Cryptocurrencies

[NviGate Systems]

Cool. Is that an IKEA shoe rack?

Well that's certainly a super-cheap and easy solution. I'd be a little concerned about leaving it running unattended though with all that combustible material. Make extra sure you're not putting too many GPUs or risers on one power cable. I've seen overloaded connectors melt, leading to shorts and eventually fire. Might be a good idea to get a coating of flame-retardant paint on the rack.

It's a storage rack from Jysk (IKEA clone).

Each Riser, HDD and GPU has a dedicated cable from PSU. The only lines that share power is some of the molex connectors that go to risers power he fans, but that is 3 fans per run, so not an issue.

Is flame retardant expensive or smelly? Actually, I have some aluminum tape, and aluminium sheets I can put down. That would stop any potential fire.

 

[Moltuae]

Is flame retardant expensive or smelly?

It is relatively expensive but you should only need a small amount. Last time I used some (a couple of years ago) it was water-based and practically odourless, like most paints these days.

For an quicker/easier/cheaper solution, you could maybe try something like this instead:
http://www.fireprotectiononline.co.uk/fire-retardant-spray.html


This was the flame retardant paint I last used (to paint the frame around an internal fire door at home). I would imagine something similar is available in the US:
https://www.rawlinspaints.com/home/...board/391-zeroflame-fire-retardant-paint.html
(Bit expensive but that's for 5L. You should need less than 1L for that rack).

 

[Moltuae]

Each Bitcoin Could Be Worth $619,047 in 10 Years: Forbes Contributor

Article Link: https://cointelegraph.com/news/each-bitcoin-could-be-worth-619047-in-10-years-forbes-contributor


Bitcoin Will Continue to Rise and its Not a Bubble – John McAfee Comments

Article Link: https://ethereumworldnews.com/bitcoin-will-continue-rise-not-bubble-john-mcafee-comments/

Video interview from the article:

 

[HCHTech]

You need to have a good story, though, and ROI plus a good upside for profit has to be part of that story. We'll see how it goes.

Spent some time searching this am, wow, that fancy extruded aluminum is expensive. I'm thinking a gorgeous frame like @Moltuae built has got to cost at least $150, maybe $200, then putting 5 1080Ti's on board is $3,600 or so, $160 for a power supply, $70 for CPU + RAM, $135 for a motherboard, $50 for the PCIe risers, and $50 for a hard drive or small SSD, and $130 for a Win10Pro license. That's $4,345 my cost. Let's assume I add $400 or $500 to build the thing, so we're at $4,745 or $4,845 retail to the customer. At $13/day current performance, we're looking at a year ROI.

If we assume a 3 year working lifetime, that's ($14,235 less electricity costs of what, $1.25 a day? that would be $1,370 so net return of $12,865) return on an investment of $4845. That's what, 55% annualized return (after hardware costs are subtracted)? Not a bad answer, but there are several ways this might go south. I think if I put enough caveats in the sales speech (past performance is no guarantee of future performance, no warranty on graphics cards, your mileage may vary, etc. etc.) it could work. Maybe have some graphs to show what return would look like if BTC halved in value, stayed even, or doubled in value, something like that. Hmm.

 

[Blues]

Every time I hear you guys moaning about chassis/frame options it makes me wish I knew how to fabricate such a solution and sell them to y'all for BTC

 

[Moltuae]

Spent some time searching this am, wow, that fancy extruded aluminum is expensive. I'm thinking a gorgeous frame like @Moltuae built has got to cost at least $150, maybe $200, then putting 5 1080Ti's on board is $3,600 or so, $160 for a power supply, $70 for CPU + RAM, $135 for a motherboard, $50 for the PCIe risers, and $50 for a hard drive or small SSD, and $130 for a Win10Pro license. That's $4,345 my cost. Let's assume I add $400 or $500 to build the thing, so we're at $4,745 or $4,845 retail to the customer. At $13/day current performance, we're looking at a year ROI.

If we assume a 3 year working lifetime, that's ($14,235 less electricity costs of what, $1.25 a day? that would be $1,370 so net return of $12,865) return on an investment of $4845. That's what, 55% annualized return (after hardware costs are subtracted)? Not a bad answer, but there are several ways this might go south. I think if I put enough caveats in the sales speech (past performance is no guarantee of future performance, no warranty on graphics cards, your mileage may vary, etc. etc.) it could work. Maybe have some graphs to show what return would look like if BTC halved in value, stayed even, or doubled in value, something like that. Hmm.

I'd say your assumptions and calculations are pretty much spot on for the most part.

The rig earns approximately $13-14 per day presently (~$5000 per year), not including electricity costs.

However, a few points to note:

• The earnings are passive: As long as you're making some profit, there's little-to-no work involved once the rigs are mining, so it's very scalable (as long as you have the resources and space).
  
• Electricity costs may be reduced (or entirely ignored) if you generate your own electricity, can make use of the heat generated or can claim most of your electricity costs as a business running expense.
• The hardware costs (in the UK at least) can usually be offset against tax as business equipment purchases (and you can often even claim for depreciation of hardware I believe).
• Most of the hardware has a good resale value. The graphics cards in particular depreciate very slowly so you may even be able to resell them for 50% or more of their original value if/when you decide to replace/decommission the rigs. If you take that into account, the non-reclaimable cost of the rig is probably closer to $3000 or less).
  
