Bitcoin Forum
June 16, 2024, 08:15:36 PM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 2 [3] 4 »  All
  Print  
Author Topic: FPGA!!!  (Read 3064 times)
Dashing
Full Member
***
Offline Offline

Activity: 138
Merit: 100


View Profile
May 26, 2013, 06:36:41 PM
 #41

true true,
Its not perfect atleast somewhat near,
just help us verify speed if you can via a screenshot of a pool Dashboard after running a few hours, with shares accepted, else it would be skewed due to time.

I don't mine on 1 pool, and that's been my strategy for long long time
I use the "load balance" option in CGminer to spread my hash rate around for many reasons,

1.
I mine on EclipseMC PPS, as you can withdraw balance anytime (no minimum amount)

2.
I mine on Ozcoin DGM, DGM = stable income

3.
I mine on BTCGuild PPLNS = shorter rounds = more $

4.
I mine on Slush pool = score based, high score = more $

and yeah, I dont' have the time to switch everything around to 1 pool just to prove a fact, the FPGAs are all connected to my RPi in my balcony in a case being cooled by natural cooling Smiley

I'm out of this topic as it seems to be going round and round again and again, and good luck finding something to mine with until much much cheaper and VIABLE ASICs come out


Pinwheel
Sr. Member
****
Offline Offline

Activity: 378
Merit: 250


View Profile
May 26, 2013, 07:05:32 PM
 #42

According to Blackarrow, a hashrate improvement to 500Mhps came from the firmware version. ie v42
The v42 also causes a reporting error which causes the speed to be reported as 1Ghps.

Although not perfect: an easy verification would be from the Web UI of the pool which shows accepted share number and speed,(btcguild or some pools report approx speed based on shares accepted) but you have run it for a few hours in the pool.

pool's reporting hash rate is not accurate. I run one blades for a day on btcguild and it reported 16 ghs which cannot be true.

they changed fan on smaller one, so with 1 ghs this board will melt down. As there is no official confirmation of 1ghs this thread is BS.

Tom Waits: We should just start as soon as possible cause we might catch a rabbit before we have our pants on. (Juxtapoz)
Dashing
Full Member
***
Offline Offline

Activity: 138
Merit: 100


View Profile
May 26, 2013, 07:30:06 PM
 #43

According to Blackarrow, a hashrate improvement to 500Mhps came from the firmware version. ie v42
The v42 also causes a reporting error which causes the speed to be reported as 1Ghps.

Although not perfect: an easy verification would be from the Web UI of the pool which shows accepted share number and speed,(btcguild or some pools report approx speed based on shares accepted) but you have run it for a few hours in the pool.

pool's reporting hash rate is not accurate. I run one blades for a day on btcguild and it reported 16 ghs which cannot be true.

they changed fan on smaller one, so with 1 ghs this board will melt down. As there is no official confirmation of 1ghs this thread is BS.

I'll be really honest, I have no idea man, cgminer reports 1 Ghash and gives good shares per minute too,


But yeah, I don't feel scammed in anyway, as Wilson never promised me 1 Ghash, he promised me cheap FPGAs and I got them  Tongue So I wouldn't call this a complete BS, as really even cgminer is just estimating against the number of shares sent per minute as cgminer cannot get the actual rate from the FPGA (this is a verified claim)
Pinwheel
Sr. Member
****
Offline Offline

Activity: 378
Merit: 250


View Profile
May 26, 2013, 07:52:53 PM
 #44

According to Blackarrow, a hashrate improvement to 500Mhps came from the firmware version. ie v42
The v42 also causes a reporting error which causes the speed to be reported as 1Ghps.

Although not perfect: an easy verification would be from the Web UI of the pool which shows accepted share number and speed,(btcguild or some pools report approx speed based on shares accepted) but you have run it for a few hours in the pool.

pool's reporting hash rate is not accurate. I run one blades for a day on btcguild and it reported 16 ghs which cannot be true.

they changed fan on smaller one, so with 1 ghs this board will melt down. As there is no official confirmation of 1ghs this thread is BS.

I'll be really honest, I have no idea man, cgminer reports 1 Ghash and gives good shares per minute too,


But yeah, I don't feel scammed in anyway, as Wilson never promised me 1 Ghash, he promised me cheap FPGAs and I got them  Tongue So I wouldn't call this a complete BS, as really even cgminer is just estimating against the number of shares sent per minute as cgminer cannot get the actual rate from the FPGA (this is a verified claim)

congratulations!! that is good machine and you for sure will make a lot of money on long run. If I did not have ASICs I will buy 10. even if it gives 500mgh it is not bad at all.

