|
From the article: "To stay on the market, companies must show that their e-cigarettes benefit public health. In practice, that means proving that adult smokers who use them are likely to quit or reduce their smoking, while teens are unlikely to get hooked on them." It just seems like that is kind of a crazy standard. Are they going to ban all nicotine vapes now? Decreasing nicotine addiction is great, but people will be addicted to nicotine for the foreseeable future. My anecdotal experience is also that people use them to quit successfully.
|
#
?
Jun 25, 2022 03:43
|
|
|
|
| # ? Jun 3, 2024 14:04 |
|
|
MayakovskyMarmite posted:From the article: "To stay on the market, companies must show that their e-cigarettes benefit public health. In practice, that means proving that adult smokers who use them are likely to quit or reduce their smoking, while teens are unlikely to get hooked on them." They’re having to prove that in the aggregate they are a societal good. That’s how good policy is made. Sometimes you die because you wore a seatbelt, doesn’t mean seat belt laws are bad, they save way more people than they kill.
|
|
#
?
Jun 25, 2022 07:08
|
|
|
Space Fish posted:When does the bull trap end so the bear rollercoaster can resume the loop-the-loops? July 1 or July 5
|
|
#
?
Jun 25, 2022 09:24
|
|
|
The acquisition of Datto was finalized the other day, and each share will be paid out $35.50 (not bad considering I paid $23). The symbol $MSP is no longer listed and the money is sitting in my ETrade account with Datto's CUSIP identifier as the symbol with the last traded price listed rather than the payout amount, which is not available as cash. I don't see any fine print in the SEC filing about a waiting period. Is there a standard wait after an acquisition or is this just something that takes a few business days to process?
|
|
#
?
Jun 26, 2022 01:03
|
|
|
GWBBQ posted:The acquisition of Datto was finalized the other day, and each share will be paid out $35.50 (not bad considering I paid $23). The symbol $MSP is no longer listed and the money is sitting in my ETrade account with Datto's CUSIP identifier as the symbol with the last traded price listed rather than the payout amount, which is not available as cash. I don't see any fine print in the SEC filing about a waiting period. Is there a standard wait after an acquisition or is this just something that takes a few business days to process? Sometimes, depending on your broker and the transaction, a corporate action might take a few days to settle. If that CUSIP hasn't turned into cash by Tuesday you should call them. Nice trade, btw.
|
|
#
?
Jun 26, 2022 02:01
|
|
|
What are the remaining good options if one wants to bet against home developers over the next year or two?
|
|
#
?
Jun 26, 2022 03:08
|
|
|
ilkhan posted:What are the remaining good options if one wants to bet against home developers over the next year or two? Shorting XHB is probably the easiest thing.
|
|
#
?
Jun 26, 2022 03:10
|
|
|
Agronox posted:Sometimes, depending on your broker and the transaction, a corporate action might take a few days to settle. If that CUSIP hasn't turned into cash by Tuesday you should call them. Thanks on the nice trade comment, I bought based on public talk about the company and how significantly undervalued all the finance sites said it was, and it turned out they were right.
|
|
#
?
Jun 26, 2022 05:04
|
|
|
ilkhan posted:What are the remaining good options if one wants to bet against home developers over the next year or two? loving poo poo home sales have been priced in for the most part, but there's always a chance it goes lower
|
|
#
?
Jun 26, 2022 05:43
|
|
|
ilkhan posted:What are the remaining good options if one wants to bet against home developers over the next year or two? ASPS is a bet on foreclosures increasing
|
|
#
?
Jun 26, 2022 18:59
|
|
|
Anybody else partial to using the OBV indicator as their go-to?
|
|
#
?
Jun 28, 2022 02:47
|
|
|
Apple MacBook M2/ARM laptop Geekbench score 8900 Thinkpad X13s Qualcomm Snapdragon/ARM laptop Geekbench score 5600 Anything above 4500 is "very very reasonably usable/borderline fast" anything above 5500 is probably faster than any of your friends laptops Modern mid to high end gaming desktop (which aren't thermally throttled, like laptops) is going to clock in about 11,000 Anyways TL;DR Intel is getting some serious competition from ARM chips, they will probably pull out of their rut, but I do not see a future where Intel is the leader over a 20 year horizon The two best laptop lines, not just consumer, but enterprise grade, both offering readily available laptop lines with ARM does not bode well for Intel long term
|
|
#
?
