Intel is shifting how it handles cache allocation in its upcoming Nova Lake processors, a change that could preserve performance in lower-tier models. Buyers looking at cutdown SKUs might find these chips retain more memory bandwidth than previously expected. This adjustment suggests Intel prioritizes cache consistency over strict core-count scaling in its high-performance mobile lineup.
Lower-tier chips retain full cache despite core reduction
The leak focuses on the Nova Lake-HX platform, which targets high-end gaming laptops and mobile workstations. According to hardware leaker Jaykihn, the architecture uses a specific method to create different product tiers. The company reportedly disables entire four-core E-core clusters to form lower-spec variants rather than disabling individual cores.
Nova Lake L3 Cache Specs
- L3 Cache (8P+12E+4 LP-E): 36MB
- L3 Cache (4P+4E+4 LP-E): 18MB
- L3 Cache (6P+12E+4 LP-E): 30MB
- Platform Variant: Nova Lake-HX
This strategy allows cutdown models to keep the L3 cache size of the fully enabled version. The 8P+12E+4 LP-E configuration now reportedly features 36MB of L3 cache, up from earlier rumors of 33MB. The 4P+4E+4 LP-E variant is expected to retain 18MB, an increase from the previously rumored 15MB. Meanwhile, the 6P+12E+4 LP-E configuration remains at 30MB.
Jaykihn explained that these disabled configurations keep the L3 cache configuration of the fully enabled variant. Intel has not confirmed these specifications or the final SKU lineup. We have been tracking Nova Lake closely — see our earlier coverage on Intel Nova Lake-S Leaked. The current data points to a cache allocation strategy dependent on how Intel creates SKUs rather than simple linear scaling.



Discussion
0 comments
Log in to join the thread with a thoughtful take, question, or correction.