Home / Component / Graphics / AMD begins to test ‘Fiji XT’ graphics card internally

AMD begins to test ‘Fiji XT’ graphics card internally

Advanced Micro Devices has begun internal testing of what is believed to be the company’s next-generation flagship graphics processing unit. Earlier this month the company’s subsidiary in India received a graphics card based on the code-named Fiji XT graphics processing unit.

According to Zauba.com database, which monitors imports and exports to and from India, where high-tech companies employ many engineers, AMD’s office in India received a “printed circuit board assembly (video graphic card) C880 Fiji XT P/N.102-C88001-00” on the 7th of November, 2014.

Videocardz web-site believes that AMD’s code-named Fiji XT graphics processing unit is the company’s next-generation flagship GPU that will power the Radeon R9 390X graphics card. It is believed that Fiji XT is based on the GCN 1.2 architecture and is made using 28nm or 20nm fabrication process.

amd_radeon_r9_390x_cooler_asetek

Given the fact that this is the first Fiji XT sample that AMD sent to its office in India, it is highly likely that commercial graphics cards based on the GPU are at least three to four months away.

AMD did not comment on the news-story.

Discuss on our Facebook page, HERE.

KitGuru Says: Introducing an all-new graphics card in the first quarter when sales are low is not the best idea, but keeping in mind that AMD needs to fight back market share from Nvidia, it looks like the Sunnyvale, California-based company will not have a choice.

Become a Patron!

Check Also

First AMD UDNA GPUs expected in 2026

AMD's unreleased UDNA GPU architecture is back in the news, with a fresh leak suggesting …

6 comments

  1. AMD’s previous CEO already clarified that nex gen GPUs would be in 20nm, other sources (which I cannot name at the moment and cannot confirm) had already clarified that the transition to 20nm has already been made in a certain chip. Hoping this gpu finally gets it right…

  2. I predict that you’ll see some staggering numbers courtesy of AMD’s use of HBM, something that will no doubt remain exclusive to AMD for some quite time (they worked on HBM in conjunction with Hynix).

  3. Yes the cake is not a lie, AMD Fan-boys can get ready to upgrade, that’s of couse you don’t mind getting the same 28nm, because TSMC still got problems, and we won’t see NO 20nm graphics crads at all come 2015. They are reporting that 2016 will be the year where we see the new out of date 16nm process. So two years we have been waiting for TSMC to get their crap together and out the door, it’ll be even longer when the time comes. Completely unbelievable if you ask me.

    PhoneyVirus
    https://twitter.com/PhoneyVirus
    https://phoneyvirus.wordpress.com/

  4. You always have so much information on every article I see you on, yet you really don’t know what the shit is you’re spewing.

  5. Would make sense if AMD was using TSMC for their 20nm chips. Considering that they’ve already confirmed using 20nm on the new cards, and that they’ll be ready by February, I don’t think TSMC is going to be getting their business for this.

  6. Only until Nvidia releases Pascal in 2016, then they’ll both be on the same memory again and things will be really interesting…