Jun 28, 2022 04:39
|
|
|
lmao at LYT
|
|
#
?
Jun 28, 2022 16:35
|
|
|
Hadlock posted:Apple MacBook M2/ARM laptop Geekbench score 8900 This is way more complicated then a Geekbench score. Among other things, ARM has a really difficult time scaling up wattage, and the way ARM silicon is being designed isn't going to be helping that any time soon. ARM is pretty explicitly targeting the 'Not Serious Compute' market, but is instead capping out around the 'pro(ish)sumer' market. The current strength of ARM chips is that they have the 10 - 35w space on lock I think. x86 implementations absolutely clean up against ARM in production workloads that scale with resources though. A 5950X is still going to clean up on compute and production today, and Zen 4 is a couple months away. For reference, my desktop clocks in a Geekbench score of 17700 on a 5950X, 32 GBs / 3800MTs, FLCK @ 1900MHZ, etc. Now that said, all this could change. I don't do chip design. I can look at the current ARM ideas and point out the obvious, but I don't really understand what the hold up on a 120w ARM part would be, other than that it seems like they don't exist. Maybe ARM will see fatter chips in the future that do awesome things, but extrapolating from the information we have today, the future looks like mobile will be ARM, and serious compute / desktop will be x86.
|
|
#
?
Jun 28, 2022 18:59
|
|
|
Canine Blues Arooo posted:This is way more complicated then a Geekbench score. I hate spec sheet racing, but the M1 Ultras do in the 23000 area as-is on 60w-ish.
|
|
#
?
Jun 28, 2022 19:10
|
|
|
Canine Blues Arooo posted:This is way more complicated then a Geekbench score. AWS and I think Google Compute both offer heavy duty servers with ARM, at a lower cost than comparable Intel or AMD x86 that benchmark same/very similar
|
|
#
?
Jun 28, 2022 19:14
|
|
|
Hadlock posted:AWS and I think Google Compute both offer heavy duty servers with ARM, at a lower cost than comparable Intel or AMD x86 that benchmark same/very similar So do Oracle Cloud and Azure. It think the landscape is changing a lot faster than some people realize.
|
|
#
?
Jun 28, 2022 19:36
|
|
|
BlackMK4 posted:I hate spec sheet racing, but the M1 Ultras do in the 23000 area as-is on 60w-ish. Yeah, it's a really strange, but compelling chip I know LTT did some benchmarks on that box with some interesting results for the Ultra as well (although the details of their bench was light). I'm really interested in how well that does against Raphael, especially as DDR5 gets better. The Ultra right now is the best argument for ARM as the future of general compute. I can't say I love the design of that chip, and I really really don't like the hardware philosophy of Apple at all, but there is something there for sure for ARM as a whole. I'm not totally convinced that it's not a somewhat contrived result, and I really want to see how it progresses with new generations, but it's certainly not dead in the water. And all that said, I might just be dead wrong. I'm not an expert in chip architecture and I understand the surface level of this at best. I don't really have to skillset to really understand the problems they have to solve or how solvable they are. Canine Blues Arooo fucked around with this message at 19:50 on Jun 28, 2022 |
|
#
?
Jun 28, 2022 19:48
|
|
|
Keep in mind that production capacity can be just as important as performance. The best price/performance chip is useless if it's 6 months backordered and you need to build your data center now. If you're prognosticating on future revenues from these competing chipmakers you can't get too narrowly focused on the technology itself. Pay attention to supply chains and production capacities and the balance sheets.
|
|
#
?
