I seem to remember the Avalon4 has an optimizer built in, where you can give it a hashrate and it'll iteratively drop the power until it finds the lowest stable point? We'll probably integrate an autotuning feature like that into it. If we can write an autotuning driver that can tune per board will be pretty great too.
I'm not sure if we're gonna have an input power measurement on there or not. Shunt measurements to take away from system efficiency - sure it's maybe a quarter percent of waste, but... yeah that's actually probably okay.
I think trying to find an optimal operating point based on power efficiency will pretty much always end you up at the top stable clock of the bottom possible voltage? If your only parameters are operating cost and hashrate you'll always end up finding the highest clock available at the highest efficiency voltage setpoint available. Being able to say "get me the most hashrate for this maximum power" or "get me the least power for this desired hashrate" are going to be different problems, but not difficult to solve.
Topic: GekkoScience BM1384 Project Development Discussion - page 59. (Read 146665 times)
yes sir now I see what you meant!
Does this make more sense?
Does this make more sense?
philipma1957
Horizontal axis on chart is C1 clock rate not USD/BTC.
Not label well, sorry for the confusion.
We're up late, burnin' the midnite oil?
Horizontal axis on chart is C1 clock rate not USD/BTC.
Not label well, sorry for the confusion.
We're up late, burnin' the midnite oil?
Just a thought. . . .
If an MCU is going to be used to program/set CoreV (that has multiple A/D GPIO's) . . .
and if there was a current shunt on the incoming power connector (a circuit trace of a known length with infinitesimal resistance ) . . .
One would be able to calculate (or approximate) real-time power consumption (P=I*V in a DC circuit).
By knowing real-time power consumption, hashing rate, and a little coding, in theory, the device could have an "auto" mode that after a few iterations and time select the most cost effective clock and CoreV values. Precisely calculating and setting optimum W/GH/s.
Like stealing a page from the solar charge controller world, MPPT.
MPPT = Maximum Power Point Tracking.
OHPT = Optimum Hash Point Tracking.
If the MCU selected had a small amount of internal NVRAM, those values (CoreV and Clock Rate) could be stored to minimize the calculations after a power cycle.
After determining the optimum W/GH/s settings (saved to NVRAM), a count down timer could be used to set the periodicity for re-evaluation (say once an hour initially). This value and the std deviation/running average could also be used to "fine tune", stored in NVRAM, the periodicity of re-evaluation. This capability would help account for manufacturing tolerances and make the device aging optimized. The longer the chip has been operated the "smarter" it would become.
Creating a "smart" hashing board/device is/would be another feature set other products lack.
Probably not a cost effective solution on a stickminer, but on an 18+ chip version . . . .
That tech may be patentable, the MCU micro code is definitely copyrightable.
And may be worth something to a larger volume board manufacturer (Bitmain, Spondoolie, Avalon, etc.) or not and maintain a competitive advantage.
Just sayin' . . .
This thought line comes from a recent rather exhaustive (and ongoing) evaluation of Bitmain C1 operating parameters with a focus on the most profitable operating parameters.
Quite revealing with some of the conclusions being of the non-intuitive nature.
"Why Waltz when you can Rock'N'Roll." -- Unknown
If an MCU is going to be used to program/set CoreV (that has multiple A/D GPIO's) . . .
and if there was a current shunt on the incoming power connector (a circuit trace of a known length with infinitesimal resistance ) . . .
One would be able to calculate (or approximate) real-time power consumption (P=I*V in a DC circuit).
By knowing real-time power consumption, hashing rate, and a little coding, in theory, the device could have an "auto" mode that after a few iterations and time select the most cost effective clock and CoreV values. Precisely calculating and setting optimum W/GH/s.
Like stealing a page from the solar charge controller world, MPPT.
MPPT = Maximum Power Point Tracking.
OHPT = Optimum Hash Point Tracking.
If the MCU selected had a small amount of internal NVRAM, those values (CoreV and Clock Rate) could be stored to minimize the calculations after a power cycle.
