The latest rumour to hit the AMD Vega mill centers around pricing , nordichardware.se yesterday reported that the upcoming graphics card will be priced close to the same bracket as a GTX 1080ti @7000SEK(~€730 ), which is the cost to retailers. Once you factor in VAT and a profit margin for retailers you could be looking at a price of 9000SEK or ~€940. Considering here at Ciaran's Customs you could get a 1080ti fitted installed and tested for Under €850 we are hoping that this price is either a very high estimate or the cost of a hybrid Air/Liquid cooled version. The swedish hardware site also is hopeful for preorders to start next week for the highly anticipated card.
On a more positive note hardocp.com claim to have been given a sort of reference version of the card to run some rudimentary tests on, unfortunately this boiled down to a very subjective and unscientific double blind hands on gaming session for some lucky people in the know; whole host of ESPorts professionals, one formerly of id Software, gaming journalists, and a few hand picked hardcore gamers, with some claim to fame. A RX Vega FreeSync vs. GTX 1080 Ti G-Sync in a 1 game only battle, DOOM(2016). The RX Vega was paired with a very nice, but pricey, ASUS MX34V FreeSync display (~€800/£717), and was pitted against a 1080ti test system paired with an even pricier ASUS PG348 G-Sync display (~€1115/£990) (Yes both of them monitors probably cost more than most people's PCs).
| "Basically what I am saying here is that all hardware and software was under our control at all times to ensure an even playing field." |
Both test systems are identical outside of GPU and Display. AMD Ryzen 7 1800X CPUs with 16GB of 2666MHz RAM. These systems were sent to hardocp fully loaded and ready to go. They immediately formatted the SSDs, and flashed the UEFI on both machines. Hardocp.com themselves installed Windows 10 64-bit OS on each machine, and installed the NVIDIA drivers. An AMD representative did install the RX Vega into the AMD test system as well as install the driver package.
Blind Game Testing Procedure:
- One player start gaming on "System 1," play until comfortable, then move to "System 2."
- Once they felt they had a good hold on gaming on each system, they ran them back through the level on each machine again.
- So a total of 4 system run throughs, all without breaks.
- After that, they got brought over and gave a quick interview and were asked about performance and value.
| I had hoped to use three games in our testing, it became quickly evident that we did not have the time and amount of equipment needed to pull that off inside six hours. 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. |
Both systems were fully shrouded and disguised so as to not be able to tell which system was which. Steps were also took to disguise the GUI from showing which system was which. The gameplay testers were not aware of any of the hardware being used. Since AMD have not released a MSRP for Vega the $300 price difference they talk about mainly comes from the cost difference of the monitors.
Ciaran's Conclusion:
Both bits of news tell us absolutely nothing really;
First, the pricing is still very much a rumour, if it is for a standard card it is a pretty steep ask, but if its going to be a AIO liquid cooled card that has higher base clocks than the reference maybe its not such bad value. Only time will tell
Secondly the hardocp blind testing. Both test systems were top of the range high end components paired with an adaptive sync technology, meaning as long as they both averaged 60-100FPS they should be very smooth. The problem I have is we got no figures no average, min or max FPS, no temperatures or clock speeds. So extremely subjective. Which admittedly is down to AMD limiting them, so im not blaming Hardocp for this one. Yes it is good news for AMD that it was hard to tell between the new Vega and a 1080ti but the problem is we know a 1080ti isnt really going to be pushed to the extremes by Doom@3440x1440, while we don't know if the RX Vega was working hard or not. I would of much rather seen the same test done @4k then maybe we would of seen a proper challenge. This is like comparing two cars in a race but telling the drivers they can't go over 100mph.
Bottom Line:
.........
(Thats the Joke, there is no bottom line, as is it's too hard to draw any conclusion from such limited info)
Anyway here is the full Hardocp video for your enjoyment.
Sources and Images hardocp.com, nordichardware.se