2012 chrysler 200 thermostat replacement cost

Context: I recently got a pair of sennheiser HD58x's.
Prior headphones: I started out on an ATH ANC7 in 2020 which I refurbished myself. Those broke somewhere around the start of 2021. Been using an AKG K361 for approx one and a half year.
Historical tonal preference: claimed to love Harman/neutral before, with +15dB subbass for bass-heavy genres.
What do I listen to? summary of my music library
Initial impressions(<10h in): Initially, when I got the 58x I thought its midbass was bloated, and it was in general quite bassy compared to the k361. Which does not seem to agree with the graphs ive seen for both. Anyways, 5-10 hours in I got acclimatized to this new sound.
Post acclimatization: I started to feel like the 58x was WAY more musical. I could just enjoy what I was listening to and feel immersed. The soundstage wasn't that much bigger, regardless I appreciated it. They are more resolving, but the sound sig isnt that neutral. I tried a few bassy songs (edm), and found the presentation to be natural, yet unenthusiastic and lacking, mostly because I really love subbass. However theyre fine for hip hop which has significantly lesser subbass. Metal and rock is where they shine at. The increased midbass is apparent with a wonderful presentation of the drums. Basically perfect as a first pair if you listen to a variety of music, or if you just listen to pop since the presentation complements normal pop mastering styles excellently.
Thoughts about cost: Its been a month since I got them. I mulled for long whether these would replace my akgs or complement them. I was very scared it would be the former, since headphones are costly and you cant exactly resell them to recoup money in India easily, so they would just lie around unused. AKGs costed me 5.8k INR (70$) from Amazon which is quite a lot here, especially as a 18 yo who doesnt earn. The 58x were 139$ I think during the December 22 drop sale, got them from a relative.
Based on my time with both of them (~1500h with the k361 and ~100h with the 58x), I think I can conclusively say:
K361 was the perfect starter headphone for me, because its neutral and it just sings with every genre I listen to (mostly everything). It can be a tad bit sibilant sometimes (depends on whether I have a premigraine). Take that with a pinch of salt. It responds to EQ amazingly, and if I had to pick one stock tonality alone, itd be the k361 hands down.
The HD58X is so much more pleasant though. I have a newfound appreciation for pop and indie, which I think is because these are not neutral. I can just close my eyes and get lost in the tracks. Forget about life. Not analyze anything. For everyday desktop use, these. Hands down. They're way more comfortable, and not at all stuffy because theyre open back. (Stuffy as in the k361 gets hot cuz closed back, not the sound).
I think I will keep the k361 with me forever because:

1. Sentiment. I adore these.
2. Closed backs.
3. for exploring new tracks with an open mind, these are perfect.
4. While not as enjoyable, its neutral sound does make critical listening way more fun. I dont get lost in a track, I can make out its features.
5. I dont know how to explain it but theyre the perfect 'lazy Sunday morning sipping coffee listening to music' headphones.

I suppose, for most people the 58x would indeed replace the k361, but for me, I'll keep both :).
Also based on the upgrade from the mi superbass wireless on ears to the anc7 (lets say 100x) to the upgrade to k361 (10x), k361 to HD58x was like 1.5x purely in terms of sound alone. I think im done with this hobby hah.
I'll lurk in these subs forever, love learning about audio and headphones, but even if I buy more headphones, they'll be <200$ midfi ones with a noticeably different signature. Neutral endgame achieved.
Tonal preference now: good headphones, open to non neutral sound signatures :)
Unless of course, someone has Focal/Abyss/HD8--/Audeze/DCA/etc out here in India, in which case I'll try them IRL before considering TOTL stuff lol. (All my buys were blind from k361, HD58x, jabra elite 2, kbear ks1, to moondrop spaceships that broke)

Hi all,
I bought a van from a dealer on 03/01/2023 Ford Custom 2016 for £10,300. van didn't seem to have any major issues. Dealer completed a MOT on the day of purchase. Yesterday, van broke down when driving (heard a knock and engine switched off) couldn't start it up again. Called for recovery and van has been towed to my house.
Called dealer whom I bought the Van from. On the invoice it says he offer up to 3months warranty or up to 3000miles.
What he told me is that he will ONLY cover repair costs up to £300 which is literally 3% of the value of the car. He also told me to get the car to the garage and send him an invoice and as mentioned above, he would cover £300. He did also ask whether I serviced the van and was trying to blame me that it's my fault etc, so I'm afraid it's not gonna be easy to deal with him, now, when I encounter problems. Unfortunately I also found out now he has many negative reviews 'stay away from this dodgy dealer, selling faulty vans and death traps '
What are my rights? Under Customer Right Act 2015 trader is obligated to repair or replace the vehicle - nothing is mentioned about the amount. I'm afraid the costs of labour will be much higher than £300. I already paid £200 for recovery yesterday.
Another thing is that warranty will run out on Monday - but I believe I should be still covered as event has happened on 31/03, and on the same day dealer has been informed. I also have an invoice for van recovery.
Please help what are my rights, and whether I should find a garage or whether the ball is in his court, now.
I was also told by the recovery guy I can contact Ford dealer directly, and ask them for their warranty period and they could help, even if I didn't buy the Van from Trusted Ford dealer - any advice on that? How willing are they to help?
Thanks!

Hi all,
I bought a van from a dealer on 03/01/2023 Ford Custom 2016 for £10,300. van didn't seem to have any major issues. Dealer completed a MOT on the day of purchase. Yesterday, van broke down when driving (heard a knock and engine switched off) couldn't start it up again. Called for recovery and van has been towed to my house.
Called dealer whom I bought the Van from. On the invoice it says he offer up to 3months warranty or up to 3000miles.
What he told me is that he will ONLY cover repair costs up to £300 which is literally 3% of the value of the car. He also told me to get the car to the garage and send him an invoice and as mentioned above, he would cover £300. He did also ask whether I serviced the van and was trying to blame me that it's my fault etc, so I'm afraid it's not gonna be easy to deal with him, when the van broke down.
What are my rights? Under Customer Right Act 2015 trader is obligated to repair or replace the vehicle - nothing is mentioned about the amount. I'm afraid the costs of labour will be much higher than £300. I already paid £200 for recovery yesterday.
Another thing is that warranty will run out on Monday - but i should be still covered as event has happened on 31/03 and on the same day trader has been informed. I also have an invoice for van recovery.
Please help what are my rights, and whether I should find a garage or whether the ball is in his court, now.
I was also told by the recovery guy I can contact Ford dealer directly, and ask them for their warranty period and they could help, even if I didn't buy the Van from Trusted Ford dealer - any advice on that? How willing are they to help?
Thanks!

My first fly rod was an Echo Base 9' 5wt. I broke the bottom section in a hilarious kayaking accident about a year and a half ago. Echo was nice enough to send me a Lift as a replacement for 60% off plus shipping. I broke the bottom section of that yesterday while casting after a taking a fall onto a rock... Wish I'd broken a different section so I could just combine the two (even though they're supposed to be different rods). Anyway, I figure it's a good time to upgrade:
I'm looking for something in the $200-$300ish range (I guess lower mid priced). Checking out the Echo Trout (the old one on clearance), St. Croix Imperial, Orvis Clearwater (have liked the ones I've casted), maybe a TFO, Redington Vice, Fenwick Aetos, etc. Thought about stretching the budget for the G Loomis IMX Pro that are on clearance right now. The two fly shops in my area don't carry most of those so test casting isn't much of an option. I'm not a fantastic caster, almost exclusively fish northern NM and southern CO, and when it comes to using the 5wt I'm often roll casting double nymph rigs with an indicator. A buddy of mine fishes a 10' 5wt Recon and I like that rod a lot so I'm considering a 9'6" or 10'. Durability / low cost / forgiving warranty is a plus given my dumbass track record.
Thanks in advance for any insight you're willing to offer!

