[ANN] Spondoolies-Tech - carrier grade, data center ready mining rigs - page 547.

zvisha

hero member

Activity: 560

Merit: 500

Quote from: Syke on June 25, 2014, 10:29:41 AM

Quote from: kendog77 on June 25, 2014, 10:27:27 AM

Right, but I think zvisha's assertion that the difficulty will level out at ~200 to 300PH is contingent upon the price of BTC remaining pretty close to where it is now. If we have another dramatic price increase, the economics of mining change completely and the difficulty will keep increasing.

Bingo! Hashrate follows price.

Yes, but then those who bought sp30 get more too. We are discussing Sp30 profitability, arent we )?

zvisha

hero member

Activity: 560

Merit: 500

Quote from: Collider on June 25, 2014, 10:20:19 AM

The basic principle isn´t wrong that there is a limit at which nobody gains ROI in the network, atleast not with new equipment.

Exactly. Then you might have some percentage of speculative miners buying equipment with negative Roi for case btc rising. Also there us momentum of asic manufecturers to stop making asics that dont roi manufacture cost. Still, network grow will stop way before 500 ph. All and all, there is not much you can improve over sp30, not more then factor of 2. Imo.

Syke

legendary

Activity: 3878

Merit: 1193

Quote from: kendog77 on June 25, 2014, 10:27:27 AM

Right, but I think zvisha's assertion that the difficulty will level out at ~200 to 300PH is contingent upon the price of BTC remaining pretty close to where it is now. If we have another dramatic price increase, the economics of mining change completely and the difficulty will keep increasing.

Bingo! Hashrate follows price.

kendog77

hero member

Activity: 742

Merit: 500

Quote from: Collider on June 25, 2014, 10:20:19 AM

The basic principle isn´t wrong that there is a limit at which nobody gains ROI in the network, atleast not with new equipment.

Right, but I think zvisha's assertion that the difficulty will level out at ~200 to 300PH is contingent upon the price of BTC remaining pretty close to where it is now. If we have another dramatic price increase, the economics of mining change completely and the difficulty will keep increasing.

Collider

hero member

Activity: 714

Merit: 500

The basic assertion isn´t wrong, as there is a limit at which nobody gains ROI in the network, atleast not with new equipment.

jamesg

vip

Activity: 1358

Merit: 1000

AKA: gigavps

Quote from: zvisha on June 25, 2014, 09:40:10 AM

Quote from: jimmothy on June 25, 2014, 12:26:01 AM

Quote from: RoadStress on June 25, 2014, 12:05:13 AM

Have you thought about the amount of money needed for last year's network hashrate jump and the money needed for next year's jump? It's a big big difference. Jumping from 150TH/s to 120PH/s is not the same as a jump from 120PH to 3600PH/s. If we assume a 0.5$/GH * 3480PH= 1.74B$ needed for this diff jump. Don't really think that there are so many money that are just waiting to be spent on bitcoin hardware.

Difficulty only needs to increase fivefold before electricity costs eat up half your earnings. So you would only need to add ~500PH or $250M.

Common, who will manufacture miners that do 0.4-0.5 when the difficulty is over 300PH? Noone.
Unless someone makes a miner that does 0.2 (which is VERY not trivial and not many of current players can do that), and then his profit is just ~200$ a month from the start - why should he invest when ROI is 2 years?
After we each 200PH-300PH mark there is no reason to create more miners, and those who have miners will make good money for a long time.

The assertion that the network will stop growing after 200-300Ph is a very very bad one. Just because retail customers might not pay $0.50/Gh doesn't mean that manufacturers cannot keep making equipment for their raw costs.

Look at who is mining today and their percentages.

Bitfury - 8-10%
KNCminer - 6%
Antpool - 4%
Cloudhashing / PeerNova - 3%

That is a full 25% of the network from just manufacturers mining. And this is only what we know about.

