Official BITMINE CoinCraft series 28nm ASIC miners thread - page 222.

Puppet

legendary

Activity: 980

Merit: 1040

Quote from: RoadStress on October 17, 2013, 04:10:03 PM

Quote from: Puppet on October 17, 2013, 03:47:37 PM

clear the "Profitability decline per year" and use the same electricity price I did. Or try a higher difficulty level. Either way, there is always a point where the non turbo mode will be more profitable, I could write you the formula if I wasnt too lazy to do it.

I don't understand how Profitability decline per year works but even with 0.01 or with 10 i still get better numbers with turbo mode and we are talking about month numbers, not year. I don't understand why should i use bigger diff. At 20 bil diff i get these numbers(numbers for 1 month):

40GH/40W:

Quote

Power cost per time frame    4.38 USD
Revenue per time frame    4.44 USD
Less power costs    0.05 USD

25GH/15W:

Quote

Power cost per time frame    1.64 USD
Revenue per time frame    2.77 USD
Less power costs    1.13 USD

As i was saying if you start being unprofitable with turbo mode you are in problem with low power mode too. Please don't be lazy and give me the "get me rich" formula.

Maybe someone else can explain me better than you.

Im not sure I understand your point then. You agree with me there is a difficulty level (and electricity cost and btc value) where the lower power mode is more profitable than the turbo mode. Thats the point of that mode. Sooner or later we will hit that point. Some miners sooner than others (due to higher electricity cost, cooling costs etc). Why would it not make sense to include it in the chip?

For someone buying batch 1 coincraft miners today, that mode may not be a big concern, because the write off of the hardware is so much bigger than the electricity cost. But it wont always be like that. In fact, it wont be for very long. Bitmine is preselling its chips at $100 right now, but for some perspective, Bitfury is selling its asics at $19 a piece and Avalon was selling complete miners at a price that corresponds to just $2 per asic. Now bitmine may not go quite that low, but even at $10-20 per chip, buying or running these chips when difficulty is 20B still makes sense, just not in turbo mode.

RoadStress

legendary

Activity: 1904

Merit: 1007

Quote from: Puppet on October 17, 2013, 03:47:37 PM

clear the "Profitability decline per year" and use the same electricity price I did. Or try a higher difficulty level. Either way, there is always a point where the non turbo mode will be more profitable, I could write you the formula if I wasnt too lazy to do it.

I don't understand how Profitability decline per year works but even with 0.01 or with 10 i still get better numbers with turbo mode and we are talking about month numbers, not year. I don't understand why should i use bigger diff. At 20 bil diff i get these numbers(numbers for 1 month):

40GH/40W:

Quote

Power cost per time frame    4.38 USD
Revenue per time frame    4.44 USD
Less power costs    0.05 USD

25GH/15W:

Quote

Power cost per time frame    1.64 USD
Revenue per time frame    2.77 USD
Less power costs    1.13 USD

As i was saying if you start being unprofitable with turbo mode you are in problem with low power mode too. Please don't be lazy and give me the "get me rich" formula.

Maybe someone else can explain me better than you.

Puppet

legendary

Activity: 980

Merit: 1040

clear the "Profitability decline per year" and use the same electricity price I did. Or try a higher difficulty level. Either way, there is always a point where the non turbo mode will be more profitable, I could write you the formula if I wasnt too lazy to do it.

RoadStress

legendary

Activity: 1904

Merit: 1007

Quote from: Puppet on October 17, 2013, 01:59:51 AM

Once difficulty reaches 10B, per month each chip:

in turbo mode, 40W 40GH, it will mine $9.17 and cost $7.2 in electricity -> $1.97 profit
in normal mode, 15W 25GH it will mine $5.73 and cost $2.7 in electricity -> $3.03 profit

Once difficulty reaches 20B in turbo mode the chip will operationally run at a loss, while in normal mode it will still be profitable.