I've been reading about how feudalism in ancient time works.
Obviously, modern democracy is better than ancient feudalism.
But is feudalism defensible under libertarian principles?
It's actually interesting. Some of it's feature is actually more compatible to libertarianism than democracy.
The feudal lord owns land. They own large amount of land. I mean I own 200 m square land and have a house on top of it. Feudal lord land is like teritory.
Feudal lords are just landlord with really really big land. A bit like Microsoft and Reddit are just really really big business. I am not happy with Reddit leftist bias but I don't say it's necessarily wrong.
Just like business can be really big, land ownership can be big too. Of course, I prefer competing small businesses than big businesses. But often economic of scale make small businesses cost ineffective.
I think you own it you rule it is a reasonable libertarian position. In fact, one points out that property right and self ownership is corner stone of libertarianism.
The problem with feudalism from libertarian perspective is not that the lord rule the land. The problem is how such ownership of such huge tract of land is obtained. As usual it's due to conquest (robbery) and seizing it from other lords.
If a feudal lord governs the land well build infrastructure and hence consensually attracts tax payers I see nothing wrong is that. It's like a landlord improving their building and attract higher paying rents.
Of course, under democracy, the role of feudal lord is simply replaced by voters. The problem with voting to me is not the voting itself. The problem is in how voting right is obtained. Why should death people lost citizenship and why should newborn babies and immigrants get free citizenship? Shouldn't ownership be heritable to children?
If citizenship and "feudal" right is transferred and obtained through capitalistic mean, I don't see how that violates NAP. What would Herman Hoppe think?
Another system is outlaws.
The king doesn't even have to coerce people to obey their law and pay taxes.
You don't have to pay taxes.
You just don't get protection either if you don't
So you're an outlaw. Anyone can kill you and rob you and rape you and it'll be legal. You're not paying tax which is a protection fee under the king, so it's not the king's business to criminalize acts against you.
How does this fit into ancap narratives?
What do you think?
As for me, anything that lower tax and increase positive freedom is fine. I am a practical libertarian.

Before I say anything, remember I’m talking about living in San Francisco where the cost of living is astronomical. Rent is $5,000+ a month for enough room for a couple and houses with enough room for two people start at $1.2 million.
Back in 2014 I got my first job where I made $200,000. I was CTO at a Series A company with a small engineering department of 4 people in San Francisco. I felt rich.
In 2022, I was working as a Chief Software Architect for a Series A company. My base salary? $200,000. The same salary despite having 8 more years of experience under my belt. Not once during the time in between had my base ever exceeded that amount. That’s a 21% pay cut due to inflation. I obviously made more since I rarely only worked one job (and we’ll cover that more in a bit) but the fact that I never seemed able to get beyond that base had really started to bother me.
I always knew I could try to go work at a FAANG or a larger corporation, but I wasn’t built for bureaucracy. I’m a builder. I once heard a story from a friend who worked at Google, who frequently told me the only hard part of his $800k/year job was the interview process. He told me he was once assigned what was described as an urgent task. He completed it in a few days and then submitted it for code review. He then asked if he should work on something else. He was told to wait until it was reviewed. So he waited. And waited. And ended up waiting three weeks……and doing nothing during that time.
Personally, I’d have gone insane. That kind of job isn’t for me, no matter the compensation. So I have always worked at small startups where I can be challenged, stay busy, and directly see the impact of your work almost daily but as you age, life becomes more expensive and the math starts to no longer add up.
Startups tell you your low salary will be made up for by equity. Well, let’s take a look at that.
As far as I can count, I have vested equity in 12 startups in my work history, not counting my own, one of which did pay out handsomely. Of those 12, only one paid out. Change:Healthcare, the very first one I worked at. I was employee #4: Director of Finance and Strategy. I actually developed the math, algorithms, and systems behind their first and premier product. I vested around 23 months of equity. So when they sold to Emdeon for $135 million in 2014, how much did I get? $5,400. Employee #4, two years of equity.
That’s the sum total of cash I have gotten from 15 years of equity working for other startups.
All the other money I earned was from salaries at or under $200,000.
So when I went looking for a new job this past January, I said I wouldn’t accept $200k any longer. I’m an extremely experienced CTO, successful CEO with an exit under my belt, and can write code and architect infrastructure with the best of them.
And what did I find? The worst job market I have ever seen. I applied to over 500 positions to get 5 interviews. Recruiters told me endlessly that my resume was absolutely top notch, but an average CTO position was getting around 1,500 applications. Even though 95-99% of those people weren’t qualified and I was, my resume was getting lost in the noise. Some of the interviews I did get did finally pay more, in the $300-400k range. But for various, and good, reasons, they took a pass on me. I worked several consulting jobs and actually lived a pretty good lifestyle in the meantime, but I was still looking for that anchor job to be stable income.
But some of the interviews were the opposite and seemed stuck in time. One Series C company that had raised $35 million reached out to me directly and showed a lot of interest in hiring me. Even better, they were an API product, my specialty. They needed a replacement CTO. Sign me up! They had a team of 55 engineers and needed someone to re-architect their software. Oh boy, this sounds fun and challenging! So how much does it pay? $225,000.
I’m sorry, can you say that again?
That’s right. You want me to go from a cushy $200k Chief Software Architect job to a full blown managing 55 people and the entire engineering department for a $25,000 raise? I had to stifle some laughter. I also didn’t explain earlier that I was working two other significant contracting positions while working that CSA job, since it took up so little time so my total take home was very high. This, of course, would leave no time for that, meaning I’d actually be taking a huge pay cut.
I gently asked them if there was any room for that figure to go up. They said no. I told them I was a pass. To date, they are still looking. Gee, I wonder why.
So enter this past week’s company, which I will leave nameless. I had two interviews with the CEO and we hit it off very well. He gave me a verbal offer for a much higher base salary that I was happier with that was contingent on some easy factors and also agreed to let me consult on the side. Seemed like a good fit. Then, through no rhyme or reason that I can figure out, he changed his mind this week. The reason he gave was “timeline” saying he couldn’t make our target April 1 start date, but of course the recruiter and I said my timeline was flexible, but the CEO would give no other reason for why he was suddenly passing on me, clearly lying for some reason.
That’s when I remembered that not only was the pay low at most of these positions……while I enjoyed the actual engineering work, the culture was generally terrible. The unending abuse from bosses who expected the impossible or for you to put in ridiculous hours while paying you (comparably) low wages is insane at startups. I have endless stories. There was the CEO who wanted me to move from one hosting provider to AWS in 24 hours. Except we had a 2TB database. When I explained to him it was physically impossible to move all of that data in that timeframe, he fired me. There was the founder who asked me in December for an estimate on his product. I said mid-March. He, of course, added dozens of features in between those dates. On March 15th, he called me up and asked, “So what time are we launching today?” Confused, I asked him what he meant. He said I had promised him that was our launch date and he had told all of his investors to expect a live website by the end of the day. I spent the rest of the day working harder than I ever have to get something out the door working.
There was the CEO who would call me repeatedly at 3am, breaking through do-not-disturb, to ask if I had completed a feature and if I hadn’t, why I wasn’t working. You can bet I didn’t stay at that job very long. No one did.
Then there was the company that hired me under false pretenses. I was hired for a very cushy, low paid job to just relax for a few years. Then my first day came and they told me I was in charge of building a bank for them, an entire new division of the company, all for $165k/yr. Over 9 months, I wrote 160,000 lines of code, worked endless 80+ hour weeks, was doing the roles of 5 people, and was denied raises dozens of times. They eventually burned me out and I quit. And if I had stayed and vested all of my equity for 4 years it would be worth exactly $14,000 today (they did IPO).
So after being lied to by this CEO and being treated unbelievably unprofessionally and disrespectfully, I’ve had it. I’m through. I’ve given 15 years of my life to working for other people’s startups with almost nothing to show for it except a lot of war wounds, PTSD, and admittedly, a lot of fantastic and valuable technical expertise.
So I’m going out on my own. I had already made this decision to some extent but I’m doubling down. My new consulting firm is now my full time job. And today, I’m proud to say we made our first hire. We now have a Senior VP of iOS Development. And he’s a real 10Xer. The best of the best. And I’m not paying him a salary but a percentage of all the revenue he generates (the vast majority in fact) and if he brings in work, he gets part of my revenue too. We’re going to be churning out top notch iOS apps soon, one after another. And no one will be under compensating me for my time or telling me what meaningless meeting I have to go to, or calling me at 3am like some inmate or firing me because they don't understand transfer rates.
I’m done being a wage slave. The only company that ever significantly paid out was my 2nd startup. Well, this isn’t a startup, but I am my own boss again, and my destiny is in my own hands. And I’m a lot more comfortable with that than it being in someone else’s because I trust me a lot more than I trust them.
edit: The most common reaction to this is that I should focus on building wealth for companies and not just my salary. Well, I have always been focused on building wealth first. I’ve added hundreds of millions if not billions of dollars to valuations of the dozen plus startups I’ve worked for. And of course, after I do, I’ve asked for more equity. And of course, it’s always been denied.
The most direct evidence I have that I’ve always been under compensated is what happens to companies when for whatever reason I leave. Quite frankly, the tech side of their business, and frequently the business itself, entirely falls apart and in some cases, the businesses have quickly failed. Others have lost so much money they’ve been forced to do down rounds, losing hundreds of millions of dollars in their valuations.
So I know exactly how much I’m worth and I have direct evidence that I’ve been under compensated at pretty much every company I’ve ever worked for.
Which is exactly why I’m not going to keep all that value for myself.

