GridSeed 5-chip USB miner voltage mod

[wolfey2014]

I can tell you that both modified miners have been running rock steady over 39 hours so far.
No errors except for a couple when first started for some reason.
Other than that, I have not seen any errors.

Wolfey2014

Seriously nice job Wolfey2014!

Thank you very much!

Feel free to PM me if you would like me to mod your pods too

Wolfey2014

[wolfey2014]

Bounty for someone making up a new CPU miner with actual hashrate and hell why not an Win GUI that would regroup all open cpuminer instances and where you could manage them from the GUI. let's dream a bit :p

Here here!
I fifty=five-ith that motion!

Wolfey2014

[jamieb81]

now it's up to find the one that could code something like that, I think people would even want to pay some alt coins for such program if it would run great

[nemercry]

Hello Miners and Friends,

we sat here the last 4 hours and soldered and tweaked and got our results now for presentation. We are a little sad that we have to say that "the cake was a lie".
The last hours we were really positive that we could achieve 600kh/s, which was stated by a user here in the forum. We tried a lot and even did go down from 1.1V PLL to 0.94V PLL (in 0.01V steps). In the end we gave up. The best, stable hashrate we could reach was 446.8kh/s at 1050mhz. We even replaced the "pencil" with some carbonglue-fluid, which acts like graphite but is more conductive and easier to use. (It doesnt fly away if you need to sneeze )


Our Final Board(2 Bridge Mod + 39k +1% resistor and "pencil mod")

Our Final PLL voltage: DONT DO ANTYHING WITH YOUR PLL VOLTAGE! OLD

Our Final Stable voltage(no HW):

And there it startet to throw HWs:

As it still is an improve in the hashrate, we would like to thank all who were being involved. Also thanks alot for even providing this public domain mod, to all users.

Best Regards,
nemercry

[jamieb81]

dont know what it is whit my chrome browers today but I cant see any pictures on the threads properly today

[Jemison]

"Thanks for the photos.  How safely can we mix and match these mods?  Right now I don't have an iron tip fine enough to do the soldering mods.  I either need to modify something or order something.  I am interested in the pencil mod and the fan mod tonight though.  Would this be safe? "

No, I don't believe the pencil mod will do any good without the other 3 mods.
Pencil mod is to adjust PLL voltage which needs to be corrected due to the offset caused by the two solder jumpers and the resistor is a timing adjustment to smooth out clock transitions at higher frequencies I believe.

You can make the 5V fan mod any time but remember, it's recommended for Scrypt only operation.
It has not been tested in SHA or SHA / Scrypt mining mode!

As usual, do these mods at your own risk!

Wolfey2014

Wolfey,

Somewhere in this thread is mention of a cgminer version that allows you to set the voltage.  I realize you're using cpuminer, but do you think all 4 mods are required if we used the voltage mod cgminer?  Value your opinion.

Thank you!

You ask a GOOD question.

Well, from what I understand - the current mods bypass having to soft set the PLL and Core Voltages.
Personally, I can't say one way or the other if soft adjustment is beneficial or not. I have heard 'rumored' that soft fine adjustment of 1000MHz  up to 1048MHz stable is possible.

I'm going to have to convert over to one of the other programs to be able to read client side hash rates.
I just don't like all the tweaking that is still going on over the course of cg and bfg during it's 'use as it evolves' phase. I am not interested, at this stage of their evolution, in becoming an unwilling guinea pig
I'm happy with cpuminer and what the pool's report in spite of their second by second inaccuracy. A bit of an understanding of odds, how they are avaraged out and common sense helps me figure out what the 'not so rough' improvement figure is for a given pod's clean hash rate.

Some sites like litecoinpool give a 24 hour average speed and that's close enough for me.
Besides and 'primarily' it only really counts what performance stats one sees pool side as that directly correlates to how much moolah it's gonna make you - the end product, the result, the win! $$$$

I guess I'm gonna have to learn one or the other 'unless there is a better alternative' because I could give a hoot what else the program does. All I want it to do is show me an accurate local hashrate as a separate program from the rest of it. And it MUST not be memory hungry like cpuminer 'isn't!' Thank you very much GridSeed!  
It would be cool if not great if GC would take back up updating / tweaking their very good and very stable program! It lives!  
IMO, cpuminer is very optimized and user friendly compared to the others IMO.
If the others were less complex and learning curve intensive, I'd of bothered to test them too by now.
Lots of other people like/love them though. So, to each their own.

