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。

Hardware for Deep Learning, Part 3: GPU

74 点作者 che_shr_cat超过 6 年前

2 条评论

jaimex2超过 6 年前
Tesla apparently designed their own chip specifically for neural net vision as GPUs were not fast enough - including nVidias Tesla chip which is currently being put in their cars.<p>Musk mentioned they got a 10x boost from doing so in their last investor call:<p>&quot;I&#x27;m a big fan of NVIDIA, they do great stuff. But using a GPU, fundamentally it&#x27;s an emulation mode, and then you also get choked on the bus. So, the transfer between the GPU and the CPU ends up being one of the constraints of the system&quot;<p>&quot;And as a rough sort of whereas the current NVIDIA&#x27;s hardware can do 200 frames a second, this is able to do over 2,000 frames a second and with full redundancy and fail-over.&quot;<p>Peter Bannon lead the development.
评论 #18083039 未加载
asparagui超过 6 年前
Great article!
评论 #18083040 未加载