Discontinued: Lancelot (Heavy Duty 400 Mega hashes, 2 x FPGA Mining Board)

[tyler26]

Any idea of when I will get an email saying they are back in stock? I went to order some this weekend, got an email saying I will receive notice when they have more available.

[LaserHorse]

Any idea of when I will get an email saying they are back in stock? I went to order some this weekend, got an email saying I will receive notice when they have more available.

Aah - I suppose that means the "In stock" label on the product page isn't quite accurate?

[Pistachio]

Another easy way to mine with these is Bitminter. It loaded right up and found all the cards.

What happens with Bitminer if you remove a Lancelot and plug it back in? How about if you add more Lancelot cards while running?

cgminer is not currently working as one would like. We have sent today a Lancelot to Kano so he can fix this.

There is an option to auto search for new com ports which recognizes new cards or current cards that are turned off and back on. But you will receive com port errors from the USB hob or some other hardware that is recognized by bitminter. It does not cause a problem but will show up every time it auto checks for new ports. I had the same problem with cgminer 3.1. However, it was my fault for adding -S com1 when it was unnecessary. I only needed com 3 through 8. The com1 error caused cgminer 3.1 to crash because it thought it was an Avalon. I did not have this problem on cgminer 2.1 or bfgminer.

[kano]

...
There is an option to auto search for new com ports which recognizes new cards or current cards that are turned off and back on. But you will receive com port errors from the USB hob or some other hardware that is recognized by bitminter. It does not cause a problem but will show up every time it auto checks for new ports. I had the same problem with cgminer 3.1. However, it was my fault for adding -S com1 when it was unnecessary. I only needed com 3 through 8. The com1 error caused cgminer 3.1 to crash because it thought it was an Avalon. I did not have this problem on cgminer 2.1 or bfgminer.

As I have already said ...
cgminer still uses the old serial-USB for the Icarus driver.
Once I get the Lancelot (today or tomorrow) I'll work on updating the Icarus driver to direct USB where my hotplug code solves the major issues of device detection.
Before direct USB, cgminer didn't do hotplug (there are various major issues attempting to do this with termios serial devices)
At the moment on cgminer it isn't direct USB so the Icarus driver won't see a hotplugged device.

[Pistachio]

As I have already said ...
cgminer still uses the old serial-USB for the Icarus driver.
Once I get the Lancelot (today or tomorrow) I'll work on updating the Icarus driver to direct USB where my hotplug code solves the major issues of device detection.
Before direct USB, cgminer didn't do hotplug (there are various major issues attempting to do this with termios serial devices)
At the moment on cgminer it isn't direct USB so the Icarus driver won't see a hotplugged device.

Thanks for working on this.

[blackarrow]

Wondering - do these boards have much resale value (for non-mining purposes)?

I'm fairly green to planet FPGA, unfamiliar with features/specs which are desirable to devs.

Yes, they are really powerful boards and selling the FPGAs alone will get you the money back.

[kano]

Just a quick post to say yep I got the board half an hour ago.
Thanks again blackarrow!

Fired up with the current driver with short timing.
Icarus 0 is an Icarus, Icarus 1 is the new Lancelot:

Code:

    [2013-05-16 17:19:29] Icarus 0 Re-estimate: Hs=2.631364e-09 W=6.278043e-03 read_count=112 fullnonce=11.308s
    [2013-05-16 17:19:47] Icarus 1 Re-estimate: Hs=2.499979e-09 W=6.630210e-03 read_count=106 fullnonce=10.744s
    [2013-05-16 17:20:33] Icarus 0 Re-estimate: Hs=2.631586e-09 W=6.193042e-03 read_count=112 fullnonce=11.309s
    [2013-05-16 17:21:21] Icarus 1 Re-estimate: Hs=2.499976e-09 W=6.668599e-03 read_count=106 fullnonce=10.744s
    [2013-05-16 17:24:42] Icarus 0 Re-estimate: Hs=2.631625e-09 W=6.103034e-03 read_count=112 fullnonce=11.309s
    [2013-05-16 17:25:02] Icarus 1 Re-estimate: Hs=2.500001e-09 W=6.790968e-03 read_count=106 fullnonce=10.744s

So clearly getting 400MH/s out of it

OK time to work on the USB code ...