Peace!
Wolfey2014

I created my own local database and query the pool api hourly to get their reported hashrate for my miners. 
On average, over 48hrs, the pool is reporting  20% more shares than cgminer, and I'm okay with that, lol. 

I use cgminer's api to query the health of my miners and to re-start them if necessary.  cgminer has the ability to run all or a selection. 
Its api can be set on different local network ports so if you have multiple instances running on one box you can query them separately. 

I've only had one hiccup with cgminer in the 7 days I've been running - but I think that was caused by my linux attempting to self-update,
which I've now disabled.

[wolfey2014]

Hello Miners and Friends,

we sat here the last 4 hours and soldered and tweaked and got our results now for presentation. We are a little sad that we have to say that "the cake was a lie".
The last hours we were really positive that we could achieve 600kh/s, which was stated by a user here in the forum. We tried a lot and even did go down from 1.1V PLL to 0.94V PLL (in 0.01V steps). In the end we gave up. The best, stable hashrate we could reach was 446.8kh/s at 1050mhz. We even replaced the "pencil" with some carbonglue-fluid, which acts like graphite but is more conductive and easier to use. (It doesnt fly away if you need to sneeze )


Our Final Board(2 Bridge Mod + 39k +1% resistor and "pencil mod")

Our Final PLL voltage:

Our Final Stable voltage(no HW):

And there it startet to throw HWs:

Nay nay - rumor mongering nay sayers! Off with your heads!  

No, really. I do believe I 'the user that said 600', stated clearly that over 600 was on POOL SIDE!
POOL SIDE! Which is variable from as little as less than 100KH/s to over 750KH/s.
Oh, and I have the screen shots to prove it.

I saw that someone else altered the facts before this happened, now look!
I knew this was going to happen!
Someone takes what I said out of context i.e. alters the facts, then 'RUMOR Monger's!' all over the place to create an effect that actually, when one knows where that kind of stuff comes from, indicates what the persons actual counter intentions are. BOO!
Do you work for the news media too?

Something else. You CAN NOT say with 100% certainty that the hash rate you see on your client side display is even accurate much less to what extent or degree! YOU CAN'T!

Your experiment was only necessary because YOU CAN'T READ, ASSIMILATE and DUPLICATE, DUMMY!! ;p

Wolfey2014

[nemercry]

Hello Miners and Friends,

we sat here the last 4 hours and soldered and tweaked and got our results now for presentation. We are a little sad that we have to say that "the cake was a lie".
The last hours we were really positive that we could achieve 600kh/s, which was stated by a user here in the forum. We tried a lot and even did go down from 1.1V PLL to 0.94V PLL (in 0.01V steps). In the end we gave up. The best, stable hashrate we could reach was 446.8kh/s at 1050mhz. We even replaced the "pencil" with some carbonglue-fluid, which acts like graphite but is more conductive and easier to use. (It doesnt fly away if you need to sneeze )


Our Final Board(2 Bridge Mod + 39k +1% resistor and "pencil mod")

Our Final PLL voltage:

Our Final Stable voltage(no HW):

And there it startet to throw HWs:

Nay nay - rumor mongering nay sayers! Off with your heads!  

No, really. I do believe I 'the user that said 600', stated clearly that over 600 was on POOL SIDE!
POOL SIDE! Which is variable from as little as less than 100KH/s to over 750KH/s.
Oh, and I have the screen shots to prove it.

I saw that someone else altered the facts before this happened, now look!
I knew this was going to happen!
Someone takes what I said out of context i.e. alters the facts, then 'RUMOR Monger's!' all over the place to create an effect that actually, when one knows where that kind of stuff comes from, indicates what the persons actual counter intentions are. BOO!
Do you work for the news media too?

Something else. You CAN NOT say with 100% certainty that the hash rate you see on your client side display is even accurate much less to what extent or degree! YOU CAN'T!

