@philipma: How did you paid for squirrelsresearch site with coins? I have option to pay only with ACH, what is not to easy from Brazil cause of hourly USD value changes.

the jury is out anyways on weather it will be better to buy next gen cards vs these fpga deals too many moving oarts imo, need new motherboards specific cpus etc , how bout i grab a gtx 1180 for $600 that hashes at stright up twice the hash rate of an 1080 and worry about fpga later when new 8 card boards with better features come out.

It makes no sense to replace hardware (mother boards and cpus) with new gear unless you are building a new system or app specific boards are released.

fuck risers

I agree. What concerns me is that we will get the ordering opened and full specs (including algos supported) all at the same time so there will be some pressure to make quick decisions

Wow lots of activity over here - sorry for not being more attentive to the forum post.

As some have mentioned, there is an extremely active (> 2500 members?) discord focused on FPGAs and a ton of information and updates have been flowing there. Let me try to address the bullet points.

1. Acorn development is deep underway, specifically the production work. A lot of you want these, and I want to make sure you can have them. The good news is that any delay in opening pre-orders hasn’t delayed actual product production. That means at best a shorter wait (and hopefully better crypto market prices) once they do open.

2. I’m not in the business of trying to sell things or push things on people. We’re building things we think are going to be very useful and helpful in the community, and getting all of the information out there before taking anyone’s money is much more important to me than tossing out half-baked hashrates or collecting FOMO money. This information is being gathered now, and includes performance stats tested against a number of host configurations, as well as most GPUs that are out there, across quite a few coins. It also includes full supply chain sourcing so we can have concrete slip-free production dates and batch sizes. As you can imagine this is a lot of work, but we’re getting close and moving as fast as we can!

3. Others have detailed the BCU-1525 project, but the logistics and support effort involved to make sure these are a success for everyone have been pretty extreme. This has had an effect of delaying information (but not production!) for Acorn. We will keep working to get everyone the details they crave.

4. My company (Squirrels, and our hardware/crypto division Squirrels Research) is working closely with a number of great people in the community to make sure all this hardware is as beneficial and well supported as possible. We stand behind everything we’re involved in, and that means working with great people to ensure all that long term support is there. Senseless/AllMine have been working very hard alongside us on the 1525 project, and are bringing a great shell + bitstreams and open development code to the community. We’ve also teamed up with OhGodAGirl (Kristy) and Mineority to help make sure all of these FPGAs are available to those that may have export restrictions or expensive power or are otherwise unable to host themselves. In addition we are sponsoring numerous developers in the community through grants and development hardware/software access to ensure there will eventually be GPU level miner and algorithm support, enabling choice in the community.

5. The BCU-1525 was limited by complete supply chain constraints, as I said early on. We are even right now working extremely hard to meet the targets we have in place.  Even with market pressure those constraints are unlikely to be relieved until late Fall. I would be extremely skeptical of anyone who suggests that they will be bringing a board to market comparable with the 1525 anytime soon, and if they do I would expect significantly lower production volumes. For reference, even building the Acorn accelerator involves massive world-wide efforts to source parts and alternatives that are in some cases fully committed until next June in factory batches.

Can anyone find a link to the Discord server mentioned in the first post?

You will continue to use 1x risers for the GPUs.
There will be special miner software that does the offloading to the accelerator.

The DE0-Nano-SoC is a great board to get into learning this stuff. I have this and a few of the Max10 50kLE boards that I occasionally use for initial RTL hardware evaluation because their just so easy to work with. Vivado and Xilinx’s tools are steeper learning curve, but definitely doable.

With that said, the Cyclone V 5CSEMA4U23C6N on that board is about 1/5th the logic of what I’m proposing and while it has high-speed DMA to the ARM cores, it doesn’t have any high speed IO such as PCIe to other peripherals. That makes it tough to use outside of single small algorithms (I.e. Keccak).  The $199 version of the M.2 has ~100 Logic Elements and the $329 version has more than 200, capable of running at higher speed. The top package also has 1GB DDR3. Peak power consumption is around 15W from on-board M.2


Here’s an update summary for those catching up on this thread:

1. A few of us have been working on algorithms on FPGAs, and they’re profitable. Whitefire990 intents to release his with miner/dev fee and others of us have decided to share with the community in various forms as well.

2. On the high end the VCU1525, based on the VU9P is a very good candidate for this. It is currently a Xilinx development board on promotion for $3995, but only very small batches are being produced and the price is set to go up. Some algorithms require connecting multiple with high speed links to achieve the best performance.

3. Senseless (and possibly self, if I can be helpful) have been organizing essentially a group buy - but with FPGAs it’s less of a group buy and more of a group build. That is expected to ramp up the VCU1525 style (some power/cooling improvements)  availblility in a similar price envelope to Xilinx dev version. Working on production in US and Europe so it can be available everywhere.

4. I’ve also decided to reveal one of the smaller FPGA options in the $200-350 price envelope that I had developed for internal use and deploy, to provide an entry level option. It has a slightly different set of capabilities from the big VCU1525, but for many things it does scale.

For both hardware offerings from the community (and possibly others) orders are expected to start in June.

These boards are not so much for Ethash (though they can be used to assist/accelerate GPUs in it) , or Equihash. They excel , can improve total system performance on, or are at an advantage on Keccak (and most SHA3 candidate) , Phi variants, NIST5, Timetravel10, Lyra/LyraRev2, etc.

Someone else can chime in if I missed anything.

I designed for Ethash originally, but things are changing there. many more algorithms are possible and I’m beginning to work on that now. Let’s continue to underpromise and overdeliver.

You can run essentially as many as you have slots or PCIe lanes for, but unlike normal GPUs on risers this actually needs the PCIe bandwidth to be beneficial.

You can use it in the multiple M.2 slots on a lot of motherboards that support it, like the popular Prime Z170/270/370  variants). You can also just get a M.2 to PCIe adapter ($10-15 on Amazon).

I’m testing using 4 in 16x bifurcated slots in host cards like the Hyper x16, but not all motherboards support that.

If your GPUs are on 1x risers and you’re not using it standalone you’ll need ~4 GPUs per M.2 for maximum benefit regardless of algorithm - because you’ll be streaming data to/from the FPGA and you want to match its 4 PCIe lanes to your other devices.

if you have one GPU in a normal 8 or 16x slot you may be able to fully utilize the accelerator paired with that GPU.

The board is capable of reconfiguring itself, so for example it can configure itself for the current timetravel10 permutation and while it is not big enough to hold all 10 algorithms at once, it can accelerate the first 2-5 of them before offloading the rest to the GPU to finish. In this use case it can provide as much as 33MH of lift. Most chain-hashing algorithms can see similar benefits.

Why is that important? As long as it's profitable, I mine at 100 pl because otherwise I don't earn as much as I could. Why leave capacity unused, even if it is somewhat less profitable above 80 pl...

Check #software or pinned #general and  #zealot-109test