Wednesday, December 13, 2017

Why Bitcoin is not a bubble... yet

Bitcoin has surged to a record breaking high of USD 17,000. Inevitably, lots of people called Bitcoin a bubble and will collapse soon. There's an interesting "Bitcoin Obituary" by 99bitcoins.com that lists (in chronological order) how many times Bitcoin has been declared dead or will die a slow and gruesome death, not unlike the Blemflarck crisis in Rick and Morty.


To talk about this issue, let's find out what a "bubble" is and some case studies:

BUBBLE

According to Investopedia, a bubble is:

"A bubble is an economic cycle characterized by rapid escalation of asset prices followed by a contraction. It is created by a surge in asset prices unwarranted by the fundamentals of the asset and driven by exuberant market behavior. When no more investors are willing to buy at the elevated price, a massive selloff occurs, causing the bubble to deflate."
Basically, a bubble is when an asset is so overvalued and it will burst when people want to cash out, it triggered an avalanche of asset selling that eventually the price plummets to almost zero.

CASE STUDY 1

The first bubble that have been recorded is the "Tulip Mania" back in the 16th century in Holland. What happened was:

The tulip bulb trade started inadvertently when a botanist brought tulip bulbs from Constantinople and planted them for his own scientific research. Neighbors then stole the bulbs and began selling them. The wealthy began to collect some of the rarer varieties as a luxury good. As their demand increased, the prices of bulbs surged with rare varieties commanding astronomical prices.
Bulbs were traded for anything with a store of value, including homes and acreage. At its peak, Tulipomania had whipped up so much of a frenzy that fortunes were made overnight. The creation of a futures exchange, where tulips were bought and sold through contracts with no actual delivery, fueled the speculative pricing.
The bubble burst when a seller arranged a big purchase with a buyer, but the buyer failed to show. The realization set in that price increases were unsustainable. This created a panic that spiraled throughout Europe, driving the worth of any tulip bulb down to a tiny fraction of its recent price. Dutch authorities stepped in to calm the panic by allowing contract holders to be freed from their contracts for 10% of the contract value. In the end, fortunes were lost by noblemen and laymen alike.

What can be learned from this event is:

1) Fundamental value. Each asset must have a real fundamental value. A tulip; whilst beautiful; have no real usage and even deteriorate over time.
2) FOMO: Fear Of Missing Out. People keep buying tulip as a "luxury goods" that have no real fundamental value.
3) Futures market and speculative pricing is a bitch. Be really careful and know what you're doing when investing in a future product. This includes oil, palm, rubber, real estate, etc.

CASE STUDY 2

The housing bubble that crippled some of the big banks in the United States in 2008. As written in thinkadvisor.com(key points bolded are from me):