https://imgur.com/a/SyBYQ4e Crap pic album but it's an album
As the title says, i wanted to build an SFF HTPC/VR PC for the living room and something small enough to move into the wifes office around the corner for when she wants to game alone or with me. Currently she uses her desk for her work laptop and monitors but has plenty of desk space to fit something small so I thought it fit well to try it for under 700 bucks and keep it SFF. Wasn't really in a rush to do it but it came together so fast after the initial bundle as sales just lined up. How did I do?
Specs/prices/purchase location are below
Cpu - 5600x
Motherboard- Asus ROG B550i Strix (has AX wifi 6 and BT)
Ram: 32gb oloy 3600
Total: $250
Purchased these 3 in a combo on hardwareswap for $250 shipped, with the knowledge of one of the usb 3.0 pins being smashed flat on the board, ended up bending back fine and both front USBs work with no issues, even installed windows through the front USB as a test :)
GPU - EVGA 3060 XC 12gb
Total: $200
Also a hardwareswap purchase for $200 shipped. card came super clean and I repasted when I got it, although probably didn't need it yet!
SSD - 2TB NVME Crucial P3
$54.36 after tax
Was one of the lucky few who caught this bestbuy deal for $51.29+tax when it went in and out of sale about 4 or 5 days ago on buildapcsales
Case, PSU, and cooler - NZXT H1 v2
Total:$211.99 after tax
Caught this case on sale for $199.99 which is 50% off i believe, and really the only reason I bought it as getting the sff case, psu, and cooling would've probably cost me more anyways and is small enough to fit anywhere. Very sharp case and was INCREDIBLY easy to build in, only took like 25 minutes start to finish. I do have a lot of experience building though.
Total spent: 716.35 after tax/shipping
I feel like especially for a $700 SFF build, this rig is pretty capable of anything I've thrown at it so far coming from my 5800x/3080 rig. Very happy with how it came out. Might try and throw some RGB in it to light it up a bit as the glass is pretty dark and, not sure an rgb fan replacement would give it enough. Maybe an RGB strip or 2 would do, but tell me below what you think or what I could've improved on!

I want to get some feedback on my financial situation and see what people think. If you need more info I will add or can make a new post. I’m in my late 20s.
Overall last year was an expensive year with paying for a small wedding, a couple of trips, and an improvement to the kitchen in my flat (here the tenant is responsible for this). With gifts we managed to brake even on the wedding and honeymoon cost. I still managed to build up my e-fund to a good level last year ($5000+) plus I have an additional $15000 in I-bonds. My only debt is student loans ($20000) which are currently on pause and may be forgiven. Now I have reworked my budget for 2023 since there are far fewer “big costs” (like wedding) to prepare for so I can go back to saving. Just want to see what others think and if it’s all in proportion.
Rough monthly income (total = $3600): + Net paycheck = 3000 (already includes health insurance and some mandatory state pension contribution) + Additional monthly income = 600
Rough expenses (needs = 1620): - Rent including utilities = 800 - Groceries for two people = 400 - Cell phone = 30 - Local public transport = 50 - Fuel for car = 100 - Additional insurances = 115 - Budget for medical costs = 25 - Buffer misc. costs = 100
Rough savings budget (1200): - Building up efund/savings = 1000 - Car replacement budget = 120
Rough wants budget (825): - couples holiday = 150 - solo holiday = 200 - hobbies = 150 (ie scuba diving) - eating out = 150 - fun money = 100 - clothing = 50 - gym membership = 25
Currently I am the main earner and tend to cover most of the “living cost” expenses (rent, food, etc).
Where I tend to overspend is taking a holiday for scuba and it being more expensive than planned, or wanting to take more scuba holidays in the year than I can afford. I also think about how for savings I actually have 1000/month “extra” in my budget which I could put to a holiday if needed. This is more of a mindset problem. I still lack some discipline. Maybe there are some items missing. A part of me definitely doesn’t want to reduce the budget for “fun money” but the most important is to have a realistic and honest overview and then go from there.
I have been in my job for several years and I’m paid below market. So that’s something I want to change by the end of this year. I want to put the 1000/month to the efund and have around 10000 there by the time I change my job. Then I can go forward with investing, etc.
But in general it’s hard to get a sense of what is a fair amount to spend. I try to cook most of the meals at home (making use of that kitchen!) and bring leftovers with me to work). Unfortunately most of my hobbies are pretty expensive but that’s also part of me enjoying my life so I try to make it work. Really glad for any feedback!

I have a 2007 Maytag top-freezer refrigerator that I got from Home Depot about 16 years ago. It's always been kind of aggressive with the defrost system and I get a LOT of freezer burn issues due to the excessive heating it does.
A few weeks ago, it started doing this weird defrost cycle thing that repeated 4-5 times a day, swinging from -5 to 32+ according to my glass thermometer and an electronic one I also added. It was destroying a lot of food due to this so I shut it down. I found an exact replacement for the defrost timer board on Amazon for about $40 and that's on its way. I also suspect the defrost thermostat so I've been looking for a replacement. The exact part is WP2321800 and they are readily available for $35 to as much as $80 but the worst part is they all take 7-10 days for shipping, even from the merchants on Amazon. OTOH, there are tons of dealers that have a replacement for WP2321799 and WP2321802 that cost $8-$10 and have overnight shipping. They appear to be identical units physically that most likely have slightly different on/off temps. I already have one of the cheap ones on the way just in case, because not having fudgecicles in the house is pissing me off. I'll test the old one with ice water before changing it, but I've seen a few random internet posts claiming that they can also fail partially or intermittently so I'm not sure how accurate the test will be.
I'm curious as to whether anyone else has run into this and successfully did a substitution with the cheaper WP2321799 part. If it has an earlier shutoff that wouldn't break my heart at all. I spent my whole life manually defrosting refrigerators every 6 months anyway (it's a good time to scrub them and throw out that stuff you're never going to eat) and I am extremely unimpressed with the automatic systems on every fridge brand I have ever used. The occasional convenience does not outweigh the amount of damaged food that results from using them. Unfortunately on modern units, you can't seem to turn off the auto mode without seriously hacking the timer board.

I currently have a Ryzen 5 1600 and Quadro M2000. This upgrade is to reduce power consumption as well as improving performance and clearing up a PCIE slot that is taken up by the M2000 (replacing it with quick sync for video transcoding). Biggest issue with power consumption is the bug that requires you to disable c-states on first gen Ryzen chips. I plan on going with the i5 13500. It performs similar to the 13600k, but is a bit cheaper and consumes less power.
Features I would like to see:

1. At least two PCIe x16 slots. Three would be nice even if one or two are actually just x8. Current expansion will only be an x8 HBA card but the future may see a 10gb/s network and/or a GPU.
2. 2.5gb/s networking. 10gb/s would be fun, but I don't need it and that increases the cost of the motherboard.
3. DDR4. I already have 64gb of DDR4 in my server, so that would save a good chunk of money. And from what I can tell the performance different is not huge.
4. Under $200. If something is just over $200 and comes with a strong recommendation, I will consider it.

I do not need Wi-Fi. I am not trying to min-max minimum power consumption, but I do not need WiFi and it does consume a little bit of power. It also adds to the cost of the motherboard.0
Motherboards I have found:
https://pcpartpicker.com/product/Z3ddnQ/gigabyte-z690-ud-ddr4-atx-lga1700-motherboard-z690-ud-ddr4
https://pcpartpicker.com/product/GvmmP6/msi-pro-z690-a-ddr4-atx-lga1700-motherboard-pro-z690-a-ddr4
https://pcpartpicker.com/product/sPLFf7/asus-prime-h670-plus-d4-atx-lga1700-motherboard-prime-h670-plus-d4

I currently have a Ryzen 5 1600 and Quadro M2000. This upgrade is to reduce power consumption as well as improving performance and clearing up a PCIE slot that is taken up by the M2000 (replacing it with quick sync for video transcoding). Biggest issue with power consumption is the bug that requires you to disable c-states on first gen Ryzen chips. I plan on going with the i5 13500. It performs similar to the 13600k, but is a bit cheaper and consumes less power.
Features I would like to see:

1. At least two PCIe x16 slots. Three would be nice though even if one or two are actually just x8. Current expansion will only be an x8 HBA card but the future may see a 10gb/s network and/or a GPU.
2. 2.5gb/s networking. 10gb/s would be fun, but I don't need it and that increases the cost of the motherboard.
3. DDR4. I already have 64gb of DDR4 in my server, so that would save a good chunk of money. And from what I can tell the performance different is not huge.
4. Under $200. If something is just over $200 and comes with a strong recommendation, I will consider it.

