<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[84gb的6000d和72gb的pro5000 差不多价格]]></title><description><![CDATA[<p dir="auto">84gb的6000d京东三年保修的第三方店50000元左右，华强北线下实体店也有，72gb的pro5000线下的价格也差不多，京东自营三年保修55000左右</p>
<p dir="auto">个人搭建家用，如果只用6000d做算力不能显示，还需要搭配一个亮机卡使用，刚好手里有4070s，另外6000d还要自己加风扇散热，这样双显卡会不会容易出bug，怕搞不定环境，为了多出来的显存去折腾6000d划算吗</p>
]]></description><link>https://lcz.me/topic/469/84gb的6000d和72gb的pro5000-差不多价格</link><generator>RSS for Node</generator><lastBuildDate>Thu, 11 Jun 2026 10:34:46 GMT</lastBuildDate><atom:link href="https://lcz.me/topic/469.rss" rel="self" type="application/rss+xml"/><pubDate>Mon, 08 Jun 2026 03:06:27 GMT</pubDate><ttl>60</ttl><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Thu, 11 Jun 2026 07:20:43 GMT]]></title><description><![CDATA[<p dir="auto">grok给的结论：RTX PRO 6000D 在 FP4/FP8/FP16/FP32 等精度下的官方/可靠规格数据确实远低于完整 PRO 6000，基本符合你说的“约 15%”水平（具体 14-15% 左右）。 以下是基于 NVIDIA 官方文档、HPE QuickSpecs 等可靠来源的准确数据（非推算），重点对比三款卡。<br />
官方 Tensor 性能对比（峰值，含 sparsity 等优化时）</p>
<p dir="auto">RTX PRO 6000 Blackwell（完整版，96GB，工作站/服务器版）：<br />
FP4 Tensor：~4 PFLOPS（4000 TOPS/AI TOPS）<br />
FP8 Tensor：~2 PFLOPS（2000 TOPS）<br />
FP16 / BF16 Tensor：~1 PFLOP（1000+ TFLOPS 范围，不同来源略有 1001 等）<br />
FP32：125 TFLOPS（单精度）</p>
<p dir="auto">RTX PRO 6000D（中国特供/84GB 版）：<br />
FP4 Tensor：593 TFLOPS（约完整 6000 的 14.8%）<br />
FP8 Tensor：296 TFLOPS（约 14.8%）<br />
FP16 / BF16 Tensor：148 TFLOPS（约 14.8%）<br />
TF32：~74 TFLOPS<br />
FP32：约 110 TFLOPS 或更低（Max-Q 版类似，完整版 125 TFLOPS 对应砍幅一致）</p>
<p dir="auto">RTX PRO 5000 Blackwell（72GB 版典型）：<br />
Tensor 核心数 ~440（完整 6000 的 ~58%）<br />
FP32：~65-74 TFLOPS<br />
FP4/FP8/FP16 等低精度：大致按核心/规格比例 ~50-60% 于完整 6000（具体 FP4 预计 2000+ TOPS 级别，但无 6000D 那么极端砍削）。实际推理中 5000 72GB 在中型负载下接近或略优于 6000D 的低精度算力。</p>
<p dir="auto">数据来源说明：完整 6000 来自 NVIDIA 官方 datasheet（FP4 4 PFLOPS 等）；6000D 来自 HPE 等 OEM QuickSpecs 的明确列出性能值；5000 来自 Exxact/NVIDIA 系列对比。Blackwell Tensor Core 低精度高度依赖核心数、时钟和带宽，6000D 的砍削主要体现在这些硬件规格上（CUDA/Tensor 核心 -17%，带宽/时钟下调，导致整体低精度算力大幅缩水）。</p>
]]></description><link>https://lcz.me/post/6277</link><guid isPermaLink="true">https://lcz.me/post/6277</guid><dc:creator><![CDATA[Rocky Law]]></dc:creator><pubDate>Thu, 11 Jun 2026 07:20:43 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Thu, 11 Jun 2026 05:22:02 GMT]]></title><description><![CDATA[<p dir="auto">你是根据什么指标得出以上结论的，Tensor core数量，缓存带宽？大家都是Blackwell，都支持nvf4</p>
]]></description><link>https://lcz.me/post/6264</link><guid isPermaLink="true">https://lcz.me/post/6264</guid><dc:creator><![CDATA[blackjack]]></dc:creator><pubDate>Thu, 11 Jun 2026 05:22:02 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Thu, 11 Jun 2026 04:27:38 GMT]]></title><description><![CDATA[<p dir="auto">研究了2天，还是别上6000d了，低精度推理算力只有6000的15%而已，算力还不如pro5000，只有5000的一半，显存你说他大吧，玩fp32高精度一张卡84g也装不了什么模型啊，组个几张就有点说法了，一张没啥用</p>
]]></description><link>https://lcz.me/post/6250</link><guid isPermaLink="true">https://lcz.me/post/6250</guid><dc:creator><![CDATA[Rocky Law]]></dc:creator><pubDate>Thu, 11 Jun 2026 04:27:38 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Tue, 09 Jun 2026 07:19:11 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/cs6" aria-label="Profile: CS6">@<bdi>CS6</bdi></a></p>
<p dir="auto">這就是對外了</p>