I don't know where did you come up with those numbers. Using http://www.bitcoinx.com/profit/ at diff 10,267,731,249 and 150 per BTC i get the following numbers:

40GH/40W:

Quote

Power cost per time frame    4.38 USD
Revenue per time frame    8.76 USD
Less power costs    4.38 USD

25GH/15W:

Quote

Power cost per time frame    1.64 USD
Revenue per time frame    5.48 USD
Less power costs    3.83 USD

So we are back at starting point.

DickMS

newbie

Activity: 28

Merit: 0

Uh, Giorgio, I noticed that you haven't answered a long-ago asked question about voltages, and it's VERY pertinent to board designers. I have a few more questions to add, as well, to finish my board design.

Your published specs don't denote a "pin" in the new package for V_IO. Also, your marketing notes say you will be running at much lower voltages than are supported by Global Foundry, at the moment. Those voltages (0.5v-0.65v) are incompatible with normal IO levels on uC's and other controllers, certainly. I need to know what voltage your SPI will be running at, and what pin I am supposed to send that voltage into the chip, please. I am currently operating on the assumption that you will be using one of the standard V_IO's that GF offers, namely 1.8V and 2.5V, and that one of the unused pins (6, 12, or 13) is where you want that voltage supplied. I suspect you will require 1.8V on pin 13, but would appreciate some real data to work from.

By the way, is this a flip chip package, that can be cooled from the top, or a normal upright installed die, that needs cooling from the back? Finally, which rail is Gnd, and which is V_internal?

Any help you can provide will make it easier to have boards ready when the chips are.... Thanks!

Dick Martin-Shorter/"DickMS"

giorgiomassa

full member

Activity: 222

Merit: 100

Quote from: ?? on ??

order #396
I think you have at least 50 emails from me, last one today...

Order #396 has been refunded as part of our compensation plan and you accepted that solution on the 23th of September, the refunded money has been wired to you on the 8th of October 2013. Your order was related to the Avalon based products, so you are kindly invited to not spam on this thread which is CoinCraft related.

Thank you!

giorgiomassa

full member

Activity: 222

Merit: 100

Quote from: ?? on ??

On my side I am in advanced talking with my lawyer to sue them in a swiss court. I'll let you know how it goes.

Wow how scaring, who are you precisely?

giorgiomassa

full member

Activity: 222

Merit: 100

Quote from: dunchy on October 17, 2013, 08:11:00 AM

So if I order today, for example, one rig + hosting is my estimated date for being online <= 01. January 2014 ?

Precisely!

dunchy

sr. member

Activity: 299

Merit: 250

So if I order today, for example, one rig + hosting is my estimated date for being online <= 01. January 2014 ?

giorgiomassa

full member

Activity: 222

Merit: 100

Quote from: masterOfDisaster on October 17, 2013, 07:04:22 AM

Quote from: giorgiomassa on October 17, 2013, 06:56:48 AM

Quote from: masterOfDisaster on October 17, 2013, 06:55:24 AM

Quote from: giorgiomassa on October 17, 2013, 06:49:53 AM

[...]

There will be a significant price drop for January deliveries but we're not there yet so, as DyslexicZombei just said, currently there's the Halloween promotion now with 15% and December delivery.

Does that mean each device that gets shipped in January will receive a kind of refund if it has already been paid?

No, that does mean that once the shipping queue will top the last shipping day for December, we will reduce the prices accordingly because the newly ordered and paid devices will ship in January.

