BlockBurner LLC - Crucible FPGA Scrypt Miner - Announcement Aug-19

[Operatr]

Hello, my developing crypto-currency business, BlockBurner (mostly a blog atm, but more to come), is researching feasibility in the generation of a real FPGA Scrypt Litecoin mining appliance, specialized for the task much like the SHA256 FPGA's and ASIC's Bitcoin is about to begin running on.

To do this, there are a few things I need to know from the crypto-currency community, please answer each question if you respond to this post-

1. Do you think the market and community is ready for FPGA Litecoin?

2. Is there definite interest in FPGA Litecoin machines? Would you buy one if the price was reasonable? What is reasonable?

3. Would you pre-order one to support first round funding for prototyping and first wave production? (I'm no one but a lowly computer specialist and budding entrepreneur burned by Big Banking already, so this would have to be crowd funded to get started without a loan shark involved. You would be relying on me to deliver the goods and not squander the investment with idiotic decisions, even still you would stand to lose your pre-order through economic issues and unseen factors, so there is a risk to early adopters as with all things, though Im thinking of other incentives for early participants once it gets off the ground, what kind of incentives come to mind to make the investment more attractive to you?)

I am quite serious here and have had correspondence with a few FPGA/ASIC specialists to case exactly how much funding would be required for prototyping and limited production run. Though it would be reckless to proceed with it without getting community input, and I want to build a device that is based on your feedback so we get exactly what you want at the price you want.

What do you think? Time to elevate Litecoin to the next level as we did with Bitcoin? Would you help BlockBurner achieve this goal though community support? Should I jump off a cliff covered in green jello?


Updated 4/25

FAQ

BlockBurner news:

BlockBurner Forums

BlockBurner subReddit


The Team (still developing):

Operatr - Administration/Operatations
Cheshyr - FPGA/Project Management
Zalfrin - FPGA Development


Project Overview

Design Goals:

Modular Scrypt FPGA system
USB Connectivity
Stand alone/Rack convertible casing for scalability
Associated open source software package


I have had a few PMs and have seen questions regarding pre-orders for this project:


On Pre-Orders

Any pre-order campaign will be associated with the current stage of development. Unlike other producers there will be no pre-orders until a certain capital requirement is met meeting the estimated costs associated with that stage. At this stage it would be in generating a working prototype device. I am taking a community approach for complete transparency, every transaction would be made public knowledge as I think if you are willing to take a chance on us, you should know exactly what your money is funding and see it develop before your eyes.

This approach minimizes risk and gives a linear progression of development that is seen by the whole community.

I don't believe it is fair to hold pre-orders in a way that in a way fakes it as if it is a real product sold online, knowing full well it does not exist. I think this practice itself is fraudulent in nature itself.

Prototype Stage

Proto-adopters would be taking the bulk of the risk, as such we would work out some other kind of benefit to funding assistance at this stage. I am open to ideas on what you would like to see if you opted to be a proto-adopter.

A known price point will be known before any pre-order campaing begins with a known cap to hit, all pre-order capital going into third-party escrow until the needed amount is reached. Otherwise it would be returned to you. This could be receiving a prototype device to help with testing or some kind of future revenue sharing.

Production Stage Once a working prototype is created, we will then move on to casing actual production costs, and much like the Proto stage, will have a certain goal needed before any capital is invested.

To do this will require a crowd-sourced effort, which would be conducted through various forums as well as things like Kickstarter campaigns and the like.


Thank you in advance for your time and support,

Operatr
BlockBurner.net

[DarkPunk]

Can it be done? Sure.  Will it be faster than GPU/CPU?  Probably.

The only problem I see here is that Scrypt is designed to be slow, because of the different conversion and hashes needed to prove the work.  This means, to get the most performance out of your FPGAs/ASICs, you would need multiple chips per card.

I'm just now looking into scrypt, and so I can't get very specific on what you would need on a card to make it substantially better than GPUs, but from a cursory glance, FPGAs stand to have a wider gap from GPUs in LTC than they do in BTC.

[Operatr]

There do exist FPGA's suitable for the job. Scrypt's main issue that it is incredibly memory intensive, where SHA256 is highly processor based and uses little memory. FPGA rigs would vastly outperform GPUs and cost much less to operate long term. GPUs work because they are designed for the memory intensity of graphics processing and delivery, which a regular processors fast cache memory is also applicable, FPGAs can follow the same path but just be much faster to hash and power efficient. Up front cost for these benefits however, is higher.

[Ryu.Hayabusa]

I'd buy one, and I'd pre-order if the time frame was reasonable.  But the the biggest risk here I think is will the price/performance ratio be worth it?