I do not need Wi-Fi. I am not trying to min-max minimum power consumption, but I do not need WiFi and it does consume a little bit of power. It also adds to the cost of the motherboard.
Motherboards I have found:
https://pcpartpicker.com/product/Z3ddnQ/gigabyte-z690-ud-ddr4-atx-lga1700-motherboard-z690-ud-ddr4
https://pcpartpicker.com/product/GvmmP6/msi-pro-z690-a-ddr4-atx-lga1700-motherboard-pro-z690-a-ddr4
https://pcpartpicker.com/product/sPLFf7/asus-prime-h670-plus-d4-atx-lga1700-motherboard-prime-h670-plus-d4

Last updated: Mar 2023

Intro

This is a long Deep Dive of Bitcoin that goes into its general topics while cutting out the bullshit.
Its target audience are crypto experts who write crypto guides for others. I've included sources inline so you can reuse them.

Bitcoin Purpose and History

History

Bitcoin was the first popular cryptocurrency. It was invented in 2008 during the 2007-2009 Financial Crisis by an anonymous entity under the pen name of Satoshi Nakamoto and launched in 2009. For the first several years, fewer than 100 supporters worked altruistically to develop its code and mine the network. It is a disinflationary cryptocurrency with a supply cap of 21M Bitcoins (2.1 quintillion Satoshis).
Core Devs
Gavin Andresen later replaced Nakamoto as the lead developer of the Bitcoin code repository and lead developer at the Bitcoin Foundation. There are currently only 5 Bitcoin Core maintainers with commit access after both Peter Wuille and Lead Developer Wladamir van der Laan left in July and August 2022.
Block size
Bitcoin's blocks were originally limited to 32MB in size but later reduced to 1MB in 2010. After the Segwit update, blocksize changed again from 1MB to 4M weight (technically it's also 1MB). In Nov 2021, the Taproot soft fork was activated, which allows for signature aggregation via Schnorr signatures.
Bitcoin is currently the most popular cryptocurrency and marketcap leader. Since cryptocurrency value is largely based on network effect and is a Keynesian Beauty Contest, it is likely to remain popular until that narrative changes.

Purpose

The original purpose of Bitcoin from Satoshi's whitepaper was to provide a "peer-to-peer electronic cash system". During the early years, the main use case for Bitcoin was black market trading on sites like the Silk Road. Many larger merchants that accepted Bitcoin for payment in the earlier years stopped due to extreme price fluctuations.
Instead, nearly all merchants now work through centralized payment systems that convert Bitcoin into fiat. Its extreme price fluctuations also prevent it from being an ideal Store of Value, and it's too slow and inefficient to be used as a Medium-of-Exchange for day-to-day transactions. Thus, the only notable purposes of Bitcoin nowadays (besides being a speculative asset) is to provide censorship-resistance and pseudonymity.
Anti-censorship: Bitcoin provides partial financial censorship-resistance against sanctions and totalitarian government restrictions. It's much harder to prevent Bitcoin transactions than it is to prevent financial transactions at a centralized bank. For example, many Russians, Iranian, and North Koreans are getting around sanctions by using Bitcoin and mixers. Legal sex workers and marijuana industries are sometimes blocked from using traditional financial services due to social stigma. Bitcoin provides those workers a way to transfer funds around that censorship.
Pseudonymous: Bitcoin's UTXO transactions can provide moderately-high levels of obscurity. A single wallet can produce a near-unlimited amount of addresses, and there's no way to link them unless they interact with each other. It's much harder to trace UTXO-based wallets than Account-based wallets because the former creates new UTXO addresses with each transaction while Account-based blockchain wallets typically reuse the same account.

Design and Consensus

Proof of Work

Bitcoin uses Proof of Work, which provides both Nakamoto Consensus and Sybil Resistance. In Proof of Work, miners compete to solve a cryptography hash puzzle that has a set number of leading zeros. Whoever figures it out is able to package a block of transactions from the mempool and submit it. PoW is very similar to picking the winning block based on a lottery where a miner's chances of winning is directly proportional to how much energy they waste. Bitcoin was originally mined by CPUs, then GPUs, and now can only be efficiently mined by specialized ASIC processors.
Slow finality
The longest chain (technically the highest-difficulty chain) is known as the canonical chain, and miners are supposed to build on that chain. However, they can decide to build on another chain and fork Bitcoin. Bitcoin is constantly being forked, sometimes intentionally and other times accidentally or due to network latency. However, only the longest chain is considered the canonical chain. Thus Bitcoin has probabilistic finality instead of deterministic finality, which means that the Bitcoin Proof of Work consensus protocol can not guarantee that transactions are final.
Block times are about 10 minutes each with 4M-weight blocks. This allows for a maximum of about 5-7 transactions per second. Block times are variable and very inconsistent. 14% of block times are longer than 20 minutes, and 5% are longer than 30 minutes [Source]. Most exchanges and wallets use 3-6 blocks for finality, which means that you should wait ~60 minutes before assuming a transaction has settled. This makes it one of the slowest popular crypto networks. Many newer Proof of Stake blockchains settle 100x faster in under 10-30 seconds.
Difficulty adjustments
The puzzle difficulty is algorithmically set so that blocks are submitted once every 10 minutes on average. Every 2 weeks, the difficulty automatically readjusts to maintain constant block times. Due to the difficulty and rarity of solving the block puzzle as an individual, miners often join mining pools where their rewards are collectively split. Miners in mining pools often get paid by the pool for solving easier puzzles (fewer leading zeros).
Block rewards
The winning miner is rewarded with a block reward, which is the sum of the block subsidy (built-in inflation on the Bitcoin network used to pay for its security) and the transaction fee (paid by the user submitting the transaction). The block subsidy halves in nominal BTC roughly once every 3.8 years, meaning that it reduces by 99% every 27 years.

UTXO Transactions

UTXO Basics
Bitcoin uses UTXO transactions, which store the unspent input and output balances of a transaction. Unlike account transactions, it is difficult to keep track of the balance of an user's account with UTXO. UTXOs are also less storage-efficient because they usually have multiple input and output UTXOs. There's usually one additional output UTXO to store the remaining change balance, and wallets automatically combine multiple inputs. Combining multiple inputs also makes them susceptible to dust attacks that reduce your privacy.
Like Ethereum smart contracts, Bitcoin can save space and fees using batch transactions, and it can do this natively using UTXO without needing smart contracts.
Transactions are submitted with a fee to the Bitcoin network. They sit in the mempool until a miner packages them into a block. The higher the fee, the quicker miners will pick up the transactions. Users can also use Replace By Fee and Child Pays For Parent to increase the transaction fee of previously-submitted transactions.
Transaction size calculations
After the Segwit update, newer transactions calculate size based on weight units instead of bytes. A vByte is equivalent to 4 weight units. To calculate weight units, the non-witness parts (including the marker and flag) of the Bitcoin transaction in bytes are multiplied by a factor of 4.
Bitcoin transactions vary in size depending on how many inputs and output UTXOs they have. Also, different versions of UTXOs vary the weight greatly. The typical 620-weight (155 byte) transaction has a throughput of 11 TPS. The typical multisig is 2x slower.
For basic transactions, Coinbase's analysis and Hasu's analysis show that the savings for batching Bitcoin output UTXOs is at maximum 78% for storage (141 vbytes for a 1:2 transfer vs 141+31n vbytes for a 1:n transfer). There are limits to savings because input and output addresses take up the majority of space in transactions. Input addresses in particular take up twice as much space (68 vbytes) as output addresses (31-34 vbytes), so batching inputs has less savings. If you filled up an entire 4M-weight block with a single batch transaction with 125k output UTXOs, you could theoretically increase effective throughput from 3.8 TPS to 54 TfrPS. However, that's an incredibly unrealistic scenario, and with the current mix of transaction types on the blockchain, the actual effective transfers is closer to 17 TfrPS even when blocks are 100% filled.
Each 4M-weight block can hold roughly 2000 transactions on average. A typical 1 UTXO input, 2 UTXO transaction uses 155 vBytes. Multisigs start at ~200 vBytes.

