The part comparing "Intel Xeon E3 1320 V6" [sic] with older "Intel Xeon E5620" as having the same memory bandwidth, same TDP, more L3 cache, etc., so having "the same performance characteristics" is just plain crazy.
1. Xeon E3 1320 never existed. Was it meant to be 1230?
2. Does DDR4-2400 really have "the same performance characteristics" as DDR3-800 (even after accounting for having 3 channels instead of 2)?
3. How can a 3.5 GHz processor be "the same performance characteristics" as 2.4 GHz one (1.5x more)? Add IPC on top of that.
3) GHz cannot be used as a speed comparison outside of the same generation of CPUs. A 10 year old 3.5GHz gets nowhere near the same compute power as a modern day 2.4GHz.
For a lot of basic operations, this is often false.
I have an application I'm working on and tested it across three different generations of CPUS going ~10 years back.
The code was written in C and the fundamental steps were to: mmap() two arrays of structs from a sata 3 SSD, compare the two, and write the intersection to a block of memory. The data size of each array was multiple gigabytes and the operation took more than a full second.
The ~10 year old low powered Xeon L5640 beat the higher-speed ~5 year old Xeon E5-2667 v2, and was only slightly beaten by the 3 year old i7-4790k. In short, they all ran at roughly the same speed.
I would say that this is because of the limits of the SSDs, but then I tested by holding that data in memory rather than disk, then ran the intersection. Same result.
This is a very basic operation that tests the performance of the CPU and memory. And for operations like these (ie: not using SSE2, AVX, etc), the perceived performance difference to the vast majority of people for a secondary home machine may be negligible.
That really depends on your load. I once feared a huge performance regression when a test took 2 minutes instead of 40s. Turned out one instance ran on a i7-6xxx machine, and the other on a i7-9x0. Normalizing for the different clock, the newer CPU had a 20% better IPC.
Disclaimer: I don't recall exact numbers, but they should be somewhat close to those given.
> I would say that this is because of the limits of the SSDs, but then I tested by holding that data in memory rather than disk, then ran the intersection. Same result.
Wouldn't it then just be testing the limits of your memory?
I think he means effective performance, ie if the machine is idle 95% of the time with the slow processor and 97% of the time with the fast one it really doesn't matter.
The Xeon E5620 mentioned is also from 2010, 8 years ago, not 11. Of course a Nehalem-Sandy Bridge era processor is still relevant. Many people are still using those today, given the Moore’s law slowdown.
> People need to do more research when they think about taking on these gig economy jobs full time.
I also think "they do" - but on the other hand, do not forget the human nature. No amount of researching, reading it or hearing it equals to living it yourself.
Last year I accepted a semi-gig job - not in the current "gig economy" sense such as Uber - but I created a combination of working for a startup for preset amount of hours (less than 1/2 month, red flag 1), working for a post-paid client on-demand (this is cool, accept or reject any project as you wish), and working for a pre-paid client on-nonrejectable-demand (red flag 2). At first, it looked great, [the hour rate] x 160 hours = excellent monthly salary!
But then you find out it isn't 160 hours, there's more. After putting in X billable hours of work, you have to issue invoices, study ever-changing laws, file taxes, yada-yada adding non-billable hours. And sometimes there aren't even those 160 hours, all three clients want their hours/projects/work packages in the same week and don't have any of the work for the rest of the month...
And then, it also isn't a salary, it is a gross revenue. So subtract all the insurances, taxes (even those that employer pays in a standard employment), your own hardware, travel costs...
-----
BUT!
Even with all the downsides, sleepless nights and opportunity costs, I would still choose to go through that again. It is like having a HDD crash or dating a borderline girlfriend.
You can do all the research in the world and still think you are the special snowflake who don't need to back up your data or that you are the captain-save-a-girl who will magically erase all the psychological issues and self-harm behavior by the sheer power of your love (true story), only to find out that:
- your HDD is suddenly not showing up in the system and you will spend 3 days of manually recovering and viewing each and every file through various obscure recovery tools; and you will ever since back up your data vigorously at geographically separated locations
- the mirroring behavior and idealization/devaluation cycle is real and while one day you are the best mate in the world, the next day, while you are riding your oxytocin high, you are called the worst incompatible personality in the world, "never see me again", spend your night crying in the corner, only to be greeted with "sometimes I say things I don't mean, get back to me immediately, or I will kill myself" on the third day; and you will ever since bail out at any of the slightest sign of cluster B personality disorder.
