Klondike - 16 chip ASIC Open Source Board - Preliminary

[Bogart]

@Bogart
Do you have a K16 board with 16 chips installed ?

I do.  Several actually.

[fasmax]

It looks like your not getting any accepted shares.
Has that timing capacitor been replaced on your boards ?
 

[Bogart]

I don't know, and I don't know how to tell.  I just asked sensei if he could go ahead and ship me the boards, minus and he needed to continue debugging.  That would have been mid-last-week he shipped them.

[fasmax]

Do you have any way to make changes when sensei finds things that need to be changed or updated ?

[cp1]

Send me one to debug   Hopefully it's just firmware.

[freeworm]

@bogart, replace C274 with a 220p, or 260p capacitor

I also have several boards with all 16 chips installed. they are working with some HW when clocked over 300MHz.
If someone is interested, please PM me.

[hf_developer]

It looks like your not getting any accepted shares.
Has that timing capacitor been replaced on your boards ?
 

The klondike design is very sensitive with the nor gates and the phase shifter in front of it. It is a matter of the manufacturer of the nor gates, as well as capacitor/resistor. For less HW errors, the Tau of R and C has to fit the propagation and switching times of the nor gate, and these are sometimes higher, sometimes lower, and changing with every type of nor gate. So very hard to make this working properly....

[freeworm]

It looks like your not getting any accepted shares.
Has that timing capacitor been replaced on your boards ?
 

The klondike design is very sensitive with the nor gates and the phase shifter in front of it. It is a matter of the manufacturer of the nor gates, as well as capacitor/resistor. For less HW errors, the Tau of R and C has to fit the propagation and switching times of the nor gate, and these are sometimes higher, sometimes lower, and changing with every type of nor gate. So very hard to make this working properly....

In this case, it would be very easy to make it work by changing the R or C to meet Tau for a particular nor gate.

[hf_developer]

It looks like your not getting any accepted shares.
Has that timing capacitor been replaced on your boards ?
 

The klondike design is very sensitive with the nor gates and the phase shifter in front of it. It is a matter of the manufacturer of the nor gates, as well as capacitor/resistor. For less HW errors, the Tau of R and C has to fit the propagation and switching times of the nor gate, and these are sometimes higher, sometimes lower, and changing with every type of nor gate. So very hard to make this working properly....

In this case, it would be very easy to make it work by changing the R or C to meet Tau for a particular nor gate.

Exactly, but this is a very empiric process and i think nobody really wants to do this for every new nor gate.

[freeworm]

It looks like your not getting any accepted shares.
Has that timing capacitor been replaced on your boards ?
 

The klondike design is very sensitive with the nor gates and the phase shifter in front of it. It is a matter of the manufacturer of the nor gates, as well as capacitor/resistor. For less HW errors, the Tau of R and C has to fit the propagation and switching times of the nor gate, and these are sometimes higher, sometimes lower, and changing with every type of nor gate. So very hard to make this working properly....

In this case, it would be very easy to make it work by changing the R or C to meet Tau for a particular nor gate.

Exactly, but this is a very empiric process and i think nobody really wants to do this for every new nor gate.

the fact is most people ordered the nor gate listed on BKK's BOM but still nobody has lowered the HW rate to a very low level by changing the Capacitor.

[kano]

...
Hopefully, there will be a working klondike cgminer soon. I think somebody is working on it.

Yes, I am.
Not sure why bogart's didn't work, but the team I'm helping a bit with, they are all getting at least some results with the klondike branch in my git.
It still needs some work, but the current code does work with the test boards in the team and gives back '=' results.

[KS]

Progress, yay!

[Bogart]

Do you have any way to make changes when sensei finds things that need to be changed or updated ?

I guess it depends what changes are needed.  I could probably change a cap without too much trouble.

@bogart, replace C274 with a 220p, or 260p capacitor

Which cap is this?  They are not labeled on my boards.

I guess I should look at the board schematics...

Is there any firmware that lives in the device in NVRAM of some sort, or is everything loaded at runtime?

[cp1]

Is there any firmware that lives in the device in NVRAM of some sort, or is everything loaded at runtime?

It's all precompiled and sits in the flash memory on the pic.

[AtomSea]

I really see this design being the precursor for an opensource board that will eventually hold an opensource chip.
And no reason that chip couldn't be 14nm or better someday.
Bitcoin needs checks and balances.
Without a strong-transparent-opensource mining community Bitcoin could descend into an old Star Trek episode where they end up on some random planet, that just went all wrong, and the weird fight music comes on, but at least Kirk gets laid.
So good work peeps - with you there is hope.

[LordTheron]

Hi guys,

I'd like to buy K16 and  2 K1 components kit without avalons. Please pm me if you have spare for sale.

Cheers,

[freeworm]

...
Hopefully, there will be a working klondike cgminer soon. I think somebody is working on it.

Yes, I am.
Not sure why bogart's didn't work, but the team I'm helping a bit with, they are all getting at least some results with the klondike branch in my git.
It still needs some work, but the current code does work with the test boards in the team and gives back '=' results.

Hi, Kano, your codes work but perform with lower hashing rate and give a lot for errors like this

---------------------------------------------------------------------
 [2013-09-26 09:52:54] KLN0: GetResults (amt=0 err=-7 ern=11)
 [2013-09-26 09:52:54] KLN1: GetResults (amt=0 err=-7 ern=11)
 [2013-09-26 09:52:54] KLN0: GetResults (amt=0 err=-7 ern=11)
---------------------------------------------------------------------

I guess it's because of the modified version of libusb in the new cgminer

[ScaryHash]

...
Hopefully, there will be a working klondike cgminer soon. I think somebody is working on it.

Yes, I am.
Not sure why bogart's didn't work, but the team I'm helping a bit with, they are all getting at least some results with the klondike branch in my git.
It still needs some work, but the current code does work with the test boards in the team and gives back '=' results.

That is wonderful news. I'm sure that if anybody can get it ironed out, it's those guys.

I'm mining more beer money for you guys with my block erupters as I type this.

Can I just compile it with the --enable-klondike --enable icarus from your git and maybe I won't get stupid usb problems anymore? Do I still need to install libusb?

I don't have a klondike (yet), but I can use the usb block erupters to see if the usb problems that I had before are still there.

[Bogart]

Mine have 5 sets of 2x2 pin headers located along the edges of the board.  The boards each came with a little PCB jumper board plugged into one of them.  Any idea what they do?

[sensei]

Mine have 5 sets of 2x2 pin headers located along the edges of the board.  The boards each came with a little PCB jumper board plugged into one of them.  Any idea what they do?

The boards can connect to each other and send data. The idea is that you can have one board connected is USB to a computer and then connect other boards in a chain. One USB instead of many.

The SW for that is not working yet. I will be looking into this soon.