The current financial crisis began when the American Dream of home ownership was planted in the hearts of millions, even those that didn't have the fiscal discipline or financial wherewithal to be homeowners.
To combat the economic aftermath of 9/11, the Federal Reserve lowered interest rates. Greenspan & Company badly miscalculated by keeping rates too low for too long.
Home prices rose anyway, reinforcing the American dream.
Artificially low interest rates created cheap mortgages which created a speculative boom in real estate.
Aggressive mortgage brokers capitalized by selling doomed subprime mortgages to unqualified borrowers. Despite a fundamental breakdown of lending standards and regulatory oversight, no one noticed. Banks and brokers collected billions.
Home prices continued to soar. From 2002 to 2007, the S&P/Case-Shiller Home Price Index of 20 metropolitan cities outpaced the stock market and jumped by 75 percent. Inflated real estate lured consumers, homebuilders and investors into a false sense of financial security.
Consumers continued to binge on real estate and home equity lines.
Instead of holding mortgages and collecting the interest on the yield from the loans, lenders pooled them into securitized burritos. The packaged loans were sold with the help of inflated ratings from Standard & Poor's and Moody's.
Wall Street's underwriters collected billions in fees and rating agencies got their cut. Earnings and profits rose, but executive compensation rose faster.
Investment banks like Bear Stearns and Lehman Brothers devolved into highly leveraged hedge fund-like entities. Both firms collapsed.
Mutual funds, hedge funds and other financial institutions that overdosed on risky mortgage securities that were once deemed safe began to falter.
Ailing mortgage borrowers set off a tsunami of defaults, which triggered a collapse in the value of mortgage-backed securities. Financial institutions were forced to write down more than $1 trillion from their balance sheets.
Other opaque financial instruments like auction rate securities and credit default swaps blew up.
Home foreclosures accelerated, which created a glut of residential real estate pushing home prices down further.
Financial giants like AIG, Freddie Mac, Fannie Mae, IndyMac, Lehman Brothers and Washington Mutual collapsed. Others followed.
The financial chaos even took its toll on "safe" investments. The $62 billion Reserve Primary Fund, one of the first money market mutual funds, was forced to liquidate.
Consumers panicked by withdrawing money from their banks and mutual funds.
FDIC increased its coverage of bank deposits to $250,000 per customer from $100,000 through 2009.
Home prices crumbled even further.
Financial stocks within the S&P 500 fell by 45 percent.
The meltdown demolished America's retirement plans by some $2 trillion.
Banks and financial institutions fought for survival by hoarding cash. They stopped lending to their clients and to each other.
Without cash flowing from banks, businesses couldn't carry on their daily operations, much less grow.
Liquidity in the financial system dried up.
Congress passed into law a $700 billion rescue package to help the U.S. financial industry.
The stock market fell and the future prospects of earnings and GDP growth dimmed.
Basically what happened was (economists help me out please):

1) Housing loan interest rates are kept low.
2) House prices rose.
3) People with no fiscal ability keep on buying houses with the intention of selling it at a better price in the future.
4) The loans were then packaged and sold to other banks.
5) When the house prices are too high, people stop buying houses and the borrowers starts to default.
6) Home foreclosures starts to drive residential home prices down.
7) Home owners are stuck with houses that are valued much less that what they had paid for.
8) Banks no longer have lending power, stopping businesses and dried up the liquidity of the financial system.

CASE STUDY 3

The Dotcom bubble that happened in early 21st century, fueled by the internet-based company. As written in Investopedia (bolded by me):

The 1990s was a period of rapid technological advancement in many areas, but it was the commercialization of the Internet that led to the greatest expansion of capital growth the country had ever seen. Although high-tech standard bearers, such as Intel, Cisco, and Oracle were driving the organic growth in the technology sector, it was the upstart dotcom companies that fueled the stock market surge that began in 1995.
The bubble that formed over the next five years was fed by cheap money, easy capital, market overconfidence and pure speculation. Venture capitalists anxious to find the next big score freely invested in any company with a “.com” after its name. Valuations were based on earnings and profits that would not occur for several years if the business model actually worked, and investors were all too willing to overlook traditional fundamentals. Companies that had yet to generate revenue, profits and, in some cases, a finished product, went to market with initial public offerings that saw their stock prices triple and quadruple in one day, creating a feeding frenzy for investors.
The NASDAQ index peaked on March 10, 2000, at 5048, nearly double over the prior year. Right at the market’s peak, several of the leading high-tech companies, such as Dell and Cisco placed huge sell orders on their stocks, sparking panic selling among investors. Within a few weeks, the stock market lost 10% of its value. As investment capital began to dry up, so did the life blood of cash-strapped dotcom companies. Dotcom companies that had reached market capitalization in the hundreds of millions of dollars became worthless within a matter of months. By the end of 2001, a majority of publicly traded dotcom companies folded, and trillions of dollars of investment capital evaporated.
Those are extreme examples of bubbles that affected millions of people. All are caused by buying craze that are not backed with proper fundamentals, and ended with people losing money without any asset to hold on to.

WHY BITCOIN IS NOT A BUBBLE

*disclaimer: this is just my opinion based on my own research. Things may go south tomorrow, or go the moon...

