Not all NVIDIA GeForce RTX 3070 laptops are created equal as early reviews show performance differences of over 30% between MSI machines with supposedly identical GPUs

[Redaktion]

The NVIDIA GeForce RTX 3070 Max-P and RTX 3070 Max-Q differ by over 30% in benchmarks, early reviews of two MSI gaming laptops have shown. Many OEMs, including MSI, are divulging the minimum information about the RTX 30 series GPUs in their machines though, leaving people in the dark over what performance to expect.

https://www.notebookcheck.net/Not-all-NVIDIA-GeForce-RTX-3070-laptops-are-created-equal-as-early-reviews-show-performance-differences-of-over-30-between-MSI-machines-with-supposedly-identical-GPUs.516987.0.html

[S.Yu]

This only makes sense when they're trying to hide the fact that they use MQ, but if they use MP they should be actively trying to promote that, so for a company that uses both it truly gets confusing.

[Joe Blowe]

if they don't list actual gpu specs, they're gonna have a ton of returns and chargebacks from retailers.
not good for business

[Gigaboly]

3070 is only around 10fps better in games compare to 2070 super maxp in the same laptop, 3070 is somewhere at 2080 super maxp level, disappointing result as expected, only desktop versions are that much faster...

[Fisherman]

This only makes sense when they're trying to hide the fact that they use MQ, but if they use MP they should be actively trying to promote that, so for a company that uses both it truly gets confusing.

I disagree, there are people (me for example), who value thermals and low fan noise as much as I value FPS in games, and I might well deliberately buy a Max-Q GPU in order to get a balanced laptop.  The specs of the GPUs should be clearly listed, in part because both low and high power parts have their markets.

[Bobby]

This only makes sense when they're trying to hide the fact that they use MQ, but if they use MP they should be actively trying to promote that, so for a company that uses both it truly gets confusing.

I disagree, there are people (me for example), who value thermals and low fan noise as much as I value FPS in games, and I might well deliberately buy a Max-Q GPU in order to get a balanced laptop.  The specs of the GPUs should be clearly listed, in part because both low and high power parts have their markets.

Max-q is not going to help you predict anything about the fan noise and thermals. And besides, you could get the max-p models and change fan noise in nvidia settings now with whispermode - just select fan noise level and it will adjust everything for you

[RTXftw]

Max-q is not going to help you predict anything about the fan noise and thermals. And besides, you could get the max-p models and change fan noise in nvidia settings now with whispermode - just select fan noise level and it will adjust everything for you

Technically that's not entirely correct, as nvidia does require oems to meet certain noise & thermal levels for max-q devices. (albeit probably not boardroom-level-quiet as they are gaming laptops afterall)

[hfm]

This only makes sense when they're trying to hide the fact that they use MQ, but if they use MP they should be actively trying to promote that, so for a company that uses both it truly gets confusing.

I disagree, there are people (me for example), who value thermals and low fan noise as much as I value FPS in games, and I might well deliberately buy a Max-Q GPU in order to get a balanced laptop.  The specs of the GPUs should be clearly listed, in part because both low and high power parts have their markets.

Max-q is not going to help you predict anything about the fan noise and thermals. And besides, you could get the max-p models and change fan noise in nvidia settings now with whispermode - just select fan noise level and it will adjust everything for you

nVidia stipulated that the laptop could not go over 40dbA when utilizing Max-Q in it's first iteration. I don't know if they are retaining that standard in this new way of defining Max-Q, but it sounds like it may still be adhering to their old standard while bringing in other things.. shrug..

[_MT_]

Technically that's not entirely correct, as nvidia does require oems to meet certain noise & thermal levels for max-q devices. (albeit probably not boardroom-level-quiet as they are gaming laptops afterall)

The question is, is that sill a thing? Given that Nvidia allows them to pick and choose what they want. I thought about it when I first read about the end of Max-Q and came to the conclusion that it's probably up to manufacturers. Otherwise, they would have to attach different requirements to different power levels/ TGPs. And there was no such information.

Nvidia snuck in a slower '1D12' variant (conned manufacturers by shipping an inferior part without their knowledge)? Or was it manufacturers that chose it and neglected to inform customers?

[t4n0n]

35 Watts? For a 3070?

Surely that's got to be a typo...

[liviu993]

35w lower than max-p version.

[alexolek]

J'ai précommandé et reçu le MSI GP76 Leopard 10UG-259FR
C'est bien une 1620Mhz donc non max Q