GekkoScience BM1384 Project Development Discussion

[sidehack]

Regarding any impersonators - I never use a handle with capital letters. Also watch for the post count and trust that no scammer's ever going to have. Also, if anyone claiming to be me asks you for money for anything at all, ever, that wasn't a reply to you specifically sending me a PM or email asking for a product or service, DO NOT BELIEVE IT. I will not solicit donations or investments or preorders or whatever else, especially not for a project in development.
Gladimor, I'm curious what the guy said to you.

Regarding a Hammer stick - yes, what Novak said. I've got most of the BM1384 breakout board PCB design done, and a refit of it basically will be a USB stick miner, so it's not that we'd be doing one or the other. The question of if anyone wants a Hammer stick (handle?) is we could look into making one ALSO, not INSTEAD.

As soon as I have some funds open up I might just buy an S5 and see what the problems are for starting at low voltages. The thing that comes to mind off the bat is the chained clock probably isn't getting through properly. I'd bet the regular comms are, since they're mostly open-drain digital IO sort of thing, but the clock I don't think actually pulls full low (at least not on the one I poked last week) so the level-shifted signal was not just shifted but also compressed. That one might not be hitting the input-low threshold or something.

[J4bberwock]

I'd have to check my notes, but I don't believe it would start at 10V. I had to start it at a higher voltage and then turn it down as it was running.
If I had to guess it's a limitation of the string design as opposed to the chip itself. The numbers Bitmain gave are likely accurate for those voltages, the S5 just won't get there. Hence why they removed the language in their thread that it could be run at 9V.

The best stable voltage I could get it to run was 11v.
Anything from 10.0v to 10.8v, the miner started but stopped after 10-30 minutes hashing.
I asked Yoshi from Bitmain if they could help understand why it failed, but got no answer on this specific subject.

Starting at 12v and lowering voltage down to 10.8v gave me the same results as starting at 10.8v: few minutes hashing, then stop.

[TheAnalogKid]

As soon as I have some funds open up I might just buy an S5 and see what the problems are for starting at low voltages. The thing that comes to mind off the bat is the chained clock probably isn't getting through properly. I'd bet the regular comms are, since they're mostly open-drain digital IO sort of thing, but the clock I don't think actually pulls full low (at least not on the one I poked last week) so the level-shifted signal was not just shifted but also compressed. That one might not be hitting the input-low threshold or something.

Sidehack, do you have any interest in blown S5's?  I've got about 6 of them that are in some sort of broken state - one started up but the fan didn't, so the "brain" of the unit is fine but the chips fried, and most of the others have at least one fried chain but still operate in some capacity.  I can't really rely on them for anything so I'd be happy to toss one or more your way, no charge.

[sidehack]

Have you considered swapping various parts from them to see how many complete units you can make?

Sure, I guess if you want to give some away I won't complain.

[TheAnalogKid]

Have you considered swapping various parts from them to see how many complete units you can make?

Sure, I guess if you want to give some away I won't complain.

Yeah, I've considered that, probably get 2, maybe 3 good units out of the bad ones after transplant surgery.  I still might attempt that, but I've deployed Spondoolies hardware to replace them so I'm not in an extreme rush.

I'll send the fried one and one half working one over to start.  I've got your details, I'll pm you.

[valkir]

Have you considered swapping various parts from them to see how many complete units you can make?

Sure, I guess if you want to give some away I won't complain.

Yeah, I've considered that, probably get 2, maybe 3 good units out of the bad ones after transplant surgery.  I still might attempt that, but I've deployed Spondoolies hardware to replace them so I'm not in an extreme rush.

I'll send the fried one and one half working one over to start.  I've got your details, I'll pm you.

Wow I which we had more person like you in the community! Thanks man! That will help a lot this project. 

[TheAnalogKid]

Have you considered swapping various parts from them to see how many complete units you can make?

Sure, I guess if you want to give some away I won't complain.

Yeah, I've considered that, probably get 2, maybe 3 good units out of the bad ones after transplant surgery.  I still might attempt that, but I've deployed Spondoolies hardware to replace them so I'm not in an extreme rush.