Do you some how think that when customers stop buying the manufacturers are going to stop producing and mining?

zvisha

hero member

Activity: 560

Merit: 500

Quote from: jimmothy on June 25, 2014, 12:26:01 AM

Quote from: RoadStress on June 25, 2014, 12:05:13 AM

Have you thought about the amount of money needed for last year's network hashrate jump and the money needed for next year's jump? It's a big big difference. Jumping from 150TH/s to 120PH/s is not the same as a jump from 120PH to 3600PH/s. If we assume a 0.5$/GH * 3480PH= 1.74B$ needed for this diff jump. Don't really think that there are so many money that are just waiting to be spent on bitcoin hardware.

Difficulty only needs to increase fivefold before electricity costs eat up half your earnings. So you would only need to add ~500PH or $250M.

Common, who will manufacture miners that do 0.4-0.5 when the difficulty is over 300PH? Noone.
Unless someone makes a miner that does 0.2 (which is VERY not trivial and not many of current players can do that), and then his profit is just ~200$ a month from the start - why should he invest when ROI is 2 years?
After we each 200PH-300PH mark there is no reason to create more miners, and those who have miners will make good money for a long time.

Collider

hero member

Activity: 714

Merit: 500

Quote from: bitcoiner23 on June 25, 2014, 05:56:14 AM

Quote from: dmeter on June 25, 2014, 04:32:50 AM

Quote from: Collider on June 25, 2014, 02:38:49 AM

16nm should allow (much) more than double the transistor density and lower leakage, while also using lower voltage...

http://forwardthinking.pcmag.com/none/322862-get-ready-for-14nm-and-16nm-chips

Just check KnC's super 20nm -30% Tongue

Process node advantages can only be compared to power advantages when basically the same mining implementation is used (specifically when optimisation on the same node is already maxed out)

Otherwise you could claim 55nm as superior to 28nm, as Bitfury had 0.8W/GH on 55nm vs 1.2W/GH on KnC 28nm.

bitcoiner23

member

Activity: 64

Merit: 10

Quote from: dmeter on June 25, 2014, 04:32:50 AM

Quote from: Collider on June 25, 2014, 02:38:49 AM

16nm should allow (much) more than double the transistor density and lower leakage, while also using lower voltage...

http://forwardthinking.pcmag.com/none/322862-get-ready-for-14nm-and-16nm-chips

Just check KnC's super 20nm -30% Tongue

dmeter

hero member

Activity: 630

Merit: 500

Quote from: Collider on June 25, 2014, 02:38:49 AM

Last i heard this wasn´t the difficulty discussion thread Wink

However, if spondoolies next chip (I mean the 2nd 28nm chip) should hit the market with a power consumption of 0.2W/GH and low overhead cost when deploying together,
you could very well get massive difficulty increases.

Does anyone per chance know the window for 16nm chip availability?

16nm should allow (much) more than double the transistor density and lower leakage, while also using lower voltage, so the chips should again be more than 2* more efficient and much cheaper per GH.

If this comes to play together with a possible bitcoin price increase (let´s say to 1500$ by April 2015) it isn´t too unlikely that those high difficulty numbers could be reached.

http://forwardthinking.pcmag.com/none/322862-get-ready-for-14nm-and-16nm-chips

RoadStress

legendary

Activity: 1904

Merit: 1007

Beastlymac will PM you to ask you for the address. I think itamidensha should do the same.

Quote from: Collider on June 25, 2014, 02:38:49 AM

Last i heard this wasn´t the difficulty discussion thread Wink

However, if spondoolies next chip (I mean the 2nd 28nm chip) should hit the market with a power consumption of 0.2W/GH and low overhead cost when deploying together,
you could very well get massive difficulty increases.

Does anyone per chance know the window for 16nm chip availability?

16nm should allow (much) more than double the transistor density and lower leakage, while also using lower voltage, so the chips should again be more than 2* more efficient and much cheaper per GH.

If this comes to play together with a possible bitcoin price increase (let´s say to 1500$ by April 2015) it isn´t too unlikely that those high difficulty numbers could be reached.

At least this discussion just ended, but yeah I think I might lose it considering the 30 June deadline. I should have asked for a 31 May deadline.

Back to hardware discussions.