After determining the optimum W/GH/s settings (saved to NVRAM), a count down timer could be used to set the periodicity for re-evaluation (say once an hour initially). This value and the std deviation/running average could also be used to "fine tune", stored in NVRAM, the periodicity of re-evaluation. This capability would help account for manufacturing tolerances and make the device aging optimized. The longer the chip has been operated the "smarter" it would become.
Creating a "smart" hashing board/device is/would be another feature set other products lack.
Probably not a cost effective solution on a stickminer, but on an 18+ chip version . . . .
That tech may be patentable, the MCU micro code is definitely copyrightable.
And may be worth something to a larger volume board manufacturer (Bitmain, Spondoolie, Avalon, etc.) or not and maintain a competitive advantage.
Just sayin' . . .
This thought line comes from a recent rather exhaustive (and ongoing) evaluation of Bitmain C1 operating parameters with a focus on the most profitable operating parameters.
Quite revealing with some of the conclusions being of the non-intuitive nature.
"Why Waltz when you can Rock'N'Roll." -- Unknown
how on earth does a c1 make less money when coins are 275 then when coins are 250 usd?
are you assuming diff varies due to price. if all variables are frozen other then price an c1 makes more if coins are 275 then when coins are 250
Just a thought. . . .
If an MCU is going to be used to program/set CoreV (that has multiple A/D GPIO's) . . .
and if there was a current shunt on the incoming power connector (a circuit trace of a known length with infinitesimal resistance ) . . .
One would be able to calculate (or approximate) real-time power consumption (P=I*V in a DC circuit).
By knowing real-time power consumption, hashing rate, and a little coding, in theory, the device could have an "auto" mode that after a few iterations and time select the most cost effective clock and CoreV values. Precisely calculating and setting optimum W/GH/s.
Like stealing a page from the solar charge controller world, MPPT.
MPPT = Maximum Power Point Tracking.
OHPT = Optimum Hash Point Tracking.
If the MCU selected had a small amount of internal NVRAM, those values (CoreV and Clock Rate) could be stored to minimize the calculations after a power cycle.
After determining the optimum W/GH/s settings (saved to NVRAM), a count down timer could be used to set the periodicity for re-evaluation (say once an hour initially). This value and the std deviation/running average could also be used to "fine tune", stored in NVRAM, the periodicity of re-evaluation. This capability would help account for manufacturing tolerances and make the device aging optimized. The longer the chip has been operated the "smarter" it would become.
Creating a "smart" hashing board/device is/would be another feature set other products lack.
Probably not a cost effective solution on a stickminer, but on an 18+ chip version . . . .
That tech may be patentable, the MCU micro code is definitely copyrightable.
And may be worth something to a larger volume board manufacturer (Bitmain, Spondoolie, Avalon, etc.) or not and maintain a competitive advantage.
Just sayin' . . .
This thought line comes from a recent rather exhaustive (and ongoing) evaluation of Bitmain C1 operating parameters with a focus on the most profitable operating parameters.
Quite revealing with some of the conclusions being of the non-intuitive nature.
"Why Waltz when you can Rock'N'Roll." -- Unknown
If an MCU is going to be used to program/set CoreV (that has multiple A/D GPIO's) . . .
and if there was a current shunt on the incoming power connector (a circuit trace of a known length with infinitesimal resistance ) . . .
One would be able to calculate (or approximate) real-time power consumption (P=I*V in a DC circuit).
By knowing real-time power consumption, hashing rate, and a little coding, in theory, the device could have an "auto" mode that after a few iterations and time select the most cost effective clock and CoreV values. Precisely calculating and setting optimum W/GH/s.
Like stealing a page from the solar charge controller world, MPPT.
MPPT = Maximum Power Point Tracking.
OHPT = Optimum Hash Point Tracking.
If the MCU selected had a small amount of internal NVRAM, those values (CoreV and Clock Rate) could be stored to minimize the calculations after a power cycle.
After determining the optimum W/GH/s settings (saved to NVRAM), a count down timer could be used to set the periodicity for re-evaluation (say once an hour initially). This value and the std deviation/running average could also be used to "fine tune", stored in NVRAM, the periodicity of re-evaluation. This capability would help account for manufacturing tolerances and make the device aging optimized. The longer the chip has been operated the "smarter" it would become.