Edit: anyone wondering what the settings for cgminer are with the current driver: --icarus-timing 2.5
To get the timing information above use: --icarus-timing short or as per FPGA-README

Edit2: and a picture or 2 ... click for 800x600

[-ck]

Well I would have posted about receiving mine today too, but I waited all day only to see the tracking change from

Code:

13 	With delivery courier 	MELBOURNE - AUSTRALIA 	13:42

to

Code:

14 	Address information needed; contact DHL 	MELBOURNE - AUSTRALIA 	16:04

When I called them, they said the driver saw "NIC" on the address and didn't know what to do with it... when clearly that was meant to read "VIC" (my state) which it had already made it to since it was in my state's capital city. Despite the fact they had the exact address and postcode, they didn't even try to deliver it. Sounds more like the driver was looking for any old excuse to not deliver the damn thing  

[Xialla]

Edit: anyone wondering what the settings for cgminer are with the current driver: --icarus-timing 2.5

thanks for hint, now mining all on my boards on 395Mh/s in average.) (previously in was around 378 without any timing settings)

[kano]

Edit: anyone wondering what the settings for cgminer are with the current driver: --icarus-timing 2.5

thanks for hint, now mining all on my boards on 395Mh/s in average.) (previously in was around 378 without any timing settings)

Yeah what was happening for you before (without that setting) is each one would go idle at the end of the nonce range for almost 0.6s if it didn't find a nonce.

2.5 comes up with 10.6s (instead of 11.2s) and of course 2.5 means 2.5ns per hash = 400MH/s

The timing code works that out and shows those numbers I have above (while it mines)
The only problem with the timing code is if your computer is busy it can mess up the numbers since it works it out based on the time to do work - which can be delayed on a busy system.

[@Zack]

Wondering - do these boards have much resale value (for non-mining purposes)?

I'm fairly green to planet FPGA, unfamiliar with features/specs which are desirable to devs.

Yes, they are really powerful boards and selling the FPGAs alone will get you the money back.

What is the resale value of this FPGA after use it 1 year ?
Is it at the same price ?

[blackarrow]

  [2013-05-16 17:19:29] Icarus 0 Re-estimate: Hs=2.631364e-09 W=6.278043e-03 read_count=112 fullnonce=11.308s
    [2013-05-16 17:19:47] Icarus 1 Re-estimate: Hs=2.499979e-09 W=6.630210e-03 read_count=106 fullnonce=10.744s
    [2013-05-16 17:20:33] Icarus 0 Re-estimate: Hs=2.631586e-09 W=6.193042e-03 read_count=112 fullnonce=11.309s
    [2013-05-16 17:21:21] Icarus 1 Re-estimate: Hs=2.499976e-09 W=6.668599e-03 read_count=106 fullnonce=10.744s
    [2013-05-16 17:24:42] Icarus 0 Re-estimate: Hs=2.631625e-09 W=6.103034e-03 read_count=112 fullnonce=11.309s
    [2013-05-16 17:25:02] Icarus 1 Re-estimate: Hs=2.500001e-09 W=6.790968e-03 read_count=106 fullnonce=10.744s

How can we see this log?

Edit: anyone wondering what the settings for cgminer are with the current driver: --icarus-timing 2.5

What would be the correct setting for even faster version of the code? When we've tried the v42 firmware we got poorer performance. Now I wonder if this was the issue.

[alfabitcoin]

To get the timing information above use: --icarus-timing short or as per FPGA-README

[FuzzyBear]

Just a quick post to say yep I got the board half an hour ago.
Thanks again blackarrow!

Fired up with the current driver with short timing.
Icarus 0 is an Icarus, Icarus 1 is the new Lancelot:

Code:

    [2013-05-16 17:19:29] Icarus 0 Re-estimate: Hs=2.631364e-09 W=6.278043e-03 read_count=112 fullnonce=11.308s
    [2013-05-16 17:19:47] Icarus 1 Re-estimate: Hs=2.499979e-09 W=6.630210e-03 read_count=106 fullnonce=10.744s
    [2013-05-16 17:20:33] Icarus 0 Re-estimate: Hs=2.631586e-09 W=6.193042e-03 read_count=112 fullnonce=11.309s
    [2013-05-16 17:21:21] Icarus 1 Re-estimate: Hs=2.499976e-09 W=6.668599e-03 read_count=106 fullnonce=10.744s
    [2013-05-16 17:24:42] Icarus 0 Re-estimate: Hs=2.631625e-09 W=6.103034e-03 read_count=112 fullnonce=11.309s
    [2013-05-16 17:25:02] Icarus 1 Re-estimate: Hs=2.500001e-09 W=6.790968e-03 read_count=106 fullnonce=10.744s

So clearly getting 400MH/s out of it

OK time to work on the USB code ...

Edit: anyone wondering what the settings for cgminer are with the current driver: --icarus-timing 2.5
To get the timing information above use: --icarus-timing short or as per FPGA-README

Edit2: and a picture or 2 ... click for 800x600

Nice, this should mine well alongside ur BFL Jally... u got one of those to right?

[daemon]

Hello
I paid 10 Lancelot board 07/05/2013
Only today received notification of sending order:
> Your order has been shipped with EMS Express.
> DHL Tracking number: EE_________CN

Why is EMS Express? This is a mistake?
With EMC Express in my country parcel may go two or three months. This is not acceptable.
In order for your website Set DHL shipping method, so I was confident that you are working with DHL.