• The Bitcoin/altcoins earned now could be worth much more in the near future so the annual return could be higher that it seems (assuming you don't sell/exchange your cryptocurrencies for fiat of course).

My last point begs the question: "Why not just buy cryptocurrencies to invest in?" ...

What you have to realise here is that by buying altcoins directly with fiat, the average tax payer is probably paying something like twice as much because the government took half of their income as tax before they could use it to purchase anything. As a business owner, buying mining rigs only increases the (largely non-taxable) business expenditure.

I don't sell the rigs I build (not yet at least). My customer buys his own hardware and I work with him to help build and maintain his rigs. I already do a lot of work for this particular customer maintaining their systems (2 large sites, several servers) so my time spent working on their rigs just goes on their bill.

 

[MudRock]

Wow. I get right angle steel from a local guy, 3/4", for $5 for 8 feet. Fairly thin, it dremels nicely and a lot cheaper.

Mining could come back up too.

Sent from my ONEPLUS A5000 using Tapatalk

 

[Moltuae]

Wow. I get right angle steel from a local guy, 3/4", for $5 for 8 feet. Fairly thin, it dremels nicely and a lot cheaper.

The aluminium extrusion isn't much more expensive than that. I'd guess around 7 USD per metre (I used 6 metres in that rig). And probably about the same cost again for all the brackets/fixings. However, there's no drilling required so the labour is less. Taking into account my hourly labour charge, aluminium extrusion may even work out cheaper.

 

[HCHTech]

Wow. I get right angle steel from a local guy, 3/4", for $5 for 8 feet. Fairly thin, it dremels nicely and a lot cheaper.

I was thinking about constructing a unit for sale. If it looks like a home-brew project, you have a strike against you right from the start. That frame @Moltuae built is stunning - I could sell that.

The idea of having them purchase the hardware themselves is interesting - but they don't have accounts with my distributors, so I would have to use all amazon or newegg or whatever for everything, raising the cost. Plus, I'm not an accountant or lawyer, but there is something that doesn't feel right about that if I am the one actually doing the buying, but, for example, using their credit card. How can I build it unless they ship it all to me? If we do that, then it sounds like a sale, which would be subject to sales tax. If they buy everything using a parts list that I provide, get the parts shipped to them, then bring them to me to assemble, it still feels weird, but is probably better. I suppose I could sell the parts list for $400 - haha. Although I have no idea if I will sell even one.

BTW, on that extruded aluminum, I found a US-based mfg here.

The aluminium extrusion isn't much more expensive than that. I'd guess around 7 USD per metre (I used 6 metres in that rig). And probably about the same cost again for all the brackets/fixings. However, there's no drilling required so the labour is less. Taking into account my hourly labour charge, aluminium extrusion may even work out cheaper.

@Moltuae , are you using this? $0.24 per inch = $9.45 per meter. Ok, that's a little better. More numbers to run, but that will have to wait - I have some real work that needs done -

 

[timeshifter]

that fancy extruded aluminum is expensive. I'm thinking a gorgeous frame like @Moltuae built has got to cost at least $150, maybe $200

If you can wait a few weeks you can have one delivered for $160.

Stackable Computer Fame 8 Graphics Card GPU USB PCI-E Cable Computer Case BTC LTC Coin Mining Case Frame Server Chassis

 

[Moltuae]

@Moltuae , are you using this? $0.24 per inch = $9.45 per meter. Ok, that's a little better. More numbers to run, but that will have to wait - I have some real work that needs done -

No, I'm using 2020. This is probably the closest from that supplier: https://8020.net/shop/20-2020.html

More precisely, I'm using this: http://uk.rs-online.com/web/p/tubing-struts/8508476/
(though I have since found an Ebay seller selling the same stuff slightly cheaper)

What I wasn't aware of when I purchased it is that not all aluminium extrusions are created equal. What I've learned is that there are two profile slot sizes; 6mm and a less common 5mm slot profile. The latter is often referred to as the 'European' profile (I think). It would seem that the 5mm slot stuff is usually cheaper and has a slightly different profile. Those with a 6mm slot seem to conform to the Bosch Rexroth design so I wonder if the higher price is due to licensing costs. Fixings for the 5mm slot extrusion can be harder to find and sometimes more expensive than the 6mm slot fixings.

Strangely, the extrusion from the supplier you linked to seems to have a 5mm-ish slot but the centre hole profile is more like the Rexroth stuff. The extrusion I used has a small round hole in the centre instead.

I'll post a parts list when I've had time to compile one, and I'll post photos and details of measurements as I build the next rig. This first rig was a trial and error construction to find the most efficient/compact way to mount everything. I took it apart and reconfigured it numerous times before I found the best compromise. For the least wastage I cut the aluminium extrusion into equal lengths. Starting with 6 x 1 Metre lengths, I cut two of them in half (500mm) and 4 or them into thirds (333mm). The 4 x 500mm pieces make the 4 uprights/legs, and the 12 x 333mm pieces are used for everything else. If you want to make it easier to lift, you can add another (13th) 333mm piece above the fans (as I have), bringing the total to 6 and 1/3 lengths per rig. Additionally, I used a piece of 40mm x 5mm flat aluminium bar (pushed into the extrusion slots) to support the risers and prevent the cards waggling from side-to-side.