Your experiment was only necessary because YOU CAN'T READ, ASSIMILATE and DUPLICATE, DUMMY!! ;p

Wolfey2014

Hey Wolfey2014,

not ment to be something against you. We really do appreciate all your work and what you have done!.
Dont take it personally. Also so far it looks like as we got the fastest miner on the market.
: )

Kind regards,
nemercry

[Xer0]

hmm... if i just want to swap the resistors for the pll,
instead of penciling, what values should i choose?

[wolfey2014]

Wolf:  If you really are making the hash you are saying, prove it.  Please post a screen shot of what your pool is reporting and what your client is reporting.  I'd love to make 600Kh on my GS.

Cool! I just love a challenge like that!
That's right. 600+KH/s pool side, running solo to that service all by itself, alone, just like I said.
Please feel free to try it out for yourself!





Now, I am not calling this proof! A lot of folks don't know the difference between proof and evidence.

I will call this evidence to support my claims! That's as truthful as it gets

Wolfey2014 ;p

[wolfey2014]

hmm... if i just want to swap the resistors for the pll,
instead of penciling, what values should i choose?

Scroll, look through the pages, read and you'll find your questions already answered!
There are even pics too!
Wolfey2014

[wolfey2014]

Hello Miners and Friends,

we sat here the last 4 hours and soldered and tweaked and got our results now for presentation. We are a little sad that we have to say that "the cake was a lie".
The last hours we were really positive that we could achieve 600kh/s, which was stated by a user here in the forum. We tried a lot and even did go down from 1.1V PLL to 0.94V PLL (in 0.01V steps). In the end we gave up. The best, stable hashrate we could reach was 446.8kh/s at 1050mhz. We even replaced the "pencil" with some carbonglue-fluid, which acts like graphite but is more conductive and easier to use. (It doesnt fly away if you need to sneeze )


Our Final Board(2 Bridge Mod + 39k +1% resistor and "pencil mod")

Our Final PLL voltage:

Our Final Stable voltage(no HW):

And there it startet to throw HWs:

Nay nay - rumor mongering nay sayers! Off with your heads!  

No, really. I do believe I 'the user that said 600', stated clearly that over 600 was on POOL SIDE!
POOL SIDE! Which is variable from as little as less than 100KH/s to over 750KH/s.
Oh, and I have the screen shots to prove it.

I saw that someone else altered the facts before this happened, now look!
I knew this was going to happen!
Someone takes what I said out of context i.e. alters the facts, then 'RUMOR Monger's!' all over the place to create an effect that actually, when one knows where that kind of stuff comes from, indicates what the persons actual counter intentions are. BOO!
Do you work for the news media too?

Something else. You CAN NOT say with 100% certainty that the hash rate you see on your client side display is even accurate much less to what extent or degree! YOU CAN'T!

Your experiment was only necessary because YOU CAN'T READ, ASSIMILATE and DUPLICATE, DUMMY!! ;p

Wolfey2014

Hey Wolfey2014,

not ment to be something against you. We really do appreciate all your work and what you have done!.
Dont take it personally. Also so far it looks like as we got the fastest miner on the market.
: )

Kind regards,
nemercry

I appreciate that you didn't mean it and your apology is accepted.
However, this does not excuse your inference that I LIED!
I never lied and so, the pie isn't a lie either.
I'd appreciate a retraction.
So, consider yourself duly corrected.
Wolfey2014

P.S.
Do try a 5% resistor. It needs the 5% drift, I think.
Not a 1%. I have also found that PLL set to 1.018ish works very well.
Anything lower and you get drifting problems.
Good luck.
Remember, you brick it, you just made yourself  a nice gold paper weight.

[wolfey2014]

hmm... if i just want to swap the resistors for the pll,
instead of penciling, what values should i choose?

NNNNNNNNNNNNNNO!
Follow the instructions - to - the - letter.

Do not squirrel this or you will get 'poor' results!

DUPLICATE correct tech for CORRECT results!
DO NOT alter them or you will fail!
And, so it goes in life!

Wolfey2014

[wolfey2014]

Wolf:  If you really are making the hash you are saying, prove it.  Please post a screen shot of what your pool is reporting and what your client is reporting.  I'd love to make 600Kh on my GS.

DONE!

Scroll down the page
DOH!
I mean, scroll up!

