The short answer is "no". The longer answer involves quoting the first post of this thread (I'll leave finding the other times this question has been answered thoroughly as an exercise for the reader):


Since the goal has always been to create boards which will run as quiet desk miners, a secondary purpose being four of them will mount on an S1 chassis, no we will not be making it compatible with the S1 controller because that either makes it unnecessarily complex (having two completely separate means of interface) or makes it useless for its primary purpose (specifically, being a simple and flexible miner on its own merits) by tethering it to very special-purpose hardware not everyone has.

My immediate family is somewhat fragmented (the four members live in four different cities in two different states) but that cousin's immediate family all grew up on the same farm and are all still very local (4 generations of them). She just got married last December too, which doesn't make things any better. The annual family reunion next weekend will probably be a more somber affair than usual.

Also, I've never thought of stuff I build as "babies". My stick miners would be terrible on a sandwich.

You should have a chat with chiguireitor; I believe he's from Venezuela and is running a fair bit of equipment. He could probably be a good resource for knowledge.

Just as an update, I probably won't have any news this entire week. I've got four proto Compacs in the wild and folks are starting to talk about them in Phil's review thread, but that's going to be it for a while. I got word on Friday that my cousin died, so between the funeral, other family stuff and other business runs I'll be out of town for all but two of the next seven days and those two days will probably be spent on power supply tasks. So I probably won't even look at miners again until the Compac and 18-board PCBs get here.

Yeah, that's about right.

Also,


...ya think?


Hey, jackass, stop glutting my financial transaction record book with your SMS transaction services. Go get your own blockchain and stop spamming ours.


Yes, I will gladly pay you to build an infrastructure readily capable of a 90% attack because it adds so much security to the blockchain. I think you miss the point of "decentralization". Sounds to me like they're wadding up all their hardware into a few locations they control and doling out subscriptions (I could be wrong but that's the impression I get from their very vague discussion) which is one step shy of the definition of "centralized" - that step being the lessees of said equipment have some tenuous, remote control over what mining gets done (until the link is severed and they use their single console). If they can actually pull off what they've been saying they're gonna do, it might not mean the end of Bitcoin but it sure as hell won't make it better.

I read some mocking for 21e6 embedded stuff and Bitfury Light Bulbs, both of which ideas I openly despise. But yeah, that was a lot of words that didn't actually say anything (or instill any confidence in them). I hope they're full of hot air and can't do what they say, because it sounds to me like they want to centralize mining almost entirely, dole out subscriptions to their gear, and glut the blockchain with billions of stupid transactions for completely unrelated things. None of which is an actually good idea. Once a few more halvings occur and transaction fees dominate the mining revenues, how many miners are still going to be in business when the blockchain is dominated by your zero-fee authentication and communication microtransactions for (id)IoT everything everywhere? How many nodes are there going to be when you're trying to transfer several gigabytes of zero-fee transactions per minute? Why should anyone think this is a good idea?

I think I'll continue to work with real people instead of overwealth'd jackasses. Thanks for the link.

My recommendation, nobody should buy until he presents a bit of proof that they exist. An offer like this from a 5 member with zero trust and no evidence of possession can't really be taken at face value.

Additionally, since this is a hardware sales thread it should be shifted to the hardware sales forum looks like it was moved over as I was typing.

Link?

I just placed an order for a couple of those meters as well. Definitely will come in handy.

The rating on the PSU is what it can safely supply, not what it's always pushing. An S5 will draw around 600W, meaning that any PSU over 600W rated should, in theory, work (though running a PSU, especially a cheap one, at or near rated load for extended periods is almost guaranteed to burn it out). The particulars of how much it actually costs to run (power bills) are based on PSU efficiency, which is usually a fairly flat curve peaking in the middle of rated load and drooping maybe 3-5% at the upper and lower ends. So to get the best possible, you'd probably want to find a Platinum-rated (peak efficiency about 93-94%) PSU rated for 1200W, so your 600W S5 will sit right in the highest efficiency. Where I live, the difference between about 85% and 90% efficient for an S5 adds something like $30 a year to the power bill.

I also recommend taking a look at server PSUs. If noise isn't a problem (which I hope it isn't since the S5 is probably louder already than most server supplies) there are some really good efficient supplies to be found in the 750-1200W range from sellers here on the forum and they love running on 220V power.

Ooh, I might have to fetch a couple of those. The one I built a bit ago isn't working quite right.

Starts hammer on the power with the U3 driver, which affects the start voltage. The code used in the S5 ramps things up a lot more slowly, so once we have something like that in place it should start on a given frequency at a lower voltage. Right now if it starts only half the time on 610mV, it might run clean (after start) with running voltage shifted down to 590mV or so.

If you have a multimeter, you can check the voltage either at the caps along the left edge below the heatsink, or on the test pads on the back of the stick. I believe from top to bottom they are SGND, Vcore, 1.8V (ASIC logic/signals), 0.9V (ASIC PLL) and PGND

The pot in the bottom corner near the USB jack allows the output voltage to adjust between about 550mV to 800mV. I believe turning clockwise increases the voltage. It should have been set to about 610mV when you got it.

At 150MHz it should converge to 8.25GH. Good to see no HW errors. You might be able to turn the voltage down a bit and it still run smooth at slightly lower power. Startup currents tend to bounce the voltage a bit so they don't always start at as low a steady-state voltage as they'll run.

At 610mV 150MHz I ran a set for 24 hours in my ~90F shop with no air movement and zero problems. Hopefully with a decent fan and cool ambient it'll go hotter better. The makeshift heatsink on there is a bit bigger than the stock which will be on the final version.

Running it at 200? What's the power like?

I could get you 750W PSUs for about $30 if you weren't picky about noise. Feel free to message if you want to talk details.

I'd look into retired server PSUs. There's I think four guys here on the forums selling various kits from 750W to 2900W, prices between about $30 and $200 depending on what you're getting. Almost unity power factor, very clean output DC, user-defined cabling and efficiencies from about 88 to 94 percent are common.

If you're not getting a string miner (Prisma, S5 etc) the voltage and ripple requirements aren't that stiff for most miners. But if you're looking for a long-term investment PSU you'll want good clean output because string topology is going to be much more common in the future.

I've been using the BEUSB creds on that pool as a low-diff test for when I wanted to see what shares were actually coming out (and yes, the one on the hub on the Pi is pointed there now). One annoying thing with the U3 driver is the timeout; it's expecting about four times the hashrate it's actually getting so it resets the device occasionally if it hasn't seen a share for 2-4 seconds (depending on frequency). With a high-diff pool, you don't often see cgminer-reported shares, so sometimes your whole window is just "timeout reset" notices when you know the thing is submitting, typically, a couple shares a second (well below vardiff). The LED flashes with each returned share, so you know it's doing something.

Thanks, but I'm still holding off on taking in funds. We've only seen one tester report that he's received the stick, and hasn't actually run it yet. I won't make any decisions until we've at least seen a couple of the proto sticks up and going outside my bench. If I have to raise money in the short term (before the final design is tested) for a batch of chips, I will, but only if I have to.

And yes, the sticks should not cost more than $25. I think that'll be the unit price if folks are okay with it.