More Cores, Higher Prices: Intel’s Server Revenue Expected to Increase

Intel’s data center business is recovering, showing healthy margins despite growing competitive pressure. 

The healthy margins result from higher average selling prices of its server chips, which are generating larger profits. 

And the prices will likely go up as the company packs more cores and accelerators into its upcoming server chips. 

“Favorable customer mix along with strong adoption of newer products with higher core density led to record Xeon ASPs in Q3,” said David Zinsner, chief financial officer for Intel, during the third-quarter earnings call held in late October. 

That could partially explain Intel’s decision last month to suddenly double the core count in its Sierra Forest dense server chips to 288 cores just a few months before its official release.  

Sierra Forest was originally configured with 144 cores, but doubling the core count will allow Intel to charge customers more for the chip. 

The server chip prices started rising with Sapphire Rapids, contributing to Intel’s healthy operating profits for the Data Center and AI group. 

The current price of the fastest fourth-gen Xeon chip (Sapphire Rapids) is approximately double that of the fastest third-gen Xeon chip (Ice Lake). 

The fastest Sapphire Rapids chip with 60 cores is priced at $17,000. The fastest third-generation Xeon with 48 cores made on the 10-nm process is currently priced at $9,359, a decline from $10,000 when it was introduced in 2021. 

Xeon chip prices will likely go even higher with next-generation chips, with core counts in next-generation Xeons — Emerald Rapids, Granite Rapids – also expected to increase.