----
I don't see accepting the "gig-economy" with low payout as not doing your research, but rather as a school fee for getting a valuable lesson in your life... You can tell that to people 100 times, but when they live it once, their eyes will open.
> have an appropriate extinguisher and a plan for quickly getting failed cells outside and onto an inert surface like concrete.
There are two problems:
a) There is no appropriate extinguisher when you have a thermal runaway on 20 kWh of chemical energy.
b) This isn't some shielded, extinguisher-nearby, one-time experiment, when you, for example, solder something off and then while yelling "see, I'm still alive!" run outside and throw it out. This is something that is intended to be running 24/7 at your home (and maybe mounted on wooden wall like the pictures in the link...). Unless you post guard duty shifts around the clock, there simply may be no person to perform the plan of getting failed cells onto concrete.
I see this as super-dangerous, especially when dealing with cells from many different vendors scavenged from thrown-out laptops.
That's fair, and you'd definitely want to balance each cell and think hard about where to put it, but these batteries are popular for a reason; they're a potentially versatile and effective power source with charge density that is currently very hard to beat, and I'm happy to see people trying to make information about doing it safely more widely available.
And the extinguisher would be for things around where the batteries used to be, in a catastrophic failure.
>There is no appropriate extinguisher when you have a thermal runaway on 20 kWh of chemical energy.
Yes there is, it's called a hose. But you're right, that 20 kWh isn't just going to up and disappear. I think it's realistically possible to do this with a modicum of safety even for a hobbyist but it needs to be passively safe. It can't rely on cells being inherently balanced with respect to one another, it needs to have some serious thought about what to do with that 20 kWh if the worst occurs but that's not an insurmountable task. Sprinkler heads are cheap and even just a garden hose will put out 25 gallons a minute. A very slight risk of water damage can be acceptable IMHO, I wouldn't have trouble sleeping with something cobbled together sitting in my garage so long as I planned for the worst along the way when piecing it together.
Well it certainly wouldn't be superheated steam and it's not like it would be a ton of steam in the grand scheme of things. It takes about 2500 kJ of energy to boil a liter of room temperature water. Someone else mentioned 20 kWh of batteries which in a catastrophic failure I'm assuming the cells might release a little more than that amount of energy but for the sake of argument since I don't know exactly what percent to tack onto that we'll just look at the nominal capacity of the cells. 20 kWh is 72 MJ so that's only enough to boil a bit less than 30 liters of water. Submerging a burning lithium ion pack in something like a bathtub full of water is a great way to absorb all of that energy.
I guess it's that people are walking since their first months after being born, so they are pretty good at it and even when you are older, unwell, intoxicated, etc., unless you are dancing on the edge of subway platform or right on the curb, you are generally pretty safe even in traffic. Compare that to jumping on and off platforms moving vertically, that's pretty new.
I had a go on this a long time ago, the building was empty on Saturday and they turned the elevator on just for us who were having a programming competition there. We rode in circles, went over the top and jokingly returned in handstands, jumped on and off on each floor...
And while they are generally pretty safe - each platform in upward shaft and in the cabin was on hinges, each top of the "door"frame and top of the cabin had a hanging piece of wood and if any of these things were moved, the elevator would stop, so your feet or other body parts won't be crushed if stuck between the cabin and the frame. But still, you can hurt yourself pretty badly (think broken bones or head injury) just from slipping or falling forward/backward if you try to step on or off too soon or too late.
Being twenty-somethings, we had a great fun on them, but if you think of people you see daily moving on escalators, in the mall, on the pavement, you can pretty easily see that an older person or a less coordinated person may have a problem with them.
Btw. For the brave out there, there is also an industrial "version" called belt manlift. Dying on them is pretty easy.
Tokyo has an average of 0.5 cars per household, which is on par with NYC. So, how will banning cars suddenly in NYC turn NYC into Tokyo, where cars aren't banned?