Forking

Soft Forks

The advantage of soft forks
All updates on the canonical Bitcoin chain are done through soft forks. This has the advantage that no one is required to update their nodes to the latest version. Everyone is allowed the freedom of choice over whether to update.
The disadvantage of soft forks
The major downside is that Soft forks require new versions of the software to maintain backwards-compatibility with older versions, which leads to a ton of TECHNICAL DEBT. Bitcoin creates new address formats every time it soft-forks to maintain backwards compatibility with old addresses so that nodes can tell them apart. This significantly slows down the adoption of new updates, which now often take 3-6 years to gain the majority.
Nowhere else in the software industry does this happen. Even Linux distributors usually drop support for LTS releases after 5 years, and major releases often break backwards-compatibility.
Due to its soft forks, the Bitcoin network has to maintain a mismatch of all sorts of different formats (P2PK, P2PKH, P2SH, P2MS, P2WPKH, Nested P2WPKH, P2PKH, P2WSH, and P2TR). All the legacy addresses are slightly less efficient than the newer ones and cost more in fees to transact.
At the start of January 2023, only 1% of transactions were using Taproot-compatible addresses while 65% were still using legacy addresses incompatible with the Segwit update from 2017. (You may notice that the percent shot up in Feb 2023, and that was due to Ordinal inscriptions using Taproot.)

• Legacy: 26%
• Nested Segwit (within Legacy): 39%
• Segwit v0: 34%
• Taproot/Bech32m: 1%

Almost no one (1%) is using addresses newer than the 2021 update because none of the major CEXs support them. Most exchanges (Binance, Coinbase, Kraken) don't support Bech32m addresses, which means they can't send to Segwit v1 and Taproot addresses, released through the BIP 350 update.
And this limitation has blocked update progress for years.
In comparison, networks that hard fork for protocol updates don't have these incompatibility issues between versions. Everyone is working on the same version, which allows for consistency. A single Ethereum address is compatible with all versions of Ethereum, Polygon, BSC, Avalanche C-Chain, Fantom, Cronos, Kava, Gnosis, Moobeam, all layer 2 networks, and hundreds of other networks.

Hard Forks

People who don't agree with Bitcoin Core protocol can hard fork it. There are many popular hard forks of Bitcoin. The largest ones are Bitcoin XT (2014), Bitcoin Cash (2017), and Bitcoin SV (2018).
The Bitcoin Cash fork is particularly notable because it was the result of a huge rift within the Bitcoin community over the size of blocks (1MB vs 8MB/32MB). Ultimately, the small-block proponents won the war, and Bitcoin kept its 1MB blocks while large-block proponents hard-forked to BCH. That's a bit ironic since Bitcoin was a 32MB-block chain for most of Satoshi's time. Much like how both mainland China (People's Republic of China) and Taiwan (Republic of China) claim themselves to be the true Republic of China, both the BTC and BCH communities tried to claim the title of "Bitcoin" even after the split. There was hot blood between them for years.

Reorgs

Reorgs are when a fork happens and the previous longest chain gets completely overwritten by a new longest chain. The new blocks in the previous chain are lost and overwritten. There have been at least 2 reorgs longer than 6 blocks: 51 blocks in Aug 2010 and 24 blocks on Mar 12, 2013 Source 1, Source 2. Both times were caused by coding bugs and had to be fixed by 51% attacks with community approval. The 2010 reorg actually caused Bitcoin to mint 184.4 billion Bitcoins, way past its 21 million cap. There have also been at least three 4-block reorgs prior to 2017. So the usual recommendation to wait 3-6 block confirmations was not guaranteed to be safe in the past. However, it has been stable for the past several years, and we haven't had any reorgs larger than 2 blocks.

Lack of Efficiency

Low throughput and slow block times

Bitcoin is a 3-4 TPS blockchain (when blocks are 100% filled) with a 30-60 minute probabilistic finality. It used to have a maximum of 7 TPS, but that has gradually fallen over the years after the Segwit update and exchanges started using batch transactions. It's much too slow to be used for point-of-sales merchant transactions. In comparison, both Avalanche's X-Chain (another UTXO network) and Algorand can reach 6000 TPS with under 5-sec deterministic finality.

High cost of security

Bitcoin is one of the least efficient cryptocurrencies. In 2021, each block cost roughly $150-300K in costs to mine, which is equivalent to $100-150 of fees per transaction. The amount of energy needed for a single Bitcoin transaction in Sept 2021, ~1800 kWh, is roughly the same as the amount of energy used by a typical US household over 62 days. The total Bitcoin network energy consumption was ~150-200 TWh / yr in 2021-2022. For comparison, the US has 92 Nuclear power plants that produced 778 TWh in 2021 source, so the Bitcoin network uses the equivalent power of 18-24 US nuclear power plants. Another way of looking at this is that Bitcoin consumes about as much energy as all data centers globally [Source].

Mempool congestion

Because of the slow transaction speed of Bitcoin, there's often a traffic jam of transactions waiting to be picked for the next block. Transactions sent to the network via gossip protocol sit in the mempool, and there were several times where the backlog ended up being greater than 100k transactions (8 hours) in 2021 and 2022. Many transactions were untouched for days until they timed out.
Due to its slow speed, Bitcoin is not suitable as a payment system for point-of-sales transactions. It would be silly to ask a customer to wait 60 minutes while the transaction finalizes.

Moderately-high Fees

Bitcoin fees vary with mempool size, congestion, and the sat/vByte ratio. Back in 2010, nearly all Bitcoin transactions had no fee. The fee has risen over time.
Bitcoin's fees are high enough that you can't use them for daily transactions. During the cheapest days of the 2022 bear market, fees fell back to $0.10 to $0.40 per transaction, and a transaction set to 1-2 sat/vB fees would go through in a couple of hours. In a bull market, fees can rise to $1-10 per transaction, and any fee set below 10 sat/vB could stay days in the mempool.
In fact, Layer 1 transfer fees even briefly rose past $50 in May 2021. That's way more than its competitors (e.g. XLM, XRP, Nano, BCH) that have average transfer fees under $0.10. And fees will rise again during the next bull run.

Unable to reach widespread global adoption

At 4 TPS, Bitcoin can only make ~345K transactions/day. There are ~8B people in the world today. If Bitcoin grows to the size of 1% of the population, each person can make an average of 1 on-chain transaction every 230 days. If Bitcoin usage grows to 10% of the population, each person can make an average of 1 on-chain transaction every 6.3 years. Bitcoin cannot achieve even 10% of world adoption unless everyone's solely using centralized exchanges and not interacting with the network itself.

Lightning Network

What is the Lightning Network?

The Lightning Network was built as a partial-Layer 2 protocol to help scale Bitcoin due to Bitcoin's slow throughput and block time limitations.
Lightning uses an interconnected network of State Channels. Two or more parties have to open a multisig payment channel using a Hash TimeLock Contract (~800 weight) and rebalance the initial state. They can do however many transactions as they want off the Bitcoin network until they run out of capacity. Once they're done, they can close the channel and receive their portions of their funds from the channel. The network links multiple of these state channels together to create the Lightning Network.

Meant for small transactions

The total value stored on public Lightning channels account for under 0.02% of Bitcoin's total locked value. Transaction fees are low, so running a Lightning Network Daemon is not particularly profitable, especially since nodes constantly have to rebalance, costing additional fees. The larger your transaction, the higher the fees you have to pay to route it through the network. As of March 2023, the average channel capacity is only 0.07 BTC, and the average node capacity is only 0.33 BTC. It's not uncommon to have a 1 BTC transaction cost $2-10 in fees to route through multiple nodes in the Lightning Network due to limited channel capacity, which can make it more expensive than L1 Bitcoin fees.

Not a true Layer 2

Similar to Plasma channels, the Lightning network is not considered a true Layer 2 because it lacks global state. There are many nodes that are not connected to the rest of the network, and onion routing issues sometimes cause nodes to be disconnected from the rest of the network. Channels only work if everyone's online. If you're offline, others can force-close your channel, leading to a 1-week wait time where the channel's funds are locked and inaccessible.
Partially-centralized, low-security layer
There are a lot of limitations to the Lightning Network, and participants have to monitor their channels constantly to make sure they aren't improperly closed or disconnected.
Most people just connect to centralized nodes in a spoke-hub network topology to gain access to high-capacity nodes. Even though average capacity is getting bigger, the number of public channels has been on the decline since 2021, meaning that Lightning is becoming more centralized.