]]></description><link>https://lcz.me/post/5954</link><guid isPermaLink="true">https://lcz.me/post/5954</guid><dc:creator><![CDATA[566656661]]></dc:creator><pubDate>Tue, 09 Jun 2026 07:19:11 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Tue, 09 Jun 2026 06:43:52 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/566656661" aria-label="Profile: 566656661">@<bdi>566656661</bdi></a> 多人在線聊.....聊天</p>
]]></description><link>https://lcz.me/post/5950</link><guid isPermaLink="true">https://lcz.me/post/5950</guid><dc:creator><![CDATA[CS6]]></dc:creator><pubDate>Tue, 09 Jun 2026 06:43:52 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Tue, 09 Jun 2026 00:41:08 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/tony-wang" aria-label="Profile: Tony-Wang">@<bdi>Tony-Wang</bdi></a> 感谢回复，我看到评测说标称是功耗600w，但是实际满载400w，比pro5000也就多100吧，散热噪音确实是个问题，要好好想想怎么弄。</p>
]]></description><link>https://lcz.me/post/5887</link><guid isPermaLink="true">https://lcz.me/post/5887</guid><dc:creator><![CDATA[Rocky Law]]></dc:creator><pubDate>Tue, 09 Jun 2026 00:41:08 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Mon, 08 Jun 2026 14:43:11 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/rocky-law" aria-label="Profile: Rocky-Law">@<bdi>Rocky-Law</bdi></a></p>
<p dir="auto">价格差不多的话, 架构又相同, 显存多肯定更划算, 而且6000D性能也比 5000 高.</p>
<p dir="auto">不过6000的功耗是5000的两倍, 散热, 噪音你也得考虑一下.</p>
]]></description><link>https://lcz.me/post/5770</link><guid isPermaLink="true">https://lcz.me/post/5770</guid><dc:creator><![CDATA[Tony Wang]]></dc:creator><pubDate>Mon, 08 Jun 2026 14:43:11 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Mon, 08 Jun 2026 14:41:29 GMT]]></title><description><![CDATA[<p dir="auto">差不多價格當然就是84gb的6000d啊</p>
<p dir="auto"><s>不過除了ComfyUI之外其實我想不到太多這麽大的VRAM要怎麽用就是了</s> <img src="https://lcz.me/assets/plugins/nodebb-plugin-emoji/emoji/android/1f602.png?v=d348ca29232" class="not-responsive emoji emoji-android emoji--joy" style="height:23px;width:auto;vertical-align:middle" title=":joy:" alt="😂" /></p>
]]></description><link>https://lcz.me/post/5767</link><guid isPermaLink="true">https://lcz.me/post/5767</guid><dc:creator><![CDATA[566656661]]></dc:creator><pubDate>Mon, 08 Jun 2026 14:41:29 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Mon, 08 Jun 2026 14:32:43 GMT]]></title><description><![CDATA[<p dir="auto">思来想去还是84gb的6000d换算，仅从AI使用角度考虑的话，大显存还是刚需</p>
]]></description><link>https://lcz.me/post/5760</link><guid isPermaLink="true">https://lcz.me/post/5760</guid><dc:creator><![CDATA[Rocky Law]]></dc:creator><pubDate>Mon, 08 Jun 2026 14:32:43 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Mon, 08 Jun 2026 12:54:02 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/stxpnet" aria-label="Profile: stxpnet">@<bdi>stxpnet</bdi></a> 84G啥时候有4.2w的价哦，pro 5000 72G都没有低于5w</p>
]]></description><link>https://lcz.me/post/5733</link><guid isPermaLink="true">https://lcz.me/post/5733</guid><dc:creator><![CDATA[fly86]]></dc:creator><pubDate>Mon, 08 Jun 2026 12:54:02 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Mon, 08 Jun 2026 10:44:30 GMT]]></title><description><![CDATA[<p dir="auto">那还行，主要现在未阉割版价格超10W了~~看来弟版还是有性价比</p>
]]></description><link>https://lcz.me/post/5720</link><guid isPermaLink="true">https://lcz.me/post/5720</guid><dc:creator><![CDATA[九龙杨生]]></dc:creator><pubDate>Mon, 08 Jun 2026 10:44:30 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Mon, 08 Jun 2026 10:39:55 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/%E4%B9%9D%E9%BE%99%E6%9D%A8%E7%94%9F" aria-label="Profile: 九龙杨生">@<bdi>九龙杨生</bdi></a> 散热简单，淘宝买个12cm或者10Cm的暴力扇加导风口</p>
