Quote from: S.Yu on December 11, 2020, 22:33:16
Quote from: Crowl on December 11, 2020, 21:20:46
Quote from: VEGGIM on December 11, 2020, 17:12:37
Where's there's not a lot they can do. What suggestion do you have for them to as i call it "lower the gap"
The obvious things they should do is bring back the M suffix to mobile gpus so that it is more immediately obvious that you are not getting a desktop variant and there should be greater clarity on the TGP of a particular card, far too few oems would make it clear which version of a mobile card you were getting with the 2060-based laptops being particularly bad for that.
Mobile gpus are always going to be slower than their desktop equivalent due to power and heat considerations, but that's fine as long as companies don't try to hide any compromises you have made from getting a mobile machine.
I wholly agree. This is the worst case scenario for the continuation of a naming scheme that's gotten rid of the M or Mobile designations under the pretext that there's minimal difference between the two versions. In fact it goes completely against it. Worse, the decision to continue this naming scheme despite lack of proper products to support it strongly suggests that the "3060Ti in disguise" will be sold at or more likely slightly above the price of the real 3070. This is simply disgusting. I didn't expect very high moral standards from Nvidia, but this too low.
Yup, 100% in agreement with the other 2 commenters. It's within Nvidia's power to do a few things here:
1) Charge for the chip, not for the badge they attach to it. No more second-tier chips selling for more than first-tier GPU on desktop just because it has the same name.
2) Name them more honestly. Top of the line would be a 3070M, to reflect that it's a downclocked variant of that chip.
3) As mentioned before, require honesty from OEMs as to how they configure the GPU's TGP - and specify a minimum performance requirement to use a given name. No more of this crap where some 2080 Max-Q implementations were getting chased by 2060 mobile devices (and utterly outclassed by the 2060 on desktop).
But the current situation works for both Nvidia (they sell more high-margin chips despite the crap performance) and the OEMs (saving on engineering costs), so they won't.