Tom Waits: We should just start as soon as possible cause we might catch a rabbit before we have our pants on. (Juxtapoz)
Dashing
Full Member
***
Offline Offline

Activity: 138
Merit: 100


View Profile
May 26, 2013, 07:56:46 PM
 #45

According to Blackarrow, a hashrate improvement to 500Mhps came from the firmware version. ie v42
The v42 also causes a reporting error which causes the speed to be reported as 1Ghps.

Although not perfect: an easy verification would be from the Web UI of the pool which shows accepted share number and speed,(btcguild or some pools report approx speed based on shares accepted) but you have run it for a few hours in the pool.

pool's reporting hash rate is not accurate. I run one blades for a day on btcguild and it reported 16 ghs which cannot be true.

they changed fan on smaller one, so with 1 ghs this board will melt down. As there is no official confirmation of 1ghs this thread is BS.

I'll be really honest, I have no idea man, cgminer reports 1 Ghash and gives good shares per minute too,


But yeah, I don't feel scammed in anyway, as Wilson never promised me 1 Ghash, he promised me cheap FPGAs and I got them  Tongue So I wouldn't call this a complete BS, as really even cgminer is just estimating against the number of shares sent per minute as cgminer cannot get the actual rate from the FPGA (this is a verified claim)

congratulations!! that is good machine and you for sure will make a lot of money on long run. If I did not have ASICs I will buy 10. even if it gives 500mgh it is not bad at all.

heh i sense a tad of sarcasm there Wink anyways, I have a few ASICs too, waiting on them though Wink
kano
Legendary
*
Offline Offline

Activity: 4522
Merit: 1844


Linux since 1997 RedHat 4


View Profile
May 27, 2013, 01:09:25 AM
Last edit: May 27, 2013, 01:21:26 AM by kano
 #46

According to Blackarrow, a hashrate improvement to 500Mhps came from the firmware version. ie v42
The v42 also causes a reporting error which causes the speed to be reported as 1Ghps.

Although not perfect: an easy verification would be from the Web UI of the pool which shows accepted share number and speed,(btcguild or some pools report approx speed based on shares accepted) but you have run it for a few hours in the pool.

pool's reporting hash rate is not accurate. I run one blades for a day on btcguild and it reported 16 ghs which cannot be true.

they changed fan on smaller one, so with 1 ghs this board will melt down. As there is no official confirmation of 1ghs this thread is BS.

I'll be really honest, I have no idea man, cgminer reports 1 Ghash and gives good shares per minute too,


But yeah, I don't feel scammed in anyway, as Wilson never promised me 1 Ghash, he promised me cheap FPGAs and I got them  Tongue So I wouldn't call this a complete BS, as really even cgminer is just estimating against the number of shares sent per minute as cgminer cannot get the actual rate from the FPGA (this is a verified claim)

congratulations!! that is good machine and you for sure will make a lot of money on long run. If I did not have ASICs I will buy 10. even if it gives 500mgh it is not bad at all.

heh i sense a tad of sarcasm there Wink anyways, I have a few ASICs too, waiting on them though Wink
It does 500MH/s
End of story - anyone thinking it does 1GH/s is being scammed by misrepresentation.

It's run time is 2 minutes so the MH/s numbers don't mean anything either.

You can fake the MH/s number up by giving the code the incorrect --icarus-timing or incorrect --icarus-options ... I know I wrote the code.

Edit: "verified claim" by who? That's bullshit. I wrote the damn code. I know what it does.
It counts the hashes done and estimates the hashes when it aborts work (that doesn't find a result or during an LP) based on the hash rate and the time when it aborts the work.
It's nothing to do with shares other than if a nonce is invalid after a hash validation, it will not count towards the MH/s work done (coz it is a hardware error)

Pool: https://kano.is - low 0.5% fee PPLNS 3 Days - Most reliable Solo with ONLY 0.5% fee   Bitcointalk thread: Forum
Discord support invite at https://kano.is/ Majority developer of the ckpool code - k for kano
The ONLY active original developer of cgminer. Original master git: https://github.com/kanoi/cgminer
Benson Samuel
Legendary
*
Offline Offline

Activity: 1890
Merit: 1000


Landscaping Bitcoin for India!


