<p>It's one thing to see the memes for the NVIDIA RTX 4000 card sizes compared to the 3000 series. It's another to see it in person.</p>
<p><img src="https://cdn.hashnode.com/res/hashnode/image/upload/v1697417558381/44480148-cead-4d48-9699-4b80e727eff3.png" alt class="image--center mx-auto" /></p>
<p><img src="https://cdn.hashnode.com/res/hashnode/image/upload/v1697417572231/c40967f6-fc12-4336-b27e-889e1da2288f.png" alt class="image--center mx-auto" /></p>
<p><img src="https://cdn.hashnode.com/res/hashnode/image/upload/v1697418547589/697e4097-dbfe-4dc5-8630-61a4ac103d7b.png" alt class="image--center mx-auto" /></p>
<p>Actual size comparison:</p>
<p><img src="https://cdn.hashnode.com/res/hashnode/image/upload/v1697417926160/f75365d8-39fd-4ca3-aa15-84b5a3ce8f68.png" alt class="image--center mx-auto" /></p>
<p><a target="_blank" href="https://flipperzero.one/">Flipper Zero</a> for scale...</p>
<p><img src="https://cdn.hashnode.com/res/hashnode/image/upload/v1697417976320/b6d3079a-cf2e-45dc-b068-dbfc21d0b68d.png" alt class="image--center mx-auto" /></p>
<p>Not to mention the cost which is still stuck at crypto-peak prices. I still spent close to $1K getting one. I couldn't rationalize spending even more money for the <code>4090</code>.</p>
<p><img src="https://cdn.hashnode.com/res/hashnode/image/upload/v1697418584470/9ea50469-8a04-4fdd-bfa7-39b9e34f5691.png" alt class="image--center mx-auto" /></p>
<p>Ran some <a target="_blank" href="https://hashcat.net/hashcat/">Hashcat</a> benchmarks on the 3080 compared to the 4080 on my rig:</p>
<p>EVGA 3080 FTW Ultimate:</p>
<div class="gist-block embed-wrapper" data-gist-show-loading="false" data-id="ee9fc320a3ddf8dab4ae88b24f1b591a"><div class="embed-loading"><div class="loadingRow"></div><div class="loadingRow"></div></div><a href="https://gist.github.com/Wind010/ee9fc320a3ddf8dab4ae88b24f1b591a" class="embed-card">https://gist.github.com/Wind010/ee9fc320a3ddf8dab4ae88b24f1b591a</a></div><p> </p>
<p>Compared with the <code>ZOTAC GAMING GeForce RTX 4080</code>:</p>
<div class="gist-block embed-wrapper" data-gist-show-loading="false" data-id="0525f85a919e803432f923cca05f8dac"><div class="embed-loading"><div class="loadingRow"></div><div class="loadingRow"></div></div><a href="https://gist.github.com/Wind010/0525f85a919e803432f923cca05f8dac" class="embed-card">https://gist.github.com/Wind010/0525f85a919e803432f923cca05f8dac</a></div>

It's one thing to see the memes for the NVIDIA RTX 4000 card sizes compared to the 3000 series. It's another to see it in person.

![](https://cdn.hashnode.com/res/hashnode/image/upload/v1697417558381/44480148-cead-4d48-9699-4b80e727eff3.png align="center")

![](https://cdn.hashnode.com/res/hashnode/image/upload/v1697417572231/c40967f6-fc12-4336-b27e-889e1da2288f.png align="center")

![](https://cdn.hashnode.com/res/hashnode/image/upload/v1697418547589/697e4097-dbfe-4dc5-8630-61a4ac103d7b.png align="center")

Actual size comparison:

![](https://cdn.hashnode.com/res/hashnode/image/upload/v1697417926160/f75365d8-39fd-4ca3-aa15-84b5a3ce8f68.png align="center")

[Flipper Zero](https://flipperzero.one/) for scale...

![](https://cdn.hashnode.com/res/hashnode/image/upload/v1697417976320/b6d3079a-cf2e-45dc-b068-dbfc21d0b68d.png align="center")

Not to mention the cost which is still stuck at crypto-peak prices. I still spent close to $1K getting one. I couldn't rationalize spending even more money for the `4090`.

![](https://cdn.hashnode.com/res/hashnode/image/upload/v1697418584470/9ea50469-8a04-4fdd-bfa7-39b9e34f5691.png align="center")

Ran some [Hashcat](https://hashcat.net/hashcat/) benchmarks on the 3080 compared to the 4080 on my rig:

EVGA 3080 FTW Ultimate:

%[https://gist.github.com/Wind010/ee9fc320a3ddf8dab4ae88b24f1b591a] 

Compared with the `ZOTAC GAMING GeForce RTX 4080`:

%[https://gist.github.com/Wind010/0525f85a919e803432f923cca05f8dac]

Hashcat Benchmarks with RTX 3080 vs RTX 4080

I am a developer in Seattle with interests in Security (cyber and IRL), machine learning, and distributed systems.  The best practice is doing.  Teaching is learning.