Jun 28, 2022 19:50
|
|
|
Yeah can't speak for the other cloud providers but ARM has been available on AWS since 2017, 5 years now, they're on CPU gen2 at least, maybe even further now Interpreted code like python,ruby, JavaScript, probably java don't need any changes to run on ARM. Containerized code can be built using a single flag and deployed to arm servers, is not like it was 10 years ago where cross compiling was a real pain in the rear end and not 100% reliable. Barriers are mostly gone it's just rate of uptake at this point If RISC-V ever takes off, it has a similar chance to upend what's left of intels market share. RISC-V is a free design out of Berkeley with a couple commercial implementations but it's a completely green sheet design independent of x86 or ARM. Deploying code to RISC-V is the same level of complexity as deploying to arm, but not all compilers support the architecture yet (all the important ones do). RISC-V demonstration boards can be bought online for $20 on eBay, Amazon etc but the architecture can scale up to fire breathing compute servers
|
|
#
?
Jun 28, 2022 19:51
|
|
|
Japan's ARM-based Fugaku only recently got knocked down to being the world's #2 supercomputer after an 18-month run in the top spot. 7nm 48-core Fujitsu chips fabbed by TSMC. https://en.m.wikipedia.org/wiki/Fujitsu_A64FX shame on an IGA fucked around with this message at 03:06 on Jun 29, 2022 |
|
#
?
Jun 29, 2022 03:04
|
|
|
Agronox posted:Sometimes, depending on your broker and the transaction, a corporate action might take a few days to settle. If that CUSIP hasn't turned into cash by Tuesday you should call them.
|
|
#
?
Jun 29, 2022 03:50
|
|
|
GWBBQ posted:It did, in fact, turn into cash today. Excellent, glad to hear everything worked out with no problems.
|
|
#
?
Jun 29, 2022 04:01
|
|
|
Talked to a buddy of mine who used to work on CPU design things of some kind for AMD. He basically said, 'x86 is on borrowed time. The genie is out of the bottle and barring some crazy breakthrough, ARM (or some kind of RISCy business, but probably ARM) is almost certainly the future. So there's that. With that said, I'm inclined to move more of my chips toward ARM.
|
|
#
?
Jun 29, 2022 19:17
|
|
|
Canine Blues Arooo posted:Talked to a buddy of mine who used to work on CPU design things of some kind for AMD. He basically said, 'x86 is on borrowed time. The genie is out of the bottle and barring some crazy breakthrough, ARM (or some kind of RISCy business, but probably ARM) is almost certainly the future. So should I try to get a job at Intel or not? A recruiter who has placed me and also placed a lot of my staff with me is telling me I should go join their rotational finance program, but I thought intel keeps getting smoked by AMD and Apple in the chip wars. Myself I prefer the Tostitos scooper chips, they're good for salsa, and to me, chips are mostly just a salsa conveyance, but to each their own.
|
|
#
?
Jun 29, 2022 19:37
|
|
|
From a job security standpoint they're a national defense strategic blah blah, and about to get a significant chunk of a $52B bailout/handout. You'll be retired before they get absorbed by some other company. They're in decline, but it's a very very long glide slope
|
|
#
?
Jun 29, 2022 19:43
|
|
|
Also Intel will eventually pivot to a new architecture if x86 is clearly dying. They have billions annually to spend on R&D.
|
|
#
?
Jun 29, 2022 19:51
|
|
|
Think of it this way: IBM is still around and probably will be until all of us reach retirement age. Intel isn't in any worse position than them.
|
|
#
?
Jun 29, 2022 19:53
|
|
|
LanceHunter posted:Think of it this way: IBM is still around and probably will be until all of us reach retirement age.
|
|
#
?
Jun 29, 2022 20:00
|
|
|
I mean Intel could have done another architecture twenty years ago (they did, Itanium, lol) but to do so they would need to compete with x86-64. They could get into ARM manufacturing, under license(lol) but then they're just another contract mfg Intel has in the last decade done GPU, memory, ssd, none of them were particularly profitable Amazingly, Intel built Itanium processors until 2020, I guess primarily for supercomputers They could contact mfg ARM, or RISC-V, or I guess design a competing product and charge a premium for it, but ARM is so good and so extensible, and RISC-V is completely free, that's not really a winning bet and even I wouldn't place money in favor of Intel at this point to do something like that, even though I'm sure they have a handful of octogenarian PhDs on staff if they ever wanted to reboot their rapidly aging R&D group
|
|
#
?
