My review of the Alienware Graphics Amplifier last summer was incomplete in that I couldn't find a capable host computer to distinguish the performance difference between AGA vs Thunderbolt 3. Through many eGPU builds with different enclosures and hosts, I've found the Alienware 15 R3 to be the perfect candidate for AGA vs TB3 testings. I used the original Razer Core because that was the same one used in the review. The graphics card is the RTX 2080.
System specs:
2017 Alienware 15 R3 - i7-7700HQ/GTX 1070 dGPU/HD Graphics 630 iGPU/32GB RAM/512GB NVMe SSD + 1TB HDD
eGPU hardware:
Razer Core + Nvidia RTX 2080 + .5m Thunderbolt 3 cable
Alienware Graphics Amplifier + Nvidia RTX 2080
Hardware pictures:
 |
 |
Installation steps:
Thunderbolt 3 eGPU was plug-and-play for this Alienware 15R3. It's an eGFX certified Thunderbolt 3 laptop and one of the very few to have direct connection from Thunderbolt 3 port to CPU. Also unique to this computer is the AGA port for the Alienware Graphics Amplifier. It's routed similarly as TB3 port, a 4 PCI lane connection directly to the CPU. As the HWiNFO screen capture below shows, Dell reserved an x8 to the discrete GTX 1070 graphics card, an x4 to the Thunderbolt 3 port, and an x4 to the AGA port. The GTX 1080 Ti was via Thunderbolt 3 and RTX 2080 was through AGA.
 |
 |
 |
Benchmarks:
I ran the RTX 2080 first through the Graphics Amplifier then Razer Core. Nvidia drivers were latest version, 411.7. The external GPU connected to an HP Z27Q 5K monitor. All benchmarks were through external monitor with internal display off. I also disabled PCI Controller 1901 [GTX 1070 dGPU] to make sure no glitches were happening. Here are the numbers.
| AW15R3 + RTX2080 | AGA FHD | TB3 FHD | AGA QHD | TB3 QHD | AGA 4K | TB3 4K | AGA 5K | TB3 5K |
|---|---|---|---|---|---|---|---|---|
| Unigine Valley | 4,637 | 4,285 | 3,315 | 3,102 | 1,574 | 1,525 | 873 | 856 |
| Unigine Heaven | 3,322 | 2,867 | 2,137 | 1,945 | 960 | 917 | 494 | 483 |
| Unigine Superposition | 18,451 | 17,348 | 13,263 | 12,310 | 6,766 | 6,522 | 4,080 | 3,963 |
| 3DMark Time Spy | 15,846 | 13,564 | 10,545 | 9,416 | 5,167 | 4,879 | 3,029 | 2,946 |
| 3DMark Fire Strike | 25,428 | 20,452 | 18,342 | 13,675 | 9,898 | 7,694 | 5,129 | 4,785 |
| Tomb Raider 2013 | 228.5 FPS | 187.7 FPS | 152.7 FPS | 135.1 FPS | 77.2 FPS | 74.7 FPS | 44.2 FPS | 43.9 FPS |
| Shadow of Mordor | 161.6 FPS | 127.3 FPS | 121.7 FPS | 99.4 FPS | 71.0 FPS | 62.6 FPS | 45.7 FPS | 42.7 FPS |
| Dirt Rally | 100.3 FPS | 119.25 FPS | 99.3 FPS | 106.0 FPS | 73.9 FPS | 74.9 FPS | 52.5 FPS | 53.3 FPS |
| Hitman | 86.3 FPS | 81.7 FPS | 77.1 FPS | 68.5 FPS | 38.6 FPS | 35.8 FPS | 21.5 FPS | 21.1 FPS |
Comments:
Here we see the performance difference between the Graphics Amplifier vs Thunderbolt 3 is similar to that of internal PCIe slot vs TB3. The gap lessens at higher resolutions. The AGA is no doubt a better performer when paired with a capable host such as this Alienware 15 R3. One thing to note is the proprietary cable. I've seen boot issues due to a damaged/kinked cable. The end connector is hard to remove and typically requires force. Limit the removal of the cable to extend its life.
• external graphics card builds
• best laptops for external GPU
• eGPU enclosure buyer's guide
2013 Mac Pro (FP D500 x2) [3rd,12C,E] + RX 5700 XT @ 16Gbps-TB2>TB3 (Mantiz Saturn Pro) + macOS 11 [build link]
how noisy would you consider this card ?
Hi
How/where did you changed your settings of your Alienware to „All benchmarks were through external monitor with internal display off. I also disabled PCI Controller 1901 [GTX 1070 dGPU]...“?
A short description and/or some screenshots would be very helpful.
Thx
To do: Create my signature with system and expected eGPU configuration information to give context to my posts. I have no builds.
.
Are you using the older razer Core, instead of the V2? I'm thinking of buying a second hand razer Core, but I'm curious if a rtx 207o fits inside it. Considering the pcie slots and all.
I'm looking to fit the gigabyte rtx 2070 windforce inside.
To do: Create my signature with system and expected eGPU configuration information to give context to my posts. I have no builds.
.@sven_breedenbach Yes I'm using the original Core and the Founders Edition RTX cards should fit just fine. Not sure about AIB cards.
• external graphics card builds
• best laptops for external GPU
• eGPU enclosure buyer's guide
2013 Mac Pro (FP D500 x2) [3rd,12C,E] + RX 5700 XT @ 16Gbps-TB2>TB3 (Mantiz Saturn Pro) + macOS 11 [build link]
Wow, I am impressed by your FHD TB3 Unigine benchmarks.
My Laptop: Razer Blade 2016 /i7-7700HQ/GTX 1060 dGPU/16GB RAM
EGPU: Razer Core X + Palit Super JetStream RTX 2080
Sorry, I don't know how to post a picture in this forum but here are my results in Unigine Heaven 4.0 (Ultra/Extreme):
FPS: 112.4
Score: 2833
Min:26.0
Max:217.1
Your benchmarks look better. Do you think this is related to less ram? Or did you overclock your 2080?
To do: Create my signature with system and expected eGPU configuration information to give context to my posts. I have no builds.
.@rainbow47 It was a stock Founders Edition RTX 2080. Are you looking at the Thunderbolt 3 results or AGA results? The TB3 FHD Unigine Heaven is almost identical to yours at FPS: 113.8 Score: 2867 Min: 26.6 Max: 216.1. The AGA FHD Unigine Heaven is much higher at FPS: 131.9 Score: 3322 Min: 28.4 Max: 261.4.
• external graphics card builds
• best laptops for external GPU
• eGPU enclosure buyer's guide
2013 Mac Pro (FP D500 x2) [3rd,12C,E] + RX 5700 XT @ 16Gbps-TB2>TB3 (Mantiz Saturn Pro) + macOS 11 [build link]
Sorry, you're right. I was looking at the FHD AGA numbers. Looks like the direct pci-e connection via the AGA really pays off. Big gap in performance. I guess that we could basically expect from a desktop PC isn't it?
To do: Create my signature with system and expected eGPU configuration information to give context to my posts. I have no builds.
.