AMD at Credit Suisse Technology Conference has not spoken about something very realistic or specific on their plans for upcoming candies, but they give us a hint on their upcoming graphics cards process node. At very first move to conference, AMD has given the clearest chit that the upcoming Fiji GPU will be at the 28nm Process node and after some time they will move to FinFETs.

AMD teases that Pirate Islands GPUs to have 28nm Node – 20nm maybe for next Gen of APUs

Not an official slide.

Not an official slide.

Previously everyone is assuming that AMD will go to the 20nm process node which is now in the production from few months. But we have also seen TSMC has not given any words on AMD production plans, although they said for Nvidia that they will be the first to use 16nm process. Also from the AMD side in the past few months we have not heard any news on 20nm process node. It may come as a shocker for many players who are waiting to see the Pirate Island GPUs on the 20nm process node. Also see the qoute from AMD which is somewhat relevant to the topic and spread more light on it.


.from an overall standpoint, today the bulk of our products are 28 nanometers. We will have certain products in 20 nanometers and then we’ll go to FinFET from there, but our partnership with GlobalFoundries is where it comes into play in terms of what’s the right point to go and intersect products with technology and introduce the parts out in the market.


Mr. Devinder Kumar mentions that the majority of our [AMD’s] offerings are on 28nm and that this majority will transition directly to FinFET. It seems like AMD is planning to rebrand their upcoming R9 300 series with the one flagship on 20nm (which is unlikely), AMD is hinting at Pirate Islands on the 28nm Node. We have already run the numbers based on initial leaks and found that the performance jump shown is very much possible on 28nm and within the 600mm^2 die limit of TSMC

If we see the whole scenario, there are lots of chances of Carrizo APU being in the 20nm process. 28nm to 16nm FF+ makes perfect sense.

You can find more about the Credit Suisse Technology Conference (Transcript) here.