> Can you use bitcoin to pay for some food? Or to get your roof fixed?
This is a bit of false argument. One of the most universally recognized value stores is gold. Yet, can you use gold to pay for some food?
Try marching into Tesco, grab some groceries, drop the gold bar on the counter and expect change...
Yes, if you overpay it 50-times (a 1 oz gold bar for 20 EUR groceries), someone (cashier, other customer, anybody) will accept it on their personal risk (maybe it is fake? who in sane mind would pay with gold at grocery store?), pay for you, take your gold and give you the groceries.
But by definition: no, you cannot pay for food with gold/Bitcoin/diamonds/stocks/land/weapons/tobacco/alcohol/other food... At most, you can barter it.
One of the most ridiculous cases I have seen lately is comparing Bitcoin fees with paying for coffee. Do you pay for coffee with gold bars? Do you pay for coffee with international non-SEPA bank transfers (expect 30 EUR fee, more than BTC, and weeks-long fulfillment)?
Yes, Bitcoin was originally intended as "pay-for-coffee" currency. No, it isn't that anymore, it shifted more towards value store, settlement money between large entities, alternative to global bank transfers.
My favorite counterexample is: Can you run from war conflict with money or gold under your shirt? You will be searched, stripped from cash and possibly detained. Or learn 20 English words, run naked and input those 20 words into the computer on the other end of the world.
Do you think this is a overinflated example applied only to the worst parts of the world, where nobody have the money with them anyway? We, in the civilized world don't have any of such oppressions, right? Try leaving European Union with more than 10 000 EUR...
During my business flights (especially to Arabic countries, where people were returning from shopping trips in Vienna etc.), I have seen quite a few folks detained just because of the cash they had on person.
...which can be electronicaly moved without giving away control of your funds to a company/government who can easily seize, devalue, or prevent you from accessing your funds.
That's the important issue that bitcoin solves. Do you understand what being decentralized & permissionless mean? And the power shift it implies on a social level?
From local laws. From physical laws they are not. As a kid long time ago, I was sitting in a Fiat 124 when a bus driver wasn't paying attention to the road ahead and hit the stationary car on intersection in full city speed, thankfully it was direct hit from behind and there was nothing in front of the car, so there were no serious and only some very light injuries, but it was ugly.
But I cannot image having a kid sitting freely in a lap nowadays just because the car in question is exempt from regulations. There doesn't even have to be a collision, emergency braking (either human or autonomous) is enough to throw the kid off against the car interior and sustain injuries (yep, had that too in the old no-rear-seat-belt-times when an ambulance without siren appeared oncoming wrong way, almost bit my tongue off).
Don't use restraint systems just when they are required by regulations. And especially in nowadays tenfold traffic.
Wow... Apart from the NPT (which is just a piece of scrap paper anyway, see Israel, India, Pakistan), last time someone thought putting nukes near adversaries on the other side of the world was a great idea, the world almost ended.
The entire incident is now known as "Cuban Missile Crisis", although it started with the great idea of placing Jupiter missiles in Italy and Turkey against USSR, continued with USSR placing missiles against USA in Cuba and ended with a single officer on a Soviet submarine, out of three required, objecting to launching a nuclear torpedo while being attacked from a USA ship above. Seeing how easy it is to spark a full war (e.g. Gleiwitz, Gulf of Tonkin), the Doomsday clock had its leap second that day.
Stationing nukes in the SK, close to both China and Russia would be such a major shift in power balance, that I would really like to see the crazyman who would go to China/Russia to even mention that.
> Go look at sci-fi from the 60s and you'll see flat screens hanging on the wall. This of course happened but flat screens were one of the critical developments for laptops and, more importantly, modern smartphones.
Star Trek: The Original Series; 1966; Electronic Clipboard / Personal Access Display Device
1. Xeon E3 1320 never existed. Was it meant to be 1230?
2. Does DDR4-2400 really have "the same performance characteristics" as DDR3-800 (even after accounting for having 3 channels instead of 2)?
3. How can a 3.5 GHz processor be "the same performance characteristics" as 2.4 GHz one (1.5x more)? Add IPC on top of that.