I'll send the fried one and one half working one over to start.  I've got your details, I'll pm you.

Wow I which we had more person like you in the community! Thanks man! That will help a lot this project. 

Well, I'd be lyin' if I said there wasn't at least some selfish reasons for doing all this... can't continue to mine without decent miners, right? 

[RadekG]

Sidehack, are you sure that info from BM about most heat goest to top is right? I don'ť understand why anyone wants to dissipate heat through plastic cover. I think bottom side is still main way of the heat dissipation, but PCB design with chain supply needs capacitors on bottom side, so bottom cooling is not possible anymore.

I am not sure if high speed miners is the way as BM has already done S2 upgrade kits with 0.3x Watt range. Not yet public for sale, only their hashnest and umisoo takes about 1Milion chips, so we will need to wait some time.

Small USB sticks are good way, but needs higher speed to be interesting. Maybe U3 style with adapter. Maybe simple PCI(e) card can be very popular, can be hidden in computer.

BTW: Hashnest do not operate 1:1 of bought miners, they use frontend to distribute BTC based on PPS rate and has nothing to do with real HW. They do not operate S2 anymore. Why they should when they have 0.3x Watt boards?

[sidehack]

If I had an already-profitable miner I'd probably run it full-bore until it hit breakeven returns, then underclock/volt it about halfway and keep running it until breakeven returns again, and then take it down to the bottom. That's just what I'd do, I haven't checked the numbers to see if it'd be better or not than running it at bottom speed the whole time.

A lot of chips dissipate most heat through the top. ASICMiner's BE300 would have been a top cooler; so is the Rockerbox and all those other big-die chips. There's no requirement for string design to have capacitors on the bottom.
I guess I don't honestly know if it's designedly a top-cooler chip, but one of the reasons for doing that (especially the chip was designed to be string'd) is because in a string design you have multiple local ground planes all at a different absolute ground potential and if they short through your heatsink things can catch on fire if your PSU doesn't trip out first. Maybe I'll test out heatsinking on our breakout board and see if one on top or one on bottom keeps the chip cooler, but that'll only matter for the USB miner. The TypeZero would have to be top-cooled so the heatsink doesn't kill the power planes.

One good thing about building a flexible miner (and by flexible I mean with software-adjustable voltage and clock) is you can make it a high-speed miner or you can make it a super-efficient miner, whichever you want. Our board should have better low-end efficiency than the S2 upgrade's fixed operating point. Actually, two of our TypeZero boards should be just about equivalent to one S2 Upgrade board (by chip count), but ours can operate at both higher speeds and lower speeds because it's not a fixed voltage. A single board can operate like a U3 (that's why we chose our board size and USB connectivity), except you'd be seeing a similar price and quite a bit more hashpower.

There is good demand for not-crappy not-crappily-priced USB sticks. We will do our best.

Also, Guy tells me they actually scrapped out all their Hammer ASICs.

[philipma1957]

well the hammer debate is gone then.

so bm1384's it is.

yeah I do agree you can sell a quality usb stick.

3 x   2 chip sticks     run balanced between 2 standard pools and 1 solo pool  . will earn back quite a bit of the out lay if they start at really low watts per gh.  and the 1 part on the solo pool could make a score.
This makes the stick a desirable item.  Just mine the 3 and nothing else.

[sidehack]

The "Hammer debate" was never whether a USB stick would be made with BM1384 XOR Hammer. It was whether, in addition to a BM1384 stick, should we ALSO dev a Hammer stick. Not necessarily because it's objectively comparable for power use or hashrate, but because market variety is a good thing and who knows it might be cheaper or more fun.