[lame.duck]

Yes i would be intrested in a nice FPGA+High speed Memory board, but  rather for use as a FPGA  than for litecoin mining. The problem is that a FPGA solution  will be hardly  less expensive compared to GPUs. Besides this, a Litecoin mining ASIC isn't that far away, there are some numbers from bitfury about  that. Besides the mining ASIC+external memory solution i would check if  a ASIC with embedded DRAM wouldnt more profitable, as it would save a lot of I/O pins, driver logic, expensive  PCBs, maybe it could even produced  in the style of topless memory Modules saving most of the chip packaging cost.

[tacotime]

Jasinlee and LaSeek apparently already have working prototypes, so you're a little late to the game.  But yes, the community interest is huge.

[jasinlee]

I would suggest looking over the open fpga project for bitcoin first. See if it will work for what your considering. If it does not look like something you would be able to easily implement (it took us many months to use it as a guide) then I would not bother. That would be a good starting point since you do not even want to touch scrypt until you know you can handle sha256 since you will most likely need to write your own module from scratch. And there is basically 0 documentation for fpga implementation of the salsa stream cipher. And the sha256 fpga documentation is incomplete and undocumented from what we found, it looks like the open fpga project was created just to get the bounty and was left almost completely undocumented except for the bits where it would effect the bounty.

[mufa23]

>1. Do you think the market and community is ready for FPGA Litecoin?
Yes.

>2. Is there definite interest in FPGA Litecoin machines? Would you buy one if the price was reasonable? What is reasonable?
Would definitely buy some. Maybe $2k worth. And a reasonable price would be $1 per 1.0 - 1.5KH/s (If power use was low)

>3. Would you pre-order one to support first round funding for prototyping and first wave production?
Probably not. One month wait at most. Just cause I've already waited once with BFL, and don't care to do it again. If you're going to hold my money, I'd rather use that money to build a GPU rig and run it for a few months. Then sell off the parts for the money to buy a working one that is ready to ship.

[jasinlee]

>1. Do you think the market and community is ready for FPGA Litecoin?
Yes.

>2. Is there definite interest in FPGA Litecoin machines? Would you buy one if the price was reasonable? What is reasonable?
Would definitely buy some. Maybe $2k worth. And a reasonable price would be $1 per 1.0 - 1.5KH/s (If power use was low)

>3. Would you pre-order one to support first round funding for prototyping and first wave production?
Probably not. One month wait at most. Just cause I've already waited once with BFL, and don't care to do it again. If you're going to hold my money, I'd rather use that money to build a GPU rig and run it for a few months. Then sell off the parts for the money to buy a working one that is ready to ship.

Which is the issue, the community has be so thoroughly burned you will end up having to fund it yourselves like we are doing. Or accept preorders and all the BS that will be thrown at you along with it. Its a double edged sword.

[actudoran]

+1
Go for it ! I'd make em in three flavors ... economy, makes sense, big $$$ ... all modular obviously. But there again lies the cost of development ... cheaper to just develop one and scale ... the gamble factor gets a bit better as these fpga can mine a broader variety of dustcoins
I'd preorder with an ETA of 30 to 45 days with a 33% deposit...

[Operatr]

Thanks for the great feedback so far guys. Keep it comin!

Jasinlee and LaSeek apparently already have working prototypes, so you're a little late to the game.  But yes, the community interest is huge.

Well damn, I better just quit then  A working proto and a solid business concept put into practice are two different things, it takes a lot to bring something new into the market. Anyone can have a design company produce something for a price. The infrastructure around it to actually ship them as a finished and refined product is the tough part. Many would be players will enter the fray only to realize this, only those who do it best will win in the end just like any other industry.


Do you think some type of modular system that scales would be desirable?  As in some type of rack system that individual boards could be added to over time?

The only problem with using existing FPGA's that are meant for Bitcoin is that they will not perform all that great with Scrypt because of the intense memory usage factor, which most FPGA boards out there are not equipped for. I've been told there do exist types that would be optimal, though so far the expense seems drastic for these (like Merrick 4/5/6 boards that have stronger memory on board).

This is the key difference between what I am attempting to do, and what is already out there for Bitcoin mining. Litecoin is a very different animal with different requirements for best performance, so that is the direction I am going with this. A truly optimized Litecoin machine with a hardware spec in mind for it over using Bitcoin hardware, which can do it though not as efficiently as it could.

>1. Do you think the market and community is ready for FPGA Litecoin?
Yes.

>2. Is there definite interest in FPGA Litecoin machines? Would you buy one if the price was reasonable? What is reasonable?
Would definitely buy some. Maybe $2k worth. And a reasonable price would be $1 per 1.0 - 1.5KH/s (If power use was low)

>3. Would you pre-order one to support first round funding for prototyping and first wave production?
Probably not. One month wait at most. Just cause I've already waited once with BFL, and don't care to do it again. If you're going to hold my money, I'd rather use that money to build a GPU rig and run it for a few months. Then sell off the parts for the money to buy a working one that is ready to ship.

