Nvidia GeForce RTX 2080 Super Max-Q Review: The GPU Efficiency King

gdavid65

Posts: 8   +3
Why do only some have comparison with RTX 2070 desktop? I think that is an important inclusion, as some may be wondering if they can ditch their desktop gaming PC for a multi-purpose gaming laptop.

Would be great to see such chips on ultra low-power half-height cards that could fit in mini-ITX cases with no external power connector. Doesn't seem like much of a trade-off in performance compared to the desktop versions from above (12% less maybe).
 
  • Like
Reactions: Reehahs

Squid Surprise

Posts: 3,151   +2,056
Why aren’t you comparing it to the “regular” 2080 super mobile chip? Wouldn’t it be valuable to know what gains you can get if you opt for a heavier laptop?
 

yeeeeman

Posts: 293   +242
Slow and wide is the name of the game in GPU department. Just look at Lakefield. It has 64EUs like 1065G7 within 7W. They had to clock it at 500Mhz, but in any case, it is more efficient than having say 24EUs at 1000+ mhz.
Still for 2500$ this is a LOT of computing power.
 

Lew Zealand

Posts: 1,381   +1,353
TechSpot Elite
Which is why mobile GPU's are such a cynical marketing con job.

Their naming convention should follow exactly that of the desktop parts, so customers ignorant of the difference aren't ripped off.
I disagree. Anyone who spends $1000-3000 on a gaming laptop without informing themselves deserves the performance they bought because they are so ignorant of how things work that they'll never know the difference anyway.

Nvidia's laptop GPU naming convention follows the desktop GPUs on all but one part, the lowest-end 1650. They have the same core count as the desktops but are clocked lower because the laptop can't handle the power requirements of the power pig desktop cards. The Max-Q versions get an even lower power limit and memory bandwidth than the regular ones to fit in smaller laptops. Still the same number of cores though.

Exceptions:

The laptop 2080 Super doesn't get the desktop 2080's special 15.5GHz memory though, but then even the 2080 Ti doesn't get that.
The 1650 laptop variant actually gets more cores than the desktop GPU, so you actually get a better part there.
 
  • Like
Reactions: Reehahs

QuantumPhysics

Posts: 2,871   +2,600
I disagree. Anyone who spends $1000-3000 on a gaming laptop without informing themselves deserves the performance they bought because they are so ignorant of how things work that they'll never know the difference anyway.

Nvidia's laptop GPU naming convention follows the desktop GPUs on all but one part, the lowest-end 1650. They have the same core count as the desktops but are clocked lower because the laptop can't handle the power requirements of the power pig desktop cards. The Max-Q versions get an even lower power limit and memory bandwidth than the regular ones to fit in smaller laptops. Still the same number of cores though.

Exceptions:

The laptop 2080 Super doesn't get the desktop 2080's special 15.5GHz memory though, but then even the 2080 Ti doesn't get that.
The 1650 laptop variant actually gets more cores than the desktop GPU, so you actually get a better part there.


Anyone who is spending this kind of money on a laptop knows exactly what they are buying.
 

scavengerspc

Posts: 350   +220
TechSpot Elite
Excellent review but when talking about laptops of this type there should also be a maximum performance laptop in the mix for comparison. A 200 watt 2080 is seriously a big omission in the article.
 

Rjmachine

Posts: 23   +8
AMD is really missing in action in the mobile GPU segment. Nvidia had owned this market for years now.

I hope AMD will use some of their Zen money on their lacking GPU department.

I really hope Big Navi will deliver. It's been many years since I have had an AMD GPU. My last was 7970 @ 1200/1600 MHz. IMO AMD's last good flagship card.