View Profile WWW
May 27, 2013, 01:47:51 AM
 #47

Suddenly a single clarification champ appears in the wild!

Wilson (OP)
Member
**
Offline Offline

Activity: 78
Merit: 10


View Profile
May 27, 2013, 01:55:33 AM
 #48

Lolz!
Just that he has bought a lot from me!
Its hard to convince people though you are right!
Proof is the most important to win the case! Tongue
Benson Samuel
Legendary
*
Offline Offline

Activity: 1890
Merit: 1000


Landscaping Bitcoin for India!


View Profile WWW
May 27, 2013, 01:57:26 AM
 #49

Alright, just saw OP edits and will pass on my 1.5 c.

OP had not posted expected hashrate. I would assume that he added
Quote
Its approx 4btc @ 1ghash!
, after checking with his customer - Dashing.

Dashing is testing the equipment that he purchased and only vouching for what he can see, accompanied by screenshots. Seems very legit and Dashing's words show that he does honestly believe the hashrate he sees.

Kano's confirmation of the same overrides what Dashing's tests have proved, unless we are talking about some other FPGA which has been customized.

In this event, I would either request a demo unit to be sent to kano to test and send back.
Or, some actual specs that can be investigated by the community.

Pinwheel
Sr. Member
****
Offline Offline

Activity: 378
Merit: 250


View Profile
May 27, 2013, 02:50:05 AM
 #50

According to Blackarrow, a hashrate improvement to 500Mhps came from the firmware version. ie v42
The v42 also causes a reporting error which causes the speed to be reported as 1Ghps.

Although not perfect: an easy verification would be from the Web UI of the pool which shows accepted share number and speed,(btcguild or some pools report approx speed based on shares accepted) but you have run it for a few hours in the pool.

pool's reporting hash rate is not accurate. I run one blades for a day on btcguild and it reported 16 ghs which cannot be true.

they changed fan on smaller one, so with 1 ghs this board will melt down. As there is no official confirmation of 1ghs this thread is BS.

I'll be really honest, I have no idea man, cgminer reports 1 Ghash and gives good shares per minute too,


But yeah, I don't feel scammed in anyway, as Wilson never promised me 1 Ghash, he promised me cheap FPGAs and I got them  Tongue So I wouldn't call this a complete BS, as really even cgminer is just estimating against the number of shares sent per minute as cgminer cannot get the actual rate from the FPGA (this is a verified claim)

congratulations!! that is good machine and you for sure will make a lot of money on long run. If I did not have ASICs I will buy 10. even if it gives 500mgh it is not bad at all.

heh i sense a tad of sarcasm there Wink anyways, I have a few ASICs too, waiting on them though Wink

sorry if you see it in that way, I really not mean any sarcasm. Why would I, it is good product and price is fair.

Tom Waits: We should just start as soon as possible cause we might catch a rabbit before we have our pants on. (Juxtapoz)
Wilson (OP)
Member
**
Offline Offline

Activity: 78
Merit: 10


View Profile
May 27, 2013, 04:29:41 AM
 #51

See...
I am not forcing anyone to buy this! I am tiered of convincing people. Those who believe may! No competition...
I have posted what i have seen.
I am not going to convince any more people!
I have made this post based on Dashing's miner performance! And what i saw was each board was at 998mhash and 1.02ghash. Avg of 994mhash. I think that's good enough and worth, as the shipping is within 20 days.
Atleast we all know that we as going to get a minimum of 800mhash each board as they 2X chips are used!
So you order 10 of them, you can be sure of getting a minimum of 9ghash as Kano said but according to Dashing, you will actually be getting the other 1ghash too!
I posted what i saw! Sounded pretty good as FPGAs can be resold once they get outdated unlike the ASICs as they are application specific!
So its up to you guyz to decide to buy or not to. I have already received a bulk order.
Anyone still interested, just PM me!


Cheers,
Wilson
rezurect
Sr. Member
****
Offline Offline

Activity: 686
Merit: 250


View Profile
May 27, 2013, 05:08:59 AM
Last edit: May 27, 2013, 05:46:21 AM by rezurect
 #52

Atleast we all know that we as going to get a minimum of 800mhash each board as they 2X chips are used!

For Wilson^^ who is still maintaining its 800Mhps. Its a 500Mhps max perboard (250 per chip)
Again,
It does 500MH/s
End of story - anyone thinking it does 1GH/s is being scammed by misrepresentation.
It's run time is 2 minutes so the MH/s numbers don't mean anything either.
You can fake the MH/s number up by giving the code the incorrect --icarus-timing or incorrect --icarus-options ... I know I wrote the code.

