Bitcoin Forum
December 12, 2017, 03:02:28 AM *
News: Latest stable version of Bitcoin Core: 0.15.1  [Torrent].
 
   Home   Help Search Donate Login Register  
Pages: « 1 ... 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 [265] 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 ... 586 »
  Print  
Author Topic: HashFast announces specs for new ASIC: 400GH/s  (Read 875111 times)
timmah
Member
**
Offline Offline

Activity: 70


View Profile
December 28, 2013, 05:38:25 AM
 #5281

Quote
Our initial modules (Rev 0) had reliability issues due to the PCB layout (Printed Circuit Boards).  We received Rev 2 PCBs this morning and they are currently being assembled.

Do you reckon they had the good sense to equip the PCBs with 4 PCIe power connectors this time around?

They'll likely be pushing about 600W into these boards...

I hope to god they learned their lesson from the likes of Bitfury. I also hope it's not 600 per board cause that would be stupidly ridiculous. I am expecting 300W per board though. If you look at this picture you'll notice two 8pin PCIE connectors - one on each side of the board.

https://hashfast.com/hashfast-announces-fastest-bitcoin-mining-chip-in-the-world/

click the first image and enlarge it to "full screen"


There was an update from them a while back saying that they "reopened the BJ (and Sierra?) cases installed upgraded power supplies", I don't see why they would have to do that if the power draw issue was resolved... or did they do that (and likely cost them more to open, replace and repack) so we could all put another upgrade board and not need to swap PSUs?  So does that mean that they are kind enough to save us the money that we would have to spend to power 2 boards?  Humm....
1513047748
Hero Member
*
Offline Offline

Posts: 1513047748

View Profile Personal Message (Offline)

Ignore
1513047748
Reply with quote  #2

1513047748
Report to moderator
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1513047748
Hero Member
*
Offline Offline

Posts: 1513047748

View Profile Personal Message (Offline)

Ignore
1513047748
Reply with quote  #2

1513047748
Report to moderator
1513047748
Hero Member
*
Offline Offline

Posts: 1513047748

View Profile Personal Message (Offline)

Ignore
1513047748
Reply with quote  #2

1513047748
Report to moderator
1513047748
Hero Member
*
Offline Offline

Posts: 1513047748

View Profile Personal Message (Offline)

Ignore
1513047748
Reply with quote  #2

1513047748
Report to moderator
aerobatic
Hero Member
*****
Offline Offline

Activity: 702


View Profile
December 28, 2013, 05:40:09 AM
 #5282

Quote
Our initial modules (Rev 0) had reliability issues due to the PCB layout (Printed Circuit Boards).  We received Rev 2 PCBs this morning and they are currently being assembled.

Do you reckon they had the good sense to equip the PCBs with 4 PCIe power connectors this time around?

They'll likely be pushing about 600W into these boards...

I hope to god they learned their lesson from the likes of Bitfury. I also hope it's not 600 per board cause that would be stupidly ridiculous. I am expecting 300W per board though. If you look at this picture you'll notice two 8pin PCIE connectors - one on each side of the board.

https://hashfast.com/hashfast-announces-fastest-bitcoin-mining-chip-in-the-world/

click the first image and enlarge it to "full screen"


There was an update from them a while back saying that they "reopened the BJ (and Sierra?) cases installed upgraded power supplies", I don't see why they would have to do that if the power draw issue was resolved... or did they do that (and likely cost them more to open, replace and repack) so we could all put another upgrade board and not need to swap PSUs?  So does that mean that they are kind enough to save us the money that we would have to spend to power 2 boards?  Humm....


that post was a joke.. someone had modified a hashfast blog entry and put a humorous edit of it up... it was indeed confusing as it didn't state anywhere that it was a joke.  usually at least one smiley is required

HarrisonS
Newbie
*
Offline Offline

Activity: 28


View Profile
December 28, 2013, 05:40:13 AM
 #5283

That "update" was a fake update written by a forum member, and not HashFast.
miaviator
Donator
Hero Member
*
Offline Offline

Activity: 672


It's for the children!


View Profile WWW
December 28, 2013, 05:45:55 AM
 #5284

This is quite a popular thread. 

Anyone want to buy my baby jet?

klondike_bar
Legendary
*
Offline Offline

Activity: 1652

ASIC wannabe


View Profile
December 28, 2013, 05:48:10 AM
 #5285

Quote
Our initial modules (Rev 0) had reliability issues due to the PCB layout (Printed Circuit Boards).  We received Rev 2 PCBs this morning and they are currently being assembled.

Do you reckon they had the good sense to equip the PCBs with 4 PCIe power connectors this time around?

They'll likely be pushing about 600W into these boards...

I hope to god they learned their lesson from the likes of Bitfury. I also hope it's not 600 per board cause that would be stupidly ridiculous. I am expecting 300W per board though. If you look at this picture you'll notice two 8pin PCIE connectors - one on each side of the board.

https://hashfast.com/hashfast-announces-fastest-bitcoin-mining-chip-in-the-world/

click the first image and enlarge it to "full screen"


There was an update from them a while back saying that they "reopened the BJ (and Sierra?) cases installed upgraded power supplies", I don't see why they would have to do that if the power draw issue was resolved... or did they do that (and likely cost them more to open, replace and repack) so we could all put another upgrade board and not need to swap PSUs?  So does that mean that they are kind enough to save us the money that we would have to spend to power 2 boards?  Humm....


2x6pin PCIe connectors are sufficient for 600W as long as they are on seperate sets of PSU leads. The problems experienced by [very few] bitfury users were almost exclusively caused by using a single lead with 2 plugs on it.

The config allows for 6+ and 6- wires. This is sufficient because in comparison, I have antminer systems running happily using only 4+ and 4- wires from the ATX leads and that is around 500W

24" PCI-E cables with 16AWG wires and stripped ends - great for server PSU mods, best prices https://bitcointalk.org/index.php?topic=563461 also selling 6" M-F-M PCIe splitters and PCIe-PCIe
No longer a wannabe - now an ASIC owner!
DPoS
Sr. Member
****
Offline Offline

Activity: 462



View Profile
December 28, 2013, 06:02:00 AM
 #5286

new postings for 15 days ago??  still open?

https://angel.co/jobs?slug=friends-family-and-founders&startup_id=271640


JOBS
Senior Software Wizard
$120k - $160k · 0.25 - 0.5%
Full Time · San Francisco · Software Engineer · Embedded Systems · C

EE Power supply designer, PCB design
$100k - $140k · 0.25 - 0.5%
Full Time · San Francisco · Hardware Engineer · Electrical Engineering

Supply Chain Management
$60k - $130k · 0.15 - 0.5%
Full Time · San Francisco · Operations · Supply Chain Management


Active 15 days ago
Refer a friend · Report this company

~~BTC~~GAMBIT~~BTC~~Play Boardgames for Bitcoins!!~~BTC~~GAMBIT~~BTC~~ Something I say help? Donate BTC! 1KN1K1xStzsgfYxdArSX4PEjFfcLEuYhid
minternj
Sr. Member
****
Offline Offline

Activity: 434


View Profile
December 28, 2013, 06:09:07 AM
 #5287

https://twitter.com/HashFast/status/416810773290430464

New update.

Warning about Nitrogensports.eu
https://bitcointalk.org/index.php?topic=709114.0
jimmothy
Hero Member
*****
Offline Offline

Activity: 770



View Profile
December 28, 2013, 06:13:02 AM
 #5288


800gh/s per baby jet? Maybe everyone doesn't have to bring out the pitchforks after all?
jspielberg
Sr. Member
****
Offline Offline

Activity: 406


View Profile
December 28, 2013, 06:31:13 AM
 #5289


800gh/s per baby jet? Maybe everyone doesn't have to bring out the pitchforks after all?

Hashrate looks a little unstable at 800GH, and we don't know the power numbers at that rate, but I agree.  800GH/s with an MPP in early February brings the limetime loss to 16BTC/BJ rather than the eye watering 27BTC/BJ.  While a 75% ROI is less than break-even, it is better than 48% we were looking at earlier.

aerobatic
Hero Member
*****
Offline Offline

Activity: 702


View Profile
December 28, 2013, 06:35:06 AM
 #5290


800gh/s per baby jet? Maybe everyone doesn't have to bring out the pitchforks after all?

Hashrate looks a little unstable at 800GH, and we don't know the power numbers at that rate, but I agree.  800GH/s with an MPP in early February brings the limetime loss to 16BTC/BJ rather than the eye watering 27BTC/BJ.  While a 75% ROI is less than break-even, it is better than 48% we were looking at earlier.

hey guys.. don't assume the 800 GH average rate is for ONE module...  

there's really little chance that a chip designed for 400 GH could be over clocked to 800 GH, so assume this is actually two GN modules hashing at 400 GH each...  (edit: they're laying claim to 700+ GH on their tweets so I'm happy to admit i was worrying unnecessarily and they may well have achieved a significant over clock performance)
ineededausername
Hero Member
*****
Offline Offline

Activity: 784


bitcoin hundred-aire


View Profile
December 28, 2013, 06:36:06 AM
 #5291


800gh/s per baby jet? Maybe everyone doesn't have to bring out the pitchforks after all?

Hashrate looks a little unstable at 800GH, and we don't know the power numbers at that rate, but I agree.  800GH/s with an MPP in early February brings the limetime loss to 16BTC/BJ rather than the eye watering 27BTC/BJ.  While a 75% ROI is less than break-even, it is better than 48% we were looking at earlier.

hey guys.. don't assume the 800 GH average rate is for ONE module... 

there's really little chance that a chip designed for 400 GH could be over clocked to 800 GH, so assume this is actually two GN modules hashing at 400 GH each...



But it was hashing at 600 GH/s earlier.... Shocked

(BFL)^2 < 0
dhenson
Legendary
*
Offline Offline

Activity: 994



View Profile
December 28, 2013, 06:37:35 AM
 #5292


800gh/s per baby jet? Maybe everyone doesn't have to bring out the pitchforks after all?

That remains to be seen.  They've just posted the update to Twitter and nothing is currently hashing to that address on Eligius.

Cautiously optimistic. Considering they were apparently having power issues with the old board, it wouldn't make sense that a single BJ would have been hashing at ~650 up until yesterday.  We very well might be looking at the history of multiple rigs.
DeathAndTaxes
Donator
Legendary
*
Offline Offline

Activity: 1218


Gerald Davis


View Profile
December 28, 2013, 06:58:40 AM
 #5293


800gh/s per baby jet? Maybe everyone doesn't have to bring out the pitchforks after all?

Hashrate looks a little unstable at 800GH, and we don't know the power numbers at that rate, but I agree.  800GH/s with an MPP in early February brings the limetime loss to 16BTC/BJ rather than the eye watering 27BTC/BJ.  While a 75% ROI is less than break-even, it is better than 48% we were looking at earlier.

hey guys.. don't assume the 800 GH average rate is for ONE module... 

there's really little chance that a chip designed for 400 GH could be over clocked to 800 GH, so assume this is actually two GN modules hashing at 400 GH each...



It may be two modules but saying there is little chance of a 100% boost over design spec is well silly since we have a prior competitor doing exactly that.

KNC design spec was 300 GH/s my Nov Jupiter runs not at 500 GH/s, not at 600 GH/s but 672 GH/s.   Is that also impossible?
dhenson
Legendary
*
Offline Offline

Activity: 994



View Profile
December 28, 2013, 07:03:25 AM
 #5294

We shall soon know, it looks like they were waiting for the old stats to clear before turning it on.  Looking good so far.

http://eligius.st/~wizkid057/newstats/userstats.php/1KSZMq1L9ZWEBKEHuW7ACXWZJYdxiTdaVt
jspielberg
Sr. Member
****
Offline Offline

Activity: 406


View Profile
December 28, 2013, 07:08:33 AM
 #5295

Their last tweet says:
Quote
Happy about the Eligius Baby Jet performance at 600+GH/s on December
24, 25, 26.  Check it out at http://ow.ly/s6MqP

Interesting that they didn't mention the 700+ on the 27th?  2 modules or just 1?

aerobatic
Hero Member
*****
Offline Offline

Activity: 702


View Profile
December 28, 2013, 07:13:45 AM
 #5296

It may be two modules but saying there is little chance of a 100% boost over design spec is well silly since we have a prior competitor doing exactly that.

KNC design spec was 300 GH/s my Nov Jupiter runs not at 500 GH/s, not at 600 GH/s but 672 GH/s.   Is that also impossible?

there's a difference.  KnC didn't have a 300 GH system (chip was 100 GH) hashing at 600 GH.   They had deliberately under quoted its intended performance so that when they knew the final performance everyone would be happy.   i.e.: they under promised and over delivered... just like they intended.   knC always knew it would perform faster than their quoted original performance...

I could be wrong.. but i honestly don't think that something designed for 400 GH (at the extremes of performance and heat dissipation) will over clock to 800 GH.   I think the previously announced 500 GH is much closer to the mark.. with a max of, say, 550-575 GH.  (edit:  hashfast seems to be saying that 600 GH is achievable)
Minor Miner
Legendary
*
Offline Offline

Activity: 1050



View Profile
December 28, 2013, 07:18:23 AM
 #5297

800gh/s per baby jet? Maybe everyone doesn't have to bring out the pitchforks after all?
Hashrate looks a little unstable at 800GH, and we don't know the power numbers at that rate, but I agree.  800GH/s with an MPP in early February brings the limetime loss to 16BTC/BJ rather than the eye watering 27BTC/BJ.  While a 75% ROI is less than break-even, it is better than 48% we were looking at earlier.
hey guys.. don't assume the 800 GH average rate is for ONE module... 
there's really little chance that a chip designed for 400 GH could be over clocked to 800 GH, so assume this is actually two GN modules hashing at 400 GH each...
But it was hashing at 600 GH/s earlier.... Shocked
Dude,
but with how much power?   you have to get the watts to those boards but can anyone find that many amps also?

minternj
Sr. Member
****
Offline Offline

Activity: 434


View Profile
December 28, 2013, 07:18:42 AM
 #5298

They said 500 was underclocked.
"HALF A TERAHASH/s (500 GH/s) on a single chip, while still running underclocked. "
https://hashfast.com/hashfast-announces-fastest-bitcoin-mining-chip-in-the-world/

Warning about Nitrogensports.eu
https://bitcointalk.org/index.php?topic=709114.0
cedivad
Legendary
*
Offline Offline

Activity: 1134



View Profile
December 28, 2013, 07:20:07 AM
 #5299

They are most probably trying to use the smoke machine. Even if it does at 800GH stably (and i don't see that happen in this world), they will delay to death the MPP. If they want me to consider not to start the process that will end on a litigation on the first of Jan i want the MPP to have a guaranteed delivery date, and i want that NOW.

I don't like people on this thread telling me what to do. I will go after a full BTC refund, i will sue them to death, i don't care. It's just the right thing to do. I will lose everything in the meanwhile? I'm prepared for that.

My anger against what is wrong in the Bitcoin community is productive:
Bitcointa.lk - Replace "Bitcointalk.org" with "Bitcointa.lk" in this url to see how this page looks like on a proper forum (Announcement Thread)
Hashfast.org - Wiki for screwed customers
Minor Miner
Legendary
*
Offline Offline

Activity: 1050



View Profile
December 28, 2013, 07:21:50 AM
 #5300

They are most probably trying to use the smoke machine. Even if it does at 800GH stably (and i don't see that happen in this world), they will delay to death the MPP. If they want me to consider not to start the process that will end on a litigation on the first of Jan i want the MPP to have a guaranteed delivery date, and i want that NOW.
MPP?Huh?
If the chip is performing at that rate your MPP complaints are moot.   You are already getting more than you ever dreamed of

Pages: « 1 ... 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 [265] 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 ... 586 »
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!