• Happy holidays, folks! Thanks to each and every one of you for being part of the Tom's Hardware community!

News Intel Core Ultra 7 258V mobile processor matches top Ryzen 'Phoenix' chips in BAPCO performance charts

Rumors indicate that the Core Ultra 7 258V will only have four P-Cores and four E-Cores, for a total of eight cores and eight threads. That means it will have fewer cores and threads than the Core Ultra 5 125H, which has four P-Cores (eight threads) and eight E-Cores.

Wrong, the Core Ultra 5 125H sports a total of 14 cores. That is 4 Performance + 8 Efficient + 2 LP E cores in total. And total thread count is 18.

So the new Core Ultra 7 258V lunar lake chip definitely has fewer cores/threads, but the entire Lunar Lake SoC lineup is targeting low-power mobile market segment. Focusing on enhancing power efficiency and optimizing performance across the board.
 
Wrong, the Core Ultra 5 125H sports a total of 14 cores. That is 4 Performance + 8 Efficient + 2 LP E cores in total. And total thread count is 18.

Yep, people always forget the Lonely Island.... err... Low Power Island.

jsI5Vp.gif
 
Yep, people always forget the Lonely Island.... err... Low Power Island.

Intel previously planned to use actual "LP E" cores on Lunar Lake as well, in place of the usual E cores which we are now getting on the lineup. But they later dropped the idea.

An old slide from Intel. Here they seem to mention 4 Low Power E core cluster.

Assuming these weren't the standard E-Cores for Lunar Lake CPUs, then Intel could have achieved even bigger generational "Performance Per Watt" gains with these thin and light series of laptop chips.


TU4A8xa.jpeg
 
  • Like
Reactions: KnightShadey
Wrong, the Core Ultra 5 125H sports a total of 14 cores. That is 4 Performance + 8 Efficient + 2 LP E cores in total. And total thread count is 18.
I thought that the LP E-cores were not visible to the end user and because of the way that they work and performance they have that the consensus of opinion is that they are not realistically counted, even by Intel as a "core" that they advertise.
 
I thought that the LP E-cores were not visible to the end user and because of the way that they work and performance they have that the consensus of opinion is that they are not realistically counted, even by Intel as a "core" that they advertise.

For Intel they actually counted as cores on the SoC tile when it came to handle deep background tasks in idle/sleep mode. When this was happening, Intel's' thread director could easily turn off the inactive CPU tile as well.
 
Why would anyone take bapco benchmarks seriously after they were found favoring intel cpus? Ftc fine intel because of this.

Amd and nvidia both left.
 
"BAPCO-being-a-joke" aside, why are they comparing Intel's 258V it to AMD's Phoenix? Phoenix is previous gen. At the very least, they should be comparing it to current-gen (Hawk Point). But realistically, given that AMD's next-gen is releasing in 17 days, a comparison to that in 17 days will make more sense...except that we won't see 258V before September.

I smell a rat
 
Why would anyone take bapco benchmarks seriously after they were found favoring intel cpus? Ftc fine intel because of this.

Amd and nvidia both left.

Nobody is taking it seriously IMO. Just a leaked benchmark spotting being reported as a news.

I guess some sort of proof that next gen parts are being tested is well worth the mention.
 
  • Like
Reactions: KnightShadey

TRENDING THREADS