|
punin (OP)
|
 |
January 17, 2016, 07:49:55 AM |
|
Neat! Hoping for passive cooling.
btw, these are 30.000x more efficient than my zTex x1.15 in 4 years. Not bad "end of moores law".
Shouldn't that be ~750x more efficient? Also a bit fairer to compare it to gen 1 Avalons, so ~140x more efficient. Ah! The good old Ztex x1.15 boards hashed @ ~200MHash/s while consuming ~10W. Did you know that Bitfury actually made a bitstream for that same FPGA that run @300Mhash/s? So those Ztex figures would be ~50J/GH vs ~0.05J/GH so ~1000 times. We can actually achieve 0.0556J/GH and I have a video that I'm editing right now 11km over Siberia where you will be able to see it.  Stay tuned. |
|
|
|
|
|
|
|
|
|
|
|
|
"There should not be any signed int. If you've found a signed int
somewhere, please tell me (within the next 25 years please) and I'll
change it to unsigned int." -- Satoshi
|
|
|
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
|
|
|
Unacceptable
Legendary
 Offline
Activity: 2212
Merit: 1001
|
|
January 17, 2016, 09:02:22 AM |
|
Neat! Hoping for passive cooling.
btw, these are 30.000x more efficient than my zTex x1.15 in 4 years. Not bad "end of moores law".
Shouldn't that be ~750x more efficient? Also a bit fairer to compare it to gen 1 Avalons, so ~140x more efficient. Ah! The good old Ztex x1.15 boards hashed @ ~200MHash/s while consuming ~10W. Did you know that Bitfury actually made a bitstream for that same FPGA that run @300Mhash/s? So those Ztex figures would be ~50J/GH vs ~0.05J/GH so ~1000 times. We can actually achieve 0.0556J/GH and I have a video that I'm editing right now 11km over Siberia where you will be able to see it. Stay tuned. I remember that,but didn't you guys charge for the 300Mhash wares  |
"If you run into an asshole in the morning, you ran into an asshole. If you run into assholes all day long, you are the asshole." -Raylan Givens Got GOXXED ?? https://www.youtube.com/watch?v=9KiqRpPiJAU&feature=youtu.be"An ASIC being late is perfectly normal, predictable, and legal..."Hashfast & BFL slogan |
|
|
|
rupy
|
|
January 17, 2016, 11:03:56 AM
Last edit: January 17, 2016, 11:19:20 AM by rupy |
|
zTex 1GH / 40W BF16 50GH / 0.06W? 40 / 0.06 = 666 (how many BF16 for 40W)666 * 50 = 33300 (each BF16 gives 50GH)Then rounded down for at the wall... what's wrong? Edit: You're right, that's completely wrong, 0.06 is per GH not per BF16 so ~666 is right... sorry!  Still impressive!  |
BANKBOOK GWT Wallet & no-FIAT Billing API
|
|
|
el_rlee
Legendary
Offline
Activity: 1600
Merit: 1014
|
|
January 17, 2016, 11:11:29 AM |
|
I remember that,but didn't you guys charge for the 300Mhash wares Could that be http://www.tricone-mining.com/ you remember? Good 'ol times  |
|
|
|
|
Unacceptable
Legendary
Offline
Activity: 2212
Merit: 1001
|
|
January 17, 2016, 12:44:57 PM |
|
Ahhh, that one  Didn't someone also sell a firmware for an FPGA miner,was it for the cairnsmore or something,god it was sooo long ago  |
"If you run into an asshole in the morning, you ran into an asshole. If you run into assholes all day long, you are the asshole." -Raylan Givens Got GOXXED ?? https://www.youtube.com/watch?v=9KiqRpPiJAU&feature=youtu.be"An ASIC being late is perfectly normal, predictable, and legal..."Hashfast & BFL slogan |
|
|
|
punin (OP)
|
|
January 20, 2016, 10:09:17 AM |
|
Neat! Hoping for passive cooling.
btw, these are 30.000x more efficient than my zTex x1.15 in 4 years. Not bad "end of moores law".
Shouldn't that be ~750x more efficient? Also a bit fairer to compare it to gen 1 Avalons, so ~140x more efficient. Ah! The good old Ztex x1.15 boards hashed @ ~200MHash/s while consuming ~10W. Did you know that Bitfury actually made a bitstream for that same FPGA that run @300Mhash/s? So those Ztex figures would be ~50J/GH vs ~0.05J/GH so ~1000 times. We can actually achieve 0.0556J/GH and I have a video that I'm editing right now 11km over Siberia where you will be able to see it. Stay tuned. I remember that,but didn't you guys charge for the 300Mhash wares We actually released the bitstream as open source. |
|
|
|
|
punin (OP)
|
|
January 20, 2016, 07:03:24 PM |
|
|
|
|
|
QuintLeo
Legendary
Offline
Activity: 1498
Merit: 1030
|
|
January 20, 2016, 08:21:35 PM |
|
Weren't you folks claiming more like 130-140 GH / chip with a heatsink?
Or is that specific to a fan-cooled heatsink setup (I notice no apparent fan blowing on that HUGE heatsink in your second video)?
|
I'm no longer legendary just in my own mind!
Like something I said? Donations gratefully accepted. LYLnTKvLefz9izJFUvEGQEZzSkz34b3N6U (Litecoin)
1GYbjMTPdCuV7dci3iCUiaRrcNuaiQrVYY (Bitcoin)
|
|
|
|
punin (OP)
|
|
January 20, 2016, 08:24:47 PM |
|
Weren't you folks claiming more like 130-140 GH / chip with a heatsink?
