NanoFury Project - Open Source Design

[Taugeran]

 

I was finding that I saw larger numbers of 'frequency drop' messages the longer that BFG ran; reducing the oscillations seemed to reduce the number, but consequently reduced the hash rate.

Using cgminer, I'm seeing a higher reported hashrate both on cgminer and my worker on Slush's pool. I'm seeing a reported 3558MH/s on the worker; but that's a combined rate for the Nanofury and two USB Block Eruptors. That figure is a definite increase over what I was seeing before; which was closer to 2800MH/s

I think I probably need to do a lot more reading on the subject, but I'm definitely observing an increased hash-rate on the worker.

hmm. interesting.  same amount of runtime?

try each at the same osc6 for a day each and report back if u dont mind plz 

[1713257525]

1713257525

[1713257525]

1713257525

[Taint]

I'll see what I can do; but I'm not currently certain what osc6 cgminer is running at. I can see a '54' listed next to the name; so I'm assuming 54.

With CGminer running on Slush, my average over 10 rounds was 3243MH/s

With BFGminer running on Slush, my average over 10 rounds was about 2600MH/s; however I'll run this again with the osc6 set at 54. If I see a big drop in my average hash rate, I'll switch back; as the Nanofury is currently my heaviest lifter. At least till my order from Hashrate store turns up...

[loshia]

I'll see what I can do; but I'm not currently certain what osc6 cgminer is running at. I can see a '54' listed next to the name; so I'm assuming 54.

With CGminer running on Slush, my average over 10 rounds was 3243MH/s

With BFGminer running on Slush, my average over 10 rounds was about 2600MH/s; however I'll run this again with the osc6 set at 54. If I see a big drop in my average hash rate, I'll switch back; as the Nanofury is currently my heaviest lifter. At least till my order from Hashrate store turns up...

Cgminer is not changing osc6 dynamically. On startup you can set it with --hexmineru-frequency
It is making 2.5gh+ stable per stick assuming everything else is ok osc6 is at 54 by default. Your score 3+ GH is a lucky one  you need at least 24 hours of mining to compare results .
If you are in doubt about hash rate you can always use accepted shares formula as suggested by Kano and double check accepted shares cgminer stats and your pool stats

[Taugeran]

I'll see what I can do; but I'm not currently certain what osc6 cgminer is running at. I can see a '54' listed next to the name; so I'm assuming 54.

With CGminer running on Slush, my average over 10 rounds was 3243MH/s

With BFGminer running on Slush, my average over 10 rounds was about 2600MH/s; however I'll run this again with the osc6 set at 54. If I see a big drop in my average hash rate, I'll switch back; as the Nanofury is currently my heaviest lifter. At least till my order from Hashrate store turns up...

Cgminer is not changing osc6 dynamically. On startup you can set it with --hexmineru-frequency
It is making 2.5gh+ stable per stick assuming everything else is ok osc6 is at 54 by default. Your score 3+ GH is a lucky one  you need at least 24 hours of mining to compare results .
If you are in doubt about hash rate you can always use accepted shares formula as suggested by Kano and double check accepted shares cgminer stats and your pool stats

There's the point I was trying to get at. Le sigh. I got too wordy :/

[rgr_rgr]

I never compared but I also think that the sticks are running faster with cgminer. But I am careful because cgminer does not report hardware errors.

And be careful two: Yesterday a psu of one of my hubs melted. It was a cheap one. It was rated to 2A and I used 4 Nanofuries with 54 Bits -> too much. Yes expectable, because with 54 Bits they are running outside of USB2.0 specs.

These beasts are really hungry for power and suck everything out.

Take care using good hubs.

[loshia]

I never compared but I also think that the sticks are running faster with cgminer. But I am careful because cgminer does not report hardware errors.

And be careful two: Yesterday a psu of one of my hubs melted. It was a cheap one. It was rated to 2A and I used 4 Nanofuries with 54 Bits -> too much. Yes expectable, because with 54 Bits they are running outside of USB2.0 specs.

These beasts are really hungry for power and suck everything out.

Take care using good hubs.

That is why I always use Psu red and black
First thing to do is to garbage hub adapter but before that I am cutting the plug

[Taugeran]

I never compared but I also think that the sticks are running faster with cgminer. But I am careful because cgminer does not report hardware errors.

And be careful two: Yesterday a psu of one of my hubs melted. It was a cheap one. It was rated to 2A and I used 4 Nanofuries with 54 Bits -> too much. Yes expectable, because with 54 Bits they are running outside of USB2.0 specs.

These beasts are really hungry for power and suck everything out.

Take care using good hubs.

That is why I always use Psu red and black
First thing to do is to garbage hub adapter but before that I am cutting the plug

if you have any 2A+ @5V i would take em

[loshia]

I never compared but I also think that the sticks are running faster with cgminer. But I am careful because cgminer does not report hardware errors.

And be careful two: Yesterday a psu of one of my hubs melted. It was a cheap one. It was rated to 2A and I used 4 Nanofuries with 54 Bits -> too much. Yes expectable, because with 54 Bits they are running outside of USB2.0 specs.

These beasts are really hungry for power and suck everything out.

Take care using good hubs.