Wolfey2014

[ZiG]

Hello Miners and Friends,

we sat here the last 4 hours and soldered and tweaked and got our results now for presentation. We are a little sad that we have to say that "the cake was a lie".
The last hours we were really positive that we could achieve 600kh/s, which was stated by a user here in the forum. We tried a lot and even did go down from 1.1V PLL to 0.94V PLL (in 0.01V steps). In the end we gave up. The best, stable hashrate we could reach was 446.8kh/s at 1050mhz. We even replaced the "pencil" with some carbonglue-fluid, which acts like graphite but is more conductive and easier to use. (It doesnt fly away if you need to sneeze )


Our Final Board(2 Bridge Mod + 39k +1% resistor and "pencil mod")

Our Final PLL voltage:

Our Final Stable voltage(no HW):

And there it startet to throw HWs:

Nay nay - rumor mongering nay sayers! Off with your heads!  

No, really. I do believe I 'the user that said 600', stated clearly that over 600 was on POOL SIDE!
POOL SIDE! Which is variable from as little as less than 100KH/s to over 750KH/s.
Oh, and I have the screen shots to prove it.

I saw that someone else altered the facts before this happened, now look!
I knew this was going to happen!
Someone takes what I said out of context i.e. alters the facts, then 'RUMOR Monger's!' all over the place to create an effect that actually, when one knows where that kind of stuff comes from, indicates what the persons actual counter intentions are. BOO!
Do you work for the news media too?

Something else. You CAN NOT say with 100% certainty that the hash rate you see on your client side display is even accurate much less to what extent or degree! YOU CAN'T!

Your experiment was only necessary because YOU CAN'T READ, ASSIMILATE and DUPLICATE, DUMMY!! ;p

Wolfey2014

Hey Wolfey...Chill down, buddy...Why so angry...?...What's the problem...?

I saw you on some other thread(s) to react the similar manner...Relax Wolfey...RELAX...Take it easy...Have a beer...or something...

[wolfey2014]

Hello Miners and Friends,

we sat here the last 4 hours and soldered and tweaked and got our results now for presentation. We are a little sad that we have to say that "the cake was a lie".
The last hours we were really positive that we could achieve 600kh/s, which was stated by a user here in the forum. We tried a lot and even did go down from 1.1V PLL to 0.94V PLL (in 0.01V steps). In the end we gave up. The best, stable hashrate we could reach was 446.8kh/s at 1050mhz. We even replaced the "pencil" with some carbonglue-fluid, which acts like graphite but is more conductive and easier to use. (It doesnt fly away if you need to sneeze )


Our Final Board(2 Bridge Mod + 39k +1% resistor and "pencil mod")

Our Final PLL voltage:

Our Final Stable voltage(no HW):

And there it startet to throw HWs:

Nay nay - rumor mongering nay sayers! Off with your heads!  

No, really. I do believe I 'the user that said 600', stated clearly that over 600 was on POOL SIDE!
POOL SIDE! Which is variable from as little as less than 100KH/s to over 750KH/s.
Oh, and I have the screen shots to prove it.

I saw that someone else altered the facts before this happened, now look!
I knew this was going to happen!
Someone takes what I said out of context i.e. alters the facts, then 'RUMOR Monger's!' all over the place to create an effect that actually, when one knows where that kind of stuff comes from, indicates what the persons actual counter intentions are. BOO!
Do you work for the news media too?

Something else. You CAN NOT say with 100% certainty that the hash rate you see on your client side display is even accurate much less to what extent or degree! YOU CAN'T!

Your experiment was only necessary because YOU CAN'T READ, ASSIMILATE and DUPLICATE, DUMMY!! ;p

Wolfey2014

Hey Wolfey...Chill down, buddy...Why so angry...?...What's the problem...?

I saw you on some other thread(s) to react the similar manner...Relax Wolfey...RELAX...Take it easy...Have a beer...or something...