That is interesting, because I ordered a device some time ago and the shipping date was already estimated for January.
So I don't really understand how someone can order a device now and can get such an early place in the shipping queue to allow for December shipping...
...what would even be more strange because of the 15% discount that is available now. That would look like punishing early buyers by giving them no such discount and shipping the devices estimatedly in January Wink

The units don't leave the factory all at the same time. They are manufactured and shipped out daily, that means: early orders paid the full price, entered the shipping queue as first and will receive (and mine with) their units before anybody else. Those ordering today, get in at the current size of the shipping queue and for this deserve a small discount. The January batch we're talking about will be shipped after the well known holidays, as such they will deserve an additional price reduction. It's pretty straightforward, actually.

masterOfDisaster

sr. member

Activity: 321

Merit: 250

Quote from: giorgiomassa on October 17, 2013, 06:56:48 AM

Quote from: masterOfDisaster on October 17, 2013, 06:55:24 AM

Quote from: giorgiomassa on October 17, 2013, 06:49:53 AM

[...]

There will be a significant price drop for January deliveries but we're not there yet so, as DyslexicZombei just said, currently there's the Halloween promotion now with 15% and December delivery.

Does that mean each device that gets shipped in January will receive a kind of refund if it has already been paid?

No, that does mean that once the shipping queue will top the last shipping day for December, we will reduce the prices accordingly because the newly ordered and paid devices will ship in January.

That is interesting, because I ordered a device some time ago and the shipping date was already estimated for January.
So I don't really understand how someone can order a device now and can get such an early place in the shipping queue to allow for December shipping...
...what would even be more strange because of the 15% discount that is available now. That would look like punishing early buyers by giving them no such discount and shipping the devices estimatedly in January Wink

giorgiomassa

full member

Activity: 222

Merit: 100

Quote from: masterOfDisaster on October 17, 2013, 06:55:24 AM

Quote from: giorgiomassa on October 17, 2013, 06:49:53 AM

[...]

There will be a significant price drop for January deliveries but we're not there yet so, as DyslexicZombei just said, currently there's the Halloween promotion now with 15% and December delivery.

Does that mean each device that gets shipped in January will receive a kind of refund if it has already been paid?

No, that does mean that once the shipping queue will top the last shipping day for December, we will reduce the prices accordingly because the newly ordered and paid devices will ship in January.

masterOfDisaster

sr. member

Activity: 321

Merit: 250

Quote from: giorgiomassa on October 17, 2013, 06:49:53 AM

[...]

There will be a significant price drop for January deliveries but we're not there yet so, as DyslexicZombei just said, currently there's the Halloween promotion now with 15% and December delivery.

Does that mean each device that gets shipped in January will receive a kind of refund if it has already been paid?

giorgiomassa

full member

Activity: 222

Merit: 100

Quote from: papamoi on October 16, 2013, 11:48:53 PM

so guys

what is your roi and when are you planning to deliver?

did you reduce your price of 40% because the difficulty have increased of the same or??

There will be a significant price drop for January deliveries but we're not there yet so, as DyslexicZombei just said, currently there's the Halloween promotion now with 15% and December delivery.

giorgiomassa

full member

Activity: 222

Merit: 100

Quote from: Puppet on October 16, 2013, 04:18:23 PM

Quote from: RoadStress on October 16, 2013, 02:50:24 PM

Am i wrong?

Not yet. But you will be wrong ~6-12 months from here when electrical efficiency will become pretty much all that matters. Bitmine certainly didnt waste that effort, once the network reaches dozens of PH, they will still have a chip thats worth buying, assuming they can hit those numbers.

Our approach was based on assumptions on how the mining difficulty curve will evolve in 2014 and the fundamental laws of any silicon design: the faster the clock, the lower the efficiency.

That said, with CoinCraft you have a chip design that can run immediately when you receive it in Turbo mode with the same efficiency as a KNC (1 W/GH/s), but when difficulty will raise further and profit margins will be even tighter, you can run your CoinCraft in nominal or even power save mode and throw away your KNC because it eats more electricity than what it mines.

giorgiomassa

full member

Activity: 222

Merit: 100

Quote from: aerobatic on October 16, 2013, 12:50:56 PM

Quote from: Puppet on October 16, 2013, 12:41:14 PM

There is no way you can cool a (tiny) 40W chip passively without an expensive monster cooler. You will need a fan. Typically anything over 10W needs a fan of some sort.

lets say you want to have a 400-500 GH unit to compete with the usual offerings from BitFury, KnC, HF, CT etc.. so youd need 12 or more of the chips running in turbo mode... generating 400-500 Watts of heat...

