Fury/Blizzard tuning and mods

[beukhof1]

What does the --zeus-nocheck-golden do?

[jekv2]

I am also power cycling my fury every day and half for an hour or so until I get 80 hours burned in. From what I read about the C2 is

Important Reminder:
Due to the unique shapes and sizes of the particles in Céramique 2, it will take a minimum of 25 hours and several thermal cycles to achieve maximum particle to particle thermal conduction and for the heatsink to CPU interface to reach maximum conductivity. (This period will be longer in a system without a fan on the heatsink.) On systems measuring actual internal core temperatures via the CPU's internal diode, the measured temperature will often drop slightly over this "break-in" period. This break-in will occur during the normal use of the computer as long as the computer is turned off from time to time and the interface is allowed to cool to room temperature. Once the break-in is complete, the computer can be left on if desired.

I've already seen a decrease in HW errors from 14% to 9.28%. SO in with the cure time I should see the magic 5%.

------------------------------------

What does the --zeus-nocheck-golden do?

https://litecointalk.org/index.php?PHPSESSID=pd562v4bbnd3pmv15jn1cmneq5&topic=16301.msg178947#msg178947

it appears that the --nocheck-golden flag overrides the initial device check and substitutes default values for the hash speed instead of measuring the actual speed).

[kevin1234a]

hello there

can we run gridseed miner and GAW miner with same Raspberry pi controller?

[jekv2]

So far the lowest I've seen is 7.87%-8.05% HW errors. @ 350Mhz.

Which minera calculates the % in which are wrong, when done by a calculator its higher than what minera states. "8.80%"

[jekv2]

So far the lowest I've seen is 7.87%-8.05% HW errors. @ 350Mhz.

Which minera calculates the % in which are wrong, when done by a calculator its higher than what minera states. "8.80%"

Now I think I am getting somewhere.

[kevin1234a]

Well, I've added the thermal paste to the back, applied heatsinks to all IC's on the top, and there is a fan blowing air through the case over the heatsinks I applied.  Ambient temp is around 73f.  Temperature probe attached to the side of the case/heatsink reads 82f.

I hadn't known about the higher speed information...however....for kicks I started it running at 350 over 2 hours ago.  Right now it's sitting at 1.39Mhs and around 4% error.

what chips settings you had used for the miners in hashra image?

i am using 2 GAW fury set with chips setting to 12 and clk speed to 250Mhz

[nst6563]

Well, I've added the thermal paste to the back, applied heatsinks to all IC's on the top, and there is a fan blowing air through the case over the heatsinks I applied.  Ambient temp is around 73f.  Temperature probe attached to the side of the case/heatsink reads 82f.

I hadn't known about the higher speed information...however....for kicks I started it running at 350 over 2 hours ago.  Right now it's sitting at 1.39Mhs and around 4% error.

what chips settings you had used for the miners in hashra image?

i am using 2 GAW fury set with chips setting to 12 and clk speed to 250Mhz

For the Fury's you should be using chip setting of 6.  It's a per-miner setting, not a cumulative setting.

[kevin1234a]

Thanks for advise

[Ntrain2k]

Thanks for advise

You also should raise that clock speed if they are the 6 chip units. At least 328. I use 340.

[icekohl]

hello there

can we run gridseed miner and GAW miner with same Raspberry pi controller?

Yes, I use Minera for this.  https://bitcointalk.org/index.php?topic=596620.0

[jekv2]

I am back up to 10% hw error rate with this piece of shit gawminer fury. Makes me fury,,,,

[BRADLEYPLOOF]

I've been rolling mine at 342 for weeks.  I'm at ~4.7% HW.  Running BFGMiner with the display of each core shows that there are generally a few that HW like mad, while the rest of them are wicked low percentages.  One out of each of the cores in the chips is always >10%.  Neat to see the stats though.  I wish there was a version of BFGminer or CGminer or something that would allow clocking of individual chips. 

[CapnBDL]

Hey all....I've been quietly watching all going on here and I have a couple of questions.