Which is the issue, the community has be so thoroughly burned you will end up having to fund it yourselves like we are doing. Or accept preorders and all the BS that will be thrown at you along with it. Its a double edged sword.

Indeed, though I think companies like BFL are looking bad simply because their core team is not cohesive on their answers to the community. Assuming there is enough interest to bootstrap the device, I would make everything public in the spirit of Bitcoin transparency as to the operating fund, what it is being spent on specifically, and hopefully realistic timeframes or at least updating on what is going on. BFL is certainly being sketchy and quickly earning themselves a bad rep for it. I don't want that.

Yes i would be intrested in a nice FPGA+High speed Memory board, but  rather for use as a FPGA  than for litecoin mining. The problem is that a FPGA solution  will be hardly  less expensive compared to GPUs. Besides this, a Litecoin mining ASIC isn't that far away, there are some numbers from bitfury about  that. Besides the mining ASIC+external memory solution i would check if  a ASIC with embedded DRAM wouldnt more profitable, as it would save a lot of I/O pins, driver logic, expensive  PCBs, maybe it could even produced  in the style of topless memory Modules saving most of the chip packaging cost.

ASIC is extremely expensive , and we're talking millions $USD to jumpstart something like that. FPGA is much easier to get into, and follows the natural progression that Bitcoin followed.

I also don't believe the other currencies are ready for ASIC power, as demonstrated by the Terracoin issue where a single ASIC device hosed the network and is still recovering. The power must scale linearly with the growth of the market cap and overall usage. It would be irresponsible to flood Litecoin with such devices at this time I think which would just drive up the difficulty, kill interest in Litecoin mining, and then kill the currency altogether. FPGA is the next logical step from GPU mining.

It is not enough for a business involved in cryptocurrency to simply make and ship product, we must be careful for the care of the network. Assuming it all goes to plan, I would certainly curtail a large amount of these things from hitting market at any one time, adhering to some kind of release schedule or batching as the new ASIC companies are doing.


Adam (Operatr)
BlockBurner.net

[Benny1985]

I'd be down for a significant investment if you could prove that they offer a serious advantage over GPUs.

[pyromaniac]

1. Do you think the market and community is ready for FPGA Litecoin?

Yes. Why not? Don't see any reasons agaist that.
 

2. Is there definite interest in FPGA Litecoin machines? Would you buy one if the price was reasonable? What is reasonable?

Yes but only if they will cheaper at 20-25% than similar bitcoin FPGAs, because they less perspective.

3. Would you pre-order one to support first round funding for prototyping and first wave production?
[/quote]
Yes. I'd want to preorder at least one sample by bitcoins (or litecoins) but only after people will write good reviews. Don't get me wrong, but here too much of scammers.

[BR0KK]

Defnetly interesting. maybe you come up with a concept to reuse old FPGAs as well. Like all those spartan boards out there (without memory) could be used in a chain to work with scrypt?! (Just guessing here, i don't know much about this)

[jasinlee]

Defnetly interesting. maybe you come up with a concept to reuse old FPGAs as well. Like all those spartan boards out there (without memory) could be used in a chain to work with scrypt?! (Just guessing here, i don't know much about this)

There are other issues involved beyond memory.....we went down that road already a few months back.

Operatr: If you need help I am happy to give any assistance. The point of the fpga is to build value into LTC, we are not out to squash any potential competition. PM me if you would like any help I can offer.

[Operatr]

Defnetly interesting. maybe you come up with a concept to reuse old FPGAs as well. Like all those spartan boards out there (without memory) could be used in a chain to work with scrypt?! (Just guessing here, i don't know much about this)

There are other issues involved beyond memory.....we went down that road already a few months back.

Operatr: If you need help I am happy to give any assistance. The point of the fpga is to build value into LTC, we are not out to squash any potential competition. PM me if you would like any help I can offer.

Awesome jasinlee I will do that. I agree, I just see Litecoin going where Bitcoin is now right along side it as a silver to Bitcoin gold. As BTC is blazing a trail through the old world financial system, LTC will start getting more attention and grow larger as a secondary economy as well probably. BTC has hit Stage 4 into ASIC, LTC should follow the same path, it seems FPGA stage is imminent.

Anyone remember the value of BTC when FPGA's first appeared off hand?

I'd be down for a significant investment if you could prove that they offer a serious advantage over GPUs.

FPGA is already proven to offer similar or better performance as GPUs without the heat and substantial power requirement once you get to a certain point adding ever more GPUs to the mix. In price it works more like a "more now but less later" long term mentality. However since the only FPGAs out there are more geared for Bitcoin SHA256 than Scrypt, I am uncertain what kind of performance an optimized board would offer. I will need to prototype to see.

1. Do you think the market and community is ready for FPGA Litecoin?

Yes. Why not? Don't see any reasons agaist that.
 