Even though Bitcoin has risen up to USD17k in the last few weeks, I still think it will rise in value in the future. Here's why:

1) Fundamental value.

Unlike tulip, artificially inflated house prices or bogus bubble companies, bitcoin is generated through it's blockchain, which is mined by physical machines which incur electricity and maintenance costs. Like gold, there will be a minimum value of bitcoin. A rough calculation indicates that:

     (the most efficient rig) Antminer S9 (14Th/s)    = RM10,000
     (electricity cost)  1.5kW/hr * 24hrs * 30days * RM0.507 (kW/hr Malaysia commercial rate)
                                                                                 = RM 550 per month
                              Internet and other maintenance = RM200 per month
                                         Reasonable ROI for rigs = 10 months,
                         Cost for rig running for 10 months = RM10,000 + RM550 (*10months) + RM200 (*10months)
                                                                                = RM17,500
So, per month of mining, the mining profit should not be less than RM1,750.

       (Current mining rate) 0.0022btc/day * 30days = 0.066btc/month
                                      Current bitcoin price, 1btc = RM70,000
                                                                  0.066btc = RM4,500 (still profitable)

Based on that calculation alone, regardless of the amount of bitcoins mined, as long as the profit exceeds RM1,750 per month, people will still mine bitcoins. Miners surely wouldn't sell bitcoins any less than that amount due to the cost of producing bitcoins. This is the fundamental value of bitcoins, the bottom price that miners will set. As the difficulty rises, less bitcoins can be mined per rig, but the production costs will stay the same. Hence, bitcoin price will only get higher in the future.

However, based on production cost, bitcoins should have a price of:
                                                                 0.066btc = RM1,750
                                                                        1btc = RM27,000

Why then, bitcoins are valued much more right now?

2) Supply and demand

Bitcoins have 21 million cap limit. Currently in circulation are around 16.7 million bitcoins. As more people are getting aware of the existence of bitcoins, more and more people are eager to get a hold of bitcoins, driving the price higher and higher. Do note that every time bitcoins reach new ridiculous price, there will always be a correction pricing that brings bitcoin value down a bit (and bitcoin death prophets to emerge).

3) Bitcoin evolution

Once upon a time, gold is processed into coins and traded for goods. As gold prices soar, they are no longer being used for daily trading, but used for asset keeping. This is because the price of gold is so high and keeping it as physical coins are very risky.
Bitcoin will eventually become virtual gold. As the transaction fees are getting higher and the price soared, it will eventually be used as asset keeping only. The value of other cryptocurrencies will be pegged to bitcoin. It will be no surprise if in the future, other coins (XMR, DASH, ZEC, ZEN, etc.) will be used for daily trading and bitcoins are kept safe in cold wallets and used for exchange/ high volume trading purposes only.

4) Blockchain technology

Blockchain technology is the future of digital asset. Regardless of bitcoin future, blockchain tech will be the standard of delivering and receiving digital goods. This alone will guarantee the future of bitcoin; the first commercial implementation of the blockchain technology. Without a doubt, there will be a disruptive technology superior to blockchain in the future, especially after quantum computing becomes commercially viable. But it is not this day.

CAUTION

A bit of caution though. Bitcoin futures market is opening. This may make bitcoin a true "bubble". Try and read this article. As bitcoin have virtually no upward price limit, thing may break loose next year as greedy whales play with bitcoin prices until the bubble bursts. Bitcoin prices can soar past 100 times it's cost value due to speculation alone and could collapse when people sell their bitcoin in a selling frenzy; not unlike the tulip bubble earlier.

Bitcoin is a decentralized currency. It exists and held value because of the users using it. No government is regulating the price and production. The price is still based on the users ability to buy and sell bitcoin freely.

I still trust my fellow human beings to make the right decision in bitcoin prices. That is; stop buying bitcoins when the price is too high.

Tuesday, October 24, 2017

Crypto mining and the elusive ROI

For those who have started mining or those who want to start mining, the usual question is "what is the ROI?". Now, for the uninitiated, ROI stands for "Return On Investment", meaning the amount of time needed to get back the initial capital spent.

