The ridiculous number would not really be a problem for end users if OEMs actually respected Nvidia's guidelines and provided clear info with the official specs. Unfortunately, very few OEMs are currently doing this, and we are seeing wild spec variations that allow lower-priced SKUs to outperform higher-priced variants.
https://www.notebookcheck.net/Buyer-beware-Nvidia-has-at-least-28-variants-prepared-for-the-mobile-RTX-3000-GPU-stack.517112.0.html
Nvidia is very consumer friendly. Not at all confusing. Which is the scalper's favorite?
That's a customer nightmare.. NVIDIA just hates us, ordinary customers.
I wasn't aware that Dynamic Boost 2 is also accounted for when it comes to TGP by SOME manifacturers...
what a mess
Do I get an achievement for picking the correct one for me? Lvl 5 Nvidiot sounds good to me.
Yawn. OEMs know better than any of you kids.
A computer is a complicated electronic device - something you wouldn't know no-thing about. So shush up. Jesus.
This looks more like an attempt to mock Nvidia than provide useful reference, but I agree with them on that approach :)
NBC, Please pin this article to the top as a PSA to all readers and potential consumers of these chips.
Quote from: Smaller Biba Hotep on January 27, 2021, 17:15:47
Yawn. OEMs know better than any of you kids.
A computer is a complicated electronic device - something you wouldn't know no-thing about. So shush up. Jesus.
Ya, we should bend over and take it from them. I pledge my conformity and vow to never question.
I'm trying to use this table as a reference, but...
In the table, a RTX 3060 FE with 170 W TDP can deliver less TFLOP than a Laptop RTX 3060 100 W right ?
I don't understand why, can someone explain ?
Thank you.
Quote from: Hentone on January 28, 2021, 09:19:17
I'm trying to use this table as a reference, but...
In the table, a RTX 3060 FE with 170 W TDP can deliver less TFLOP than a Laptop RTX 3060 100 W right ?
I don't understand why, can someone explain ?
Thank you.
The RTX 3060 FE actually has 3584 CUDA cores. I think the relative performance is where it should be, but VideoCardz just specified the wrong number of CUDA cores there.
Just a Q: hypothetically, if we have a RTX 3080 with 100W vs 150W, this would translate in the 150W model to deliver closer to 25% more performance because you have diminishing returns for wattage due to heat, no?