2. Is there definite interest in FPGA Litecoin machines? Would you buy one if the price was reasonable? What is reasonable?

Yes but only if they will cheaper at 20-25% than similar bitcoin FPGAs, because they less perspective.

3. Would you pre-order one to support first round funding for prototyping and first wave production?

Yes. I'd want to preorder at least one sample by bitcoins (or litecoins) but only after people will write good reviews. Don't get me wrong, but here too much of scammers.
[/quote]

Unfortunately these would be similar in price to Bitcoin FPGAs, simply because hardware cost is hardware cost, it doesn't care what you are doing with it per se.

Yes there are a lot of scammers out there to be wary of, I wouldn't expect a startup like mine to have any credibility until something is actually visible, and delivered as promised. Though this is why I am asking the community before anything else to build validity and everyone can see this start from the very beginning in detail. These sites like gxminer.com spring up out of nowhere (blatant BLF copy scam) without a shred of proof about their products or production even exist, or even who they really are, and are there just to snare a few suckers buying into Bitcoin hype. Even still legitimate manufacturers are so new to this emerging industry it is hard to tell sometimes.

I have no intention of scamming anyone, those people only hurt Bitcoin at large. My vision is a long term one, where many seem to be fairly short sighted when it comes to cryptocurrency. I believe its the future, and I want to help. Money is nice but only as a byproduct of the real reason Im getting into this, which is simply I believe we need it now in a world so utterly broken. BlockBurner exists for the betterment of mankind and nothing more. It feels like the right thing to do, and given my economic status personally, the only thing I can do that will make a difference in the world and in my own life. My faith is derived from being one of the many shut out of the banks system for good, which means also like many Im pretty much stuck exactly where I am, even my last IT position's salary was a sick joke, and I'm tired of it. With all my energy and resources, I will assist in taking that system out of power forever. Bitcoin is how we do that.  I am down to build machines with the direct purpose to siphon power back to the people, as they did to us.

Hardware is only one of the elements of Bitcoin business I intend to tap going down the road, which will give alternative sustainability to the hardware side.

I have an ASIC on order with BFL, though Im feeling a little iffy like the rest given inconsistent updates from the BFL crew. Unlike them Im not out to dick anyone around here. Just to keep building this dream with the rest of you.

[jasinlee]

Well first step would be comb through the old open fpga bitcoin project, that will give you most of the stepping stones. But you will need to read through the code yourself. It is a mess and extremely undocumented so if you cant find something let me know I might be able to point you to the resources we found along the way. Maybe you will be able to put the pieces in less than 5 months like it took us.

[fpgaminer]

Anyone remember the value of BTC when FPGA's first appeared off hand?

I started open development of FPGA mining designs in May 2011; BTC was about $5USD at the time.  The first commercial FPGA mining products first appeared on August 18th, 2011, and BTC was about $10USD then.

Food for thought: scrypt as it is used in litecoin is not a memory-hard algorithm.  Rather, it's an algorithm with a space-time trade-off skewed towards using more space, given the current relative costs of memory fetches versus computations.  You can implement litecoin's scrypt with only a little over 2 kilobits of memory, but it runs ~512 times slower.

It's likely that a performant implementation will use a sparse LUT to cut down on fetches.  For example, with a 512 element LUT (instead of 1024), the algorithm performs 50% less fetches, and requires only one extra salsa round 50% of the time.

[phk]

Anyone remember the value of BTC when FPGA's first appeared off hand?

I started open development of FPGA mining designs in May 2011; BTC was about $5USD at the time.  The first commercial FPGA mining products first appeared on August 18th, 2011, and BTC was about $10USD then.

Food for thought: scrypt as it is used in litecoin is not a memory-hard algorithm.  Rather, it's an algorithm with a space-time trade-off skewed towards using more space, given the current relative costs of memory fetches versus computations.  You can implement litecoin's scrypt with only a little over 2 kilobits of memory, but it runs ~512 times slower.

It's likely that a performant implementation will use a sparse LUT to cut down on fetches.  For example, with a 512 element LUT (instead of 1024), the algorithm performs 50% less fetches, and requires only one extra salsa round 50% of the time.

I think a simplistic port of the existing scrypt.c from the litecoin source code would just use 128KB of block RAM, which is readily available on most of the FPGA's already in circulation.   I can see how it might increase the area/cost of an ASIC, but for an FPGA  it's a non-issue.

It's not clear why people are even discussing external memory requirements for this algorithm and getting ready to write-off their purchased bitcoin boards.

Overall, it sounds like we should expect an FPGA implementation of litecoin to perform at (1/1024) the rate of equivalent bitcoin hash.  And this is due to the sequential loops, and not the "memory-hard" aspect of the algorithm.

[Timzim103]

I'm very interested. I am on the fence about pre-ordering though.