tgtweak

24 ms·

331.
by tgtweak 3y ago
Highly doubt this was the initial idea for nvidia given they were graphics-only for a very long time. CUDA definitely felt like more of a value-add for the first 5-6 years than a concerted effort to build accelerators and to fund that with
332.
by tgtweak 3y ago
I suspect these can be patched in as well - most of these functions have implementations in CUDA implying they should be able to run on the hardware even without dedicated instructions.
333.
by tgtweak 3y ago
It's not the license expiration date is just the annual date that you need to pay your license renewal fees on.
334.
by tgtweak 3y ago
I have a friend who was born on Feb 29, and in Quebec your driver's license fees must be paid on or before your birthday or your license is effectively revoked (It's a convenient reminder). He was on his way to pay them on the 29
335.
by tgtweak 3y ago
at $28,000/MT it's rare enough to worry about - for context it looks like a model 3 has about 4.5KG of cobalt in it's entire pack, or around ~$130 at current spot rate.
336.
by tgtweak 3y ago
I guess if you're a massive recycling facility and working on 30,000+MT of raw batteries as input, the 2% extra could equate to a fair amount, certainly when the spot price is $28,000/MT. Considering spot prices for cobalt are dro
337.
by tgtweak 3y ago
I don't really get the impression that current recycling techniques are lacking... typically the battery packs are immersion-shred and separated into the metal, plastic and the cathode "paste" which is a mix of lithium, manga
338.
by tgtweak 3y ago
I don't think there's a way they can enforce that reasonably. There's no connection to the mothership to report back what version is being used or license keys at runtime... Seems more like a "if we discover something u
339.
by tgtweak 3y ago
What market were they in? What did the growth and earnings look like?
340.
by tgtweak 3y ago
It's a good idea for a feature - many other out-of-band solutions have this. I think the issue comes mostly in that you would need to unplug the server to add (and remove...) some kind of relay-controlled power interrupt. I would assu
341.
by tgtweak 3y ago
That is true, and you would assume that if he sold he would either work retained and draw a salary or hire someone at a fair cost. I think the company is too early to realistically sell - but I don't think the value today is zero - it&
342.
by tgtweak 3y ago
You don't need to sell it - you can lend against it or sell part of it. I don't think it's a toxic mindset but a financially literate mindset.
343.
by tgtweak 3y ago
The reality, and this is how a private party, PE firm, traditional bank or public market will determine this: The company is generating profits currently at this rate Y, and is growing at this rate X, that means that over 3 years it should
344.
by tgtweak 3y ago
on $10/day max spend you'll get 1-2 people that will visit the page, and you can put a "coming soon, leave an email and we'll let you know" then you can email them when it is deployed as well to recoup the adspend.
345.
by tgtweak 3y ago
10x ebit is not 10x rev - I think hardware startups that can demonstrate economies of scale and straight line growth on advertising + additional product segments and markets are catching very healthy valuations still - not saas levels but
346.
by tgtweak 3y ago
I don't think there were any insinuations that that was after-tax income in his personal bank account - just that the company earned that much in the previous year and it could be paid out (dividends or otherwise) after much in the way
347.
by tgtweak 3y ago
I think people (and the founder) are focusing on yearly profits as their remuneration and comparing it to a salary... but the reality is you're creating a company that should be valued (and eventually sell) for 7-15X Earnings - and yo
348.
by tgtweak 3y ago
>com.google.android.apps.bard is it though... Also, no availability outside of the US? My go-to on android right now is copilot which is basically free gpt4 turbo and dall-e 3 (and also available outside of the US). https://pl
349.
by tgtweak 3y ago
Intel was literally at $25 in February last year, that's the 10-years-ago stock price. Intel's P/E was about 50 at that time, AMD was over 1000. That's "throwing in the towel" in as pointed of a way that the m
350.
by tgtweak 3y ago
You can't compare optimized clock frequencies on 2 year mature process with first run on a new process... AMD and Nvidia both improve stable frequencies with process improvements at TSMC even on the same nodes over time (RTX 4060 TI, t
351.
by tgtweak 3y ago
Gelsinger put a $4BN compensation package in effect for securing and retaining talent within his first 6 months of taking the role, one of the first things noted was brain drain to competitors. >Intel Poaches Head Apple Silicon Architect
352.
by tgtweak 3y ago
They bought the stock on the principle that it was going to pay a consistent 5% dividend every year and weren't looking for moonshots at the cost of that consistent revenue.
353.
by tgtweak 3y ago
Licking County (next-gen, post-18A) has already broken ground and is in assembly, Magdeburg and Ireland (18A) also well underway and in production. Arizona's 20A facility (Fab 52 and Fab 62) have been done for half a year and are alre
354.
by tgtweak 3y ago
Intel is still behind TSMC at this point in terms of raw process efficiency, but that rate of change is moving quickly and I posit that the products released later this year will have a process efficiency advantage over AMD's offerings
355.
by tgtweak 3y ago
If you think the concern over China and Taiwan is understated I think you'd do well to look at how both the US and China are putting insane amounts of resources behind this.
356.
by tgtweak 3y ago
Intel is recipient #1 of CHIPS and similar EU initiatives - and the government may pressure nvidia and other US companies (ie: apple) to move their procurement domestically. Intel being the only player outside of taiwan and south korea to
357.
by tgtweak 3y ago
I'm kind of bullish on Intel right now. They've moved up so many process nodes so quickly and have made some earnest headway in being an actual fab. Let's ignore the elephant in the room which is taiwan and it's sover
358.
by tgtweak 3y ago
Seems pretty trivial to support Python serverside. Python is very easy to run in a lightweight isolated container (and you can compile it to static executable). The game data that the engine makes available to the server side code looks pre
359.
by tgtweak 3y ago
Was more referring to the process and professionalism of the rail employees and procedures that I was notified right away vs the fact nothing was stolen - but I noted it since they opened up the bag to find my ID inside and reach out.
360.
by tgtweak 3y ago
DGX GH200 is next gen, but afaik not available to anyone aside from internal partners. All the quoted data here is for several of them together in a chassis. >Each NVIDIA Grace Hopper Superchip in NVIDIA DGX GH200 has 480 GB LPDDR5 CPU

More ›