[FlappySocks]

Got my notification yesterday:
"Your order has been shipped with DHL Express."

Been stuck on "Shipment information received" for almost 24 hours.

[kano]

...
Nice, this should mine well alongside ur BFL Jally... u got one of those to right?

I'm still working on the code, but have 3 ICA devices connected with the Jalapeno waiting to test the USB code:

cgminer version 3.1.1f - Started: [2013-05-16 22:17:02]
--------------------------------------------------------------------------------
 (5s):7.031G (avg):6.522Gh/s | A:22  R:0  HW:1  U:0.5/m  WU:89.8/m
 ST: 2  SS: 0  NB: 6  LW: 5092  GF: 0  RF: 0
 Connected to au.ozco.in diff 256 with stratum as user
 Block: 0157b0dd63f5d36d...  Diff:11.2M  Started: [22:53:28]  Best share: 39.2K
--------------------------------------------------------------------------------
 [P]ool management [S ]ettings [D]isplay options [Q]uit
 ICA 0:                | 335.5M/337.0Mh/s | A: 1 R:0 HW:0 U: 0.02/m
 ICA 1:                | 379.7M/379.2Mh/s | A: 2 R:0 HW:0 U: 0.04/m
 ICA 2:                | 399.4M/399.1Mh/s | A: 2 R:0 HW:0 U: 0.04/m
 BAJ 0:  max 45C 3.48V | 5.574G/5.407Gh/s | A:17 R:0 HW:1 U: 0.36/m
--------------------------------------------------------------------------------
.
.
.
 [2013-05-16 22:45:27] Icarus 1 Re-estimate: Hs=2.631610e-09 W=6.303242e-03 read_count=112 fullnonce=11.309s
 [2013-05-16 22:47:41] Icarus 0 Re-estimate: Hs=2.976137e-09 W=6.770804e-03 read_count=126 fullnonce=12.789s
 [2013-05-16 22:48:36] Icarus 2 Re-estimate: Hs=2.500001e-09 W=6.720939e-03 read_count=106 fullnonce=10.744s

The ICA are: 0=AsicminerUSB 1=Icarus 2=Lancelot - the fastest

I'm planning on running everything on a single rig, once I get this working, to see how it all goes together.
I also have 1xBFLFPGA, 1xModMinerQuad and 1x6950

[kano]

  [2013-05-16 17:19:29] Icarus 0 Re-estimate: Hs=2.631364e-09 W=6.278043e-03 read_count=112 fullnonce=11.308s
    [2013-05-16 17:19:47] Icarus 1 Re-estimate: Hs=2.499979e-09 W=6.630210e-03 read_count=106 fullnonce=10.744s
    [2013-05-16 17:20:33] Icarus 0 Re-estimate: Hs=2.631586e-09 W=6.193042e-03 read_count=112 fullnonce=11.309s
    [2013-05-16 17:21:21] Icarus 1 Re-estimate: Hs=2.499976e-09 W=6.668599e-03 read_count=106 fullnonce=10.744s
    [2013-05-16 17:24:42] Icarus 0 Re-estimate: Hs=2.631625e-09 W=6.103034e-03 read_count=112 fullnonce=11.309s
    [2013-05-16 17:25:02] Icarus 1 Re-estimate: Hs=2.500001e-09 W=6.790968e-03 read_count=106 fullnonce=10.744s

How can we see this log?

Edit: anyone wondering what the settings for cgminer are with the current driver: --icarus-timing 2.5

What would be the correct setting for even faster version of the code? When we've tried the v42 firmware we got poorer performance. Now I wonder if this was the issue.

Yes as mentioned above --icarus-timing short
That's the easiest way to get maximum performance of a device you aren't sure of the correct numbers.

However, it also requires the computer it's running on to be stable - i.e. cgminer can't be swapped out ever or the CPU be unavailable ever, since that will affect the timing code.
short is only run for about an hour and uses the answer at that point onwards until you stop cgminer.
long keeps recalculating the answer forever.

As mentioned, there's more details in FPGA-README.

[blackarrow]

Why is EMS Express? This is a mistake?
With EMC Express in my country parcel may go two or three months. This is not acceptable.
In order for your website Set DHL shipping method, so I was confident that you are working with DHL.

We had to ship with EMS. Since 2 or 3 years ago all courier companies stopped working with Russia. You are welcome to check this information if you do not believe us.

[phm]

Hi guys, I received my order today, only single unit but I'm very happy with it. When it comes to mining I'm a total newbie, but everything looks good, so far I earned 0.00162255 BTC . Good work, Black Arrow Software.