Frankly speaking, for me ROI is an elusive term. Cryptos are governed by several things, namely the hashrate, difficulty, and price volatility. Your coin shares received per day are determined as per below:

Hashrate:- the more mining power added to the network, the less shares you get.
Difficulty:- the higher the difficulty level, the less shares you get.

The more people start to mine that particular coin, the lesser coin you get per day. Considering that the electricity bill and the coin value is the same, you would actually lose your profit. However, if you somehow found the "gold mine" before other people, you could actually make much more profit than just holding coins.

Story 1: when I started mining Ethereum back in April 2016, with my RM30,000 rig (around 8,000USD), until December 2016, I got around 233ETH. Using the ROI calculation back in April-May '16, with 1ETH = USD10-15, it would take a year or two for me achieve my ROI. However, during the price jump on May '17, where 1ETH = USD400, I got my ROI and got much more in profit. Nowadays, using the same rig, I can get only around 0.4ETH monthly.

Story 2: in December '16 I started mining ZEC using the same rig. Managed to get around 30ZEC up till April '17. Lo and behold, ZEC price jumped up to USD400 in May '17, earning me much profit for my mining. Nowadays, I can only mine around 1-2ZEC per month compared to 1 ZEC per day due to the difficulty spike.

ZEC difficulty over a 3 month period

Story 3: In July '17, I started noticing a new coin, ZENcash, which is a derivative of ZEC but with better technology implementation. Using my rigs, I switched to ZEN and managed to mine around 750ZEN (got around 16ZEN per day) till early Oct '17. The price then was around 5USD per ZEN. Later this month, ZEN managed to climb up to 25USD per ZEN, earning me around 18,000USD (RM75,000). However I'll continue to hold on to this coin because I think it'll be much worth than 25USD. The difficulty has also risen exponentially due to miners switching from ZEC to ZEN (refer to ZEC difficulty chart above) and also Ethereum miners switching algorithms to ZEN. Nowadays I can only get around 2ZEN per day...

ZEN difficulty spike
So there. Over 1 year I managed to get not only my ROI, but extra profit from mining those coins. The bad thing is much of my coins I had sold to pay my bills, or I would've gained much more. :(

The whole point of this post is that if you want to be a coin miner, don't get stuck in "ROI" or fast cash thinking. Mining is work. Mining is a means to say that you support the whole cryptocurrency ecosystem. If it's profitable, good for you. If it ain't, and you want to purely make profit from cryptocurrencies, it's better that you buy some bitcoins (BTC) and hold them. Mind you, 8,000USD (RM30,000) in bitcoins will earn me around 17BTCs (1BTC = 450USD in April '16), which will actually get me 93,500USD (around RM400,000) in BTC value nowadays. Duduk rumah goyang kaki can get you more than 1000% ROI.

If you are a computer person, like to handle computer parts, fiddling with softwares, supporting the whole crypto ecosystem, then mining is for you. Don't give up. Find the coin that you have confidence in. Mine the shit out of that coin before other people realize them. Support the system. Just remember: MINING IS WORK.

*hint: Bitcoin Gold can be mined using Equihash algorithm (ZEC, ZCL, HUSH, ZEN...), will it be the next goldmine?*

Thursday, July 6, 2017

Menggunakan exchanger Bittrex.

Yo yo yo.

Recently ramai orang nak tahu macam mana nak mine altcoin (particularly ZEC) dan tak mau bergantung kepada Nicehash. Sini bagi aku ajar sikit camne nak main-main dengan exchange nih. Personally aku guna Bittrex, tapi exchanger lain sepatutnya lebih kurang je modus operandi dia...

Tak tau apa tu aku tengah sembang apa? Baca post sebelum ni:
Part 1
Part 2
Part 3
Part 4

Ok let's get straight to the topic. First of all, silalah sign up dengan Bittrex ni. Make sure register masuk guna email yang lazim digunakan sebab setiap transaction keluar dia akan mintak confirmation thru email kita kelak.

