model. Not 1200W.
Why? Because gaming the system would fluctuate between 400 and 650 watts, rarely peaking above that.
1200W PSUs putting out 100W aren't efficient - even Platinum and Titanium models. Why is the trend to buy PSUs that are 50-100% larger than they need to be? It just wastes power at idle, which is a lot of the time for most PCs.
The only reason I can think of for this nonsense is that a long time ago, 400W PSUs would be able to put out 200Won 12V and 200W on 5+3.3V.
But now most PSUs can put 95+% of their rated power output on 12V rails. So get over it! Stop buying huge power supplies that aren't needed!!!
Yes - IF TDP being kept the same, more surface area would mean lower thermal density.
No -
From TPU's own E-cores only 12900k test, 8 E-cores would do 70W of power consumption.
And it is unrealistic to assume Intel could magically squeeze out 70W of headroom out of the same manufacturing process while keeping the same TDP.
So something has to be sacrificed
Per core performance , or the 'Actual power consumption'
I think the 13900k would be both.
In stock locked TDP mode, its per core performance suffers
In Unlocked mode, your electricity bill suffers (and your room temperature).
If the user isn't doing tile-based rendering ( Cinebench ) or something similar (Editing software), he/she shouldn't bother with the 13900k and should go 13700k if all they need is 8 P-cores
Gaming and normal day applications will never utilize 16 e-cores.
Heavy multicore applications like virtualization (e.g. VMware) aren't friendly to this Hybrid Architecture anyway.
So Yes, only those doing Cinebench ( something similar ) as the sole purpose of the machine would actually need this CPU.
This thing just isn't made for anything else.
Intel be like if we can beat Nvidia in GPU ll try to match their GPU TDP with our silicon xD
GGWP with overclocking folks with the thermals - I can feel the pain in the ass while doing OC to a 13900k or ks.
Also, 13900ks might come 6.0Ghz out of the box :)
Yes - not everyone care about value. There are people buying 5950x when all they do are gaming.
No -
Most of the crowd DO care about value and are significantly more than those who don't.
Everyone else DO care about value, power consumption and heat.
And those factors are heavily hindering the potential of the 13900k.
65 Watt CPU is all you need for gaming anyways.
It's not catastrophic, but intel will probably need the price hike they announced to keep their margin at the same level.
But those number start to get really low. To put that in perspective, AMD on a process with similar defect rate would get 87% and 706 good ccd per wafer.
That is just the CCD and they also need the I/O die and all that. but that clearly show the potential of smaller chiplets for yields.
For the I/O die, they would have 78% yield and 359 dies per wafer. So for 2 wafer, and assuming all chips meet the clock, intel would get 260 CPUs and AMD would get 350 possible 7950x CPU.
But for power consumption. I think in game, where most people will use it, the power consumption difference between Zen 4 and RPL will be marginal.
And if you need a full core load, the power consumption / performance time might be worth the tradeoff.
I am more concern with the GPU power since I game most of the time on my PC and the Power usage is sustained.