X11 FPGA is here?: Betarigs Mining rig #1713 3x Phaethon PCI-E FPGA board

[true-asset]

Is this real?


https://www.betarigs.com/rig/1713

https://www.betarigs.com/list?miner_type=5&rig_size=4%2C5&rent_duration=3%2C6%2C12%2C24%2C48%2C72%2C168&support=0&quality=0

3x Phaethon PCI-E FPGA board   27 Mh/s   0.01512 BTC    0.0006 BTC   Rent now!

Mining rig #1713 3x Phaethon PCI-E FPGA board
Experimental 3x Phaeton PCI-E FPGA card worked at dedicated server

[anatolikostis]

Neat...but what is the price?

[goodluck0319]

where do I buy the board?

[Amph]

so 1 give you 9MH/s, dunno about the price, but i don't think it can beat a 750ti, with the upcoming boost(2.8MH/s)

[true-asset]

Where do you buy one? Cannot find and info on google.

[dspair]

Waiting for DRK bagholders saying all FPGAs are fake and chinese farmers with 2GH/s don't exist.

[djm34]

so 1 give you 9MH/s, dunno about the price, but i don't think it can beat a 750ti, with the upcoming boost(2.8MH/s)

considering price/performance (didn't find phaeton fpga, so based on pci-e fpga board) it is about 2x 780ti for somewhat less than the price of one (addind the cost of the code would certainly add to the price though... it won't be free to do shitcoins).
Not bad but not exceptional either (don't know if you can put more chip per board or anything like that)

edit: Now if we compare to the performance of the forthcoming gtx880 (5x more core than the 750ti and assuming linear scaling) a gtx880 would beat it with around 10~11mh/s for about the same price tag and a better resell value... higher power usage though)

edit2: actually if those three are in india, it might be those gtx880 engineering samples  

[djm34]

Waiting for DRK bagholders saying all FPGAs are fake and chinese farmers with 2GH/s don't exist.

That's still doesn't explain anything... I mean the high hash rate is certainly more related to the rig renting than the fpga by themselve

[majeis]

Where do you buy one? Cannot find and info on google.

You know how to use google?

Maybe you should google us the email addresses of the four corporate heads who back your project?

Or maybe google their websites since you only linked one of them?

What kind of multi-million dollar business doesn't have a website?




But about the FPGA, it's FUD.

[Amph]

Where do you buy one? Cannot find and info on google.

You know how to use google?

Maybe you should google us the email addresses of the four corporate heads who back your project?

Or maybe google their websites since you only linked one of them?

What kind of multi-million dollar business doesn't have a website?




But about the FPGA, it's FUD.

it's not fud, it's crap, it don't bring any advantage compared to a 750ti

[goodluck0319]

I need to buy at least 3 of this fpga board in my rig if it is avialble.

but there is no where to be found.

[smolen]

Numbers out of thin air for FPGA route  - I'd expect 300 MH/s X11 algo board with 6 Spartan 6 LX150 chips to be developed in half of year with 50000 USD development cost, production price would be something like 2000 USD per board. (Anyone with actual field knowledge is welcome to correct me!)

[djm34]

Numbers out of thin air for FPGA route  - I'd expect 300 MH/s X11 algo board with 6 Spartan 6 LX150 chips to be developed in half of year with 50000 USD development cost, production price would be something like 2000 USD per board. (Anyone with actual field knowledge is welcome to correct me!)

would it be possible in one or two month time ? (the X11 didn't really start before that)

[smolen]

would it be possible in one or two month time ? (the X11 didn't really start before that)

That's just my wild guess, I'm not an expert.
If someone already has multi-FPGA board and this board can pump data with 512 bits * 300 MHz rate between chained FPGAs, then development time would be weeks, not months. I expect unrolled implementations of most hash functions to produce one hash value per cycle, and LX150 can run at 300 MHz doing crypto tasks despite high toggle rate, as been proved by Bitfury. Another question - is such data rate between FPGAs possible at all? Well, data transfer doesn't need to be 100% reliable, implementing some error correction code shouldn't take too much LUTs, hashing engine and i/o can sit on different clocks, looks like there are enough diffio pairs in top LX150 variants. Again - anyone with actual field knowledge is welcome to correct me!

[SpeedDemon13]

9 mh/s per board, still looks like they are still in the development process. But that's just a little more efficient than a 750ti, I would have expected in the ball park of 20~30 mh/s per board. No matter what, this is signalling that within a year to year and a half, that an ASIC more likely will surface. This is just my assumption.

[Amph]

one year is plenty of time(even 6 months are enough) to move on another algo, and repeat this process for every asic that will come

[Burninj]

You won't need to wait a year to see x11 asics.

[sartech]

Waiting for DRK bagholders saying all FPGAs are fake and chinese farmers with 2GH/s don't exist.

they know already...  check darkcointalk forum.

[therealbigcoin]

Is there any source for anything? Who is developing this?

[kingscrown]

I heard long ago the x11 FPGAs exist in China. This just shows it was propably true