How come when I load up my Zeus/Fury miners /w the latest (4.3.5, I think) CGminer, I get only about a third of the end hashrate? By end hashrate I mean 'at the pool end'. Sure, the GUI for it says I'm right on the money. At a 340 clock speed it shows a 1.7Mh. But that sure isn't what the pool rate says, and that's the one that counts. I even back it off (clock) and still the pool hash is soooo much smaller.   
Here's the kicker...if I go back to the old 3.1.1 curse CGMiner....works fine and I get about what I should. So...WTF? Am I missing something in the translation? I'm no pro, but not a newb either. I've tweaked the .conf file, etc., and it ain't right. Has anyone else been seeing this and if so, how the heck did ya fix it? PLEASE.....tell me!     I need to know. I've lost sleep. I've been up ALL night.

Second question; Has anyone seen a difference in the hashrates between the GenA releases and the latest? I have a GenA Fury and a newer, recent order one and, WOW. What a difference in those. Can anyone tell me WHY? And Yes...before anyone asks...I've ensured that my informal testing has been on a level playing field. Same drivers, software, etc.

Oh..I should say: Win8.1 64, i7, 16G ram, nVidia, yada, yada. Further, I have 'no clue' how to compile or write code. I work with what's out there and tweak the sh$t out of it. No hardware mods, etc. Out of the box stuff only.

Thanks for any and all help, kind people. It is very much appreciated.   

[Borsht]

I posted my volt mod results @ http://www.techunboxed.com/2014/08/gaw-miners-fury-mods-voltage-mod.html

the short story is about a 14% increase in hash rate over a reasonable overclock of a stock fury but a 23% increase in power consumption.

[CapnBDL]

Well, I'm back at it again. Sorry to say...the results really suck. I really liked the way that CGMiner 4.3.5 gave me a slight bit more control over these pieces of crud, but it just isn't cutting the mustard. Not about to start doing a voltage mod. Not worth the trouble or time. Someone tell me they have had better luck! I really don't wish to use 3.1.1curses. It's almost as bad as the Furys.

Talk to me peeps....

Thanks

[BRADLEYPLOOF]

BFGMiner 4.6 or .7 work great in terms of being able to see every core in one of these things so you can find out which ones are crapping out.  I run mine at 342 with 4.8% HW and 1.42 MH/s.  Everything on it is stock and all I did was put a fan on the front to blow through the whole thing.  Cooling is key and so is adding better heat sink goo.  I'm using an R.Pi. 

[jeezy]

I've been rolling mine at 342 for weeks.  I'm at ~4.7% HW.  Running BFGMiner with the display of each core shows that there are generally a few that HW like mad, while the rest of them are wicked low percentages.  One out of each of the cores in the chips is always >10%.  Neat to see the stats though.  I wish there was a version of BFGminer or CGminer or something that would allow clocking of individual chips. 

Yeah I had the same with my Zeus Blizzards. Would be cool to implement chip wise clock setting. So with the new HW Chip display you could easily isolate the underperforming chips and set them to a lower MHz. Would be awesome.

[CapnBDL]

Hey all....I've been quietly watching all going on here and I have a couple of questions.

How come when I load up my Zeus/Fury miners /w the latest (4.3.5, I think) CGminer, I get only about a third of the end hashrate? By end hashrate I mean 'at the pool end'. Sure, the GUI for it says I'm right on the money. At a 340 clock speed it shows a 1.7Mh. But that sure isn't what the pool rate says, and that's the one that counts. I even back it off (clock) and still the pool hash is soooo much smaller.  
Here's the kicker...if I go back to the old 3.1.1 curse CGMiner....works fine and I get about what I should. So...WTF? Am I missing something in the translation? I'm no pro, but not a newb either. I've tweaked the .conf file, etc., and it ain't right. Has anyone else been seeing this and if so, how the heck did ya fix it? PLEASE.....tell me!    I need to know. I've lost sleep. I've been up ALL night.

Second question; Has anyone seen a difference in the hashrates between the GenA releases and the latest? I have a GenA Fury and a newer, recent order one and, WOW. What a difference in those. Can anyone tell me WHY? And Yes...before anyone asks...I've ensured that my informal testing has been on a level playing field. Same drivers, software, etc.

Oh..I should say: Win8.1 64, i7, 16G ram, nVidia, yada, yada. Further, I have 'no clue' how to compile or write code. I work with what's out there and tweak the sh$t out of it. No hardware mods, etc. Out of the box stuff only.

Thanks for any and all help, kind people. It is very much appreciated.  

Arrrgghhh! My hair, my hair!! I'm about to quit and go back to 3.1.1.

[nst6563]

Hey all....I've been quietly watching all going on here and I have a couple of questions.

