Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: combjelly who wrote (1571113)	11/10/2025 3:06:38 PM
From: Eric	1 Recommendation Recommended By Goose94 Respond to of 1572295

Yep, Just like The Donald.. Trying to buy his way out of serious trouble.

To: combjelly who wrote (1571113)	11/10/2025 3:12:01 PM
From: Tenchusatsu	3 Recommendations Recommended By Bonefish Goose94 pocotrader Respond to of 1572295

CJ, dumbass MAGA cultists think liberals are trying to usher in The Great Reset, where you will "own nothing and be happy." Yet they cheer on this idea of a 50-year-mortgage where you will end up doing nothing but paying tax-deductible rent money to the bank. We're all socialists now ... Tenchusatsu

To: combjelly who wrote (1571113)	11/11/2025 2:17:28 PM
From: Tenchusatsu	Read Replies (2) \| Respond to of 1572295

Hey CJ, check this out ... Tesla's FSD chips: Integer vs. Floating-point At Tesla's shareholder meeting Elon said that, "integer operations are fundamentally more efficient than floating-point operations...integer is much more power efficient, much more silicon efficient, but you actually have to train… pic.twitter.com/TUKPrCVYVZ — Cern Basher (@CernBasher) November 9, 2025 This is interesting, but I also think it's another instance of Musk believing he's smarter than everyone else when it comes to topics that he doesn't specialize in. For example, he posted the following: As the number of bits drops, the difference between floating point and integer decreases until they are the same thing at 1 bit. “Floating point” is not real. It is emulated with 2 integers and a lot of complexity. — Elon Musk (@elonmusk) November 9, 2025 Like what does that mean, "Floating point is not real?" Of course it's real, and no, it's not the result of emulation with "2 integers." Instead, it's the result of the IEEE 754 standard. As for Tesla focusing more on integer (i.e. fixed-point) math over floating-point, I'm skeptical. Yes, CPUs can process fixed-point math more efficiently than floating-point, but what's the cost to the AI inference model when all the data points are represented by fixed-point numbers instead of floating-point? I actually don't know that much about AI training and AI inference. But the mathematic operations, I believe, are similar to that of 3D graphics, and I know a thing or two about that. The whole "integer vs. floating point" thing was already resolved back in the late 90's when the 3D video game Quake was getting popular. Floating point is better, period. It was also the topic of debate between the Intel Pentium, with its vastly superior FPU, and the AMD K5. AMD had to beef up its FPU with the K6 in order to compete (and edge out) the Pentium II. Back to the Muskperger. This guy may be intelligent, but part of intelligence involves knowing what areas you're not that smart person in. And I guarantee anyone that Musk doesn't know WTF he's talking about WRT integer vs. floating-point processing. Either that, or he's collectively smarter than every single engineer in nVidia and AMD's GPU departments, which is a load of crap. Tenchusatsu