AMD’s Azor: No Ampere-Like Shortages for RX 6000

BY admin September 27, 2020 Technology 2 views

Nowadays, the other big topic besides Nvidia’s GeForce RTX 30-series (codename Ampere) graphics card is their lack of availability at retailers. The GeForce RTX 3080 and RTX 3090 might be two of the fastest gaming graphics cards right now, but they’re worthless if consumers can’t buy them.

(Image credit: AMD)

 

Nowadays, the other big topic besides Nvidia’s GeForce RTX 30-series (codename Ampere) graphics card is their lack of availability at retailers. The GeForce RTX 3080 and RTX 3090 might be two of the fastest gaming graphics cards right now, but they’re worthless if consumers can’t buy them.

We suspect there was a short supply for Ampere in the first place, and to make matters even worse, scalpers were using bots to pick up every GeForce RTX 3080 in sight to later flip them on eBay for outrageous prices. You know the shortage is real when Nvidia came out yesterday and confessed that the GeForce RTX 3090’s supply was going to be scarce as well. 

Frank Azor, Chief Architect of Gaming Solutions at AMD, took to Twitter to have some fun. Twitter user @AndreElijah wrote “Welp… not being able to pick up a 3090 today means my work is going to be [bleeped] for the next little bit. Can they at least release a new Quadro so I can get my work done? $10 says AMD will be a paper launch too.” Azor snapped back with a witty remark saying “I look forward to taking your $10 :)”

20200927.AMD-Azor-No-Ampere-Like-Shortages-for-RX-6000-01.PNG

(Image credit: Twitter)

 

Azor’s statement implies that supply for AMD’s next-generation Radeon RX 6000 (popularly known as “Big Navi“) won’t be an issue, and we believe him. Nvidia took a big gamble by picking Samsung to fabricate the Ampere dies for the GeForce RTX 30-series graphics cards. Samsung basically tailored its 8nm manufacturing process to Nvidia’s requirements. The 8N Nvidia process node is technically new, and we suspect the yields might not be that great, meaning the process could require time to mature. 

AMD, on the other hand, intelligently taps into TSMC’s proven 7nm FinFET manufacturing process. Although it’s not confirmed yet, the Radeon RX 6000 should still be based on the 7nm process node with a last reported defect density of 0.09 defects per square centimeter. Rumors have it that the Navi 21 silicon, the one that’s speculated to power the Radeon RX 6900, could have a die size of 505mm square. Assuming that TSMC’s defect density has improved since the last report and the die size for Navi 21 is accurate, AMD could be reaping a yield rate close to 65% for Navi 21, which isn’t too bad.

The fight between the Radeon RX 6000-series and GeForce RTX 30-series will surely be interesting. If Big Navi is able to match Ampere in both performance and pricing, availability will ultimately determine the final winner.

Comments

write your comment.

Your email address will not be published.