Jun 29, 2022 20:08
|
|
|
I work in the fabs for Intel (and have for 23 years), and all I can say is we are still hiring and installing a lot of new equipment. Can't speak for the design side of things but at my level things are very busy.
|
|
#
?
Jun 29, 2022 20:28
|
|
|
Investment wise INTC is still priced as a growing company, if x86-64 becomes legacy (and we wont go full stop cut over) they are likely to stay floating for 50+ years. Look at IBM. INTC will go down in price if their future looks flat instead of an upward slope because part of the price is future value.
|
|
#
?
Jun 29, 2022 20:46
|
|
|
Intel's domestic design and growing domestic fabrication surely carries some value, no? TSMC may be run from Beijing in fifteen years.
|
|
#
?
Jun 29, 2022 21:15
|
|
|
I used a lot of ifs there. I think x86-64 is here to stay still. Willing to be wrong winds are saying ARM has a shot, discussing what would happen if that happened isn't wrong. You should have a plan.
|
|
#
?
Jun 29, 2022 21:32
|
|
|
LanceHunter posted:Think of it this way: IBM is still around and probably will be until all of us reach retirement age. Intel isn't in any worse position than them. Same reason I felt like having Pfizer is a no-brainer.
|
|
#
?
Jun 29, 2022 21:48
|
|
|
What metric prices INTC as a growing company? Their p/e is in the single digits, their p/s is below 2 and their p/b is under 1.5. They're actually priced under IBM in that regard.
|
|
#
?
Jun 29, 2022 22:00
|
|
|
ranbo das posted:What metric prices INTC as a growing company? Their p/e is in the single digits, their p/s is below 2 and their p/b is under 1.5. They're actually priced under IBM in that regard. Yeah, they are priced like a company that is stagnant at best. Growth companies usually trade at high multiples based on expectations of higher profits in the future. Leperflesh posted:Also Intel will eventually pivot to a new architecture if x86 is clearly dying. They have billions annually to spend on R&D. Interestingly the first CPU they made on Intel 4 was a SiFive RISC-V CPU: https://www.sifive.com/press/sifive-partners-with-intel-to-spark-innovation-in-high-performance. Small cores like this make sense for kicking the tires on a new process since they will yield a usable number of dies a lot sooner than a big x86 CPU. drk fucked around with this message at 22:19 on Jun 29, 2022 |
|
#
?
Jun 29, 2022 22:16
|
|
|
ranbo das posted:What metric prices INTC as a growing company? Their p/e is in the single digits, their p/s is below 2 and their p/b is under 1.5. They're actually priced under IBM in that regard. I think Intel's biggest bet is on their fabs. Right now, TSMC is running away with the title of 'best' in the market, but not everything needs to be on the latest node. I could very much see a space where Intel uses it's fabs to supply chips for 3rd parties, and making that part of their core business. This is doubly true of Xi decides that they actually want to take Taiwan. The other thing as mentioned before is that Intel doesn't *need* to be married to x86. They could start designing their own ARM stuff (and probably are) as a successor to their current mainstream processors. That is all to say that while I think that x86 might not be the Chosen One anymore, I don't think Intel is doomed to regression because of it.
|
|
#
?
Jun 29, 2022 23:20
|
|
|
Intel also has hybrid designs out now that are pretty good at scaling, and I bet they could plop some arm cores onto a die without too much trouble if their atom designs are facing a bunch of headwinds.
|
|
#
?
Jun 29, 2022 23:35
|
|
|
|
| # ? Jun 3, 2024 14:04 |
|
|
ilkhan posted:Intel also has hybrid designs out now that are pretty good at scaling, and I bet they could plop some arm cores onto a die without too much trouble if their atom designs are facing a bunch of headwinds. Those are x86/x86 hybrids, though. An OS needs to be able to move a process between different types of cores and thats not a thing if they aren't the same architecture. The Atom/Efficiency cores are actually pretty good now - around Skylake in performance.
|
|
#
?
Jun 29, 2022 23:49
|
|