I told you : Difficulty to 358 885 145 before November
Next difficulty jump will be above 358 885 145
Topic: 2302TH added by KNCMiner, difficulty to 358 885 145 before November (Read 6972 times)
October 19, 2013, 02:03:41 PM
I told you : Difficulty to 358 885 145 before November
Next difficulty jump will be above 358 885 145
August 23, 2013, 03:56:50 PM
Honest question about BTCGarden: I don't understand how their 192 GH/s miner is worth $10271(USD) when you can get 400+ GH/s for $6995(KNC) and $5600(HF). Are they shipping in the next few days or something?
Can you? Let me know, I'll come to pick it up tomorrow.
August 23, 2013, 02:28:58 PM
Honest question about BTCGarden: I don't understand how their 192 GH/s miner is worth $10271(USD) when you can get 400+ GH/s for $6995(KNC) and $5600(HF). Are they shipping in the next few days or something?
August 23, 2013, 11:11:10 AM
Can someone please try to make a counterpoint to HyperMega's estimate?
3Ph/s*600/2^32 ~ 400*10^6 difficulty
Which sorta makes my Jupiter unprofitable even before all machines are shipped...
So need to consider all angles here.
this is exactly what I was talking about here https://bitcointalksearch.org/topic/picture-looks-way-too-pessimistic-this-fall-279363
So... what is the point of mining this fall?? the KnC miners I buy, the more losses I take... Even if I get 100 of them. Even if KnC delivers perfectly on time So what do we do? Mine litecoins?
Mine another sha256 based altcoin?
August 23, 2013, 03:35:48 AM
Can someone please try to make a counterpoint to HyperMega's estimate?
3Ph/s*600/2^32 ~ 400*10^6 difficulty
Which sorta makes my Jupiter unprofitable even before all machines are shipped...
So need to consider all angles here.
this is exactly what I was talking about here https://bitcointalksearch.org/topic/picture-looks-way-too-pessimistic-this-fall-279363
So... what is the point of mining this fall?? the KnC miners I buy, the more losses I take... Even if I get 100 of them. Even if KnC delivers perfectly on time So what do we do? Mine litecoins?
August 22, 2013, 08:13:33 PM
Can someone please try to make a counterpoint to HyperMega's estimate?
3Ph/s*600/2^32 ~ 400*10^6 difficulty
Which sorta makes my Jupiter unprofitable even before all machines are shipped...
So need to consider all angles here.
August 02, 2013, 01:09:24 PM
AFox: Just noticed how close your estimate is to this one, reached from an entirely different perspective:
I doubt that KnC or OrSOC is seen as a big customer for this project, because they probably will not order more than 50 wafers in the first year (which would be already about 3 PH).
3 PetaHash/s ? That's a very interesting number. Could you expand the calculation for us? If I am seeing correctly that would be 10 times the present total network hashing power.
Is 50 wafers some sort of bulk discount number or some sort of standard order? Maybe there;s some minimum amount of wafers?
Thanks!
KnC ASIC will have about 100 GH/s realised with 192 engine IPs (see KnC R&D news from 7/19/2013).
I assume that an engine IP is equivalent to a pipelined hash core. To realise 100 GH/s they must run at a little bit more than 500 MHz, what is feasible in 28nm.
Based on technology scaling (2x more logic on same area form technology node to next technology node) we get based on BFL hash core size (estimated based on BFL die including 16 cores removing spare area and pad frame overhead) from 65(55)nm -> 45(40)nm -> 32(28)nm an estimated area per hash core of 0.6 mm2 in 28nm.
This results in an overall KnC die area of about 115 mm2 (0.6mm2 x 192).
Adding some area for supporting logic, I would say max 120 mm2 would be a good target.
An 28nm 300mm wafer has an area of about 70000 mm2. So we get 583 dies per wafer (assuming 100% yield, not realistic I know, but you can scale it yourself to your yield assumptions).
For 50 wafer:
583 dies/wafer -> 58.3 TH/wafer
50 wafer -> 2915 TH
Minimum ordering at foundries is normally one lot (25 wafer). Maybe also half lots are possible. But keep in mind, additional wafers costs nothing compared to the initial mask costs.
Thanks again to HyperMega!
His numbers don't seem to have hit criticism so far. Posting this here because of course it interests me greatly and trying to not contribute with speculative numbers on the original thread.
I do agree that the rule of 3 in this case should lead to a too high estimate but taking the current orders and the 25 multiples, the 50 wafer count does sound very much like a possible order for this year's KNC needs if they want to cover both the existing orders and the ones to come up to October or so.
Well, I wish all the best to KNC despite my reduced prospect for long term ROI
If they deliver this ambitious hardware on time, I will definitely consider them again next year for the next gen.
I doubt that KnC or OrSOC is seen as a big customer for this project, because they probably will not order more than 50 wafers in the first year (which would be already about 3 PH).
3 PetaHash/s ? That's a very interesting number. Could you expand the calculation for us? If I am seeing correctly that would be 10 times the present total network hashing power.
Is 50 wafers some sort of bulk discount number or some sort of standard order? Maybe there;s some minimum amount of wafers?
Thanks!
KnC ASIC will have about 100 GH/s realised with 192 engine IPs (see KnC R&D news from 7/19/2013).
I assume that an engine IP is equivalent to a pipelined hash core. To realise 100 GH/s they must run at a little bit more than 500 MHz, what is feasible in 28nm.
