Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Intel's Lion Cove P-Core and Gaming Workloads

chipsandcheese.com

Post reply on HN
Print view View original

51–60 of 126 posts

Re: Intel's Lion Cove P-Core and Gaming Workloads

#51

Post by Havoc »

masklinn wrote: post #43
Havoc wrote: post #35

Earlier quoted context omitted.

Yeah intel either needs to ensure much better day 1 support for E / P core split or drop it entirely. People see them as an active negative right now rather than how intel pitches them

Scheduling is the OS’s decision not the CPU’s. Although Intel’s configurations are unhelpful to say the least: it’s hard to make good scheduling decisions when Intel sells a 2P, 8E, 2LE as a 12 cores to the user.

It is the OS fault as you say but consumers don't care who's fault it is only that it's broken

Re: Intel's Lion Cove P-Core and Gaming Workloads

#52

Post by pjmlp »

dandellion wrote: post #50
pjmlp wrote: post #14

Earlier quoted context omitted.

I think many haven't yet grasped the future is heterogeneous computing, especially, when many desktops are actually laptops nowadays. Software working poorly in such setup means no effort was made to actually make it perform well in first place. Games requiring desktop cases looking like a rainbow aquarium with top everything will become a niche, in today's mobile computing world, and with diminishing sales and atten…

People have been making the same argument for more than a decade now, meanwhile PC gaming has only kept growing. If that was the case, why hasn't it happened yet?

PC gaming has been growing since MS-DOS 3.3, when I was playing Defender of the Crown, in case people haven't been paying attention.

Especially in Europe where consoles were largely irrelevant during the 8 and 16 bit days.

Re: Intel's Lion Cove P-Core and Gaming Workloads

#54

Post by dwattttt »

saati wrote: post #37
raverbashing wrote: post #27

Earlier quoted context omitted.

Which is funny because the pushback the PS3 got from developers was too much. Maybe it was "too heterogeneous" But I guess the future was already set.

That was very different, here the problem is the game's threads get scheduled on the weak E-cores and it doesn't like that for some reasons, with the PS3 that would have been impossible, SPEs had a different ISA, and didn't even have direct access to memory, the problem was developers had to write very different code for the SPEs to unlock most of the performance of the Cell.

Even more weirdly: affinity was set to the P cores, so it wasn't being scheduled on E cores at all.

Maybe it was spawning more threads than P cores, because it expected to be able to use all cores.

Re: Intel's Lion Cove P-Core and Gaming Workloads

#55

Post by mathiaspoint »

pjmlp wrote: post #48
flohofwoe wrote: post #32

Earlier quoted context omitted.

> Software working poorly in such setup means no effort was made to actually make it perform well in first place. How do you optimize for a system architecture that doesn't exist yet? (e.g. CoD could probably be fixed quite easily, but you can't do that before the hardware where that problem manifests even exists - I'd say it's much more likely that the problem is in the Windows scheduler though) > Games requiring de…

Laptops are PC gaming, the only ones declaring them dead are the console generation that never played 8 and 16 bit home computers. For a game that gets yearly releases, I guess a decade would be long enough, given the existence of heterogeneous core programming across consoles and desktop systems.

Is windows even usable on laptops still? Between the waking from sleep to check notifications and the intensity of all the new background tasks it likely isn't pleasant.

Re: Intel's Lion Cove P-Core and Gaming Workloads

#56

Post by p_ing »

gruturo wrote: post #40
pjmlp wrote: post #14

Earlier quoted context omitted.

I think many haven't yet grasped the future is heterogeneous computing, especially, when many desktops are actually laptops nowadays. Software working poorly in such setup means no effort was made to actually make it perform well in first place. Games requiring desktop cases looking like a rainbow aquarium with top everything will become a niche, in today's mobile computing world, and with diminishing sales and atten…

There's a giant pile of software - decades worth of it, literally - which was already written and released, much of it now unmaintained and/or closed source, where the effort you cite is not a possibility. By all means anything released in 2025 doesn't have that excuse - but I can't fault the authors of 15+ years old programs for not having a crystal ball and accounting for something which didn't exist at the time. I…

AMD and Intel do work with Microsoft (and I'm sure provide code to the Linux scheduler) to optimize the NT scheduler appropriately.

Sometimes it doesn't happen Day 1 which is only a problem for the consumers that notice.

Re: Intel's Lion Cove P-Core and Gaming Workloads

#57

Post by p_ing »

vladvasiliu wrote: post #47
dur-randir wrote: post #41

Earlier quoted context omitted.

It doesn't matter to the public's perception.

Right, it certainly doesn't. But since we're on a technical forum where my initial comment's parent seems to argue this issue is Intel's fault, I think it's interesting to determine whether that's actually the case. Of course, had Intel not released such a CPU, we wouldn't be having this conversation. But since they have released similar ones a few years ago now, is it still (only?) their fault?

OS scheduler is responsible; AMD/Intel work with Microsoft on the NT scheduler and likely contribute code on the Linux scheduler.

Re: Intel's Lion Cove P-Core and Gaming Workloads

#58

Post by HelloNurse »

adrian_b wrote: post #4
samrus wrote: post #2

122 points and no comments? Is this being botted or something?

Such articles are very interesting for many people, because nowadays all CPU vendors are under-documenting their products. Most people do not have enough time or knowledge (or money to buy CPU samples that may prove to be not useful) to run extensive sets of benchmarks to discover how the CPUs really work, so they appreciate when others do this and publish their results. Besides learning useful details about the stre…

It is an interesting but particularly non-actionable analysis: only a handful of engineers at Intel are in a position to design an improved Lion Cove, while the main takeaway for the game programmers who care about game workload performance is that nothing performs too badly and therefore general performance improvement techniques like accessing less memory are a good fit for this processor.

Re: Intel's Lion Cove P-Core and Gaming Workloads

#59

Post by Retric »

dandellion wrote: post #50
pjmlp wrote: post #14

Earlier quoted context omitted.

I think many haven't yet grasped the future is heterogeneous computing, especially, when many desktops are actually laptops nowadays. Software working poorly in such setup means no effort was made to actually make it perform well in first place. Games requiring desktop cases looking like a rainbow aquarium with top everything will become a niche, in today's mobile computing world, and with diminishing sales and atten…

People have been making the same argument for more than a decade now, meanwhile PC gaming has only kept growing. If that was the case, why hasn't it happened yet?

It’s arguing desktop gaming is dying not PC gaming. Q4 2024 shipped 13.6 million desktops (down 1.8%) vs 53.2 million laptops up (6.8%).

That trend of declining Desktops vs increasing laptops has been steady since 2010. There’s still plenty of desktops to justify major investments by gaming companies, but that’s not guaranteed long term.

Re: Intel's Lion Cove P-Core and Gaming Workloads

#60

Post by CoastalCoder »

pjmlp wrote: post #49
CoastalCoder wrote: post #46

Earlier quoted context omitted.

I did some programming for the Cell, and it definitely wasn't in assembly. It did require using intrinsics though, so I guess it was a little assembly-like.

I did mention C was supported, though if you really wanted to push them to the limit, and given the limited memory as well, Assembly was the way to go.

Fair. At the time I packed the necessary understanding to know of the GCC-emitted object code was significantly suboptimal.
Post reply on HN