nitrolife

Part of my last minig farm, if you interesting.

and yes, I also mined bitcoins on the CPU. back when you could get near 200 bitcoins per hour of work. After that I mined altcoins on video cards, mined through nicehash. what I have not mined. and I can say - it’s all thrown money. much, much more profitable just to buy cryptocurrency and hold on cold wallet.

if you don’t want to tell me that people are setting up mining farms in Somalia (because it’s pointless from a risk perspective), I don’t think electricity rates are particularly lower anywhere else.

In any case, you can make money from anything. It’s just that the risks associated with mining don’t justify the investment. That’s all.

I think you have the wrong units. The average fee is 1.5 USD.

https://bitinfocharts.com/ru/comparison/bitcoin-fee_to_reward.html#3m

Percents you can calculate self. That’s not for all blocks, but in a day.

UPD: This is not for every block, of course, but for the whole day. I was wrong about that. Maybe really 2%

This is quite quick. Last time I looked the it was around 3 years. Most of the cost comes from buying the hardware.

my calculations were made without taking into account the growth of the network’s complexity. So, when I tried it last time, the network’s complexity had increased so much in a year and a half that the equipment was not bringing in much, and it was not worth the risk of investing. However, things may be different now, and I may be mistaken.

UPD: Now I just buy Bitcoin on exchanges, and it brings me the same % of income as mining. But I don’t have to deal with equipment, follow ridiculous laws, or waste electricity. =) That’s why I say that many peole just tkabe bitclin to cold wallets. Less bitcloin exists on exchage then grow price.

There have been no new Bitcoins for a long time. Everything that miners mine is just a transaction tax. In fact, to describe the reason for bitcoin’s growth, you need to understand what money is all about. Not just crypto money, but in general. In short, the price is rising because many (including miners) believe that it will rise and do not spend bitcoins. In a normal economy (except Japan), you could just print more money and the price would drop because the currency unit would depreciate. But bitcoin is a mathematical model, and it has a limit. You will not be able to create more Bitcoins than you have already created in any way. Therefore, the belief in the growth and retention of the currency reduces turnover and the price increases. If any of the whales withdraw their entire stock in one day, the market will fall for many years.

UPD: Excuse me, I really made a mistake. You can still mine 3 bitcoins per block… but to be honest, 3 bitcoins for a whole pool is only an eighth of the original 25 bitcoins per person. In general, mining has not compensated for mining for a long time.

UPD: I checked just in case. The average commission payment is now 1.5 bitcoins. almost half of the reward

UPD: I will reveal my thought even more. An ASIC at 1160 Th/s costs 33k dollars and consumes 11 kW. Even in my region with a low-cost light (only 5 cents per kW), such an asic will be able to bring only 58 dollars per day. And it will pay off only in 1.7 years. This is the moment when the miner will FINALLY stop working at a loss. And this is in ideal conditions without increasing the complexity of the network and other things. So all the miners who don’t buy huge amounts in bulk barely pay for their business.

There is a limited amount of Bitcoin, and some of it is lost in forgotten wallets, so the total volume is constantly falling. This may partially increase the price.

But in reality, as in any speculative market, the price of bitcoin depends mainly on faith in it and speculation about world events (some kind of cataclysms, regular statements of this or that person about cryptocurrency, etc.)

The main real value can only be found in countries that are disconnected from SWIFT. However, almost no one appreciates this because there are only 5 officially disconnected countries. However, if this list continues to grow, cryptocurrencies (including Bitcoin) will become more prevalent in international transactions.

In the initial stage of shared library support, everything was exactly the same. Let’s look at it in 5 years… When some soft will archived and die, some stop maintaining, some new crated and brakes old dependencies…

runtime have versions too. If one runtime version use only one flatpack than exactly same as just static linking binary. Flatpack have just docker layeredfs and firejail in base.

id: org.gnome.Dictionary runtime: org.gnome.Platform runtime-version: '45' <- here sdk: org.gnome.Sdk command: gnome-dictionary

They don’t have to! Flat pack doesn’t remove all other ways to install software. But for 95% of use cases, it will do just fine.

Tell this to canonical, they even firefox put in the snap. You know that when choosing “quickly compile something for a flatpack” and “support 10+ distributions”, the developers will choose a flatpack. Which in general looks fine, until you realize that everything is just scored on the mainline of libraries and molded on anything. The most striking example of this is Linphone. just try to compile it…

Provided that flatpack has a common parent container, which is not always the case. More precisely, it almost never does. Because someone updates flatpack to new versions of the parent containers, and someone else does not.

this is a system for work tasks. Of course, I understand what the developers are going for. that is Android. And it’s really nice to read the Internet on android. But try to do something more complicated than that and you’ll realize that it’s hell. However, I don’t mind if such distributions appear. Why not? I just don’t understand people who voluntarily limit their abilities. And why you don’t just install Android 64?

The flatpack approach automatically remove everything low-level from the equation. Do you want to write directly to the graphics card buffer? Read the input? Do I set the fan rotation parameters directly in the /proc? All these applications will never work in flat pack.

On the other hand, flatpack is superfluous and for convenience. You can simply build an executable file without dependencies and configure firejail for it yourself… That’s all. Or run the file from another user. That is so popular exactly bacause RedHat pushed them. Literaly like Canonical pushed snap.

However, the extent of the damage is limited by flatpak and whatever permissions you have set, and, if I understand it correctly, you cannot attack one flatpak through the other unless they share access to some files.

there is a problem here that permissions are also set by the packages developers. User in most cases click accept all and alll done.

On an unrelated note: apparently, there is finally some Russian Lemmy instance? That’s a welcome change.

Well… Appeared 2 years ago. It’s just that practically no one needs it. =)

Times are changing, and memory constraints for most programs are generally not relevant anymore.

But there are gaps in the libraries that, unlike distributions with dependencies, can no longer be managed. And all the security of your system depends on a small flatpack access control, which 99% of users do not understand at all and, with any problems simply opens access to the entire home directory.

It’s not the 80s, and I can save a few megabytes to keep my system running smoothly and well-managed.

And then it turns out that you have 18 libssl libraries in diffirent fpatpacks, and half of them contain a critical vulnerability that any website on the Internet can use to hack your PC. How much do you trust the limitations of flatpack apps? are you sure that a random hacker won’t hack your OBS web plugin and encrypt your entire fpatpack partition (which some “very smart” distributions even stuff office into, and your work files will be hidden there). People have come up with external dependencies for a reason.

I’ve been working on Linux for 15 years now and I perfectly remember the origin of many concepts. If you look at it through time, what would it be like:

1. We can build applications with external dependencies or a single binary, what should we choose?
2. The community is abandoning a single binary due to the increased weight of applications and memory consumption and libraries problems
3. Dependency hell is coming …
4. Snap, flatpack, appimage and other strange solutions are inventing something, which are essentially a single binary, but with an overlay (if the developer has hands from the right place, which is often not the case)
5. Someone on lemmy says that he literally doesn’t care if the application is built in a single binary, consumes extra memory and have libraries problems. Just close all permissions for that application…

Well, all I can say about this is just assemble a single binary for all applications, stop doing nonsense with a flatpack/snap/etc.

UPD: or if you really want to break all the conventions, just use nixos. You don’t need snap/flatpack/etc.

I’ll finish my house, clear out the old tree debris, and complete all the physical work I’ve planned for years to come.