I don’t think this is an “us” so much as “you”. Not a criticism, just I think that your particular position on the matter is a little more thought out than others
This is what I was referring to when I said “scrapped” before. I don’t think this is happened, but I don’t think it’s absolutely impossible.
As far as I can tell, this is the rationale here. Please tell me if I’m missing something
- Drake is referred to in two places (a leak from kopite7kimi and once in the Nvidia hack)
- Dane appears awful big, going by the hack.
- No new info about release since September
- Chip shortage and Switch success make Pro a bad idea economically right now
- Ergo, Dane is an beefier version of Drake targeting “successor” status.
I’ve considered this possibility but here is why I don’t buy it
- paying the kill fee on broken contracts with 3rd party devs drives up costs
- There are likely dedicated porting engineers for new hardware who would be let go
- The japanese version of Mochi’s article includes devs stating that delaying the revision would kill trust in Nintendo
- Killing Pro would likely cause more leaks than “staying on target”
- And probably costs more than releasing revision that, due to capacity problems “only” sells 20 million units.
- The current switch is already throttled by the chip shortage. Wildly enough, a redesign is their best opportunity to reduce their need for non-SoC chips and expand production
The stronger financial pressure on Pro is likely inflation, a far more variable situation, and one which might be hitting Nvidia harder than Nintendo if the prices for the silicon were set 2 years ago
I’ve also considered that “yes a pro is coming but Drake isn’t it - Drake is the beefy successor that is also in development.” That feels more possible to me, just as a gut check. But if they’re really doing that and targeting a 7+ year generation, and 5nm, then why rework Ampere/Orin? Why not just build on Ada/Atlan?