Angry? Nope.
A bit annoyed, yes!
To answer your question. READ THE POST! I give my reasons there in!
Geesh!
Wolfey2014

[ZiG]

Hello Miners and Friends,

we sat here the last 4 hours and soldered and tweaked and got our results now for presentation. We are a little sad that we have to say that "the cake was a lie".
The last hours we were really positive that we could achieve 600kh/s, which was stated by a user here in the forum. We tried a lot and even did go down from 1.1V PLL to 0.94V PLL (in 0.01V steps). In the end we gave up. The best, stable hashrate we could reach was 446.8kh/s at 1050mhz. We even replaced the "pencil" with some carbonglue-fluid, which acts like graphite but is more conductive and easier to use. (It doesnt fly away if you need to sneeze )


Our Final Board(2 Bridge Mod + 39k +1% resistor and "pencil mod")

Our Final PLL voltage:

Our Final Stable voltage(no HW):

And there it startet to throw HWs:

Nay nay - rumor mongering nay sayers! Off with your heads!  

No, really. I do believe I 'the user that said 600', stated clearly that over 600 was on POOL SIDE!
POOL SIDE! Which is variable from as little as less than 100KH/s to over 750KH/s.
Oh, and I have the screen shots to prove it.

I saw that someone else altered the facts before this happened, now look!
I knew this was going to happen!
Someone takes what I said out of context i.e. alters the facts, then 'RUMOR Monger's!' all over the place to create an effect that actually, when one knows where that kind of stuff comes from, indicates what the persons actual counter intentions are. BOO!
Do you work for the news media too?

Something else. You CAN NOT say with 100% certainty that the hash rate you see on your client side display is even accurate much less to what extent or degree! YOU CAN'T!

Your experiment was only necessary because YOU CAN'T READ, ASSIMILATE and DUPLICATE, DUMMY!! ;p

Wolfey2014

Hey Wolfey...Chill down, buddy...Why so angry...?...What's the problem...?

I saw you on some other thread(s) to react the similar manner...Relax Wolfey...RELAX...Take it easy...Have a beer...or something...

Angry? Nope.
A bit annoyed, yes!
To answer your question. READ THE POST! I give my reasons there in!
Geesh!
Wolfey2014

Hey Wolfey...I don't have impression that you are 20 years old...to act like a moody teenager...All this "Geesh", etc...Kapish...?...Capisco...?...

On the other hand I also ask you if you can post "POOL side" pics...this will end the discussion...in your favor...and will bring you more appreciation...and probably more customers for the mod...that you could imagine...including myself...  

Take it easy...

[Andareed]

I'm pretty skeptical that these are running at 600 kh/s at 1000-1100 MHz. Hash rate should scale roughly linear, so you would need to be running closer to 1400 MHz to achieve those hash rates. It's possible you just go really lucky, or the pool is incorrectly calculating your hashrate.

[wolfey2014]

Hello Miners and Friends,

we sat here the last 4 hours and soldered and tweaked and got our results now for presentation. We are a little sad that we have to say that "the cake was a lie".
The last hours we were really positive that we could achieve 600kh/s, which was stated by a user here in the forum. We tried a lot and even did go down from 1.1V PLL to 0.94V PLL (in 0.01V steps). In the end we gave up. The best, stable hashrate we could reach was 446.8kh/s at 1050mhz. We even replaced the "pencil" with some carbonglue-fluid, which acts like graphite but is more conductive and easier to use. (It doesnt fly away if you need to sneeze )


Our Final Board(2 Bridge Mod + 39k +1% resistor and "pencil mod")

Our Final PLL voltage:

Our Final Stable voltage(no HW):

And there it startet to throw HWs:

Nay nay - rumor mongering nay sayers! Off with your heads!  

No, really. I do believe I 'the user that said 600', stated clearly that over 600 was on POOL SIDE!
POOL SIDE! Which is variable from as little as less than 100KH/s to over 750KH/s.
Oh, and I have the screen shots to prove it.

I saw that someone else altered the facts before this happened, now look!
I knew this was going to happen!
Someone takes what I said out of context i.e. alters the facts, then 'RUMOR Monger's!' all over the place to create an effect that actually, when one knows where that kind of stuff comes from, indicates what the persons actual counter intentions are. BOO!
Do you work for the news media too?

Something else. You CAN NOT say with 100% certainty that the hash rate you see on your client side display is even accurate much less to what extent or degree! YOU CAN'T!

Your experiment was only necessary because YOU CAN'T READ, ASSIMILATE and DUPLICATE, DUMMY!! ;p

Wolfey2014

Hey Wolfey...Chill down, buddy...Why so angry...?...What's the problem...?

