Device ID 687F:C1 surfaces in AOTS - has VEGA 10 aka Radeon RX 490 been spotted ?
Over the weekend, well just before it actually a new entry in the Ashes of the Singularity database appeared, it's coded under number 687F:C1, and that does point a little to the product that should slay the GTX 1080, the Radeon RX 490.
Hardware ID's in the AOTS database show up if the driver has not gotten a proper name lined up with the Device ID. It is not the first time that something like this happens. The hardware ID 687F:C1 is definitely AMD, no doubt there and in fact is a device ID close to Polaris. The more interesting fact is that the results are faster, much faster then the Radeon RX 480. So if it's true, it might be the Radeon RX 490 (I mean what else could it be?). A lot of things are unknown about the Radeon RX 490, but it is rumored to get 8GB HBM2 memory and would be based on a GPU with 4096 shader processors. That means 64 Compute Units, multiply that with 64 shader processors per cluster and you'll get to the 4096 shader processors. The GPU will be paired with either 8 or 16 GB HBM2 memory at an up-to 512 GB/s memory bandwidth, logic would dictate 8GB though.
A few months ago AMD's Radeon Technologies Group (RTG) chief Raja Koduri tweeted that development of "Vega10" had just crossed a milestone, although it's a long way to go before you can see it". A little while ago an RX 490 entry was shortly listed on AMDs own website as well.
The product with ID 687F:C1 scores an average framerate of 88 FPS. In comparison, with the same settings a GeForce GTX 1080 would be in the 70~75 FPS range with the factory overclocked parts in that same 80~85 FPS region. Intersting times.
Check the thumbnails for result.
Senior Member
Posts: 410
Joined: 2016-06-08
I can see how Nvidia is just waiting for AMD to drop that RX 490 then they will surprise the whole world with the GTX 1080ti. Seriously I personally think AMD missed this round completely due to not hitting the performance targets in a similar time to what Nvidia has hit. I mean all this time has passed and the best they can do is a card that performs similar or slightly better than the 1080??? Nvidia must be laughing at this point.
This is just my view guys and nothing to get upset about and yes I know we still have to wait for the true results of the RX 490, but hey where there is smoke, there is fire.
Senior Member
Posts: 6484
Joined: 2012-11-10
I can see how Nvidia is just waiting for AMD to drop that RX 490 then they will surprise the whole world with the GTX 1080ti. Seriously I personally think AMD missed this round completely due to not hitting the performance targets in a similar time to what Nvidia has hit. I mean all this time has passed and the best they can do is a card that performs similar or slightly better than the 1080??? Nvidia must be laughing at this point.
This is just my view guys and nothing to get upset about and yes I know we still have to wait for the true results of the RX 490, but hey where there is smoke, there is fire.
So... you make a bold assumption that a 1080Ti will exist in the first place (yet conveniently forgets about Fury), make an even bolder assumption about the performance of the 490, make a pretty antagonistic opinion about the whole situation, and say "nothing to get upset about".
Right...
Senior Member
Posts: 410
Joined: 2016-06-08
So... you make a bold assumption that a 1080Ti will exist in the first place (yet conveniently forgets about Fury), make an even bolder assumption about the performance of the 490, make a pretty antagonistic opinion about the whole situation, and say "nothing to get upset about".
Right...
Yes, exactly. Let's see how this turns out.
Unregistered
This better not be Vega.. It scored lower than 3 other 1080 results with a score of 8500 and which this unknown card, only scored 8400. How much more driver optimization could the card have that blows the 1080 out of the water? I am sure the 1080ti would beat that, albiet it would cost more money I suppose..

Senior Member
Posts: 3929
Joined: 2014-12-09
Déjà vu.
1 - March 2016:
AMD joined 2 Fury X and called it PRO DUO.
2 - June 2016:
When marketing shows that 2 480s used at misleading 51% (effectively one...) beat a single 1080 at 98.7% usage in a "fair" AOTS benchmark we should believe it.
Fast forward to the present.
3(?¿?) - December 2016:
AMD join 2 480 in a single card and call it 490.
I want to be completely wrong on 3.