facebook rss twitter

AMD RX Vega Vs Nvidia GTX 1080 Ti blind test video published

by Mark Tyson on 28 July 2017, 13:31

Tags: AMD (NYSE:AMD), NVIDIA (NASDAQ:NVDA)

Quick Link: HEXUS.net/qadj34

Add to My Vault: x

Earlier this week we saw the first photos published of an AMD Radeon RX Vega consumer graphics card (reference model) in the flesh. The respected source was Kyle Bennett, the editor in chief at HardOCP. Now Mr Bennett has delivered, as promised, the full video recorded comparison on two PC systems, identical apart from their GPUs and displays.

HardOCP sought to make its blind test as scientifically design as is possible with such a test, which is after all based upon subjectivity. To start with Bennett personally wiped, checked, re-instated and verified the two PC systems. They were based on AMD Ryzen 7 1800X CPUs with 16GB of 2666MHz RAM. Their identical SSDs had a fresh install of Windows 10 64-bit.

For the monitors, an ASUS MX34V FreeSync display ($720) and an ASUS PG348 G-Sync display ($1300) was chosen for the respective AMD and Nvidia systems. Both are 100MHz panels. To finish off the test setup the systems were "fully shrouded and disguised so as to not be able to tell which system was which," that includes monitor markings, and any software Windows/GUI indicators.

Now onto the panel of 'judges', Bennett summoned a number of eSports professionals, one formerly of id Software, gaming journalists, and "a few 'kids' that I know are all hardcore computer gamers". All subjects are active gamers who regularly splash lots of cash on PC hardware.

During the testing all the subjects got to play DOOM (Vulkan) at 3440×1440 on the first system (Nvidia) until they felt comfortable enough to have gauged its performance, then they tested the same game on the second system (AMD) and later got to try both systems again. Bennett explained that "I chose to go with DOOM because I knew this was a title that all our game testers would be comfortable with and the fact that I invited primarily FPS players." Unfortunately time was against the plan to test further games (3 in total were lined up) with the 10 judges.

At the end of the test day at HardOCP HQ in Dallas, six of the testers thought the systems were basically equal. One person judged the Nvidia system to be superior and three thought the AMD system was the best. Interestingly, of those whom preferred the AMD RX Vega system, they commonly talked about objects on screen being easier to track and the system being 'snappier'.



HEXUS Forums :: 19 Comments

Login with Forum Account

Don't have an account? Register today!
Blind tests are great and all but people really are going to be focusing on numbers as that's what 99.9% of all reviews are based on. It may cut the mustard in usability but systems are built and bought on specs :/
One thing just to note,is that one monitor used a VA panel and the other an IPS panel.
Price,price,price.

That's all that matters with Vega, the performance metrics won't be a surprise to anyone, but the price certainly will.

It just remains to see whether it's a pleasant surprise or a card killing one.
Though even reviews are starting to gather metrics that are important in your experiences with the product, like frame times and 1% and 0.1% lows, for e.g., that can be jarring if they're particularly bad. Both hard figures and subjective experiences are useful information. Still, I hope this blind test doesn't go to AMD's head when it comes to pricing RX Vega, they're going to have to undercut the GTX 1080 by a good bit if they want to have a chance.
People don't normally buy hardware based on subjective judgments.