Rebalancing issues

One of the biggest problems with opening channels is that they start out with zero incoming liquidity. Anyone who opens a channel starts out with a metaphorical full cup of water. They can't receive any more water until they first empty the cup a little. And they can only receive additional water equivalent to the amount they removed. Similarly, people who open new channels to the Lightning network need to find a way to spend their Sats safely so that they can have incoming liquidity. Merchants and Lightning node providers often have a lack of incoming-liquidity while consumers who only spend usually run out of outbound liquidity. They have to pay up to $1 to rebalance each $1000.
There are several ways to get incoming liquidity. You can't just send Sats to another one of your accounts because that will just create a private channel disconnected from the rest of the Lightning network. You have to do it with another node on the Lightning network that has large incoming capacity, and that costs money because that uses up their capacity.
Most methods involve some variation of either 1) paying for a service (as expensive as $1 fee per $1000 of liquidity) or 2) finding other trusted members to take your funds on Lightning and then send you back the funds off-chain or on mainnet. Merchants who only receive funds have to constantly rebalance their channel (or more likely pay some centralized 3rd-party provider to do it). While it's not anywhere as expensive as the 2% that credit card companies charge, it's an ongoing cost and annoyance. Some newer methods available for initial setup are Dual Funding, which is only available for certain nodes like C-Lightning, and liquidity triangles.

Limits to adoption

Not even the Lightning Network could scale Bitcoin beyond 10% global adoption because opening and closing a channel requires 2 on-chain transactions. Each Lightning channel has directional capacity, and whenever that gets exceeded (varying times depending on usage, e.g. every 2-4 weeks), it will need to be rebalanced. This usually means someone on the network is closing and reopening with new capacity. You can't expect people to store months of funds on a single channel. Half of the US is living paycheck to paycheck and gets new checks biweekly. Merchant stores typically close their accounts at the end of the day. If even 1% of the world used the Lightning Network and opened/closed channels twice a year, the Bitcoin Network would become completely congested.
The only way Bitcoin and the Lightning Network could grow to 1% global adoption is if most of the users are only interacting on centralized exchanges/nodes and settling on the Bitcoin network directly no more than twice a year.

Other Weaknesses

Lack of Features

Bitcoin is very basic. It only supports 1 token: Bitcoin. The scripting language it uses, Bitcoin Script, is also rudimentary. Most miners will refuse to run anything beyond the few known basic scripts that have been whitelisted for Bitcoin use. This includes multi-signature and time-locks. These are scripts, but they're too basic to be considered actual smart contracts.

Mining Pool Centralization

The top 3 mining pools own 60% of the network [Source]. Mining is not something the average crypto user can do by themselves unless they join a mining pool because the chance of winning the block is astronomically low. You need an expensive and specialized high-end ASIC miner for SHA-256 mining. To prevent miners from stealing the block, mining pool servers do not provide enough information to miners for them to tell when an attack is coming. They will only find out if they're running full nodes and paying attention, and only after the attack has been committed. Individual miners have no financial incentive to run full nodes, so it's rare for them to be auditing their pool operators.
This could be fixed with Stratum v2, but that's not available yet. And we don't even know if mining pool will allow it. There are multiple configurations of Stratum v2, and only 2 out of 4 give control of block production to miners.

Lack of Client Diversity

Everyone is running some version of Bitcoin Core, which is developed by a single skeleton crew. All documentation on how to run a node point to Bitcoin Core, and if you search for "how to install a node" on Bitcoin Talk, they all use Bitcoin Core.
In addition, the largest mining pools (AntPool, Foundry USA, and F2Pool, and Binance Pool) all use Stratum v1 clients, which gives full control of block production to operators. News about Stratum v2 has been slow, and it's uncertain if the pools will even want to give up their control and switch to it.
In comparison, Etheruem has at least 5 consensus clients and 4 execution clients. And their community is constantly encouraging others to switch to minority clients.

Security Issues

Lack of sustainability

Bitcoin pays its Proof of Work miners with a block reward for providing security to the network. The block reward is the sum of:

1. a fixed block subsidy (currently 6.25 BTC) paid through inflation of BTC, and
2. a variable transaction fee from customers

Currently, that block subsidy is about $200K per block and it accounts for [97-99%]((https://bitinfocharts.com/comparison/bitcoin-fee_to_reward.html)) of the block reward. Thus the current subsidy ($80 per transaction) is over 50x higher than current transaction fees.
As halvings continue and BTC's value can no longer keep doubling (due to exceeding the value of all assets in the world), the total block reward will keep decreasing. Some combination of the following must happen:

1. transaction costs increase to replace the block subsidy
2. miners drop out, leading to a decrease in Bitcoin's security

I don't expect consumers to pay for $50 transaction fees. Instead, there's a very high chance that Bitcoin will experience an ice age where all miners drop out except for the few miners who can acquire cheap ASIC rigs and run at the cheapest energy costs, leading to more centralization. This has been discussed many times before as the Tragedy of the Commons for Bitcoin since 2011. At that point, it will be extremely profitable to perform 51% attacks.

Susceptible to 51% attacks in the distant future

Proof of Work networks are inherently weak to 51% attacks.
Many smaller PoW have been successfully-attacked by mining pools from larger networks. Some PoW networks like Bitcoin Cash have introduced checkpoints to limit damage from attacks. Bitcoin Cash is actually quite resistant to 51% attacks due to that checkpoint, at the cost of having a longer finality time.
Bitcoin lacks finality checkpoints. It only takes $5-10B of mining equipment to compromise the Bitcoin network, and many billionaires and nation states easily have the funds to do this. Even poor countries like Nigeria have a $400B GDP. What's preventing others from attacking Bitcoin isn't the monetary cost--it's because it's hard to acquire that many mining rigs. But as halvings continue, if the price of Bitcoin doesn't double every 4 years, miners will eventually sell their equipment on secondary markets. Some nation state or billionaire could easily buy them, short Bitcoin, and then 51%-attack the network. They don't even need to go through the trouble of stealing funds.
Simply producing empty blocks is enough to grind the network to a halt. And they still get paid the block subsidy.
List of PROs (below): https://np.reddit.com/MPlankton/comments/127ztpv/bitcoin_research_mar_2023/jegk1nh/
List of CONs (below): https://np.reddit.com/MPlankton/comments/127ztpv/bitcoin_research_mar_2023/jegk6oh/