SIGNUP

Dah?

Senang je nak signup. Guna email address je.

WALLET CREATION

Ok, next step, klik kat "wallet" tab atas tu, untuk checkout coins apa yang Bittrex support. Note that tak semua exchanger support semua coins. Contoh macam Hush yang masih belum disenaraikan di Bittrex, tapi ada di Cryptopia exchanger, dan ZEN yang ada di Bittrex, tapi tiada di exchanger lain. So kalau nak mining any altcoins, make sure exchanger yang dipilih ada senaraikan coin tersebut. Sebab by the end of the day, highly likely kita akan convert all the altcoins kepada bitcoin untuk liquidity.


Wallet tab.

Dah klik kat situ, akan keluar interface wallet list yang kite ada. Kalau tak nampak coin yang kita nak, just "search" je kat tepi tu.

Wallet interface

Kat setiap coin tu, belah kiri dia ada simbol "+" dengan "-", untuk tambah coins tu dari sumber luar, atau withdraw coins tu ke wallet luar. Kalau tak pernah ada wallet coin berkenaan, tekan "+" akan keluar interface wallet generation seperti di bawah:

Wallet creation.

Klik je kat new address tu, pastu korang akan ada satu wallet address khas untuk coin tersebut; contoh di atas untuk Litecoin (LTC). Setiap kali korang nak buat transaction, dia akan guna wallet address tersebut. Make sure tulis address yang dijana dan simpan di tempat yang selamat agar senang nak berurusan kelak.

HINT: jangan lupa generate wallet bitcoin juga ye. Sebab endgame kite nak ada bitcoin. Tak sure pulak apa jadi kalau convert ke bitcoin tanpa generate wallet dulu, tapi lebih baik elak untuk selamat.

DEPOSIT COIN

Dah ready wallet, boleh la masukkan coin pulak dalam wallet tu. Sama jugak macam proses tadi, tekan "+", tapi kali ni dah ada wallet address. Copy paste (atau salin balik secara manual) wallet address tu dan bagi kat orang yang nak bagi coin tersebut.

Wallet address tu juga boleh guna sebagai mining address kepada mereka yang nak melombong coin tersebut.

HINT: Make sure baca dulu mining pool yang berkaitan, adakah mereka menerima exchanger address sebagai mining wallet  address. Kalau tak, kena generate wallet kat tempat lain untuk dijadikan mining wallet address.

WITHDRAW COIN

Kalau nak keluarkan coin tu kepada wallet lain pulak, tekan "-", pastu isi details yang dia suruh:

Withdrawal interface.

Masukkan address yang nak dikeluarkan tu, masuk kuantiti yang nak transfer, pastu klik "Withdrawal". "Tx Fee" tu akan dikira secara default, so tak perlu risau.

HINT: Kalau ada 1btc dalam wallet, tak boleh withdraw tepat 1btc jugak sebab ada transaction fee. Selalunya minimal je, tapi kalau nak transfer tepat-tepat, boleh jadi sangkut. Rule of thumb dia adalah kalau nak transfer any amount, make sure dalam wallet ada lebih 1/100 dari jumlah yang nak transfer tu. Kalau nak transfer 1btc, make sure ada at least 1.01btc dalam wallet. Camtulah. Tapi kalau amount besar (100btc+), tx fee takdela sampai 1btc... haha.

Bila semua dah dibuat, akan ada "pending transaction" kat bawah tu menunjukkan berapa confirmations yang perlu sebelum transaksi itu di verified. Tak perlu risau. Dalam 10-30 minit akan settle kecuali ada masalah besar kat network.

Pending transaction
Oh. Kalau nak withdraw, jangan lupa check email anda. Bittrex akan hantar email confirmation pasal transaction ni, kena klik "authorize transaction" kat email, baru dia lepaskan. Security feature.


EXCHANGING

Dah test urusan keluar masuk duit? Now kita pergi ke good stuff. Jual beli matawang.