Edit to add: The Acrylic sheet started as two 500mm x 500mm sheets. The supplier will cut-to-order so I asked them to cut the sheets into two 500mm by 370mm sheets (for the sides) and then cut the offcuts down to 370mm too, making two 370mm x 130mm panels for the front and back. If you're using a single PSU, you could even make use of the remaining 130mm x 130mm offcuts to make small panels for either side of the PSU, but I haven't bothered to do that.

 

[Moltuae]

50 Blocks to SegWit: Bitcoin's Coming Upgrade Is a Blockchain Game-Changer

Article Link: https://www.coindesk.com/50-blocks-segwit-bitcoins-coming-upgrade-blockchain-game-changer/

Block 481,824.

That's when Segregated Witness (SegWit) will activate on bitcoin late today or tomorrow morning, bringing the years-long debate over the contentious code upgrade to an official close. At that time, users will finally be able to take advantage of the long-promised technology.

Segregated Witness (SegWit) to Activate Within 24 Hours: How Bitcoin Will Change

Article Link: https://www.cryptocoinsnews.com/seg...vate-within-24-hours-how-bitcoin-will-change/

 

[Moltuae]

A long but interesting article for anyone who wants to better understand what SegWit is and how it came about ...

The Long Road to SegWit: How Bitcoin’s Biggest Protocol Upgrade Became Reality
Article Link: https://bitcoinmagazine.com/article...oins-biggest-protocol-upgrade-became-reality/

Segregated Witness (SegWit) has activated on Bitcoin. As of today, all SegWit-ready nodes on the Bitcoin network are enforcing the new rules, marking Bitcoin’s biggest protocol upgrade to date.

But activation did not come easy, and it did not come fast.

This is a look back at the long road to SegWit.

 

[MudRock]

A long but interesting article for anyone who wants to better understand what SegWit is and how it came about ...

The Long Road to SegWit: How Bitcoin’s Biggest Protocol Upgrade Became Reality
Article Link: https://bitcoinmagazine.com/article...oins-biggest-protocol-upgrade-became-reality/

I wonder how long before transaction fee will drop...

 

[Moltuae]

I wonder how long before transaction fee will drop...

As I understand it, that depends on the time taken to implement it, especially by wallet developers.
At the end of the article it states:

The final step for Segregated Witness is, of course, actual user adoption. Since SegWit has only just activated at the time of publication of this article, it’s impossible to know how quickly and how much the upgrade will actually be used. Some critics, perhaps most notably Garzik, predict that widespread adoption could take up to a year or even longer. Others, including a number of wallet and library developers, think they can utilize the feature within weeks, or they are prepared already. And other technologies that depend on the upgrade, such as the Lightning Network, but also Merkelized Abstract Syntax Trees (MAST), atomic swaps, faster transaction signing for hardware wallets, the more efficient Schnorr signature algorithm, and TumbleBit in payment processor mode, are in various stages of development as well.

See also: Bitcoin’s Capacity Increase from Segregated Witness Could Kick in Fast
https://bitcoinmagazine.com/articles/bitcoins-capacity-increase-segregated-witness-could-kick-fast/

Why Segregated Witness Is Not an Instant Block Size Limit Increase

The reason that it was originally thought that Segregated Witness would not offer much in terms of increased capacity right out of the gate is that wallets, exchanges and other creators of transactions must also complete their own upgrades before the benefits of SegWit could be seen on the network.

Since each user must effectively opt in to the increased capacity offered by SegWit, the growth in capacity would be seen when users start using SegWit transactions rather than when the protocol change is initially activated.

 

[MudRock]

Ah okay. So unlike the August 1st fork, this is more an 'optional' upgrade. An available but not mandatory perk. I don't know why wallets wouldn't jump on it though.

 

[Moltuae]

I expect they will.

Yeah, SegWit was the gentler approach. A User Activated Soft Fork, or UASF (not to be confused with the USAF --- don't you just love the hats! haha). It's arguably the safer solution, developed by the core team, using well-proven and tested code that was also successfully deployed and proven on the Litecoin network a few months ago. Not everyone agreed that SegWit was the way to go however, hence the recent Bitcoin Cash hard fork.

 

[Moltuae]

On the subject of scalability and performance issues, Andreas posted this:

Bitcoin Q&A: The Internet will eat your business

 

[timeshifter]

I just watched that video too (I'm a subscriber to his channel). Very good points made.

 

[MudRock]

On the subject of scalability and performance issues, Andreas posted this:

Bitcoin Q&A: The Internet will eat your business

I think the issue at hand, is while it is decentralized, we have players in play already able to manipulate it. BCC/BCH for example. Or how long before segwit was used. We could be leapyears ahead (And obvious when you compare vs technologies in altcoins.) But if Bitcoin doesn't pick it up, someone/somealt else will.