Bitcoin Forum
May 31, 2024, 01:20:30 AM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 2 [3] 4 5 »  All
  Print  
Author Topic: ASIC power consumption estimates  (Read 15389 times)
Bogart
Legendary
*
Offline Offline

Activity: 966
Merit: 1000


View Profile
October 01, 2012, 06:59:15 PM
 #41

You (and Tom) are dreaming if you think it's going to run at 60w.  The reason he doesn't want to post his power specs is because he knows they are going to be high.

Why is it attainable for your company but only dream-able for others?

Because BFL is using a 65nm process and Tom is using 130nm? That's a 4x difference in power use right there.

Actually, I'm not sure what Tom is using. Maybe I'm thinking of Avalon. Anyone know for sure?

3, this ASIC will manufacture by SMIC or TSMC.  0.11 or 0.13 process.

There the Avalon feature size has been stated.  I don't think I have seen either BFL or Tom/cablepair mention their process feature size.

"All safe deposit boxes in banks or financial institutions have been sealed... and may only be opened in the presence of an agent of the I.R.S." - President F.D. Roosevelt, 1933
Shadow383
Sr. Member
****
Offline Offline

Activity: 336
Merit: 250


View Profile
October 01, 2012, 07:02:59 PM
 #42

Anyone remember when the BFL singles were going to use 20W? Good times  Wink

Yeeeah, I'll believe anyone's power figures when I see the device plugged into a power meter, not before  Cheesy
MrTeal
Legendary
*
Offline Offline

Activity: 1274
Merit: 1004


View Profile
October 01, 2012, 07:03:35 PM
 #43

You (and Tom) are dreaming if you think it's going to run at 60w.  The reason he doesn't want to post his power specs is because he knows they are going to be high.

Why is it attainable for your company but only dream-able for others?

Because BFL is using a 65nm process and Tom is using 130nm? That's a 4x difference in power use right there.

Actually, I'm not sure what Tom is using. Maybe I'm thinking of Avalon. Anyone know for sure?

Has anyone from BFL confirmed that they're on a 65nm node?
Shadow383
Sr. Member
****
Offline Offline

Activity: 336
Merit: 250


View Profile
October 01, 2012, 07:05:56 PM
 #44

You (and Tom) are dreaming if you think it's going to run at 60w.  The reason he doesn't want to post his power specs is because he knows they are going to be high.

Why is it attainable for your company but only dream-able for others?

Because BFL is using a 65nm process and Tom is using 130nm? That's a 4x difference in power use right there.

Actually, I'm not sure what Tom is using. Maybe I'm thinking of Avalon. Anyone know for sure?

Has anyone from BFL confirmed that they're on a 65nm node?
No, but they have shown the same solidworks model with two different extrusions  Wink
Xfinity
Full Member
***
Offline Offline

Activity: 127
Merit: 100



View Profile
October 01, 2012, 08:38:53 PM
 #45

Thanks Gigavps for this thread, I will watch this thread closely!

Donations (BTC): 1L8EcGAuaDNK4kNxAbEBawo8ZaeFZvibgj
jamesg (OP)
VIP
Legendary
*
Offline Offline

Activity: 1358
Merit: 1000


AKA: gigavps


View Profile
October 01, 2012, 09:05:52 PM
 #46

Thanks Gigavps for this thread, I will watch this thread closely!

You are quite welcome. Hopefully some more manufacturers will be posting estimates or actual results soon.
mrb
Legendary
*
Offline Offline

Activity: 1512
Merit: 1027


View Profile WWW
October 02, 2012, 02:12:32 AM
 #47

  • BFL - 1 watt/Gh +- 10% source
  • Avalon - 2-10 watts/Gh source
  • ASICMINER - 6 watts/Gh waiting on source
  • bASIC -???/Gh
  • DeepBit "Reclaimer" -???/Gh

The ASICMINER source says 4.2 Joule/Ghash:
https://bitcointalk.org/index.php?topic=91173.msg1215501#msg1215501

(BTW, fix your units, it's either Watt/Ghash/s or Joule/Ghash  Tongue </science-nazi>)
jamesg (OP)
VIP
Legendary
*
Offline Offline

Activity: 1358
Merit: 1000


AKA: gigavps


View Profile
October 02, 2012, 10:28:23 AM
 #48

The ASICMINER source says 4.2 Joule/Ghash:
https://bitcointalk.org/index.php?topic=91173.msg1215501#msg1215501

(BTW, fix your units, it's either Watt/Ghash/s or Joule/Ghash  Tongue </science-nazi>)

I've fixed the OP. I'm still debating whether or not to put the bASIC claims from the BFL thread in the OP. Anyone have an opinion on this?
Xfinity
Full Member
***
Offline Offline

Activity: 127
Merit: 100



View Profile
October 02, 2012, 10:31:57 AM
 #49

The ASICMINER source says 4.2 Joule/Ghash:
https://bitcointalk.org/index.php?topic=91173.msg1215501#msg1215501

(BTW, fix your units, it's either Watt/Ghash/s or Joule/Ghash  Tongue </science-nazi>)

I've fixed the OP. I'm still debating whether or not to put the bASIC claims from the BFL thread in the OP. Anyone have an opinion on this?

