Original Post
Written by
Edited July 30, 2021
at 08:30 AM
by
Newegg [newegg.com] has GIGABYTE AERO 15 OLED KD 15.6" UHD 4K AMOLED, i7 -11800H, NVIDIA GeForce RTX 3060, 6GB GDDR6, 16GB RAM, 512GB SSD Gaming Laptop on sale for
$1249 after code and rebate.
Code: 82BTSWKN474
82 Comments
Your comment cannot be blank.
Featured Comments
Numbers aren't everything. Driver and developer support still count
It's AMD's time and there's almost zero argument other than the on par Single core performance. Other than that, AMD just demolishes Intel in almost every Productivity test.
Don't get me wrong, I used to say the opposite ~5 years ago when Intel was clearly on the lead.
Sign up for a Slickdeals account to remove this ad.
So you aren't overpaying.
Long Battery life for a laptop is crucial. Also, being cool on your lap is again an advantage.
But then again, you will ignore all of these advantages and will still stick on the inferior CPU because you already made up your mind
How can I trust you if I don't even know you though?
11800h is pretty much on par with the Ryzen 7 5800h so it's actually a good creator choice despite what others think. It may not be as power efficient performance per watt but raw power is there and competitive.
- AMD Zen, Zen+. Manufactured on Global Foundaries 16/12 nm process. 36.7 million transistors per square mm (MT/mm^2).
- Intel Broadwell - Rocket Lake. Manufactured on Intel's 14nm process. 37.5 MT/mm^2
- AMD Zen 2, Zen 3.. Manufactured on TSMC's 7nm process. 96.5 MT/mm^2 in early versions. Currently 114 MT/mm^2.
- This Intel (Tiger Lake). Manufactured on Intel's 10nm process. 100.8 MT/mm^2.
- Zen 4. Scheduled for TSMC's 5nm process. 173 MT/mm^2
When Intel was up against AMD's Zen and Zen+, their density was about the same. Higher transistor density directly translates into lower power consumption. So they were on equal footing, and Intel performed slightly better due to a superior architecture.AMD's Zen 2 switched to TSMC's 7nm, while Intel ran into problems with its 10nm process, forcing it to stay on 14nm. That gave AMD a nearly 3:1 density advantage, which translated into a massive power advantage (either lower power, or higher speeds at the same power). That's what gave Intel poor battery life, and allowed AMD to outperform Intel. It's nothing inherent about the two companies' architectures. Just that AMD has been enjoying a massive density advantage.
That difference is mostly gone with Intel's Tiger Lake (which is made on Intel's 10nm but confusingly is given the i7-11xxx numbering, same as Rocket Lake which is made on Intel 14nm). AMD opted to stick with TSMC's 7nm process for Zen 3, so the density of these newest Intel and AMD processors are roughly equivalent. 114 vs 101 MT/mm^2. Consequently, I would not expect much difference in power consumption.
Zen 4 is scheduled to be manufactured on TSMC 5nm, which is 173 MT/mm^2, so AMD will have an advantage again in 2022, though not as large as it enjoyed with Zen 2. If/when Intel gets its 7nm process working (estimated to be about 200 MT/mm^2, that advantage should vanish again. (The different companies are measuring different structures, so the nm figures were never comparable between companies.)
Similarly, Apple has been buying out TSMC"s capacity on its newer processes, which is what's been giving it an advantage in cell phone SOCs and its M1. Their SOCs were consistently being manufactured one process smaller than Qualcomm's (both use TSMC). When TSMC's rate of advancement slowed and both ended up manufactured on the same process, the performance was much closer, with Qualcomm actually beating out Apple on a few benchmarks. The M1 is currently manufactured on TSMC's 5nm (173 MT/mm^2), which is why the M1 performs so well. When Apple said the M1 ran x86 code in emulation faster than Intel, that wasn't at all surprising. They were comparing their 173 MT/mm^2 processor against Intel's 37.5 MT/mm^2 processor. The Intel CPU was at a massive power disadvantage, so had to be clocked a lot slower/have fewer cores to work in a laptop. Apple has already bought out TSMC"s initial 3nm capacity (rumored to be around 250 MT/mm^2).
If you remember way back when Nvidia's Maxwell GPU release stumbled out the gate, it was for the same reason. Nvidia was expecting to be able to manufacture Maxwell on TSMC's 14nm process, so designed it assuming lower power consumption (heat generation). But then Apple bought all of TSMC's 14nm capacity, forcing Nvidia to manufacture Maxwell on TSMC's 28nm process. The higher heat meant only their mobile versions of their Maxwell GPUs (the Nvidia 8xx series) were viable. They had to redesign their Maxwell GPU line for 28nm, which eventually became the 9xx series. That's why there were no Nvidia desktop 8xx GPUs.
Sign up for a Slickdeals account to remove this ad.