Collider

hero member

Activity: 714

Merit: 500

Last i heard this wasn´t the difficulty discussion thread Wink

However, if spondoolies next chip (I mean the 2nd 28nm chip) should hit the market with a power consumption of 0.2W/GH and low overhead cost when deploying together,
you could very well get massive difficulty increases.

Does anyone per chance know the window for 16nm chip availability?

16nm should allow (much) more than double the transistor density and lower leakage, while also using lower voltage, so the chips should again be more than 2* more efficient and much cheaper per GH.

If this comes to play together with a possible bitcoin price increase (let´s say to 1500$ by April 2015) it isn´t too unlikely that those high difficulty numbers could be reached.

itamidensha

member

Activity: 75

Merit: 10

Quote from: RoadStress on June 25, 2014, 01:33:14 AM

So if the difficulty is equal or more than 300 billion you win. If it's less than 300 billion then I win. We have a deal! Winner pays 0.1 BTC to the loser.

I will accept any escrow even if for this low amount I don't think it's needed.

Beastlymac has kindly offered to do it for free! If it's ok with you let's send him the funds up front and the bet can be settled at midnight (let's say GMT) June 30, 2015.

Quote from: Beastlymac on June 25, 2014, 01:47:05 AM

I will escrow for free if you would like.

Thank you!

Beastlymac

hero member

Activity: 630

Merit: 501

Miner Setup And Reviews. WASP Rep.

Quote from: RoadStress on June 25, 2014, 01:33:14 AM

Quote from: itamidensha on June 25, 2014, 01:15:16 AM

Quote from: RoadStress on June 25, 2014, 12:45:56 AM

Sure. Let's have a 0.1 BTC bet. You bet on your desired difficulty at the end of June and I win if it's less and you win if it's more.

If the bitcoin difficulty is more than 300 billion at midnight on June 30, 2015 - I win.
Otherwise - you win.
Escrow? winner pays?

EDIT: accidentally a word

So if the difficulty is equal or more than 300 billion you win. If it's less than 300 billion then I win. We have a deal! Winner pays 0.1 BTC to the loser.

I will accept any escrow even if for this low amount I don't think it's needed.

I will escrow for free if you would like.

RoadStress

legendary

Activity: 1904

Merit: 1007

Quote from: itamidensha on June 25, 2014, 01:15:16 AM

Quote from: RoadStress on June 25, 2014, 12:45:56 AM

Sure. Let's have a 0.1 BTC bet. You bet on your desired difficulty at the end of June and I win if it's less and you win if it's more.

If the bitcoin difficulty is more than 300 billion at midnight on June 30, 2015 - I win.
Otherwise - you win.
Escrow? winner pays?

EDIT: accidentally a word

So if the difficulty is equal or more than 300 billion you win. If it's less than 300 billion then I win. We have a deal! Winner pays 0.1 BTC to the loser.

I will accept any escrow even if for this low amount I don't think it's needed.

itamidensha

member

Activity: 75

Merit: 10

Quote from: RoadStress on June 25, 2014, 12:45:56 AM

Sure. Let's have a 0.1 BTC bet. You bet on your desired difficulty at the end of June and I win if it's less and you win if it's more.

If the bitcoin difficulty is more than 300 billion at midnight on June 30, 2015 - I win.
Otherwise - you win.
Escrow? winner pays?

EDIT: accidentally a word

RoadStress

legendary

Activity: 1904

Merit: 1007

Quote from: jimmothy on June 25, 2014, 12:26:01 AM

Quote from: RoadStress on June 25, 2014, 12:05:13 AM

Have you thought about the amount of money needed for last year's network hashrate jump and the money needed for next year's jump? It's a big big difference. Jumping from 150TH/s to 120PH/s is not the same as a jump from 120PH to 3600PH/s. If we assume a 0.5$/GH * 3480PH= 1.74B$ needed for this diff jump. Don't really think that there are so many money that are just waiting to be spent on bitcoin hardware.

Difficulty only needs to increase fivefold before electricity costs eat up half your earnings. So you would only need to add ~500PH or $250M.