How come when I load up my Zeus/Fury miners /w the latest (4.3.5, I think) CGminer, I get only about a third of the end hashrate? By end hashrate I mean 'at the pool end'. Sure, the GUI for it says I'm right on the money. At a 340 clock speed it shows a 1.7Mh. But that sure isn't what the pool rate says, and that's the one that counts. I even back it off (clock) and still the pool hash is soooo much smaller.  
Here's the kicker...if I go back to the old 3.1.1 curse CGMiner....works fine and I get about what I should. So...WTF? Am I missing something in the translation? I'm no pro, but not a newb either. I've tweaked the .conf file, etc., and it ain't right. Has anyone else been seeing this and if so, how the heck did ya fix it? PLEASE.....tell me!    I need to know. I've lost sleep. I've been up ALL night.

Second question; Has anyone seen a difference in the hashrates between the GenA releases and the latest? I have a GenA Fury and a newer, recent order one and, WOW. What a difference in those. Can anyone tell me WHY? And Yes...before anyone asks...I've ensured that my informal testing has been on a level playing field. Same drivers, software, etc.

Oh..I should say: Win8.1 64, i7, 16G ram, nVidia, yada, yada. Further, I have 'no clue' how to compile or write code. I work with what's out there and tweak the sh$t out of it. No hardware mods, etc. Out of the box stuff only.

Thanks for any and all help, kind people. It is very much appreciated.  

Arrrgghhh! My hair, my hair!! I'm about to quit and go back to 3.1.1.

Who's cgminer fork are you using?  If you're not using Dmaxl's fork I would suggest trying that (https://github.com/dmaxl/cgminer).  I strictly mine using linux based controllers (rpi or PC)...so if you're on Windows I'm not sure what to tell you.  Maybe get an rpi?  It's wicked easy to get things going on the rpi platform.  You can even try rpi mining firmware like Starminer, Minera, Minepeon, etc. 

[CapnBDL]

Hey all....I've been quietly watching all going on here and I have a couple of questions.

How come when I load up my Zeus/Fury miners /w the latest (4.3.5, I think) CGminer, I get only about a third of the end hashrate? By end hashrate I mean 'at the pool end'. Sure, the GUI for it says I'm right on the money. At a 340 clock speed it shows a 1.7Mh. But that sure isn't what the pool rate says, and that's the one that counts. I even back it off (clock) and still the pool hash is soooo much smaller.  
Here's the kicker...if I go back to the old 3.1.1 curse CGMiner....works fine and I get about what I should. So...WTF? Am I missing something in the translation? I'm no pro, but not a newb either. I've tweaked the .conf file, etc., and it ain't right. Has anyone else been seeing this and if so, how the heck did ya fix it? PLEASE.....tell me!    I need to know. I've lost sleep. I've been up ALL night.

Second question; Has anyone seen a difference in the hashrates between the GenA releases and the latest? I have a GenA Fury and a newer, recent order one and, WOW. What a difference in those. Can anyone tell me WHY? And Yes...before anyone asks...I've ensured that my informal testing has been on a level playing field. Same drivers, software, etc.

Oh..I should say: Win8.1 64, i7, 16G ram, nVidia, yada, yada. Further, I have 'no clue' how to compile or write code. I work with what's out there and tweak the sh$t out of it. No hardware mods, etc. Out of the box stuff only.

Thanks for any and all help, kind people. It is very much appreciated.  

Arrrgghhh! My hair, my hair!! I'm about to quit and go back to 3.1.1.

Who's cgminer fork are you using?  If you're not using Dmaxl's fork I would suggest trying that (https://github.com/dmaxl/cgminer).  I strictly mine using linux based controllers (rpi or PC)...so if you're on Windows I'm not sure what to tell you.  Maybe get an rpi?  It's wicked easy to get things going on the rpi platform.  You can even try rpi mining firmware like Starminer, Minera, Minepeon, etc.  

Dang....I have 2 rPis and never set them up. I don't have a monitor for them. Mine is VGA and I don't have an HDMI adapter. I did order one but it doesn't seem to decode the signal. I think it might have to do /w the nVidia card in this. Not sure. Anybody have any ideas. I know I really should get these running and let my 'puter rest. As for the correct fork...yup, tried it out but my 2 Furies really slowed down (see quote). It didn't work for me /w Windows. Maybe a Pi controller would be better. At this point...I just don't know. I'm sooo confused. Arrhhggg.