​

Imagine living in a world where the gas station has trouble providing gasoline.
Every few times a driver fills up, something goes haywire — the gas doesn’t flow, or it flows fast for a while and then slows to a trickle. Other times, the credit card payment is mysteriously rejected or the screen is blank.
If the consumer wants a helping hand, too bad. In this world, the gas station has no human, and the only option is a 1-800 number. The gas pumps are alone in the middle of a big parking lot.
Swap the word “gasoline” for “electricity,” and this is a realistic description of what happens every day at electric-vehicle charging stations across the United States. The high-tech, high-speed highway fueling system that America is building to power its EVs and replace the gas station is riddled with glitches that are proving difficult to stamp out.
Individually, they are hiccups, but collectively, their consequences could be profound.
“It adds to the non-EV driver’s view of the world that EV charging is painful,” said Bill Ferro, a software expert and founder of EVSession, an EV charger analytics firm. “People feeling that it’s a risk to buy an EV because the fast-charging infrastructure stinks is going to slow down EV adoption.”
The problems are experienced by those who use fast chargers on the go and who aren’t driving Teslas. Studies and innumerable anecdotes describe the strange stumbles they encounter: a blank screen, a broken plug, a credit card payment that fails, sessions that abort without warning, electric current that flows fast this moment and slowly the next (Energywire, March 21).
Behind the snafus are a daunting set of structural problems. They are tied to the peculiar way that EV chargers have evolved, and the fact that wires and batteries are way more complicated than what happens at the gas station.
“It’s a harder problem than pumping fuel from one reservoir into another,” Ferro said.
The problems are persisting even as billions of dollars pour into the charging sector from the federal and state governments, network operators and automakers.
Several recent studies of the charging system have found discouraging results.
Last year, researchers visited every public fast charger in the San Francisco Bay Area and found that almost 23 percent of them had “unresponsive or unavailable screens, payment system failures, charge initiation failures, network failures, or broken connectors.” And in a survey of EV drivers, the auto consultancy J.D. Power found the public charging network “plagued with non-functioning stations.” One in five sessions failed to deliver a charge. Almost three-quarters of those failures involved a station that malfunctioned or was offline.
Realizing the urgency of a fix, a variety of public and private players are trying out solutions.
The Biden administration, for example, set standards for “uptime,” or the percentage of time a charger is operational. California is launching a major inquiry into how to improve the customer experience. Automaker Ford Motor Co. last year deployed its own squad of station auditors (Energywire, Jan. 4, 2022). The largest public network, Electrify America, is replacing a fifth of its stations with newer models.
But many of these actions work at the edges of a black hole.
No one can define what it means for an EV driver to have a satisfactory charging experience. No underlying data exists. As hundreds of thousands of Americans buy EVs and start traveling the highways, this lack of a yardstick means that no one is accountable. Without accountability, problems are likely to persist.
The concern for industry is that the swelling ranks of EV drivers will tell their friends that highway charging is a little buggy, a little annoying — just enough of a hindrance that those millions of friends hold off from going electric, while the planet steadily warms.
Government steps in
Difficulties with EV fueling aren’t encountered everywhere. But they’re common in areas where the government is spending the most money and has staked its claim on improving the charging experience.
Home charging generally goes off without a hitch. The same goes for other “slow,” or Level 2, chargers that are sprinkled at workplaces and in parking lots and fill the battery by sipping on electrons for hours.
The bugs are most common with fast chargers, known as direct-current fast chargers, or DCFC. These are the charging solution when a driver is on the go and needs to refill the battery in 15 or 30 minutes.
These super-outlets exist near highway stops, in urban cores and at suburban crossroads. Building a network of them is the top priority of the Biden administration as it spends $7.5 billion for EV-charging infrastructure that Congress approved in the bipartisan infrastructure law. The first funding round is intended to blanket the country with DCFC chargers at 50-mile intervals along major highways.
Besides just building the stations, the government’s goal has been to create high standards, including better customer satisfaction.
The Federal Highway Administration, which is administering the program, said it “intends to incentivize charging station operators to improve reliability not just for chargers purchased with [government] funds but for all charging stations in the country.”
The irony is that a reliable, national charging network with high customer satisfaction already exists. It’s called the Supercharger network, built and run by Tesla Inc.
Users and experts are in wide agreement that Tesla has generally solved the problems that dog other networks. Its drivers plug in, pay and charge with few fails. Superchargers were exclusively for Tesla drivers until last month, when Tesla started opening parts of the network to other EV models (Energywire, Feb. 15, 2023).
Tesla also pioneered and excels at other practices that competing networks struggle to match — and that gas stations can’t do at all.
For example, punch in a destination on the Tesla app or the dashboard screen, and you get a route of Superchargers to stop at, with reliable information on which plugs are working and whether or not they are currently occupied.
A prime reason for Tesla’s success, experts say, is that Tesla owns and controls the entire ecosystem and all of its data. It built the cars, runs the chargers and manages the payments. If something breaks at the station, it’s Tesla’s job to repair it. There’s zero doubt who’s responsible.
The reason other, non-Tesla networks are having such troubles is that the public charging system has a lot more actors. They include a panoply of automakers, charging network operators, route-finding tools and now the government.
None has stepped up to take Tesla’s level of responsibility, and it’s not clear anyone will.
“Who’s owning this experience for me?” said Matt Teske, the founder and CEO of Chargeway, a EV-network software platform, putting himself in the position of an EV driver. “The answer is no one.”
A product not for customers
One reason today’s charging stations don’t work very well is their strange evolution as a consumer product.
Like EVs themselves, charging stations first arrived on the roads not because customers sought them out, but because regulators required them.
In the early 2000s, the California Air Resources Board demanded that automakers sell EVs to participate in the state’s auto market. Charging stations followed a parallel path.
One of today’s leading networks, EVgo Inc., was born from a 2012 legal settlement between NRG Energy Inc. and the California Public Utility Commission to resolve the electricity-market power crisis of the early 2000s. Another, Electrify America, is an entity that automaker Volkswagen AG was forced to create in 2016 as a penalty for cheating on its diesel emissions.
Tasked with satisfying regulators — and not customers — these companies saved money with a particular set of solutions.
“It was treated like it was really simple,” Teske said.
The charging station wouldn’t work like a gas station, with an employee in a nearby kiosk. Instead, it would operate without human intervention, like an ATM or a vending machine, but one that sells high-voltage electricity instead of Cokes.
But, unlike the ATM or vending box, the charging station wouldn’t get the full customer treatment. It wouldn’t be located under an awning, or be particularly well-lit or observed by security cameras. Instead, the electric car would be treated more like, well, a car. Stations were sited out in the elements, in the middle or at the edges of a parking lots, far from watchful eyes and easy targets for vandals.
Finally, the early networks didn’t offer the option of paying by credit card, although Electrify America was required to. Charging networks preferred to avoid the fees and complexity of Visa or Mastercard. Rather, payments would go to the network directly through a membership card.
What goes wrong
In the early days, none of these decisions much mattered. Early EV buyers were true believers who shrugged off the inconvenience of a dark parking lot or a frustrating charging session.
But as EVs not called Tesla have started to sell in earnest, the cost-trimming decisions of yesteryear have contributed to today’s outbreak of snafus.
A charging station, it turns out, is vastly more complicated and breakable than a vending machine.
“If you’re looking for a checklist of what could go wrong,” Teske said, “it’s a long list.”
Inside the kiosk’s metal shell sit sophisticated power electronics. It sends dangerous levels of electric current through a heavy-duty cable and to a connector, which can be easily disabled with a wad of gum. The video screen — crucial for communicating information to the driver, like the prices and how long a charging session will last — can be defaced or broken. Any number of problems can befall a driver trying to tap, insert or swipe a payment card.
And then there’s the computer code.
“It’s all software, at the end of the day,” said Ken Tennyson, the director of quality and conformance at Electrify America, which has 800 or so stations around the United States.
The recipient of the cash from an ATM or the Snickers from a vending machine is a human hand. The recipient of the electrons from a charging session is an electric vehicle — or better put, an ever-changing, ever-widening array of electric vehicles, each of which communicates with the charging station with its own version of software.
While software protocols exist for the charging industry, “adherence to those standards is not perfect, and the standards themselves are not perfect,” Tennyson said.
Some of these software problems are the growing pains that come with any young and high-tech industry that hasn’t worked out the kinks. This one, however, is complicated by the sheer number of systems involved.
A satisfying charging session is an orchestra. The charging station, the network operator, the vehicle and the payments all work together seamlessly. But today, the orchestra is out of tune.
These disjunctions create problems that to the driver are inexplicable.
For example, the kilowatts delivered at any particular moment during a charging session — the rough equivalent of the amount of gasoline tumbling out of a pump — can fluctuate up and down without explanation. This is often due to faulty communication between the station’s electronics and the car’s, as they try to speak the same software language but fail.
All these handoffs also make it hard for networks to emulate Tesla’s navigational prowess. Tesla competitors, like General Motors Co.’s Bolt or Ford’s Lightning F-150, might be able to tell you that the next station is working and has room — or it might not. Software incompatibility makes things janky.
“It’s hard because people are bundling all the pieces together and there’s not one owner of the process,” said Ferro of EVSession.
A solution, of sorts
To improve reliability amid this confusion, industry and governments have turned to a measurement of dubious value.
That metric is uptime. Applied to machinery or systems, uptime is a simple measurement. It is the amount of time in a given period that a machine is operational, stated as a percentage.
Uptime is a binary. A machine is either up, or it’s down. A charging station is considered “up” if its operator pings it and gets a positive response. But that narrow definition can collapse with an experience as complex as charging.
Uptime “doesn’t measure whether the connector is broken, or there’s payment-processing issues, or the parking space is ICE’d,” said Loren McDonald, an independent EV-charging analyst. ICE refers to a EV-charging parking space being occupied by an internal-combustion engine vehicle, another bane of EV drivers.
The metric fails, McDonald added, because it doesn’t answer the key question: “Can I charge or not?”
“Uptime doesn’t capture that,” he said.
Nonetheless, uptime has become the foundation on which federal and state governments are measuring performance as they dole out billions of public dollars.
For example, the Biden administration decreed last year that charging stations receiving federal money from the bipartisan infrastructure law must achieve better than 97 percent uptime.
It’s not alone in that target range. Colorado, New York and Vermont have adopted a 97 percent standard for their charger funding, while the state of Maine has adopted a lower 95 percent. Some electric utilities, like Louisville Gas & Electric in Kentucky and Consolidated Edison in New York, have aimed for as high as 99 percent.
Narrow as it is, the federal standard is at least a start. “It’s really helpful to have that federal guidance out there,” said Jesse Way, a clean transportation policy advisor at Nescaum, a nonprofit association of air-quality agencies in the Northeast.
Whether a 97 percent uptime is rigorous or not depends on whom you ask.
Tennyson of Electrify America considers it quite rigorous because it leaves a charging station with little room for error. A breakdown requires repairs that can consume days. In a year, he said, “You don’t have a tolerance for more than two or three failures.”
The reverse is argued by McDonald, the EV analyst. He points out that 97 percent uptime means three percent downtime. In the course of a year, that’s downtime of 11 days.
“Which is a really low bar if you ask me,” he said. “Could you imagine Amazon Web Services” — Amazon.com Inc.’s cloud product for businesses — “being down 11 days a year?”
A data void
One reason the Biden administration may have seized the narrow solution of uptime to solve the broad problems of EV-charger reliability is that no other options exist.
“The uptime calculation does not address all categories of failure or ways that chargers may fail to provide a satisfying customer experience,” the Federal Highway Administration conceded in guidance it released to states earlier this year.
Why not do better? “Insufficient data are available,” FHWA said.
That lack of data is a key gap, experts say. No independent, third-party source of charging data exists in the U.S. today. If a charging network claims to achieve 97 percent uptime — and many do — there’s no way to check out the claim.
That’s a worry for states that are entrusted with spending millions of dollars of federal money to build charging networks. The feds require them to achieve 97 percent uptime. But Teske, of the company Chargeway, pointed out that states who are vetting the companies to build those networks are “taking the sellers at their word.”
Solutions may be coming, but they will take time.
As part of the bipartisan infrastructure bill, the Biden administration gained the authority to gather data from the charging stations it funds. States are required to start sending data along a year from now. That information will, FHWA says, become “a national database and analytics platform” with “a public-facing dashboard.”
A major structural solution may be in the works in California.
Last year, California lawmakers passed a bill that requires new record-keeping and reporting standards for charging stations that get state money. The rules, to be drafted by the California Energy Commission (CEC), are due by 2024. They could eventually be copied by the numerous of states that align themselves with California’s transportation emissions policies.
“It’s very complicated, we don’t have all the answers,” said Patty Monahan, a CEC commissioner, at a Bloomberg New Energy Finance conference earlier this year. “This is a huge inquiry.”
In the meantime, the gritted teeth of those using highway fast chargers are unlikely to relax anytime soon. The next wave of EV adoption will, in part, continue to be a story of inert plugs and frowning drivers, posted on Instagram for all to see.
“I see this is a problem for the next five years,” said Ferro, the EV charging expert. “Either Tesla will take over the entire charging network of the U.S., or everyone else will get their act together, or a little bit of both.”

