Topic: Swedish ASIC miner company kncminer.com - page 2069. (Read 3049521 times)

I hope you realize that is very much easier, cheaper and faster to develop an FPGA miner that an ASIC miner.

yeah like, What the crap Spidey Senses

Prolly didn't explain myself very well.   At the point that they have a working FPGA (which seems like it's imminent), I feel like saying to them, here's my FPGA money, only don't build the FPGA, just build the damn ASIC, and give me a juicy discount as an early adopter. Does that make sense?

Prolly didn't explain myself very well.   At the point that they have a working FPGA (which seems like it's imminent), I feel like saying to them, here's my FPGA money, only don't build the FPGA, just build the damn ASIC, and give me a juicy discount as an early adopter. Does that make sense?

Perhaps we are in furious agreement?  

I understand the plan, I'm just questioning whether it's going to bite them having to divert effort doing the FPGA run to raise the funds to do the ASIC run. I could be completely wrong, I hope I am.

Dude, you had to post there first didn't you? A full 67 seconds after the OP! You have antennas or sth?

Dude, you had to post there first didn't you? A full 67 seconds after the OP! You have antennas or sth?

Dude, I don't get your point, that's exactly what they are doing. Proving the design with a small FPGA run based on investors helping out to validate the design. Then moving straight on to producing ASICs. They haven't had VC, they are crowdsourcing from the community, same as all the other companies.

Thank you for that, I already understood the difference between the two and how they relate. My point remains - they are still going to end up doing 2 production runs of similar, yes, but still different products. My original point was that if they had investors helping out they could have finished producing the FPGA in small quantities to validate the design, then move straight on to producing ASICs. The path they've chosen is going to cost them time. I presume they've chosen it because they have to, if they had VC offered to them they would have been mad not to take it.

No, an FPGA naturally prequels the ASiC.

With FPGAs, engineers do not customize anything on the chips physically. They simply use software to program the FPGAs to interconnect their elements as desired.

With ASICs, engineers customize only the last few mask layers. These layers interconnect elements ASIC manufacturers predefine in standard base wafers.

The FPGA is suited to small batch prototype and trial run of sub 10,000 chips. FPGAs are expensive, but they allow you to refine your design prior to committing it to ASIC.

Yeah, a lot closer together makes my point. They are going to do 2 productions runs of different products close together.

Possibly the unannounced Avalon Gen 2?

Well, let me put it this way: if you asked me to discuss the finances of my company, I would not answer you. My business is my business.

I cannot fault them for not willing to disclose the way they run their business, but given the few erroneous statements they made, their vague answers or dismissal of the questions, one might want more transparency in the process. They are avoiding the questions probably because they do not want to show how thin their company is. I'm sure they would have no problem disclosing the numbers, or some of them, if they had lots of money in the bank. Nothing new here. I just wish they'd come out and say it: "we're broke, but, pretty please, send us your money so we can finish this project". I value honesty a lot, alas it is not very common in the business world.

There is really nothing new going on, I'm waiting for the "open house" visit report but I wish they would disclose advancement a little bit. Also, I'd be wary about the lead times from the wafer factory. They haven't asked for money yet, meaning they are not about to order any wafers or FPGA's, so the late September/October date for the finished products seem to be slipping already.

I don't know their build capacity, but given the low numbers of the first batch, I doubt they can do both products shotgun style. It will probably be one after the other (again, speculating here). Avalon has a lot more money and they're still not shipping hundreds of products per month, because it's hard to setup a production line and size it properly to your market (and now they just won't do it, only sell chips).

The FPGA is programable using software, the ASIC isn't. You use the FPGA to develop and refine a small batch and then copy the final design over to an ASIC. The ASIC is a semi-customised chip.

BFL, and Avalon, both had FPGAs prior, this is just a lot closer together.

Price too high?

350gh/s for $7,000 is $20 x GHs. Please let me know ASAP about a single ASIC product with a better cost per GHs.

I think the project is quite intriguing, but the price might just be too high.

Even in the very unlikely event that knc can ship a 350 gh/s device with a power consumption of 1-2 kw, difficulty will explode in late summer when all the Avalon-chip based devices come online.

Of course nobody can predict where the btc price will be at that point, but if it stays around 120$, I don’t think it'll be trivial to make that money back...

I am interested in this company, but it seems that a model of being vague with customers when you have no assets and no product is not a winning strategy. They would have been better off seeking early-stage investment to get the ASIC to beta-stage rather than luring customers in with an FGPA miner, which will divert their efforts away from getting the ASIC right. Having said that, I did put my name down for a Mars the other day and if they do get that working to spec I'll buy it. But I fear it's costing them time.