At least it is in correct the Off Topic area `cause it's just about as far from crypto as ya can get
An amusing break from reality though...

Hopefully you will find a way out of this and will have learned to change your ways. If not.... Just another case of Darwin at work...

Um, any PC PSU over 100w that has at least 1 6-pin PCIe connector...

Oh btw: If you insist on using a connector for supplying Vcore then check out Anderson Powerpole connectors like found at https://www.amazon.com/anderson-power-pole/s?k=anderson+power+pole  They are specifically made for low voltage use (but rated up to 600v) and start at 15A with up to 350A capacity. Rather inexpensive as well The companies page is at https://www.andersonpower.com/us/en/resources/PowerPoleResourcesPage.html

The' Best' all depends on the buyer - not some random person's thoughts... Most hardware guides here are pretty clear on that: 1st pick your criteria and choose from either speed, efficiency or, service life and go from there. There are no miners that are the top in all 3 areas.

Personally I use service life as my #1 and that means any of Canaaan's Avalon miners.

No, it is NOT a 'joke'.
Just a different version of the more common pumped-storage solutions that use water which is pumped into reservoirs and released through generators when peak loads need it. Yes it takes a fair bit of energy input to do the pumping but typical power-in to power out is still a bit over 50%. The point of it is to suck up excess power when available by using massive electrically driven pumps or compressors to fill the potential-energy 'bank' (the compressed air or elevated water reservoirs) and then on very short notice release that stored energy to be able to put back into the grid a large % of the power put into the system to handle short-term peak loads.

A perfect example of a massive one built in the 1960's is here It um also had massive failure in 2005 when one of the side walls gave away due to water level sensor failures and design faults. 1 billion US gallons (3,800,000 m3) of water was released in 12 minutes....

Ja it makes zero sense to use either method for small-scale operations but on the Utility scale both are very practical.

The US and other countries have had scaled up variants for well over 10 or more years. They use sealed underground caverns, smallish mines, and tapped out salt domes. Pressures usually top out at less than 100bar but when you have millions of cu/ft of storage that represents a shitload of stored energy. Just google "compressed air energy storage" or ref https://energystorage.org/why-energy-storage/technologies/compressed-air-energy-storage-caes/

Again it must be said: When building mining platforms one MUST follow local industrial wiring codes for distributing the power. If local codes are non-existent then follow international ones.. The pics above are just more proof of what WILL happen when ya 'just run a lot of wires' without knowing what you are doing...

NO!
For the umpteenth time of someone new to the Forum and BTC world asking ' would it is a good idea to' the answer is: Doing that is called just creating yet another fork aka an altcoin that would have zero relationship to BTC.

From its inception by Satoshi it has always been known that technology would push difficulty higher until they balance out. He foresaw that better tech starting with GPU's then FPGA's and now ASIC's - would show up making the older tech useless. Period. End of story.

This thread and the link you gave are from the very early years of BTC and were long ago proven to be irrelevant when compared to using ASIC's to just brute-force the process.

Ja as I pointed out in the thread in Development & Technical Discussion thread using on-board Vcore regulator is a must. At the low voltage and high currents involved only a properly bypassed bolted cable connection with remote sensing would work when using an external low voltage PSU. Any other kind of connector will simply have too much voltage drop unless it's huge and - be way too expensive.
TI has several good links regarding proper Vcore regulator design that I posted in that above link along with Those TI links also giving parts sourcing and availability links.

As for finding parts & sourcing them - use Octopart They link to every reputable component mfgr on the planet to show current and historical parts availability, full data sheets, Factory Approved distributors, 3D component models, etc.

And given all of that, this now belongs in the Altcoin (Mining) area - not here which is only for BTC topics.

Email sent at 3:42p EST
Thank you very much, it will be interesting to see what it is 

Considering we are nearing the end of 2022, a little late eh?

It's not that sha256 is not suited for GPU's - it's that the algo is very easily implemented in 'hard wired' ASIC's without any of the other circuits that a GPU has in it. Just 1 of the chips used in Sidehacks Compac-F has 8k SHA256 cores in it. Compare that to how many cores (spread over several chips) even a top-end GPU has...

Current BTC hash rate is several 100's of ExaH/s so yes, its difficulty is *very* high.

No!
The voltage applied to the strings of ASIC chips is controlled to very specific levels and is set by the miner controller which talks to the PSU. The voltage applied to each chip is called Vcore and has a very large effect on how fast it can run and how much current it pulls at any given speed.

As for cooling... The water flowing through the miner MUST be very clean because there are several very small passages it flows through. You cannot just suck in water from the river/lake/ocean and pump it through the miner.

And using a CPU for it means that you are NOT mining BTC but some other sha256 altcoin. There is literally a HUGE difference in difficulty between them...

Yes, pools do set a work difficulty for the shares they send to a miner but that has very little to do with current BTC block difficulty. They are 2 different things ref https://kano.is/index.php?k=workdiff

Ja. Do remember that using your 5v 3A target, that is of course 15W. To pull that using a Vcore of 400mv means supplying 37.5A to the chip(s). A good place to start for resources is TI.
https://www.ti.com/design-resources/design-tools-simulation/processor-fpga-power/overview.html
https://www.ti.com/design-resources/design-tools-simulation/power-stage-designer.html

AFAIK Vcore for Sidehacks single-chip Compac-F is 1.48V which helps a bit with the current (10.14A for 15W) but more to the point allows for more stable operation at high speeds. Can't say I've bothered to look at mine but I believe he uses a buck controller driving MOSFET's. And yes, TI has some rather nice controllers as well as matched power stages to work with them.

Personally I'd lose the idea of powering a 2-chip design through the USB connector and use 12v in via a barrel connector. Most can handle up to 5A or so all day long though pushing them to their typical rating of >8A continuously is a bad idea. The higher input makes the power stages a fair bit easier.

Ja. the fact the boards do not have any conformal coating on them to seal the solder joints and other things from moisture coupled with the incredible amount of dust/dirt that the miners inhale leads to dendrites (tin whiskers) growing and those soon cause short circuits.

More efficient and faster - yes there is still a slight amount improvement that is possible by using the 3nm node provided that Biitmain, Canaan, et al think it is worth the multi-millions of $$$ it would cost them to develop the chips for it.

Use less power (which = less heat) -- no. Miner manufacturers have zero incentive to make lower power miners as their biggest customer base
 - farms - are perfectly happy with ~3-5kw per miner. So, the makers will just pack more chips in the miners to hit that power usage point just they have always done.