Kalau kat bittrex ni, market dia hanya *pairing bitcoin/altcoin sahaja. Pairing altcoin/altcoin takde. So kalau nak beli ZEC menggunakan ETH yang kita ada, kena convert ETH tu ke BTC dulu sebelum beli ZEC.


Klik kat "symbol" coin yang nak di trade, nanti masuk ke trading page seperti di bawah:


Order filled.

Aku cuba terangkan satu-satu:

Graf:- menunjukkan perubahan harga dari masa ke semasa. Bawah graf tu menunjukkan masa, belah kanan tu menunjukkan harga satu altcoin tersebut dengan harga btc. Contohnya kalau dia tulis 0.111, bermaksud 1ZEC boleh membeli 0.111BTC. Kalau nak tahu 1BTC boleh beli berapa ZEC, kira 1 bahagi dengan nilai 0.111 tu (1BTC = 9.009ZEC), simple math. 

Perubahan exchange rate ini reflected dari perubahan harga pasaran BTC dan ZEC. Semasa article ni ditulis, 1BTC = 2600USD dan 1ZEC = 288.6USD. (288.6/2600 = 0.111)

Senario lazim adalah kalau ZEC naik harga dan/atau BTC turun harga, exchange rate akan meningkat. Sebaliknya juga adalah benar iaitu kalau ZEC turun harga dan/atau BTC naik harga, exchange rate akan menurun.

Kalau nak main exchange, boleh la study graf tu dan set harga mana kita nak jual/beli altcoin. Sebagai contoh, tengok graf tu balik, tengok rate paling rendah dan rate paling tinggi. Dalam kes ini, paling rendah di sekitar 0.106 dan paling tinggi di 0.112. So kalau nak try main exchange, boleh set jual ZEC ke BTC di nilai tinggi (0.112) dan beli ZEC di nilai rendah (0.106).
Over time, graf akan berubah secara total. Checkout graf di bawah sejak awal Jun 2017 sehingga sekarang:

June - July 2017

Pada pertengahan Jun haritu adalah waktu yang baik untuk convert ZEC ke BTC sebab rate tinggi.

Kalau nak mendalami ilmu timing untuk buy and sell ni boleh la belajar dari sifu-sifu forex. Ok?

Trading:- Bawah graf tu pulak ada ruang trading. Buy Zcash dengan Sell Zcash. Kat situ kita boleh set berapa amount yang nak trade. Buy ZCash bermaksud menukar/membeli ZEC menggunakan berapa amount BTC yang kita ada. Sell pula menukar/membeli BTC menggunakan berapa amount ZEC yang kita ada. Price pula adalah exchange rate yang kita nak buy/sell, boleh masukkan secara manual atau klik untuk letak price "Last", "Bid" atau "Ask" untuk tengok price default yang sedia ada dalam sistem. Contoh kat graf atas tadi, aku dah jual 0.5ZEC dengan rate 0.11079999 per unit (guna rate "Last") dan tunggu order tadi dibeli orang lain. Bila order dah filled (maksudnya ada orang bersetuju dengan rate yang aku letak tadi), maka aku pun dapatlah BTC sewajarnya, iaitu 0.51265*0.11079=0.0567BTC.

HINT: kalau letak rate rendah, order akan cepat fill, tapi rugi la sikit. Kalau letak rate tinggi, lambat sikit nak fill. Tapi kalau konfiden rate akan naik tinggi, takde masalah. Just tunggu harga semasa sampai ke rate yang kita nak, dan order akan automatically fill. Takyah ngadap pc sampai selesai. Dah letak order, boleh pergi tido.

Contoh kat atas untuk ZEC, tapi hampir semua coin macam tu. Monero (XMR) lain sikit, ada Payment Id dengan Base Address, tapi basically SOP sama je dengan yang lain.

