If you have been following the latest tech news and Nvidia’s roadmap since past few months or so, then you might have heard rumors that Nvidia was allegedly prepping to launch more SKUs in the current Ampere GPU lineup.
The following SKUs were on the roadmap according to previous rumors, the RTX 3080 20 GB, RTX 3070 16 GB, & RTX 3060 Ti/Super, respectively. Even GALAX’s roadmap and Gigabyte’s Watch Dogs Legion code redeeming website also hinted that Nvidia plans to release more SKUs based on the 8nm Ampere architecture.
Now two independent sources have told Videocardz that Nvidia has apparently cancelled the following Ampere SKUs, the GeForce RTX 3080 20GB and RTX 3070 16GB, respectively. The AIB custom board partners have been given this information, but since we can’t verify the legitimacy of this rumor/source, you should take this leak with a grain of salt.
Nvidia has made no official announcement or given any statement on this issue yet. These cards were planned for a launch this December. Do note that now the sources claim that these two SKUs have been cancelled, and not postponed. The RTX 3070 Ti model was also cancelled before.
PSU manufacturer Silverstone had also listed some list of supported GPUs before, but later on removed the RTX 3070 Ti model entry from the list.
— 188? (@momomo_us) October 20, 2020
We were expecting Nvidia to release the RTX 3080 20GB variant in response to AMD’s NAVI 21 GPU lineup, but there is no official reason given for this cancellation. One of the source claims that the RTX 3080 20GB might have been cancelled due to low GDDR6X memory yield issues, or poor availability.
The RTX 3070 16GB variant uses the GDDR6 memory type, but the reason behind the cancellation is also unknown.
But according to me, one major reason could be due to low yield issues of the current SAMSUNG 8 nm fabrication process node, rather than the Memory. We also know that the GeForce RTX 30-series Ampere lineup currently has a very tight stock, and the demand is also very high.
And, the GeForce RTX 3080 and 3090 GPU shortages will also last until 2021, according to CEO Jensen Huang. Even Komachi claims that the issue is with the Samsung’s 8nm/8N process node yields rather than the GDDR6X memory availability.
— ????@Komachi (@KOMACHI_ENSAKA) October 21, 2020
There are also rumors circulating on the web that Nvidia might tap TSMC’s N7 process node for the Ampere 8nm refresh/successor lineup in 2021. This could also be a major deciding factor, though nothing can be confirmed as of now.
According to Tom’s Hardware, there’s no question that TSMC’s N7 process is superior to Samsung’s 8N in nearly every important metric, except price and available production capacity. Besides N7, TSMC also has N7P, N7+, N6, and N5 nodes all in active production.
In other words, there are a lot of options available for Nvidia to consider, any one of which should be better than Samsung 8N, provided TSMC and Nvidia can come to some sort of agreement on price.
We also have to assume that any GPUs manufactured on one of TSMC’s advanced nodes would perform better than the current GA102/GA104 SKUs, so it makes sense to drop Samsung’s 8nm process node.
Stay tuned for more!