So... can you just plop a heatsink on each of the 12 coincraft chips, and then run a fan or two at the end of the enclosure to move the air past all the chips...? OR... do you need active cooling, ie: a Fan on a heatsink on each of the 12 chips? (thats starting to get BIG).

40 Watts is an awkward amount of heat. its not so small, like a bitfury chip.. where you dont need anything... and its not so large, like an Intel chip (or more likely, a KnC chip) that you need a heatsink with fan strapped to it... so its somewhere in the middle.. and the cooling solution for the coincraft chips will be interesting because it doesnt seem to fit with either the 'lots of cool chips' nor the 'a few big hot chips' bracket. it seems somewhere in the middle... and thats why im curious about the cooling solution

Exactly, that's what we call a "sweet spot", we found a die size that gets the best of each of the two extremes you just mentioned.

DyslexicZombei

sr. member

Activity: 364

Merit: 250

I just got out of a Skype Chat with Giorgio.

2 things:

1) Bitmine is still on track for December deliveries for new orders according to Giorgio.

2) Bitmine is having a new Halloween Sale. It's 15% off until the end of the month.

It's likely that I'm the first person outside of Bitmine with this new info because I was Johnny on the spot.

For those interested in partial ownership of Bitmine Rig, we already have a Bitmine Rig ordered for our co-op and we're attempting to raise funds for another 800GH/s in a 2nd Bitmine fundraising round, with very little to no markup to the GB shares: https://bitcointalksearch.org/topic/closed-r8b-bitmine-rig-600ghs-at-costhost-70-10-15ghsups-310217

Newar

legendary

Activity: 1358

Merit: 1001

https://gliph.me/hUF

Quote from: papamoi on October 16, 2013, 11:48:53 PM

[..]

did you reduce your price of 40% because the difficulty have increased of the same or??

Huh, did I miss anything?

Puppet

legendary

Activity: 980

Merit: 1040

Quote from: zefir on October 17, 2013, 02:41:07 AM

Quote from: Puppet on October 17, 2013, 01:59:51 AM

Quote from: RoadStress on October 16, 2013, 05:09:15 PM

I would like more than one opinion.

Who needs opinions when you can have math?
[...]

While looking at profitability only from the cost perspective is perfectly fine for the home user, in larger scale mining the available power is most often the limiting factor.

Say you operate a mining farm that has 16kW available and want to fill it up with Bitmine 40-chip units. Would you prefer running 10 units in turbo mode, or 26 units in nominal mode - 16THps vs. 26THps?

It is not all that black and white if you start considering those factors.

My point is merely that it makes sense for bitmine to provide different performance/efficiency modes. RoadStress thought it was a waste of time. Besides, not much time will have been spent on that, every chip works in that way; increase voltage and you increase the maximum stable clockspeed (up to a point), but you almost always decrease effiency. When you plot that relationship, you get whats called a shmoo plot:

zefir

donator

Activity: 919

Merit: 1000

Quote from: Puppet on October 17, 2013, 01:59:51 AM

Quote from: RoadStress on October 16, 2013, 05:09:15 PM

I would like more than one opinion.

Who needs opinions when you can have math?
[...]

While looking at profitability only from the cost perspective is perfectly fine for the home user, in larger scale mining the available power is most often the limiting factor.

Say you operate a mining farm that has 16kW available and want to fill it up with Bitmine 40-chip units. Would you prefer running 10 units in turbo mode, or 26 units in nominal mode - 16THps vs. 26THps?

It is not all that black and white if you start considering those factors.

Topic: Official BITMINE CoinCraft series 28nm ASIC miners thread - page 222. (Read 565236 times)