Intel and AMD compete with each other, shipping raster GPUs.
AMD also ships CDNA, which is a legitimate competitor to Nvidia's hardware (albeit not a CUDA replacement). Breaking up Nvidia into B2C and B2B portions would probably end up killing CUDA in the consumer GPUs (which enthusiasts rely on), and send the value of datacenter GPUs skyrocketing even further. Nvidia's competitors wouldn't benefit from that.
I treat this as a sign of the inevitable terminal phase of any private cooperation in US: they start to invest their resources to stretch into areas that are irrelevant to their core mission, but they have to, because their core mission is accomplished already, and their business, which is built on top of the fruits of their core mission, needs protection beyond direct means relevant to their core mission.
When pursing a core mission, many corps fail, either because lack of resources, or mission itself is not meaningful, execution failure, etc.; the fortunate few, might be in growth, while their mission keeps self-reinforce into bigger and bigger scope, that's the wonderful time of growth driven by meaningful mission.
When the core mission is done. The corp resets.
Vast majority of them make profit that is small enough that the normal corporate logic prevail, and they behave normally, primarily to survive.
Extremely rarely, when the core mission is so meaningful, the corp reaps so large a profit, and so many stakeholders are attracted, that they no longer focus on core mission, they start to focus on profit.
That's why NVDA is now buying assets across the whole stack of AI, while still increase price to maintain its profit. NVDA wants to maximize profit.
The inherent flaw is that private corp in US system, has no meaningful end game when they are in this state of astronomical profit.
Their biggest customers are struggling to build out locations to use their products. This is nothing more than an attempt to engage in regulatory capture to overrule local politics on behalf of their customers.
> Their biggest customers are struggling to build out locations to use their products.
That's true.
Although, the fact remains that any large corp in US, eventually becomes too big to make sense to pursue their core mission.
Essentially, NVDA does not need this much profit to continue its core mission or try to expand the scope into more meaningful ventures. But profit seeking inevitably would make them to go blindly.
They'll find reasons to do that, as profit itself is non-binding to any particular person, ideology, principles etc.
Shame that the local residents that are angry at data centers being built near them voted for politicians that… don’t give a shit if data centers are built near them.
The company first invests in engineering/technology/product, in order to actually build a product, find the market fit, and innovate.
Then the company invests in operations to scale the success they are having.
Then the company invests in accounting and finance to squeeze costs and move numbers around, so they can make line go up without physically doing anything.
Then the company invests in lawyers to attack competition and defend their business from attacks.
Finally, the company invests in lobbying to shape the rules such that they don't even have to do any of the above four to make line go up.
Is the problem here NVIDIA or is the problem that our government doesn't function unless there's a pac funneling money into decision-making?
Hardly surprising to see a company with strategic interests strategically try to achieve them.
A business exists to make money, and if pulling policy levers is required to make money, you'll see them try to pull the levers.
Reverse Citizens United and you solve these kind of problems at the source.
So the first step down the lobbying-is-good-for-governance that both sides embraced was when they defunded the congressional research office. So a vital way in which congress "informed themselves" was basically minimized to the point that the only way for congress to now inform themselves is via lobbying and th eonly way to write complex legislation is via those lobbyings.
This is the endpoint of that effort and the people who pushed it most likely knew what they were doing.
PACs are an investment. You'd be stupid (principles be damned) not to invest. (Not defending this--this is the system that promotes wealth hoarding and theft of government funds)
The mundane solution is to give congressional offices more money for in-house staff with expertise on technical issues.
A lot of the "lobbying" is not really pressure groups, it's sort of a corporate-financed knowledge base congressmen can ask questions when they want to understand like, how to expand the semiconductor manufacturing base in their state. Of course it's not all altruistic on the PAC's part, but it is seen positively by individual congressmen because it solves a real need on their side they can't meet with underpaid 24 year old staffers.
Hopefully Huang will ask for tariffs on solar panels & batteries to be lifted. After several years of tariffs, there is no American solar panel industry anywhere near as good as China's. Open the floodgates & reap clean, silent energy at 5-6 cents per kWh. There's plenty of land for those panels.
AMD also ships CDNA, which is a legitimate competitor to Nvidia's hardware (albeit not a CUDA replacement). Breaking up Nvidia into B2C and B2B portions would probably end up killing CUDA in the consumer GPUs (which enthusiasts rely on), and send the value of datacenter GPUs skyrocketing even further. Nvidia's competitors wouldn't benefit from that.
When pursing a core mission, many corps fail, either because lack of resources, or mission itself is not meaningful, execution failure, etc.; the fortunate few, might be in growth, while their mission keeps self-reinforce into bigger and bigger scope, that's the wonderful time of growth driven by meaningful mission.
When the core mission is done. The corp resets.
Vast majority of them make profit that is small enough that the normal corporate logic prevail, and they behave normally, primarily to survive.
Extremely rarely, when the core mission is so meaningful, the corp reaps so large a profit, and so many stakeholders are attracted, that they no longer focus on core mission, they start to focus on profit.
That's why NVDA is now buying assets across the whole stack of AI, while still increase price to maintain its profit. NVDA wants to maximize profit.
The inherent flaw is that private corp in US system, has no meaningful end game when they are in this state of astronomical profit.
That's true.
Although, the fact remains that any large corp in US, eventually becomes too big to make sense to pursue their core mission.
Essentially, NVDA does not need this much profit to continue its core mission or try to expand the scope into more meaningful ventures. But profit seeking inevitably would make them to go blindly.
They'll find reasons to do that, as profit itself is non-binding to any particular person, ideology, principles etc.
Then the company invests in operations to scale the success they are having.
Then the company invests in accounting and finance to squeeze costs and move numbers around, so they can make line go up without physically doing anything.
Then the company invests in lawyers to attack competition and defend their business from attacks.
Finally, the company invests in lobbying to shape the rules such that they don't even have to do any of the above four to make line go up.
Hardly surprising to see a company with strategic interests strategically try to achieve them. A business exists to make money, and if pulling policy levers is required to make money, you'll see them try to pull the levers.
Reverse Citizens United and you solve these kind of problems at the source.
This is the endpoint of that effort and the people who pushed it most likely knew what they were doing.
I think he laid it out very clearly, but never found traction to try to change this (I don't think US Congress wants to be changed, though).
I mean you can dislike PACs but it's pretty delusional to assume most people think they are categorically bad. Clearly there is an echo chamber here.
And to be clear, NVIDIA started a PAC, not a super pac.
PAC approval (by Public Affairs Council Pulse):
[0] https://pac.org/public-affairs-pulse-survey-2020[1] https://pac.org/impact/democrats-pro-business-republicans
[2] https://pac.org/poll-americans-concerned-about-honesty-openn...
That is the antithesis of how politics should function (people are the interest, not moneyed companies)
https://archive.fo/Zd1LP
https://news.ycombinator.com/item?id=41765734
A lot of the "lobbying" is not really pressure groups, it's sort of a corporate-financed knowledge base congressmen can ask questions when they want to understand like, how to expand the semiconductor manufacturing base in their state. Of course it's not all altruistic on the PAC's part, but it is seen positively by individual congressmen because it solves a real need on their side they can't meet with underpaid 24 year old staffers.
Mark Hanna was a political consultant in the late 1800's, in the US.
Source: https://quote.org/quote/there-are-two-things-that-are-import...