Do it and add a disclaimer on those figures until we get the correct/real numbers.

Donations (BTC): 1L8EcGAuaDNK4kNxAbEBawo8ZaeFZvibgj
jamesg (OP)
VIP
Legendary
*
Offline Offline

Activity: 1358
Merit: 1000


AKA: gigavps


View Profile
October 02, 2012, 07:56:43 PM
 #50

Just updated the OP with new info from the ASICMINER thread. ASICMINER's estimates that they are going to be 4.2x the power usage of BFL. That would mean 4.2kW per Terahash! So, hypothetically, if you compared 1Th of asic miner to 1Th of BFL:

Code:
             Terahash    Power    Monthly Usage    Avg. Cost kW    Cost per Month
ASICMINER       1Th/s    4.2kW            3,066           $0.11           $337.26
BFL             1Th/s    1.0kW              730           $0.11            $80.30

I don't know about you guys, but that BFL equipment, if delivered as expected, is going to be able to run a lot longer with increasing difficulty compared to ASICMINER.
squid
Member
**
Offline Offline

Activity: 112
Merit: 10


View Profile
October 02, 2012, 09:32:13 PM
 #51

Just updated the OP with new info from the ASICMINER thread. ASICMINER's estimates that they are going to be 4.2x the power usage of BFL. That would mean 4.2kW per Terahash! So, hypothetically, if you compared 1Th of asic miner to 1Th of BFL:

Code:
             Terahash    Power    Monthly Usage    Avg. Cost kW    Cost per Month
ASICMINER       1Th/s    4.2kW            3,066           $0.11           $337.26
BFL             1Th/s    1.0kW              730           $0.11            $80.30

I don't know about you guys, but that BFL equipment, if delivered as expected, is going to be able to run a lot longer with increasing difficulty compared to ASICMINER.

I am still calling bs on all power ratings until I see actual prototypes performing at predicted rates.
kano
Legendary
*
Offline Offline

Activity: 4508
Merit: 1821


Linux since 1997 RedHat 4


View Profile
October 02, 2012, 09:48:34 PM
 #52

Isn't the BFL device 1.65kW? ... which also can pose problems for some with 110V ... since it's 15A ... well above 10A

Pool: https://kano.is - low 0.5% fee PPLNS 3 Days - Most reliable Solo with ONLY 0.5% fee   Bitcointalk thread: Forum
Discord support invite at https://kano.is/ Majority developer of the ckpool code - k for kano
The ONLY active original developer of cgminer. Original master git: https://github.com/kanoi/cgminer
nomnomnom
Sr. Member
****
Offline Offline

Activity: 313
Merit: 250



View Profile
October 02, 2012, 09:56:49 PM
 #53

Just updated the OP with new info from the ASICMINER thread. ASICMINER's estimates that they are going to be 4.2x the power usage of BFL. That would mean 4.2kW per Terahash! So, hypothetically, if you compared 1Th of asic miner to 1Th of BFL:

Code:
             Terahash    Power    Monthly Usage    Avg. Cost kW    Cost per Month
ASICMINER       1Th/s    4.2kW            3,066           $0.11           $337.26
BFL             1Th/s    1.0kW              730           $0.11            $80.30

I don't know about you guys, but that BFL equipment, if delivered as expected, is going to be able to run a lot longer with increasing difficulty compared to ASICMINER.

If I read this correct they use 135nm chips, so pretty old tech. So I think it is expected to be slower,
but they are located in china and maybe can make up for that with cheaper product prices.
Jutarul
Donator
Legendary
*
Offline Offline

Activity: 994
Merit: 1000



View Profile
October 02, 2012, 10:48:20 PM
 #54

Just updated the OP with new info from the ASICMINER thread. ASICMINER's estimates that they are going to be 4.2x the power usage of BFL. That would mean 4.2kW per Terahash! So, hypothetically, if you compared 1Th of asic miner to 1Th of BFL:

Code:
             Terahash    Power    Monthly Usage    Avg. Cost kW    Cost per Month
ASICMINER       1Th/s    4.2kW            3,066           $0.11           $337.26
BFL             1Th/s    1.0kW              730           $0.11            $80.30

I don't know about you guys, but that BFL equipment, if delivered as expected, is going to be able to run a lot longer with increasing difficulty compared to ASICMINER.

If I read this correct they use 135nm chips, so pretty old tech. So I think it is expected to be slower,
but they are located in china and maybe can make up for that with cheaper product prices.


The electricity costs are only important long-term. ASICMINER consciously decided to go for an older, more reliable technology, because it minimizes the risk of the chip being DOA, which can happen with newer tech.