Based on technology scaling (2x more logic on same area form technology node to next technology node) we get based on BFL hash core size (estimated based on BFL die including 16 cores removing spare area and pad frame overhead) from 65(55)nm -> 45(40)nm -> 32(28)nm an estimated area per hash core of 0.6 mm2 in 28nm.
This results in an overall KnC die area of about 115 mm2 (0.6mm2 x 192).
Adding some area for supporting logic, I would say max 120 mm2 would be a good target.
An 28nm 300mm wafer has an area of about 70000 mm2. So we get 583 dies per wafer (assuming 100% yield, not realistic I know, but you can scale it yourself to your yield assumptions).
For 50 wafer:
583 dies/wafer -> 58.3 TH/wafer
50 wafer -> 2915 TH
Minimum ordering at foundries is normally one lot (25 wafer). Maybe also half lots are possible. But keep in mind, additional wafers costs nothing compared to the initial mask costs.
Thanks again to HyperMega!
His numbers don't seem to have hit criticism so far. Posting this here because of course it interests me greatly and trying to not contribute with speculative numbers on the original thread.
I do agree that the rule of 3 in this case should lead to a too high estimate but taking the current orders and the 25 multiples, the 50 wafer count does sound very much like a possible order for this year's KNC needs if they want to cover both the existing orders and the ones to come up to October or so.
Well, I wish all the best to KNC despite my reduced prospect for long term ROI
If they deliver this ambitious hardware on time, I will definitely consider them again next year for the next gen. July 29, 2013, 08:44:54 AM
Well, since this seems to be on topic, here's a mega estimate for Avalon also (+500% increase):
http://thegenesisblock.com/latest-shipment-of-avalon-asics-could-increase-network-hashrate-by-500/
http://thegenesisblock.com/latest-shipment-of-avalon-asics-could-increase-network-hashrate-by-500/
July 29, 2013, 02:52:32 AM
Allegedly.
July 26, 2013, 03:32:00 AM
If not by KnC, then by these guys, who look like they mean serious business:
https://bitcointalksearch.org/topic/hashfast-announces-specs-for-new-asic-400ghs-262052
https://bitcointalksearch.org/topic/hashfast-announces-specs-for-new-asic-400ghs-262052
July 25, 2013, 08:04:43 PM
Could you rework with just the known confirmed orders?
Already done:AFox, can you rework your calculation, but only with the known orders and update the OP?
With just the known orders, we're looking at 58,400 TH/s added.TBH, I think applying the rule of 3rds is providing an estimate that is way off:
July 25, 2013, 04:12:33 PM
Could you rework with just the known confirmed orders?
July 24, 2013, 09:13:47 PM
Plus there are people who actively create fake accounts and constantly order units without paying just to increase the perceived perception of orders. This means potential buyers might not order and therefore lessening hash rate increase.
July 24, 2013, 08:54:12 PM
On the 4100 orders, there is an estimated total of 4573 Jupiters, 2207 Saturns and 315 Mercury ordered.
What?
So on the 4100 orders there are 7095 units...? You think the average order is 1.7 units?
July 24, 2013, 07:42:18 PM
I simply applied the rule of three.
It gives these numbers :
On the 4100 orders, there is an estimated total of 4573 Jupiters, 2207 Saturns and 315 Mercury ordered.
That is a total of 2302TH ! And it will only go up until KNCMiner blocks the orders or says that they will deliver new orders in November or later.
That's over $40 million USD in orders!!!
I don't believe the rule of 3 is applicable to this situation.
July 24, 2013, 06:01:40 PM
AFox, can you rework your calculation, but only with the known orders and update the OP?
With just the known orders, we're looking at 58,400 TH/s added.TBH, I think applying the rule of 3rds is providing an estimate that is way off:
Keep in mind: each time someone checks out a miner on kncminer's website (no payment needed), it counts as an order. 4100 is not the number of paid orders.
You also included group buys as one order. 45/116 of the Jupiters are from 4 orders, and 10/56 Saturns are from 1 order, which isn't a very good sample.
This is spot on. You can't extrapolate known orders to find out total orders in this case - it just doesn't work. No one besides the company really has a clue how much TH/s is actually on pre-order. You also included group buys as one order. 45/116 of the Jupiters are from 4 orders, and 10/56 Saturns are from 1 order, which isn't a very good sample.
July 24, 2013, 05:48:05 PM
AFox, can you rework your calculation, but only with the known orders and update the OP?
July 24, 2013, 05:44:37 PM
Don't forget Bitfury...
There's also BTCGarden and labcoin
However, I doubt there are actually 4000 orders actually placed at KnC. Avalon batch #2 had orders in the 3000s and they, obviously, only had 600 actual units to ship. Order numbers can be generated by the site all the time even if no one actually buys anything.
July 24, 2013, 02:12:44 PM
Keep in mind: each time someone checks out a miner on kncminer's website (no payment needed), it counts as an order. 4100 is not the number of paid orders.
You also included group buys as one order. 45/116 of the Jupiters are from 4 orders, and 10/56 Saturns are from 1 order, which isn't a very good sample.
You also included group buys as one order. 45/116 of the Jupiters are from 4 orders, and 10/56 Saturns are from 1 order, which isn't a very good sample.
July 24, 2013, 01:45:26 PM
Don't forget Bitfury...
I don't know how many devices have been sold by them.If you count them as well (and all the other companies) the difficulty will be above 358 885 145.
For me, by November, we will be above 400 000 000.
Jump to: