[ANN] First 500Gh/s BFL unit up and running!

[Hippie Tech]

It's been up and down numerous times as I've been chipping away at the code to make it work at its theoretical maximum 480GH.

It's mostly working now. Here's what I've got:

Code:

 cgminer version 3.2.2 - Started: [2013-06-19 02:17:06]
--------------------------------------------------------------------------------
 (5s):477.6G (avg):472.1Gh/s | A:16  R:0  HW:193  U:10.8/m  WU:6575.3/m
 ST: 2  SS: 0  NB: 2  LW: 11012  GF: 0  RF: 0
 Connected to XXX diff 500
 Block: 00179d4404ca1359...  Diff:19.3M  Started: [02:18:28]  Best share: 9.56K
--------------------------------------------------------------------------------
 [P]ool management [S]ettings [D]isplay options [Q]uit
 BAS 0:  max 68C 3.28V | 56.23G/58.22Gh/s | A:1 R:0 HW:18 U: 0.67/m
 BAS 1:  max 69C 3.27V | 58.92G/60.48Gh/s | A:1 R:0 HW:23 U: 0.67/m
 BAS 2:  max 74C 3.27V | 60.53G/59.85Gh/s | A:4 R:0 HW:19 U: 2.69/m
 BAS 3:  max 80C 3.28V | 62.52G/61.58Gh/s | A:1 R:0 HW:20 U: 0.67/m
 BAS 4:  max 86C 3.27V | 74.56G/59.90Gh/s | A:4 R:0 HW:39 U: 2.69/m
 BAS 5:  max 71C 3.28V | 64.25G/58.79Gh/s | A:1 R:0 HW:25 U: 0.67/m
 BAS 6:  max 81C 3.27V | 59.93G/58.98Gh/s | A:2 R:0 HW:25 U: 1.34/m
 BAS 7:  max 72C 3.27V | 43.22G/60.81Gh/s | A:2 R:0 HW:25 U: 1.34/m
--------------------------------------------------------------------------------

Note the hardware error count may appear very high, but bear in mind that it's mining at diff 500

That's just too hot, 86C degree??? Delta fans needed  

it may not be accurate because there are a lot of units in that room heating up the room temp so who knows what it would when you have it

Considering it just started.. notice the share count.. thats pretty damn hot.

[ultrix]

It's been up and down numerous times as I've been chipping away at the code to make it work at its theoretical maximum 480GH.

It's mostly working now. Here's what I've got:

Code:

 cgminer version 3.2.2 - Started: [2013-06-19 02:17:06]
--------------------------------------------------------------------------------
 (5s):477.6G (avg):472.1Gh/s | A:16  R:0  HW:193  U:10.8/m  WU:6575.3/m
 ST: 2  SS: 0  NB: 2  LW: 11012  GF: 0  RF: 0
 Connected to XXX diff 500
 Block: 00179d4404ca1359...  Diff:19.3M  Started: [02:18:28]  Best share: 9.56K
--------------------------------------------------------------------------------
 [P]ool management [S]ettings [D]isplay options [Q]uit
 BAS 0:  max 68C 3.28V | 56.23G/58.22Gh/s | A:1 R:0 HW:18 U: 0.67/m
 BAS 1:  max 69C 3.27V | 58.92G/60.48Gh/s | A:1 R:0 HW:23 U: 0.67/m
 BAS 2:  max 74C 3.27V | 60.53G/59.85Gh/s | A:4 R:0 HW:19 U: 2.69/m
 BAS 3:  max 80C 3.28V | 62.52G/61.58Gh/s | A:1 R:0 HW:20 U: 0.67/m
 BAS 4:  max 86C 3.27V | 74.56G/59.90Gh/s | A:4 R:0 HW:39 U: 2.69/m
 BAS 5:  max 71C 3.28V | 64.25G/58.79Gh/s | A:1 R:0 HW:25 U: 0.67/m
 BAS 6:  max 81C 3.27V | 59.93G/58.98Gh/s | A:2 R:0 HW:25 U: 1.34/m
 BAS 7:  max 72C 3.27V | 43.22G/60.81Gh/s | A:2 R:0 HW:25 U: 1.34/m
--------------------------------------------------------------------------------

Note the hardware error count may appear very high, but bear in mind that it's mining at diff 500

That's just too hot, 86C degree??? Delta fans needed  

it may not be accurate because there are a lot of units in that room heating up the room temp so who knows what it would when you have it

Considering it just started.. notice the share count.. thats pretty damn hot.

Pretty bad when hardware error rate exceeds accepted count.

[Sitarow]

Fellow Bitcoiners,

It has been a long road to get to this point but I am pleased to announce that the first 500Gh/s BFL units have been delivered and are happily hashing away as I type this. This is a huge milestone for BFL and shows their continued efforts to fulfill their promises to customers.



http://kcmo2.gigamining.com/

The picture above shows multiple FPGA minirigs along with 1 ASIC minirig.

The rig will be up and down for the evening while conman works to get cgminer fine tuned with the new firmware.

Best,
James

Congrats.

[PuertoLibre]

It's been up and down numerous times as I've been chipping away at the code to make it work at its theoretical maximum 480GH.

Why 480 and not 500?

It behaves like 8 60GH singles instead of some separate unit. I have no idea about their claims and whatnot.

Nah, don't worry about it. I am just wondering [rhetorically] why they didn't spec test the unit before sending it out.

Keep us posted on any developments.

[PuertoLibre]

I will buy it right now for 100.000 BitBar! I hear they are worth 7 BTC each and come with Scrypt FPGA so what do you say? 

Tell us more?

[crazyates]

It's been up and down numerous times as I've been chipping away at the code to make it work at its theoretical maximum 480GH.

It's mostly working now. Here's what I've got:

Code:

 cgminer version 3.2.2 - Started: [2013-06-19 02:17:06]
--------------------------------------------------------------------------------
 (5s):477.6G (avg):472.1Gh/s | A:16  R:0  HW:193  U:10.8/m  WU:6575.3/m
 ST: 2  SS: 0  NB: 2  LW: 11012  GF: 0  RF: 0
 Connected to XXX diff 500
 Block: 00179d4404ca1359...  Diff:19.3M  Started: [02:18:28]  Best share: 9.56K
--------------------------------------------------------------------------------
 [P]ool management [S]ettings [D]isplay options [Q]uit
 BAS 0:  max 68C 3.28V | 56.23G/58.22Gh/s | A:1 R:0 HW:18 U: 0.67/m
 BAS 1:  max 69C 3.27V | 58.92G/60.48Gh/s | A:1 R:0 HW:23 U: 0.67/m
 BAS 2:  max 74C 3.27V | 60.53G/59.85Gh/s | A:4 R:0 HW:19 U: 2.69/m
 BAS 3:  max 80C 3.28V | 62.52G/61.58Gh/s | A:1 R:0 HW:20 U: 0.67/m
 BAS 4:  max 86C 3.27V | 74.56G/59.90Gh/s | A:4 R:0 HW:39 U: 2.69/m
 BAS 5:  max 71C 3.28V | 64.25G/58.79Gh/s | A:1 R:0 HW:25 U: 0.67/m
 BAS 6:  max 81C 3.27V | 59.93G/58.98Gh/s | A:2 R:0 HW:25 U: 1.34/m
 BAS 7:  max 72C 3.27V | 43.22G/60.81Gh/s | A:2 R:0 HW:25 U: 1.34/m
--------------------------------------------------------------------------------

Note the hardware error count may appear very high, but bear in mind that it's mining at diff 500

Pretty bad when hardware error rate exceeds accepted count.

I want you to switch your miners to a diff=500, and see how many shares you submit to your pool. Hint, my FPGA Single would get less than 40 shares a day at that diff.

[PuertoLibre]

Fellow Bitcoiners,

It has been a long road to get to this point but I am pleased to announce that the first 500Gh/s BFL units have been delivered and are happily hashing away as I type this. This is a huge milestone for BFL and shows their continued efforts to fulfill their promises to customers.



http://kcmo2.gigamining.com/

The picture above shows multiple FPGA minirigs along with 1 ASIC minirig.

The rig will be up and down for the evening while conman works to get cgminer fine tuned with the new firmware.

Best,
James

Congrats.

Oh,

I get the joke!

I was thinking they were all ASIC miners....

Now I finally understand that only one of them is the latest generation of ASIC while the others that look similar are just the older FPGA version from the previous generation.

I wonder if GigaVPS made the pic that way to give a false impression?

Which one is of the three or four is the actual current generation ASIC?

[papaminer]

im so jealous...

i have seen gigavps mining rigs before... and here it is again....

a new set of money making machines..

[Luke-Jr]

I wonder if GigaVPS made the pic that way to give a false impression?

How is it gigavps's fault that you can't read??  

[fran2k]

Pretty bad when hardware error rate exceeds accepted count.

Only 6% accepted. wtf!

[Syke]

"Came through"?! It was supposed to be 1500 GH/s @ 1500 watts in October '12. Now it's 500 GH/s at 2400 watts in June '13. They didn't "come through". They blew it.

A) They will be compensating customers with 3x the number of units, so you will be getting the same hashrate as what you paid for.

B) Lets see you do any better.

2000 BTC invested in a BFL pre-order 1 year ago now generates 13 BTC a day! Wow!!! This *might* break even someday.

100 BTC invested in an Avalon pre-order 9 months ago has already made back its investment several times over.

[juhakall]

Pretty bad when hardware error rate exceeds accepted count.

Only 6% accepted. wtf!

http://kcmo2.gigamining.com/?ref=60&rig=2

Compare difficulty accepted with rejected+hardware errors.

[roy7]

Pretty bad when hardware error rate exceeds accepted count.

How does difficulty play into that though? It's set at 500. So it is really 16*500 shares vs 193 errors?

[crazyates]

Pretty bad when hardware error rate exceeds accepted count.

How does difficulty play into that though? It's set at 500. So it is really 16*500 shares vs 193 errors?

Exactly. Which turns out to be ~3%.

[PuertoLibre]

[PuertoLibre]

I wonder if GigaVPS made the pic that way to give a false impression?

How is it gigavps's fault that you can't read??  

Ah, luke:

« Last Edit: June 18, 2013, 07:17:33 PM by gigavps »

The first OP post was edited after it was made.

Also, I understood there was three components to a BFL [ASIC] minirig. I saw several so I assumed GigaVPS was showing more than 1/3.

So far, (from looking at the plugs) I only see 1/3rd of a minirig being pictured. Where are the other two units?

[Syke]

So far, (from looking at the plugs) I only see 1/3rd of a minirig being pictured. Where are the other two units?

In Inaba's farm.

[PuertoLibre]

So far, (from looking at the plugs) I only see 1/3rd of a minirig being pictured. Where are the other two units?

In Inaba's farm.

That explains this message from him then.

https://bitcointalk.org/index.php?topic=89902.msg2456884#msg2456884

Ok folks... we are at a critical juncture here and I'm looking for input on how to move forward.

I am not able to host all the singles, given the new power requirements.  I would have to split it into separate DC's and even if I were to keep them in the current locations, the increased power cost far exceeds the 10% held back on the units. In short, I simply can't operate this number of singles even at break even on my part.

Obviously, this is an untenable situation for me.  I'm considering invoking the original clause:

Should mining become unprofitable or other circumstances cause the operator to be unable or unwilling to perform his duties, sale of all hardware and equal distribution of funds will commence according to the number of shares held by each individual.

I am obviously not willing to operate at a continual loss.  If anyone has suggestions on moving forward, I'm open to them.

[-ck]

Pretty bad when hardware error rate exceeds accepted count.

How does difficulty play into that though? It's set at 500. So it is really 16*500 shares vs 193 errors?

Exactly. Which turns out to be ~3%.

Yes and as always, when you first start it, it looks worse. Here it is after running for ~2hrs without changing the code.

Code:

 cgminer version 3.2.2 - Started: [2013-06-19 03:21:34]
--------------------------------------------------------------------------------
 (5s):479.5G (avg):478.1Gh/s | A:1502  R:0  HW:5182  U:12.7/m  WU:6662.8/m
 ST: 2  SS: 0  NB: 14  LW: 824569  GF: 0  RF: 0
 Connected to XXX diff 500
 Block: 005c9eedec8c830c...  Diff:19.3M  Started: [05:12:05]  Best share: 914K
--------------------------------------------------------------------------------
 [P]ool management [S]ettings [D]isplay options [Q]uit
 BAS 0:  max 68C 3.28V | 47.90G/58.20Gh/s | A:186 R:0 HW: 337 U: 1.58/m
 BAS 1:  max 70C 3.28V | 67.61G/60.23Gh/s | A:180 R:0 HW: 674 U: 1.53/m
 BAS 2:  max 76C 3.27V | 68.62G/59.84Gh/s | A:200 R:0 HW: 550 U: 1.70/m
 BAS 3:  max 80C 3.27V | 46.95G/61.31Gh/s | A:198 R:0 HW: 644 U: 1.68/m
 BAS 4:  max 86C 3.27V | 56.42G/60.44Gh/s | A:178 R:0 HW:1274 U: 1.51/m
 BAS 5:  max 71C 3.28V | 48.24G/58.86Gh/s | A:183 R:0 HW: 542 U: 1.55/m
 BAS 6:  max 82C 3.27V | 56.12G/58.79Gh/s | A:193 R:0 HW: 347 U: 1.64/m
 BAS 7:  max 77C 3.26V | 60.53G/60.45Gh/s | A:184 R:0 HW: 815 U: 1.56/m
--------------------------------------------------------------------------------

Accepted is 1502, so 1502 * 500 = 751000 accepted diff 1 shares, so HW errors are 5182 or 5182 / 751000 * 100 ~0.7%

Number 4 definitely runs  hotter, and consequently has the highest HW error rate.

[jamesg]

I wonder if GigaVPS made the pic that way to give a false impression?

Really? 

I took a picture of my setup in the datacenter to share and you think I'm trying to give "false impressions"?   

Also, I edited the OP to add the link to the stats page.