Sure. Been there done that already, don't worry. By the time the network will add another 500PH I will have better and more efficient equipment. Since I don't see AMs 60PH being sold so fast as you expected I'm sure that it will take some time before adding another 500PH.

Quote from: itamidensha on June 25, 2014, 12:37:48 AM

Good point. At today's BTC price, there's only about half that $1.74B of value to be mined in the coming year.
You can't assume $0.5/GH will be the going hardware rate for a whole year though.

I would be interested in making a small wager about what the difficulty will be at the end of June 2015!

Sure. Let's have a 0.1 BTC bet. You bet on your desired difficulty at the end of June and I win if it's less and you win if it's more.

itamidensha

member

Activity: 75

Merit: 10

Quote from: RoadStress on June 25, 2014, 12:05:13 AM

Quote from: itamidensha on June 24, 2014, 11:50:00 PM

Quote from: RoadStress on June 24, 2014, 11:17:51 PM

Quote from: itamidensha on June 24, 2014, 09:47:41 PM

What am I missing here?

You are missing that difficulty can't rise forever with 15% per increase.

I agree, it can't rise forever. The rises will taper off and we'll reach some rough equilibrium of hashrate.

It doesn't have to rise forever at 15% though. Only for the next 12 months,
or less if there's a couple of bigger leaps early on, or if your miner arrives at the end of Sept. instead of the beginning.

This time last year the network hashrate was 150TH/s. Today it's >600x that.
You don't think we can see a 30x rise over this coming year with all the manufacturers now on the scene?

Have you thought about the amount of money needed for last year's network hashrate jump and the money needed for next year's jump? It's a big big difference. Jumping from 150TH/s to 120PH/s is not the same as a jump from 120PH to 3600PH/s. If we assume a 0.5$/GH * 3480PH= 1.74B$ needed for this diff jump. Don't really think that there are so many money that are just waiting to be spent on bitcoin hardware.

Good point. At today's BTC price, there's only about half that $1.74B of value to be mined in the coming year.
You can't assume $0.5/GH will be the going hardware rate for a whole year though.

I would be interested in making a small wager about what the difficulty will be at the end of June 2015!

jimmothy

hero member

Activity: 770

Merit: 509

Quote from: RoadStress on June 25, 2014, 12:05:13 AM

Have you thought about the amount of money needed for last year's network hashrate jump and the money needed for next year's jump? It's a big big difference. Jumping from 150TH/s to 120PH/s is not the same as a jump from 120PH to 3600PH/s. If we assume a 0.5$/GH * 3480PH= 1.74B$ needed for this diff jump. Don't really think that there are so many money that are just waiting to be spent on bitcoin hardware.

Difficulty only needs to increase fivefold before electricity costs eat up half your earnings. So you would only need to add ~500PH or $250M.

RoadStress

legendary

Activity: 1904

Merit: 1007

Quote from: itamidensha on June 24, 2014, 11:50:00 PM

Quote from: RoadStress on June 24, 2014, 11:17:51 PM

Quote from: itamidensha on June 24, 2014, 09:47:41 PM

What am I missing here?

You are missing that difficulty can't rise forever with 15% per increase.

I agree, it can't rise forever. The rises will taper off and we'll reach some rough equilibrium of hashrate.

It doesn't have to rise forever at 15% though. Only for the next 12 months,
or less if there's a couple of bigger leaps early on, or if your miner arrives at the end of Sept. instead of the beginning.

This time last year the network hashrate was 150TH/s. Today it's >600x that.
You don't think we can see a 30x rise over this coming year with all the manufacturers now on the scene?

Have you thought about the amount of money needed for last year's network hashrate jump and the money needed for next year's jump? It's a big big difference. Jumping from 150TH/s to 120PH/s is not the same as a jump from 120PH to 3600PH/s. If we assume a 0.5$/GH * 3480PH= 1.74B$ needed for this diff jump. Don't really think that there are so many money that are just waiting to be spent on bitcoin hardware.

Topic: [ANN] Spondoolies-Tech - carrier grade, data center ready mining rigs - page 547. (Read 1260354 times)