Or is that specific to a fan-cooled heatsink setup (I notice no apparent fan blowing on that HUGE heatsink in your second video)?
We could push it that high with some modifications, but we're not marketing the chip with those specs. The silicon has demonstrated capability of up to 184GH/s with some trickery, but it's unachievable with traditional methods. |
|
|
|
|
Mitak
|
|
January 20, 2016, 10:29:33 PM |
|
Great work,
Hope to be able to purchase engineering samples soon and play with it
|
|
|
|
|
wttbs
Legendary
Offline
Activity: 2210
Merit: 1109
|
|
January 20, 2016, 10:40:44 PM |
|
Great work,
Hope to be able to purchase engineering samples soon and play with it
For now they are not selling sample chips |
|
|
|
|
dogie
Legendary
Offline
Activity: 1666
Merit: 1183
dogiecoin.com
|
|
January 20, 2016, 10:50:50 PM |
|
Heatsinked
500mV
104.5GH
0.107 J/GH
Weren't you folks claiming more like 130-140 GH / chip with a heatsink?
Or is that specific to a fan-cooled heatsink setup (I notice no apparent fan blowing on that HUGE heatsink in your second video)?
We could push it that high with some modifications, but we're not marketing the chip with those specs. The silicon has demonstrated capability of up to 184GH/s with some trickery, but it's unachievable with traditional methods. Is ~110GH the expected nominal in the reference design? |
|
|
|
QuintLeo
Legendary
Offline
Activity: 1498
Merit: 1030
|
|
January 21, 2016, 07:51:28 AM |
|
The silicon has demonstrated capability of up to 184GH/s with some trickery, but it's unachievable with traditional methods.
Immersion cooling, I presume, to get that high. |
I'm no longer legendary just in my own mind!
Like something I said? Donations gratefully accepted. LYLnTKvLefz9izJFUvEGQEZzSkz34b3N6U (Litecoin)
1GYbjMTPdCuV7dci3iCUiaRrcNuaiQrVYY (Bitcoin)
|
|
|
sidehack
Legendary
Offline
Activity: 3318
Merit: 1847
Curmudgeonly hardware guy
|
|
January 21, 2016, 09:49:41 AM |
|
That's what the initial press release said - 184GH from immersion, 140 max on air cooling. I figured 11W 100GH would be a good practical setpoint for a production miner; gives you room to play up or down if voltage is adjustable.
|
|
|
|
|
one4many
|
|
January 22, 2016, 12:13:18 PM |
|
That's what the initial press release said - 184GH from immersion, 140 max on air cooling. I figured 11W 100GH would be a good practical setpoint for a production miner; gives you room to play up or down if voltage is adjustable.
100 GH/s * 50000 (the minimal order for getting those puppies done at a reasonable cost) means 5 PH/s for the network within 1 to 2 months after you can buy them. Over the course of 1/2 year and more chips, let's assume just 1.000.000 of those we have hash rate of around 100 PH/s more in the net. We are going to crack the ExaHash/s this year easily. |
|
|
|
|
|
KNK
|
|
January 22, 2016, 12:29:16 PM |
|
We are going to crack the ExaHash/s this year easily. I think ExaHash is guaranteed even before the block halving, not just the end of the year and probably even without Bitfury. |
|
|
|
|
carlosmnk
|
|
January 22, 2016, 01:14:44 PM |
|
That's what the initial press release said - 184GH from immersion, 140 max on air cooling. I figured 11W 100GH would be a good practical setpoint for a production miner; gives you room to play up or down if voltage is adjustable.
100 GH/s * 50000 (the minimal order for getting those puppies done at a reasonable cost) means 5 PH/s for the network within 1 to 2 months after you can buy them. Over the course of 1/2 year and more chips, let's assume just 1.000.000 of those we have hash rate of around 100 PH/s more in the net. We are going to crack the ExaHash/s this year easily. We will catch the 2 ExaHash before next 12 months... |
Played with: USB RedFury - BlackArrows Prospero X1/X3 - Butterfly Monarch- Spondoolies SP20E - Avalon 6 - Antminer U3/S3/S3+/C1/S5/S7 Batches 3-7-8 - Sfards SF100 - Innosilicon A2 Terminator - Alcheminer 96/256 - KNC Titan - Etherum Rigs
|
|
|
|
dimke_yu
|
|
January 23, 2016, 05:09:56 PM |
|
Quite possible, congrats on new chip! Also waiting to see possibility to purchase and test out...
|
|
|
|
Rabinovitch
Legendary
Offline
Activity: 2030
Merit: 1076
BTCLife.global participant
|
|
May 04, 2017, 07:04:27 AM |
|
|
|
|
|
Unacceptable
Legendary
Offline
Activity: 2212
Merit: 1001
|
|
May 04, 2017, 11:14:51 AM |
|
Only way you'll hear about it.....is if you got millions....& your Russian |
"If you run into an asshole in the morning, you ran into an asshole. If you run into assholes all day long, you are the asshole." -Raylan Givens Got GOXXED ?? https://www.youtube.com/watch?v=9KiqRpPiJAU&feature=youtu.be"An ASIC being late is perfectly normal, predictable, and legal..."Hashfast & BFL slogan |
|
|
|
