AMD has only announced its upcoming Polaris GPU architecture last week and already several Polaris GPUs have been spotted in transit, gettingshippedacross AMD's different facilities. Polaris is the code name AMD has given it's next generation family of 14nm FinFET based graphics chips. These chips are scheduled to launch by mid 2016 and will succeed AMD's current Radeon 300 series and Fury series.

So far AMD, has only demoed the smallest member of the Polaris family and the demonstrated performance and power efficiency results were by far the most impressive we had seen from one generation to the next to date. Apart from this tiny Polaris chip which AMD showed us in a private press event back in December, we hadn't seen any other Polaris GPU. However, AMD does acknowledge that there will be multiple GPUs based on the Polaris architecture to address the entry level, mid-range and high-end segments of the discrete graphics market. And as our luck has it, we might have just spotted the second GPU belonging to the Polaris - Radeon 400 series - family.

Two Potential AMD Polaris Graphics CardsSpotted In Shipping Database

Let's start with the C98 chip. This graphics board made its first appearance in the Zauba shipping database last month. This particular part holds a per unit value that exceeds AMD's Hawaii chips that were listed in the same database by 33%. So we're potentially looking at a high-end graphics cardhere. An opinion that's shared byvideocardzas well.

Then we have the C91/C92 chips. These have shipping dates that go as far back as January of 2015. However the (FOC) suffix which we believe stands for Full Operational Capacity has only been added in August of 2015, a few months beforePolaris was demoed to the press for the very first time. Sowe're possibly looking at the baby Polaris GPU that AMD demoed at CES and a month prior in the private press event that I'd mentioned earlier.

What makes all of this information very interesting is that it indicates that AMD has had Polaris chips up and running in the lab for a while and certainly well ahead ofwhat the company has made publicly known. It's obviousthat thecompany chose to remain tight lipped all this time for competitive reasons. Nvidia's decision to announceitsPascal based Drive PX2 and showcase itequipped withGTX980 MXM chips instead of actual Pascal GPUs was certainly an eye brow raiser.

This decision could very well play into the premise that the FinFET GPU transition has beenkinder toone vendor versus the other. Historically it's been the case that AMD and even ATi before itare the ones tofair better in theirtransitions to new process nodes as opposed to Nvidia. But there's just too much cloak-&-dagger that takes placein the industry to know with certaintyif this is truly the caseright now.