Ok? Dah semangat nak trading? Good luck, have fun! Any persoalan boleh letak kat komen section atau masuk group https://www.facebook.com/groups/BitcoinMalaysia/

Satu je nak pesan, kalau boleh, trade dengan duit/coin yang boleh let go sahaja. Sebab risiko sama macam forex. High risk, high gain, high loss. Jangan guna duit bayar sewa rumah pulak. Kang kalau jadi apa-apa, satu hal pulak nak cekau duit sewa. Huhu.

Wednesday, June 14, 2017

Perbandingan GPU untuk melombong cryptocurrency.

Post seterusnya berkenaan melombong cryptocurrency ini adalah berkisar akan perbandingan GPU yang sesuai untuk mining, terutamanya alt-coins. Bitcoin dah tak kira sebab dengan adanya ASIC, GPU dah tak efficient untuk dilombong.

Kalau anda terlepas, ni post sebelum:
Part1: Bitcoin sebagai matawang.
Part2: Cara-cara mendapatkan bitcoin.
Part3: Melombong ZCash

So apa parameters yang kita tengok sebelum mulakan mining? First of all, kita tengok komponen yang ada untuk satu rig:

Contoh 1 rig. Credit to here
1) Rig chassis: memandangkan mining ni masih dalam kategori niche market, tiada lagi supplier commercial yang jual chassis yang standard. Biasanya miners akan build sendiri dari kayu, besi atau aluminium. Price range varies, tapi kalau nak beli dari orang, expect around RM500.

2) PCIE Risers: ada sesetengah kedai di lowyat dah start jual risers ni. Kalau nak beli banyak, order online pun bagus jugak. Harga around RM30-RM50 satu riser. So for benchmarking purposes, kita letak RM400 (Rm50x8) per rig. Letak lebih 2 unit sebab riser faulty merupakan masalah yang tak best, dan kalau takde yang faulty boleh pakai lagi 2 riser tu untuk next rig.

3) Motherboard: sekarang ni motherboard yang popular adalah Biostar Pro-BTC TB250 atau ASUS Prime Z270-P. Both model Intel socket 1151 sebab boleh dapat CPU murah lagi kat situ. Harga pulak between RM500-RM700. For benchmarking purposes, kita letak RM800 agar ada lebih margin.
Motherboard Pro-BTC AMD socket AM4 pun ada jugak, tapi AMD Ryzen masih lagi agak mahal. Maybe Julai nanti dia keluar Ryzen versi marhaen boleh consider beli socket AM4.

4) PSU: part ni tricky sikit. Ada mazhab suruh beli seketul 1500w PSU. Ada mazhab ajak beli 2 ketul 1000w PSU pastu cantum. Mazhab aku pulak menyuruh mining guna GTX1060 dan pasang 2 bijik 650w PSU sahaja. Sebab apa? GTX1060 makan power kurang banding dengan card AMD. Dan juga price beza gile. 1500w boleh cecah RM2000, manakala 650w boleh dapat around RM400 seketul. Takpe pasal power nanti checkout nak beli card apa dan try cari appropriate price range.

5) CPU and RAM and HDD: for mining purposes, tak perlu la beli CPU dan RAM yang imba. Cukup syarat je. Kalau ikut atas, beli Intel Pentium (socket LGA1151) dan RAM cap ayam 4Gb dah cukup. CPU around RM250, RAM harga around RM150. HDD around RM200 boleh dapat dah. Total benchmarking price RM600 dah cukup kat sini.

So far kita dah listdown the backbone of the rig. Total cost RM2300 (tanpa PSU). Sekarang kita cerita GPU. For "simplicity" purpose, kita hanya akan sembang 8 jenis GPU yang paling top rated untuk mining, yang akan dikategorikan oleh power consumption, hashrate, harga kos dan mining efficiency. Mining pulak banding dengan 2 coins sahaja iaitu ETH dan ZEC sebab harga diorang tengah imba sekarang ni. Kalau nak tahu capabilities mining coin lain, boleh google. Jumlah VRAM (memory GPU tu) juga memainkan peranan penting juga, dan beza performance around 10% antara model yang sama. So kalau ada option nak beli, beli yang VRAM dia lebih tinggi, ok?