Creating a "smart" hashing board/device is/would be another feature set other products lack.
Probably not a cost effective solution on a stickminer, but on an 18+ chip version . . . .
That tech may be patentable, the MCU micro code is definitely copyrightable.
And may be worth something to a larger volume board manufacturer (Bitmain, Spondoolie, Avalon, etc.) or not and maintain a competitive advantage.
Just sayin' . . .
This thought line comes from a recent rather exhaustive (and ongoing) evaluation of Bitmain C1 operating parameters with a focus on the most profitable operating parameters.
Quite revealing with some of the conclusions being of the non-intuitive nature.
"Why Waltz when you can Rock'N'Roll." -- Unknown
Right, wasn't really suggesting that, thus the winky face.
I use one of these: https://www.coolgear.com/product/metal-7-port-usb-2-0-hub-w-din-rail-mounting-kit-japan-nec-chip
About the most robust usb hub I've seen.
I use one of these: https://www.coolgear.com/product/metal-7-port-usb-2-0-hub-w-din-rail-mounting-kit-japan-nec-chip
About the most robust usb hub I've seen.
Man, that thing looks serious! I just have a couple little 10 port rosewill 3.5A hubs. Had a ton of them that I used to use for Gridseed pods way back in the day, sold most but kept a couple. I figure I can probably run 2-3 sticks per hub, the adapter is 3.5A.
I use a 3-port hub. Just enough to fan one USB miner.
Right, wasn't really suggesting that, thus the winky face.
I use one of these: https://www.coolgear.com/product/metal-7-port-usb-2-0-hub-w-din-rail-mounting-kit-japan-nec-chip
About the most robust usb hub I've seen.
I use one of these: https://www.coolgear.com/product/metal-7-port-usb-2-0-hub-w-din-rail-mounting-kit-japan-nec-chip
About the most robust usb hub I've seen.
Man, that thing looks serious! I just have a couple little 10 port rosewill 3.5A hubs. Had a ton of them that I used to use for Gridseed pods way back in the day, sold most but kept a couple. I figure I can probably run 2-3 sticks per hub, the adapter is 3.5A.
Right, wasn't really suggesting that, thus the winky face.
I use one of these: https://www.coolgear.com/product/metal-7-port-usb-2-0-hub-w-din-rail-mounting-kit-japan-nec-chip
About the most robust usb hub I've seen.
I use one of these: https://www.coolgear.com/product/metal-7-port-usb-2-0-hub-w-din-rail-mounting-kit-japan-nec-chip
About the most robust usb hub I've seen.
And make an option for a built-in fan for the sticks while you're at it. 
No, because then it becomes special-purpose hardware.
Also, I mentioned somewhere else but Novak and I are gonna spend some time looking into building our own snazzy flexible powered hub, probably 7 port. We're looking at an internal regulator that can take in a range about 6V to 24V and steps it down to a clean 5V. The 5V rail would be tied to a barrel jack which could be used as a power input or output. We'd build it to handle at least 1.5A per socket (by which I mean every socket could deliver 1.5A simultaneously and it'd still work).
And make an option for a built-in fan for the sticks while you're at it. Sounds cool, part of the fun of playing with stick miners is the usb hub.
Sorry about your loss.
May you have fair winds and following seas on your journeys.
If you need something shout.
May you have fair winds and following seas on your journeys.
If you need something shout.
I'm reusing one of the hubs I originally jacked up for Block Erupters about two years ago (https://bitcointalksearch.org/topic/m.3343826). I pulled 5V in from my bench supply and used that to power the Pi and five sticks without any trouble. Doesn't take much for a crappy powered hub to be a not crappy powered hub.
Also, I mentioned somewhere else but Novak and I are gonna spend some time looking into building our own snazzy flexible powered hub, probably 7 port. We're looking at an internal regulator that can take in a range about 6V to 24V and steps it down to a clean 5V. The 5V rail would be tied to a barrel jack which could be used as a power input or output. We'd build it to handle at least 1.5A per socket (by which I mean every socket could deliver 1.5A simultaneously and it'd still work). I'm not sure yet if that'll be a fused limit, an implied limit, or a hardware-switched limit. But something like that would be pretty nifty for running sticks on, and also pretty nifty for hooking up to solar or a car battery or whatever else.