​
https://www.eenews.net/articles/why-americas-ev-chargers-keep-breaking/

I've budgeted for the last 10 years but now I'm not sure we're allocating our money well in proportion to our income and setting ourselves up well for retirement. We started making good money ($60k) 5 years ago, before that we were making poverty income. It has gone up a good amount every year (now $155k)
My main concern is that our budget after retirement contributions is close to our monthly income, and I wanted to boost our emergency fund to 1 year (currently have about 5 months worth)
My wife and I are 35 & 40, and behind on retirement savings ($70k saved). Maxing out the 401k + match should put us at 4-5x our annual income by 50 years old. So I'm hoping that should cover it. Is that a good assumption?
I plan on using my annual bonus to max out maybe 1 IRA at least, the rest for emergency savings and vacations. I don't like planning on that income since the bonus isn't guaranteed.
Also I'm not sure if our need to catch up on retirement means we should cut some from our budget and limit our travel. We planned on using it to fly overseas to visit my wife's family.
For a family of 4, low/mid cost living area
Household income - $155k
$700 - HSA (just started contributing) $660 - insurance $2778 - taxes $1660 - 401k (maxed, not including 5% match + 5% match not vested for 3 years, so 10% total -> $35500 annual contribution in total )
= $6100 net monthly takehome (averaged over 26 checks) +10-20% annual bonus of total income
Assets $350,000 - house $10,000 - vehicles $33,000 - checking, savings, emergency fund $70,000 - 401k, IRAs
Debts $210,000 - mortgage, 3.5% $33,000 - student loans, 3-5% rates
Monthly expenses:
Home (3400 sqft) $1770 - mortgage, taxes, insurance $350 - utilities (average over 3 years) $500 - saved for repairs and upkeep $230 - squirreled away for furnishings, replacing appliances Auto (2 vehicles, 10 & 12 years old) $70 - insurance $100 - gas $200 - repairs and service (averaging out for expected future repairs, new tires every 5 years, etc)
Utilities $50 - internet $130 - cell phones (3 lines, Google Fi - should go down to $80 in a year)
Entertainment $150 - streaming services, family activities $350 - kids activities (karate, gymnastics, swim lessons) $100 - pets (dog medical insurance, food, vet bills)
Shopping $150 - clothing $300 - everything else (gifts, things for the house, etc)
Food $950 - groceries, eating out 1-2x a month $75 - personal care items
Health $300 - doctor visits, pharmacy, ootc products, vitamins etc. I plan on not taking from the HSA for this, and using the HSA as a retirement account. Not sure if its a good idea or not.
Student loans $250 - I've been paying this since covid forbearance to pay down the balance, not sure what it will be later this year
Total: ~5900

To be fair, as a matter of transparancy, the killing of Credit Suisse is still in the news cycle over there. On the money front consensus seems to be, that from the 259 billion involved, the first fifty billions, the tranquilizer from the SNB for CS, does not count as monetary poison for the surviving UBS. the 9 billion is an actual loss guarantee from the state to UBS. the remaining 200 billion involve, again, the SNB. A casual observer understood it is 100 billion for each of the two banks. The explicit political aproval of providing a credit line is only for the surviving UBS. The other 100 billion are not talked about. It's the SNB dealing in the shadow with the rotten corpse of CS, one is tempted to believe. The credit line provision comes at a cost to UBS of a quarter percent (dead ones tell no tales). Funny enough some narrative spins this as a boon for the public coffers, the old horsedroppings theory for the sparrows.
In politics, the parliament gets involved. the first action was to nod full support for the executive emergency actions by the commissars. Talk of a truth finding commission to establish a well thought through narrative is on the table. We want a public investigation from knowlegeable people, not your handpicked experts. Parliamentary procedures are scheduled for next week. One more thing to the numbers, of course it was also admitted, that at least the tranquilizer was intended to be high dosage, bazooka central bank style, you know, and that's probably a reason the numbers are so outrageous, too. There's a safety margin with the bonus of boasting, boooh! no, come on, man. Oh, wich brings me to another clown show, the PR management of the gnomes replaced the UBS CEO with a déjà vu guy, known as the George Clooney of Paradeplatz. Probably quite an ambitious guy. Media went orgastic on this. The other gnome was more than happy to step down, pretty sure about that.
The SNB also said they have been monitoring the situation for months, and the sunday night last minute action was the best solution, merging the two too big to failures. The international market ramifications are often mentioned as an important factor to adjust the law ad hoc, but nobody makes the link to the rotten financial system underneath. Let alone say something critical. The old finance minister said, what do you wnt to do if everybody around you is building mountains of debt? As you know, the swiss banks wanted to play front line in that game. The SNB backed them, politics too, this is the second time they would go bust. Changing the law of the game is also known as cheating. Some people believe they didn't get the truth about the vax-op, and some people don't believe they will ever come clean about anything. This has nothing to do with good or bad, it's just the perspective you find yourself in.

My apartment has a 'high security' key that locksmiths have been unable to copy. I went to the original maker of the key (Bulger) and they said it costs like $15 for a replacement, just that I need to make the order through an 'authorized' contact (apartment ownemanager).
My manager said that this would cost me $200. More than 10x what the key costs. I asked if they would change the lock, and they said no, they're just gonna give me a copy. This 'clause' (about replacement cost for a key) is in my lease, which I signed before I moved to the US (I'm an international student). I didn't realise that getting key copies made was much cheaper, and that I was being overcharged.
Is there any way around this?

All,
I'm looking to replace the thermostat in my 2012 JK, as I have the P0128 code indicating a thermostat issue.. And there's a nationwide parts shortage on the OEM housing. The only replacement I've found around me is Napa's "OE" housing. Has anyone tried one of these? It's like $22 so it's not the end of the world if it can't clear the engine code, but preferably I don't want to have to swap thermostats over and over again on the jeep if I can help it.
​
If you have any other brands/specs outside of the OEM, that would be helpful too..