I saw you on some other thread(s) to react the similar manner...Relax Wolfey...RELAX...Take it easy...Have a beer...or something...

Angry? Nope.
A bit annoyed, yes!
To answer your question. READ THE POST! I give my reasons there in!
Geesh!
Wolfey2014

Hey Wolfey...I don't have impression that you are 20 years old...to act like a moody teenager...All this "Geesh", etc...Kapish...?...Capisco...?...

On the other hand I also ask you if you can post "POOL side" pics...this will end the discussion...in your favor...and will bring you more appreciation...and probably more customers for the mod...that you could imagine...including myself...  

Take it easy...

Sure, what ever you say, fella.  

Anyway, check page 15 , scroll down just over half way and you'll see MY pool side results in living color.
Proof, as you call it.

Wolfey2014

[wolfey2014]

"Thanks for the photos.  How safely can we mix and match these mods?  Right now I don't have an iron tip fine enough to do the soldering mods.  I either need to modify something or order something.  I am interested in the pencil mod and the fan mod tonight though.  Would this be safe? "

No, I don't believe the pencil mod will do any good without the other 3 mods.
Pencil mod is to adjust PLL voltage which needs to be corrected due to the offset caused by the two solder jumpers and the resistor is a timing adjustment to smooth out clock transitions at higher frequencies I believe.

You can make the 5V fan mod any time but remember, it's recommended for Scrypt only operation.
It has not been tested in SHA or SHA / Scrypt mining mode!

As usual, do these mods at your own risk!

Wolfey2014

Wolfey,

Somewhere in this thread is mention of a cgminer version that allows you to set the voltage.  I realize you're using cpuminer, but do you think all 4 mods are required if we used the voltage mod cgminer?  Value your opinion.

Thank you!

You ask a GOOD question.

Well, from what I understand - the current mods bypass having to soft set the PLL and Core Voltages.
Personally, I can't say one way or the other if soft adjustment is beneficial or not. I have heard 'rumored' that soft fine adjustment of 1000MHz  up to 1048MHz stable is possible.

I'm going to have to convert over to one of the other programs to be able to read client side hash rates.
I just don't like all the tweaking that is still going on over the course of cg and bfg during it's 'use as it evolves' phase. I am not interested, at this stage of their evolution, in becoming an unwilling guinea pig
I'm happy with cpuminer and what the pool's report in spite of their second by second inaccuracy. A bit of an understanding of odds, how they are avaraged out and common sense helps me figure out what the 'not so rough' improvement figure is for a given pod's clean hash rate.

Some sites like litecoinpool give a 24 hour average speed and that's close enough for me.
Besides and 'primarily' it only really counts what performance stats one sees pool side as that directly correlates to how much moolah it's gonna make you - the end product, the result, the win! $$$$

I guess I'm gonna have to learn one or the other 'unless there is a better alternative' because I could give a hoot what else the program does. All I want it to do is show me an accurate local hashrate as a separate program from the rest of it. And it MUST not be memory hungry like cpuminer 'isn't!' Thank you very much GridSeed!  
It would be cool if not great if GC would take back up updating / tweaking their very good and very stable program! It lives!  
IMO, cpuminer is very optimized and user friendly compared to the others IMO.
If the others were less complex and learning curve intensive, I'd of bothered to test them too by now.
Lots of other people like/love them though. So, to each their own.

Peace!
Wolfey2014

I created my own local database and query the pool api hourly to get their reported hashrate for my miners. 
On average, over 48hrs, the pool is reporting  20% more shares than cgminer, and I'm okay with that, lol. 

I use cgminer's api to query the health of my miners and to re-start them if necessary.  cgminer has the ability to run all or a selection. 
Its api can be set on different local network ports so if you have multiple instances running on one box you can query them separately. 

I've only had one hiccup with cgminer in the 7 days I've been running - but I think that was caused by my linux attempting to self-update,
which I've now disabled.

Thanks, that's good data to know.
I'm considering setting up a separate laptop to run Linux Ubuntu.
I'll have to learn it from scratch since the last low level language I messed with years ago was Assembly to program a PIC
Wolfey2014