I told them politely to run it for hours, instead of putting it straight in their face, they don't take heed. They needed the dev to come here personally and say it right in their face that's its a "fudged" figure, to make them understand.


Oh, and before I forget - you can get them in bulk at $260 each.  Roll Eyes

I wonder what brought Kano here  Grin
Wilson (OP)
Member
**
Offline Offline

Activity: 78
Merit: 10


View Profile
May 27, 2013, 05:19:47 AM
 #53

Lolz! I myself am wondering...   Tongue
Benson Samuel
Legendary
*
Offline Offline

Activity: 1890
Merit: 1000


Landscaping Bitcoin for India!


View Profile WWW
May 27, 2013, 05:20:27 AM
 #54

Quote
I wonder what brought Kano here  

Check the OP. One of the Global Moderators added an edit and might have pinged Kano.

rezurect
Sr. Member
****
Offline Offline

Activity: 686
Merit: 250


View Profile
May 27, 2013, 05:22:54 AM
 #55

or somebody could have posted it on the Lancelot main thread calling for Kanos attention, you know somebody
Benson Samuel
Legendary
*
Offline Offline

Activity: 1890
Merit: 1000


Landscaping Bitcoin for India!


View Profile WWW
May 27, 2013, 05:29:27 AM
 #56

or somebody could have posted it on the Lancelot main thread calling for Kanos attention, you know somebody

Hahahah, just saw that.

Dashing
Full Member
***
Offline Offline

Activity: 138
Merit: 100


View Profile
May 27, 2013, 05:42:27 AM
 #57

Alright, just saw OP edits and will pass on my 1.5 c.

OP had not posted expected hashrate. I would assume that he added
Quote
Its approx 4btc @ 1ghash!
, after checking with his customer - Dashing.

Dashing is testing the equipment that he purchased and only vouching for what he can see, accompanied by screenshots. Seems very legit and Dashing's words show that he does honestly believe the hashrate he sees.

Kano's confirmation of the same overrides what Dashing's tests have proved, unless we are talking about some other FPGA which has been customized.

In this event, I would either request a demo unit to be sent to kano to test and send back.
Or, some actual specs that can be investigated by the community.

Hmm, i think this is what might have happened :/ I got the boards from Wilson (he never promised me 1 Ghash) and then was surprised to see the hash rate on the first 2 boards and told him about the hashrate, hence the reason why he probably offered it at 1 ghash

I never fudged/played with any stats,
I just use --ircrus-timing short as usual and nothing else, I don't know why its reporting that, i'll probably recompile cgminer agian from git Smiley
Benson Samuel
Legendary
*
Offline Offline

Activity: 1890
Merit: 1000


Landscaping Bitcoin for India!


View Profile WWW
May 27, 2013, 05:45:41 AM
 #58

Quote
Hmm, i think this is what might have happened :/ I got the boards from Wilson (he never promised me 1 Ghash) and then was surprised to see the hash rate on the first 2 boards and told him about the hashrate, hence the reason why he probably offered it at 1 ghash

I never fudged/played with any stats,
I just use --ircrus-timing short as usual and nothing else, I don't know why its reporting that, i'll probably recompile cgminer agian from git

Exactly what i thought too :p

Wilson: May be good if you could add this to the posts and clear it up.

Dashing
Full Member
***
Offline Offline

Activity: 138
Merit: 100


View Profile
May 27, 2013, 05:46:46 AM
 #59

Quote
Hmm, i think this is what might have happened :/ I got the boards from Wilson (he never promised me 1 Ghash) and then was surprised to see the hash rate on the first 2 boards and told him about the hashrate, hence the reason why he probably offered it at 1 ghash

I never fudged/played with any stats,
I just use --ircrus-timing short as usual and nothing else, I don't know why its reporting that, i'll probably recompile cgminer agian from git

Exactly what i thought too :p

Wilson: May be good if you could add this to the posts and clear it up.


Yep, looks like this stupid confusion around caused a lot of trouble for Wilson :/
rezurect
Sr. Member
****
Offline Offline

Activity: 686
Merit: 250


View Profile
May 27, 2013, 05:52:59 AM
 #60

I did tell you guys about results being "skewed by time".
No matter how many times you recompile cgminer, the error will stay till the devs fix it in a new version.
Pages: « 1 2 [3] 4 »  All
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!