The ASICMINER Project https://bitcointalk.org/index.php?topic=99497.0
"The way you solve things is by making it politically profitable for the wrong people to do the right thing.", Milton Friedman
squid
Member
**
Offline Offline

Activity: 112
Merit: 10


View Profile
October 02, 2012, 10:53:42 PM
 #55

Just updated the OP with new info from the ASICMINER thread. ASICMINER's estimates that they are going to be 4.2x the power usage of BFL. That would mean 4.2kW per Terahash! So, hypothetically, if you compared 1Th of asic miner to 1Th of BFL:

Code:
             Terahash    Power    Monthly Usage    Avg. Cost kW    Cost per Month
ASICMINER       1Th/s    4.2kW            3,066           $0.11           $337.26
BFL             1Th/s    1.0kW              730           $0.11            $80.30

I don't know about you guys, but that BFL equipment, if delivered as expected, is going to be able to run a lot longer with increasing difficulty compared to ASICMINER.

If I read this correct they use 135nm chips, so pretty old tech. So I think it is expected to be slower,
but they are located in china and maybe can make up for that with cheaper product prices.


The electricity costs are only important long-term. ASICMINER consciously decided to go for an older, more reliable technology, because it minimizes the risk of the chip being DOA, which can happen with newer tech.

I'm going to assume the initial costs are also significantly cheaper. But we don't have word for bfl or btcfpga on what type of technology (device size) they will be using yet.

Please correct me (with evidence) if I am wrong.
Bogart
Legendary
*
Offline Offline

Activity: 966
Merit: 1000


View Profile
October 02, 2012, 11:10:35 PM
 #56

Isn't the BFL device 1.65kW? ... which also can pose problems for some with 110V ... since it's 15A ... well above 10A

I have yet to encounter a US house with outlet circuits rated for anything less than 15A.  Maybe it's different in other parts of the world that run on 120V (or 110V).

1500W consumption seems to be the upper limit for commonly encountered 110V/120V devices, I guess because 15A is all the makers count on being available at an outlet.

People wired for 220/240V power are at a definite advantage when it comes to being able to easily run higher wattage appliances.

"All safe deposit boxes in banks or financial institutions have been sealed... and may only be opened in the presence of an agent of the I.R.S." - President F.D. Roosevelt, 1933
jamesg (OP)
VIP
Legendary
*
Offline Offline

Activity: 1358
Merit: 1000


AKA: gigavps


View Profile
October 02, 2012, 11:14:38 PM
 #57

Isn't the BFL device 1.65kW? ... which also can pose problems for some with 110V ... since it's 15A ... well above 10A

I was using 1Th as a comparison, not as actual device figures...  Tongue
Shadow383
Sr. Member
****
Offline Offline

Activity: 336
Merit: 250


View Profile
October 03, 2012, 12:16:12 AM
 #58

Just updated the OP with new info from the ASICMINER thread. ASICMINER's estimates that they are going to be 4.2x the power usage of BFL. That would mean 4.2kW per Terahash! So, hypothetically, if you compared 1Th of asic miner to 1Th of BFL:

Code:
             Terahash    Power    Monthly Usage    Avg. Cost kW    Cost per Month
ASICMINER       1Th/s    4.2kW            3,066           $0.11           $337.26
BFL             1Th/s    1.0kW              730           $0.11            $80.30

I don't know about you guys, but that BFL equipment, if delivered as expected, is going to be able to run a lot longer with increasing difficulty compared to ASICMINER.
The singles ended up using 4 times as much power per Megahash as BFL claimed they would initially. What's to stop that happening again?

However, I guess this time around there's unlikely to be any quantum leap in hashing rate on the horizon... what really comes after ASIC?  Cheesy
On cheap power you're hopefully looking at a multi-year service life on these.
jamesg (OP)
VIP
Legendary
*
Offline Offline

Activity: 1358
Merit: 1000


AKA: gigavps


View Profile
October 03, 2012, 12:24:06 AM
 #59

The singles ended up using 4 times as much power per Megahash as BFL claimed they would initially. What's to stop that happening again?

I usually learn lessons the hard way, but once they are learned, they are not repeated. I hope it is the same with BFL.

what really comes after ASIC?  Cheesy

Moore's Law and higher nm processes.
Shadow383
Sr. Member
****
Offline Offline

Activity: 336
Merit: 250


View Profile
October 03, 2012, 12:30:48 AM
 #60

Moore's Law and higher nm processes.
True, but say you had a basic $1200 budget:

2010: ~10-15Mh/s
2011: ~1400Mh/s
2012: ~2200Mh/s (achievable with cheap GPUs)
2013: >50000Mh/s

As much as Moore's law etc will have an impact, I don't think we'll ever see hashrate/$ go up >20x in a single year again. Hope not anyway  Cheesy
Pages: « 1 2 [3] 4 5 »  All
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!