RTX 30 laptops have 28 variants, meaning an RTX 3080 Max-Q may be slower than an RTX 3060

Written by Chad Norton on Fri, Jan 29, 2021 5:30 PM

Nvidia’s CES 2021 keynote revealed two major things for the company: first of all, the RTX 3060 was officially announced and launching sometime in February for $329 (if you can find it at that price that is). Secondly, they unveiled their new generation of laptops with next-gen mobile performance.

However, one thing that has recently been revealed is that there are over 28 different variants of the RTX 30 series mobile GPUs, resulting in wildly different specs that will ultimately make it much harder for customers to choose the right one if they want the best performance. What makes matters worse, is that most notebook manufacturers do not list the official specs on their own websites.

So you could go out and buy an RTX 3060 laptop, but depending on which one you buy, you could be getting a worse or better performing laptop due to unlisted and wildly different specifications. In fact, this problem is so prevalent that looking at the actual specs and TFLOP counts, you can actually find an RTX 3080 Max-Q laptop that technically performs worse than an RTX 3060 non-Max-Q.

There’s also the problem with Nvidia’s Dynamic Boost 2.0 technology, which uses AI in order to balance power load between the GPU, CPU, and GPU memory. This adds anywhere between 5W to 20W of power to the graphics card, making the specs even more confusing.

Thankfully we have a full list of every RTX 30 series mobile GPU and their specs, but whichever one the laptop manufacturer wants to use is up to them. So unless you get lucky and manage to find the official specs from the manufacturer themselves, this will be a very tricky process.

Here’s the official specs of the 28 variants for the RTX 30 mobile series, split between RTX 3060, RTX 3070, and RTX 3080 GPU specs as well as a final chart that displays the TFLOP count for each GPU, theoretically creating a hierarchy of performance. The specs listed below do not take into account the extra TDP that Dynamic Boost 2.0 adds.

RTX 3060 laptop GPU specs

GPU TDP (W) CUDA Cores Base Clock (MHz) Boost Clock (MHz) Bus width Memory speed TFLOPS
RTX 3060 Mobile 115 3840 1387 1702 192-bit 14Gbps 10.7
110 1342 1680 10.3
105 1305 1642 10
100 1267 1605 9.7
95 1215 1567 9.3
90 1163 1530 8.9
85 1035 1485 7.9
80 900 1425 6.9
RTX 3060 Max-Q 70 1050 1402 12Gbps 8.1
65 975 1357 7.5
60 817 1282 6.3

RTX 3070 laptop GPU specs

GPU TDP (W) CUDA Cores Base Clock (MHz) Boost Clock (MHz) Bus width Memory speed TFLOPS
RTX 3070 Mobile 125 5120 1215 1620 256-bit 14Gbps 12.4
120 1170 1590 12
115 1110 1560 11.4
RTX 3070 Max-Q 90 930 1410 12Gbps 9.5
85 855 1365 8.8
80 780 1290 8

RTX 3080 laptop GPU specs

GPU TDP (W) CUDA Cores Base Clock (MHz) Boost Clock (MHz) Bus width Memory speed TFLOPS
RTX 3080 Mobile 150 6144 1350 1710 256-bit 14Gbps 16.6
145 1320 1695 16.2
140 1275 1665 15.7
135 1260 1665 15.5
130 1230 1635 15.1
125 1185 1605 14.6
120 1155 1575 14.2
115 1110 1545 13.6
RTX 3080 Max-Q 90 930 1365 12Gbps 11.4
85 870 1320 10.7
80 780 1245 9.6

RTX 30 mobile series GPU performance hierarchy

So there you have it, the RTX 30 mobile series is a confusing mess, but hopefully if you were thinking of buying one this will help you to be more vigilant and keep an eye out for official specs.

What do you think? Were you interested in an RTX 30 series laptop? Will this list help you make a more informed decision? Which notebook GPU were you planning on getting? Let us know!

Are you interested in a new RTX 30 gaming laptop?

Will this help you make a more informed decision?

Our favourite comments:

An absolute mess.


Login or Register to join the debate

11:26 Jan-31-2021

Nvidia casts confusion...Its super effective!

senior admin badge
11:01 Jan-30-2021

absolutely P A T H E T I C

10:30 Jan-30-2021

I have been saying ti again and again. The only thing Nvidia ships is buyer's remorse, not GPUs.

09:40 Jan-30-2021

This in itself is just confusing. The big mess will probably be when the retailers won't advertise or falsely advertise the TDP. I'm skipping the mobile 30 and will wait for the 40xx, but if I were on the market for one, this would be a huge turn off. Look at the graph, the best 3070 is almost on par with the lowest 3080 and in real life, if it's purely marketed as 3080 vs 3070, you'd spend at least $200-350 to upgrade to the 3080 for little return.

12:48 Jan-30-2021

XMG is fully transparent with their TDP's. Others not so much...

admin approved badge
22:52 Jan-29-2021

Well, that in a way didn't disappoint. Ampere isn't very power efficient, also it is not gaming optimized architecture. It is compute architecture designed for heavy work, not power efficiency. So in a way, this didn't disappoint, since I kind of expected it, even if I didn't know to what extent. But yeah, that is nVidia, god forbid their post Pascal card won't be mining efficient,...

admin approved badge
22:54 Jan-29-2021

... because ironically, nVidia is one of few gaming companies which doesn't want to be gaming company anymore, it wants to be datacenter, mining and professional work company, so they can overcharge even more and get away with it. They just can't ditch gaming as easily. But they really did everything by the book so Polaris vs Pascal won't happen in next mining boom and it didn't.

admin approved badge
21:42 Jan-29-2021

An absolute mess.

18:59 Jan-29-2021

Wow, talk about confusing...


Can They Run... |

| 60FPS, Low, 1080p
Core i7-9750H 6-Core 2.6GHz GeForce GTX 1650 Mobile 16GB
| 60FPS, Medium, 1080p
Core i5-4690 3.5GHz Radeon RX 580 Sapphire Nitro+ 8GB 16GB
| 60FPS, High, 1080p
Ryzen 5 2600 GeForce RTX 2060 Super 8GB 16GB
Ryzen 5 2600X 6-Core 3.6GHz GeForce RTX 2060 6GB 16GB
75% Yes [4 votes]
| 30FPS, Low, 720p
Core i7-7500U 2-Core 2.7GHz GeForce 940MX 2GB 16GB
100% Yes [2 votes]
Core i5-4590 3.3GHz GeForce GTX 1650 12GB
100% Yes [1 votes]
| 60FPS, Ultra, 1440p
Core i5-10400F 6-Core 2.90GHz GeForce GTX 1080 Ti Inno3D Twin X2 11GB 16GB
33.3333% Yes [3 votes]