TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Current x86 vs. Apple M1 Performance Measurements Are Flawed

2 点作者 feelix超过 4 年前

2 条评论

seek3r超过 4 年前
“The “flaw” we’re going to talk about isn’t a problem with any specific benchmark or reviewer. It’s a difference in how the Apple M1 allocates and assigns resources versus how x86 CPUs work.”
stncls超过 4 年前
tl;dr Extreme tech argues that, because x86 typically has SMT and M1 does not, single-threaded benchmarks are &quot;flawed&quot; (they can&#x27;t take advantage of SMT).<p>Nonsense. The most prominently advertised benchmarks are generally the multi-threaded ones, which is fair. Then, single-threaded tests are often added because some applications don&#x27;t parallelize as well (or at all). From an end-user perspective, it is useful information. They are arguing for performance-per-core tests. The only fair way to do that would be 2-threaded workloads. Sure, why not. But it does not invalidate the use of single-threaded workloads as a (maybe niche? but) relevant data point.