Please explain "3 x   2 chip sticks". Are you expecting a two-chip BM1384 stick? Because that's not going to happen unless you have some 1.5A-per-socket USB hubs. Though it might be fun to work up a secondary design that uses two chips string'd under a higher voltage of the same buck circuit; it'll cost more, but honestly not a lot more since 80% of the parts wouldn't change and the only added BOM would be the second ASIC, level-shifting components and a bigger heatsink. It'd certainly be overkill since the two chips could be cranked up to about 40GH and drink 20W, but a 5W 16GH stick might be fun to play with for folks with beefy hubs. Dangit that's so cool. Dangit Phil. Stop making me think about awesome stuff and adding to my workload.

[AJRGale]

You're playing with hardware and not gimmie pics? i want my Porn!

[chiguireitor]

Hey sidehack! nice idea you got there. Have you considered creating an SPI motherboard with daisy chain pluggable daughters with some BM1384s? if i had the SMT skills i would experiment with them, but i'm afraid my skills aren't up to par.

It would be nice to have expansion packs you can buy to expand onto your motherboard chaining them togheter.

Also, a miner developed exclusively for high density, oil tank setups would be great too

[sidehack]

Yep, I haven't taken a picture of anything in a couple weeks. I don't even remember what it was now.

Hey sidehack! nice idea you got there. Have you considered creating an SPI motherboard with daisy chain pluggable daughters with some BM1384s? if i had the SMT skills i would experiment with them, but i'm afraid my skills aren't up to par.

It would be nice to have expansion packs you can buy to expand onto your motherboard chaining them togheter.

Also, a miner developed exclusively for high density, oil tank setups would be great too

The short answer is, no I haven't. You're basically asking for an S6?

Also, a miner developed exclusively for high density oil tank setups might be nice if it's a ~1KW unit with easily configurable wifi that someone can use as a radiant heater in the wintertime. We're not here to build industrial stuff for million-dollar customers. The big guys are already doing that. We're here to take care of the regular folks that the big manufacturers are ignoring and/or ripping off.

[Bicknellski]

Yep, I haven't taken a picture of anything in a couple weeks. I don't even remember what it was now.

Hey sidehack! nice idea you got there. Have you considered creating an SPI motherboard with daisy chain pluggable daughters with some BM1384s? if i had the SMT skills i would experiment with them, but i'm afraid my skills aren't up to par.

It would be nice to have expansion packs you can buy to expand onto your motherboard chaining them togheter.

Also, a miner developed exclusively for high density, oil tank setups would be great too

The short answer is, no I haven't. You're basically asking for an S6?

Also, a miner developed exclusively for high density oil tank setups might be nice if it's a ~1KW unit with easily configurable wifi that someone can use as a radiant heater in the wintertime. We're not here to build industrial stuff for million-dollar customers. The big guys are already doing that. We're here to take care of the regular folks that the big manufacturers are ignoring and/or ripping off.

Some of these units could be build low cost and be used in system that could provide supplemental or primary heat for home use depends on those putting together the system. Interesting options.

[chiguireitor]

Imagine all that hot oil reflowed through a water immersed radiator, used for home use.... wouldn't that be REALLY nice?

[sidehack]

Maybe. But I can only handle so many projects at once, especially with approximately zero budget.

[valkir]

Philipma1957 propose it many time.

Building a heater :


But using heat from the chip could be really nice. People will use it to mine during winter and then paid less for there heat cost.  

Edit: https://blockchain.info/tx/9e0a6a9d662916eb44d90b008a758d6dd5996554fb24f6ae4b52f68929358513

More sandwich on the way! 

[chiguireitor]

Maybe. But I can only handle so many projects at once, especially with approximately zero budget.

I feel ya... wish i had the time too to invest on the ideas i'm talking

[sidehack]

I'm gonna try and finish the breakout PCB tonight. Lost all day yesterday for a 21-hour drive to fetch PSUs, we got orders to make you know. Tomorrow will be all manufacturing so what I don't get done today will probably not get looked at until Thursday or so. I want to get a head start on the stickminer PCB so I can get an idea of heatsinking and send some specs to a manufacturer to get some samples run out.

I may post the breakout PCB when it's done. It won't be tested until we get prototypes populated and played with, so mostly it's just a proof of hey we're doing something to earn that sandwich money.