]]></description><link>https://lcz.me/post/5719</link><guid isPermaLink="true">https://lcz.me/post/5719</guid><dc:creator><![CDATA[stxpnet]]></dc:creator><pubDate>Mon, 08 Jun 2026 10:39:55 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Mon, 08 Jun 2026 10:37:40 GMT]]></title><description><![CDATA[<p dir="auto">京东84G 6000D 5月初才4万2呢？ 啥时候能降价啊啊啊啊啊啊.</p>
]]></description><link>https://lcz.me/post/5718</link><guid isPermaLink="true">https://lcz.me/post/5718</guid><dc:creator><![CDATA[stxpnet]]></dc:creator><pubDate>Mon, 08 Jun 2026 10:37:40 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Mon, 08 Jun 2026 10:17:57 GMT]]></title><description><![CDATA[<p dir="auto">他这种可能还是要改造散热对个人玩家比较麻烦吧，我看了96G RTX6000的拆解，感觉个人玩家很难搞啊</p>
]]></description><link>https://lcz.me/post/5716</link><guid isPermaLink="true">https://lcz.me/post/5716</guid><dc:creator><![CDATA[九龙杨生]]></dc:creator><pubDate>Mon, 08 Jun 2026 10:17:57 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Mon, 08 Jun 2026 04:36:20 GMT]]></title><description><![CDATA[<p dir="auto">我自己的M B是 Asus ProArt B850-Creator WiFi NEO 就有支持x8 *2</p>
]]></description><link>https://lcz.me/post/5667</link><guid isPermaLink="true">https://lcz.me/post/5667</guid><dc:creator><![CDATA[CS6]]></dc:creator><pubDate>Mon, 08 Jun 2026 04:36:20 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Mon, 08 Jun 2026 03:55:08 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/kop-wang" aria-label="Profile: kop-wang">@<bdi>kop-wang</bdi></a> 这个应该问题不大吧，x4的位置就放算力卡，反正推理都在显存里面运行，不进行数据交换，可能也就是第一次加载模型影响点速度</p>
]]></description><link>https://lcz.me/post/5661</link><guid isPermaLink="true">https://lcz.me/post/5661</guid><dc:creator><![CDATA[Rocky Law]]></dc:creator><pubDate>Mon, 08 Jun 2026 03:55:08 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Mon, 08 Jun 2026 03:50:35 GMT]]></title><description><![CDATA[<p dir="auto"><a href="https://item.taobao.com/item.htm?id=1052983962058&amp;mi_id=0000lNeDCs1ITyBXqagtBhfvgmJlX1bXFlOOdgIwkca7pyw&amp;spm=a21xtw.29178619.0.0&amp;xxc=shop&amp;skuId=6093376673767" rel="nofollow ugc">https://item.taobao.com/item.htm?id=1052983962058&amp;mi_id=0000lNeDCs1ITyBXqagtBhfvgmJlX1bXFlOOdgIwkca7pyw&amp;spm=a21xtw.29178619.0.0&amp;xxc=shop&amp;skuId=6093376673767</a><br />
看到這個..... 可惜人在台灣</p>
]]></description><link>https://lcz.me/post/5660</link><guid isPermaLink="true">https://lcz.me/post/5660</guid><dc:creator><![CDATA[CS6]]></dc:creator><pubDate>Mon, 08 Jun 2026 03:50:35 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Mon, 08 Jun 2026 03:21:38 GMT]]></title><description><![CDATA[<p dir="auto">如果去掉其他风险（改散热、无显示能力、店铺不是京东自营等），6000D肯定是划算的。<br />
你担心的兼容问题一般来讲不存在。</p>
<p dir="auto">但是，一般的主板都不提供两个满血pcie插槽，甚至两个pcie对等x8都没有，一般都是x16+x4的组合。</p>
]]></description><link>https://lcz.me/post/5653</link><guid isPermaLink="true">https://lcz.me/post/5653</guid><dc:creator><![CDATA[kop wang]]></dc:creator><pubDate>Mon, 08 Jun 2026 03:21:38 GMT</pubDate></item><item><title><![CDATA[Reply to 84gb的6000d和72gb的pro5000 差不多价格 on Mon, 08 Jun 2026 03:13:05 GMT]]></title><description><![CDATA[<p dir="auto">感覺主板/cpu 才是關鍵</p>
]]></description><link>https://lcz.me/post/5648</link><guid isPermaLink="true">https://lcz.me/post/5648</guid><dc:creator><![CDATA[CS6]]></dc:creator><pubDate>Mon, 08 Jun 2026 03:13:05 GMT</pubDate></item></channel></rss>