The rest of y'all, I'll get back to you sometime.
Also, I mentioned somewhere else but Novak and I are gonna spend some time looking into building our own snazzy flexible powered hub, probably 7 port. We're looking at an internal regulator that can take in a range about 6V to 24V and steps it down to a clean 5V. The 5V rail would be tied to a barrel jack which could be used as a power input or output. We'd build it to handle at least 1.5A per socket (by which I mean every socket could deliver 1.5A simultaneously and it'd still work). I'm not sure yet if that'll be a fused limit, an implied limit, or a hardware-switched limit. But something like that would be pretty nifty for running sticks on, and also pretty nifty for hooking up to solar or a car battery or whatever else.
The rest of y'all, I'll get back to you sometime.
I have not tried that. Any reason to choose that instead of a powered Hub, with a solid power supply?
Yes, as philipma1957 said, I own one and so it is free - but my experience with a powered hub was not good - of course the spacing was too close so I could only plug in 3 Antminer U2/U2+ and the device dropped so much voltage that I had to feed it with 5.75 Volts to get 4.25 out - so as you say the hub needs to be capable of supplying current in excess of the USB2 spec and if you look on ebay they seldom give that information and maybe you have other problems if you use a USB3 hub.
The points made by alh are a valid criticism. I had a cable to hand for a hard drive that took a normal male USB connector so that wasn't a problem for me and I used one of those Buck converters with a variable output one can buy from China or Hong Kong on ebay as it was before I found the Vinsic device (which as I suggested is quite nice for powering a number of PIs).
So, update.
I just got final prices from Bitmain on chips. We're organizing a two-part order, 1200 chips for me and 800 for the guy in Germany. With 1200 chips I can use 1000 as intended earlier for Compacs and Amitas (probably 600 and 200, respectively) and have 200 chips left which will allow me to do a full set of 4 of both 18-chip (half S1) and 30-chip (Prisma) boards for testing.
The portion I'll need to put up for chips and shipping is about 16.6BTC using current values. I'll be shifting funds into the 1BURGER address for now. Being as I'll be all over three states until Thursday I don't expect to take care of final business and submit payment until the end of the week. I don't want to lean on y'all for footing any of that bill, but the way things have been going, I'd probably get lynched if I didn't at least alert folks to the opportunity.
I just got final prices from Bitmain on chips. We're organizing a two-part order, 1200 chips for me and 800 for the guy in Germany. With 1200 chips I can use 1000 as intended earlier for Compacs and Amitas (probably 600 and 200, respectively) and have 200 chips left which will allow me to do a full set of 4 of both 18-chip (half S1) and 30-chip (Prisma) boards for testing.
The portion I'll need to put up for chips and shipping is about 16.6BTC using current values. I'll be shifting funds into the 1BURGER address for now. Being as I'll be all over three states until Thursday I don't expect to take care of final business and submit payment until the end of the week. I don't want to lean on y'all for footing any of that bill, but the way things have been going, I'd probably get lynched if I didn't at least alert folks to the opportunity.
I will be happy to support development of this as well. PM'd sidehack.
I also PM'd sidehack way earlier in the development about getting some dev board design files to play with but it seems I was the only one who was interested in that.
Naively (and correct me if I am wrong) but I think that a Pi with its 4 USB ports could handle 4 stick miners - subject to the following conditions:
Each stick would be attached to the pi by one of those USB cables that has an auxiliary power connector.
The auxiliary connectors would plug into a USB charger type device.
As an example I have a Vinsic USB power adapter that is specified as 40W and has 5 USB power outlets - so that a Pi and 4 sticks could be powered from the device.
http://www.amazon.com/Vinsic-5-Ports-Charger-Motorola-Cellphone/dp/B00L1XQBBM
I quite like this device which I have used to power a couple of PIs and charge a phone and a battery at the same time.
Whether this is sensible of course is another matter.
Each stick would be attached to the pi by one of those USB cables that has an auxiliary power connector.
The auxiliary connectors would plug into a USB charger type device.