That is why I always use Psu red and black
First thing to do is to garbage hub adapter but before that I am cutting the plug

if you have any 2A+ @5V i would take em

Unfortunately all are thrown away. But I will start to save them

[Mudbankkeith]

I never compared but I also think that the sticks are running faster with cgminer. But I am careful because cgminer does not report hardware errors.

And be careful two: Yesterday a psu of one of my hubs melted. It was a cheap one. It was rated to 2A and I used 4 Nanofuries with 54 Bits -> too much. Yes expectable, because with 54 Bits they are running outside of USB2.0 specs.

These beasts are really hungry for power and suck everything out.

Take care using good hubs.

That is why I always use Psu red and black
First thing to do is to garbage hub adapter but before that I am cutting the plug

if you have any 2A+ @5V i would take em

Unfortunately all are thrown away. But I will start to save them

No need to cut the cable...........http://www.ebay.co.uk/itm/4pcs-lot-Male-12V-DC-Power-Jack-Connectors-Cable-Adapter-CCTV-Camera-CAM-DVR-/221288401417?pt=UK_Sound_Vision_Other&hash=item3385d10a09

[loshia]

I never compared but I also think that the sticks are running faster with cgminer. But I am careful because cgminer does not report hardware errors.

And be careful two: Yesterday a psu of one of my hubs melted. It was a cheap one. It was rated to 2A and I used 4 Nanofuries with 54 Bits -> too much. Yes expectable, because with 54 Bits they are running outside of USB2.0 specs.

These beasts are really hungry for power and suck everything out.

Take care using good hubs.

That is why I always use Psu red and black
First thing to do is to garbage hub adapter but before that I am cutting the plug

if you have any 2A+ @5V i would take em

Unfortunately all are thrown away. But I will start to save them

No need to cut the cable...........http://www.ebay.co.uk/itm/4pcs-lot-Male-12V-DC-Power-Jack-Connectors-Cable-Adapter-CCTV-Camera-CAM-DVR-/221288401417?pt=UK_Sound_Vision_Other&hash=item3385d10a09

Then you will need to cut psu red and black
Anyway I am the last owner of my psu's so I am cutting them all then I spike and solder the wires quick easy and secure. Nothing melts anymore
In the beginning I used various extenders and crimp type connectors
My math showed me that I was spending for this as much as psu cost itself and it was less secure

[pauljbl]

on v0.7 how do you choose the footprint for the MCP2210 I/SO or I/SS when making gerber files.

I would like to use the MCP2210 I/SO


thanks

[vs3]

on v0.7 how do you choose the footprint for the MCP2210 I/SO or I/SS when making gerber files.

I would like to use the MCP2210 I/SO

thanks

The gerber files will have both footprints. Later during the assembly process you will solder it at the footprint that matches the chip (e.g. if you got I/SO chips you would put them at that place on the PCB).

[pauljbl]

I see


the holes that are under the bitfury chip is there solder coming throught them or does thermal paste go through them for heat

[Taugeran]

I see


the holes that are under the bitfury chip is there solder coming throught them or does thermal paste go through them for heat

most likely a small amount of wicked solder. helps with thermal conductivity

[vs3]

I see


the holes that are under the bitfury chip is there solder coming throught them or does thermal paste go through them for heat

most likely a small amount of wicked solder. helps with thermal conductivity

Yes and Yes
Small amount does go through and that's specifically to help with conductivity (in addition to the copper coating of the holes). Unfortunately there is no way to control the amount of solder that goes through - sometimes very little, sometimes quite a lot .. and in the second case that makes a little bump on the back side, so you either need to process it a bit further if you want an extra-smooth surface, or just use a thicker stick-on thermal tape (that can tolerate some minor roughness).

[3devilred]

hello to all. sorry but you can ' know where you can find the components to produce the product ? thanks

[Taugeran]

hello to all. sorry but you can ' know where you can find the components to produce the product ? thanks

should be a BOM somewhere in the git repo: https://github.com/nanofury/NanoFury

[3devilred]

hello to all. sorry but you can ' know where you can find the components to produce the product ? thanks

should be a BOM somewhere in the git repo: https://github.com/nanofury/NanoFury

I would like to achieve at least 10, but me and ' hard to find parts , even if you know some store on ebay that sells them , thank you

I would like to know if there 'to plan some component , or you just need to solder them ? ? Thank you.

P.S. sorry if I do so many questions, but in the Italian translation many words are unclear .

[Taint]

Ok, reporting back with my findings.

After over 24 hours running on BFGminer against Slush pool, my average hashrate over 10 rounds was 2779MH/s; which is a combined figure including two Block Eruptors ~335Mh/s each.

Compared to my 24 hours+ on cgminer at 3243Mh/s (including the two Block Eruptors) I have to say it definitely appears that cgminer is faster.

In both instances, this the reported hash rate at Slush's end; not the hash rate reported in the miner.

The VM was rebooted between the uses of CG and BFG; but the host was not. Aside from changing the miner, no other settings were changed and no patches installed. Both were run at 54 osc6. Currently there's nothing else running on the host and the USB hub has a 12V 4A power supply; so I don't think it's a power issue.

I do realise that there are other factors that can influence the hashrate, but it looks a lot like the Nanofury is under-performing while running under BFG. Can anyone corroborate my findings?

[bigbeninlondon]

I didn't notice a difference between the two as far as hashing, but I did notice that cgminer was less stable for me over long periods than bfgminer.