6) GPU listing

NAMA/CARD MODEL
Estimated price: harga yang dianggarkan pada Jun 2017. Mungkin naik atau turun di masa depan.
Power consumption: berapa banyak elektrik yang digunakan oleh GPU tersebut ketika mining.
Dagger hashimoto (ETH) hashrate: nilai kalau mining ETH. Lagi tinggi lagi bagus.
ETH mining efficiency: nilai hash per watt mining. Lagi tinggi lagi bagus.
Equihash (ZEC) hashrate: nilai kalau mining ZEC. Lagi tinggi lagi bagus.
ZEC mining efficiency: nilai hash per watt mining. Lagi tinggi lagi bagus.


GPU
AMD
NVIDIA
RX470
RX480
RX570
RX580 (4gb)
RX580 (8gb)
GTX1060 (6gb)
GTX1070
GTX1080
GTX1080Ti
Est. Price (RM)
1,200
1,600
1,200
1,400
1,600
1,300
2,000
3,000
3,500
Power Consumption
(watt)
120
160
160
150
190
100

150
180
250
DAG (ETH) hashrate
(Mh/s)
25
29
24
22
26
20
28
27
35
Eth mining efficiency
(Mh/watt)
0.208
0.181
0.150
0.146
0.137
0.200
0.186
0.15
0.14
Equihash (ZEC) hashrate
(Sol/s)
280
305
311
280
310
281
435
470
630
ZEC mining efficiency
(Sol/watt)
2.33
1.90
1.94
1.86
1.63
2.81
2.9
2.61
2.52


Seperti yang dijadualkan, warna hijau menggambarkan skor yang paling bagus dan biru merupakan skor yang kedua bagus untuk pemilihan GPU mining korang. How to make sense of this table? Simple.

First of all, kalau bajet korang paling penting, tengok top row. Estimated price. Tu harga satu GPU. Tambah dengan total cost kat atas tadi, boleh bajet dah satu rig korang harga berapa.

Contoh:
Nak buat rig RX570. Backbone + GPU est price + PSU = RM2300 + RM1200*6 + PSU = RM9,600 + PSU.

Now, kalau nak dapat mining dengan pantas, consider pilih GPU yang tinggi hashrate dia. Tapi kalau nak consider bill elektrik sekali, pilih GPU yang paling efficient. Part ni boleh debat lagi sebab considering harga coins duk naik, mungkin dapatkan coin dengan pantas adalah lebih berbaloi dari bayar bill elektrik rendah. Entah. Korang sila consider.

7) PSU choice

Ok now dah tau nak pakai GPU apa, skang masa untuk memilih PSU yang bersesuaian. Tengok balik power rating setiap card tadi, tambah 200w untuk backbone requirements, pastu bahagi 80, darab 100. Sebab apa? Max power requirement kita haruslah berada di sekitar 80% daripada output PSU kita. Nak run 24 jam sehari. kalau power usage tinggi sangat, PSU cepat burn. Juga consider kalau nak pasang 2 PSU, make sure agihan kuasa tu seimbang dengan PSU capabilities. Kalau imba nanti satu cepat burn.

PSU Power = (GPU wattage * cards total) + 200 [backbone requirement] * (100/80)

Contoh:
Nak pakai RX480 (160watt) rig. (160*6 + 200)*(100/80) = 1,450watt. So boleh la cari PSU 1500watt atau 750watt x2.

Harga PSU pulak berlainan ikut rating. Bronze, Silver, Gold dan Platinum rating. Ada jugak PSU yang takde rating. Kalau nak buat mining rig ni, consider la pakai PSU yang ada rating. At least bronze. Quality, money saving dan tahan lama sket.
Power Rating



Ok. Tu je kot. Ada pape soalan boleh roger kat ruangan komen, atau roger terus kat facebook group https://www.facebook.com/groups/BitcoinMalaysia/