As an example I have a Vinsic USB power adapter that is specified as 40W and has 5 USB power outlets - so that a Pi and 4 sticks could be powered from the device.
http://www.amazon.com/Vinsic-5-Ports-Charger-Motorola-Cellphone/dp/B00L1XQBBM
I quite like this device which I have used to power a couple of PIs and charge a phone and a battery at the same time.
Whether this is sensible of course is another matter.
This might work, if you can find the right kind of "3-connector USB Aux cable". My experience with the USB Aux cables was that they would 2 "A" ends for plugging into two sockets on the computer, and the 3rd was a mini-USB connector to plug into a portable hard drive. In this application, instead of the mini-USB, you would need a "Female A" socket, as if it was a USB extension cable. I haven't seen such a cable, but I haven't ever looked.
I would think for most folks that a powered USB Hub would be much simpler and less complicated with fewer cables involved.
As for the Hub that Phil mentioned above, I too found that despite it's 10 USB sockets, I could never get more than 6 to work consistently and reliably. It looked great in terms of socket spacing and such, and I was willing to supply more 12V than the original power brick, but It always choked when I tried to add a 7th stick. I ultimately just used it as a "widely spaced" 5 port hub where it worked just great. I was running U1, U2, and Red Fury stick miners at the time.
I have not tried that. Any reason to choose that instead of a powered Hub, with a solid power supply?
he said he owned one. so it is free.
I read reviews on it I am not sure it would work I could not find anyone using it for data.
but if you have it on hand the worse that happens it it does not work or you burn it out.
Now for those that do not have a hub you can do this a few ways on the cheap.
http://www.amazon.com/Etekcity-Adapter-Charging-Protector-Indicators/dp/B00ELPSEVW/ref=sr_1_15?
I used this cheap hub it ran this 1 stick from side hack on a pc.
I had lots of these hubs I would sell usb stick kits on ebay 3 ant miner u-1's 1 fan 1 usb stick with miner software.
so this usb hub was fool proof on the 3 U-1 stick level. and freq 175.
the power brick was 12 volts and 3 amps.
once you got to 6 sticks you ran into trouble. So if you want 2 or 3 of sidehacks stick this should be okay.
If you want 6 to 10 sticks I don't think it will work
I have not tried that. Any reason to choose that instead of a powered Hub, with a solid power supply?
Naively (and correct me if I am wrong) but I think that a Pi with its 4 USB ports could handle 4 stick miners - subject to the following conditions:
Each stick would be attached to the pi by one of those USB cables that has an auxiliary power connector.
The auxiliary connectors would plug into a USB charger type device.
As an example I have a Vinsic USB power adapter that is specified as 40W and has 5 USB power outlets - so that a Pi and 4 sticks could be powered from the device.
http://www.amazon.com/Vinsic-5-Ports-Charger-Motorola-Cellphone/dp/B00L1XQBBM
I quite like this device which I have used to power a couple of PIs and charge a phone and a battery at the same time.
Whether this is sensible of course is another matter.
Each stick would be attached to the pi by one of those USB cables that has an auxiliary power connector.
The auxiliary connectors would plug into a USB charger type device.
As an example I have a Vinsic USB power adapter that is specified as 40W and has 5 USB power outlets - so that a Pi and 4 sticks could be powered from the device.
http://www.amazon.com/Vinsic-5-Ports-Charger-Motorola-Cellphone/dp/B00L1XQBBM
I quite like this device which I have used to power a couple of PIs and charge a phone and a battery at the same time.
Whether this is sensible of course is another matter.
First, sorry for your loss 
I don't want to lean on y'all for footing any of that bill, but the way things have been going, I'd probably get lynched if I didn't at least alert folks to the opportunity.
If it's taken as just donations or if it's offset to the cost of the final product(s), that shouldn't be a problem. Pre-orders in general suck, but traditionally these are for highly theoretical products or (in the past) based on supplies that are on shaky grounds. Neither should apply here. You know I'm in for several of the sticks anyway.I pm'd a pledge for support.
I would be looking to buy sticks and 18 chip boards down the road. I don't mind leaving some funds in sidehack's hands.
Same here. sidehack, PM me with details.
Jump to: