I am saying it makes no financial sense. The major GPU coins already have asics, with ETH being the biggest. And those coins that have no asic on it will have asics on it before the FPGA will come close to ever breaking even.
And everytime an asic takes over an alternate coin, all the GPUs and FPGAs that were mining that coin will go to a non-asic coin. That means the remaining non-asic coins will have difficulty skyrocket and guess wat, that means your GPUs and FPGAs mining make less money. Everytime a new asic comes out, the FPGAs make less money. So, there is no niche. There is a limited amount to be earned from mining u see. If the asics keep appearing to take some of it, the switchers (FPGAs/GPUs) will have less and less.
Why not u calculate how much it would cost to make an FPGA that can do the equivalent of mining 504mh like the L3+ or 14TH like the S9.
Run your numbers and see the gap. While I dont have the numbers, I dont think it can work.
Anyways, good luck in your endeavors. Maybe I am wrong.
Just my 2 cents
While your point of view is true in general, there are a couple of exceptions: mainly space and power constraints. Lets be very specific in our comparison to avoid meaningless comparisons, XCVU9P ($4,000 FPGA) vs GTX 1080 Ti ($800 GPU).
Now lets take the Phi1612 algorithm and give the FPGA a more realistic 5x performance advantage against the GPU at a power consumption rate of 150 watts (0.150 kWh). I would rather run a 100 FPGA farm (17.5 kWh) vs a 500 GPU farm (87.5 kWh) any day and heres why. Lower overall costs if you believe the 5x advantage ascribed to the FPGA.
4x FPGAs and components approx. ($16K + $1.2K) in server chassis and add 100 watts for overhead.
4x GPUs and components approx. ($3.2K + $0.65K) in frame and add 100 watts for overhead.
$430K 100 FPGA farm vs $481.25K 500 GPU farm. Can you at least agree that this makes financial sense?
No that makes no sense. 481.25k/430k = 12% better only LOL.
---------------------
Rough Comparison:
----------------------
And while the E3 does 180mh for 800watts which is comparable to some GPUs, bitmain can sell it for as little as 800 USD if they want to. Thats the price they sold on first batch. To do 180mh, u will need to spend about 2500 on a rig. The asic can be 3 times cheaper
For 300watts, u can get 10,000 sol on asic or for gpu, u get say 1,100 sol? Thats 9 times more efficient and the asic is also far cheaper on a per sol basis.
If u consider the gap between using gpu to mine bitcoin/litecoin, vs asics, the gap is even bigger.
Doing it 12% beter on FPGA isnt a game-changer at all.
---------------------
Why FPGA cant win
-----------------------
There is a limited amount of money that can be made from mining.
Lets say, for illustration, there are 100k of coins released for miners to earn every 10 mins
In the past, 10% of all coins are mined by ASICS and 90% by GPUs or FPGAs (aka switchers)
Time passes and asics mine 30% and then 50% and then 70% of all total mined coins.
Thus, as time pass, the switchers (GPUs and FPGAs), now mine only the remaining 70% and then 50% and now 30% of coins.
Hence, as time passes, there will be more and more switchers sharing a smaller and smaller percentage of the mining PIE. The switchers mining profits will just decrease and decrease as the total % of all coins they mine decrease
And if a new major coin appears, the FPGA wont mine it long before an asic takes over. The remaining switchers will be sharing a smaller and smaller pool of coins. Thus, they will earn less and less as the remaining pool of coins have their difficulties skyrocket
Thus, u gotta be abit mad to think u should invest in more expensive equipment just for that.
--------------------
Speed of Asics being developed
---------------------
I mean just look at this year. It is only early May and we have a decred, sia, eth, cryptonight and now zcash asic. Thats 5 major algo and at least 5 major coins being taken down by asics. It is like one new algo takes down one major coin a month. Furthermore there might be other asics algo already made, just not released yet. And lets face it, a new major alt coin appears, it will be taken down by asics soon.
-------------------
Difficulty wave coming
-------------------
Ok, so in June 2018, the E3s will be shipped. ETH difficulty will skyrocket and alot of GPUs will move from ETH to other coins. Expect other coins difficulty to increase. The good news is that E3 price was increased from 800 to 2,100 usd. This means slighty lesser E3 purchased and lesser difficulty increase compared to if it were 800 USD. Do note that E3 price may go down again in future down which leads to more E3 purchases and higher difficulty. The antminer L3+ went down in price and wat occured was a massive litecoin difficulty increase. If the E3 goes down in price, it is gonna take down GPU/FPGA mining even more. Remember, it can be priced 3 times lower than a gpu rig and a far stronger E3 may appear as well. U are pretty much at the whim of bitmain here. ETH is actually the major gpu coin. E3 price down to 800 USD and goodbye to all gpu because all gpu will move to other coins and all other coins difficulty will skyrocket. This will probably occur eventually.
The Z9 mini will take down zec/btc-gold gpu mining, period. SIA and Decred are asic owned.
Thus, I just dont see how it will work. By right, it was over for gpu when E3 appeared since eth is the major gpu coin. Does anyone know how much % of GPU mining is ETH. Is it over 50%?
Asics have clearly won the war. They may very well push GPUs to have 5-10% of total coins being mined in future.
I believe u are letting your feelings cloud your judgement. Anyways, good luck to u buddy.
Just my 2 cents
Your 2 cents are quite interesting but miss the point.
Basically, choosing hashing algorithm for which developping ASIC cost a lot and therefore renders mass production unattractive.
This was the main point of Cryptonight in 2013/2014, choosing high memory intensive algorithm, to put a physical economical barrier to ASIC developpment (back in the time memory integration on ASIC was very expensive).
Given technological progress, in 2017, it seems memory unit are more affordable for ASIC, and we see ASIC of Cryptonight publicly released in ferbruary 2018, and we suspect them to be already mining since july 2017.
The major pitfall of ASIC is staticness. Which means that the work it is designed for cannot be updated or changed after. So the new strategy for beeing asic resistant, if to often change POW algorithm. It's a cat and mouse game.
So when coin dev choose to change their pow algorithm, like Monero dev did, those expensive and efficient piece of hardware are becoming useless except for blocking door (or targeting minor CN coins for less profit).
Some other coin dev choose to not change their pow like BTC did not after ASIC appearance. Like zcash seems to choose also.
In that case, it is clearly obvious that ASIC will soonner or later dominate the mining of those coins, any other kind of hardware are far less efficient and more energy consumming.
So where is the market for FPGA mining: Basicaly all coins which developpers want to keep ASIC mining away. Most of cryptonight variant for instance, Even if monero change pow every 6 months, every 6 months developpers can release updated fpga file to adapt to new pow.
Also all coin which are not attractive enough to invest huge money into ASIC developpment based on exotic POW which can fit nicely in FPGA. Bringing better h/w ratio than gpu/cpu mining. Of course when an ASIC is existing for a coin, and developper dont want to hardfork POW, there is no more room for cpu/gpu/fpga mining.