Pages:
Author

Topic: High Efficiency FPGA & ASIC Bitcoin Mining Devices https://BTCFPGA.com - page 45. (Read 218469 times)

legendary
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
Thanks AmDD and Plazzman for the info - much appreciated.

After 6 more hours of running it's got down to 1.3% HW and 837MH/s
It's 9:50am here and will be interesting to see how it handles the day getting warm here
- though it is in my basement garage, the temp down there does change, but more slowly, so the day peek is always much lower than outside

It's set to change less the longer it runs i.e. early faster changing at startup (minutes) but slower changes up or down as time proceeds (up to an hour/hours) - that's controlled by a set of constants, but I may make them --modminer-options (something like --icarus-options) so people who have boards that perform better (or worse) or are in more (or less) stable environments can possibly change some of them (within certain limits) at runtime

kano, to what extent will you be supporting the bASIC and how well do you foresee cgminer running on Ubuntu 12.04?
ckolivas and I are 2 of the developers getting a dev bASIC for ensuring early cgminer support.

This work on the MMQ (mainly the rethreading part I'll get to next) is also very much in line with supporting the bASIC (and other ASIC) due to the major performance issues they will represent.

cgminer already runs on 12.04
I also have a 12.04 VM setup, but idle, since I don't need it while ckolvias binaries support and run on 12.04
... I created it in case anyone ever asked me for a non GPU 12.04 binary Smiley

I also have the 11.04 VM that I make the 'Version'a release each time in the cgminer thread since there is no 11.04 binary released by ckolivas (and I need it Tongue)
That will continue while GPU mining is still (in my opinion) an option, and I will start creating 12.04 binaries (in my downloads) if they are ever needed for a different version to what ckolivas creates if required

Hopefully that answers your questions Smiley
legendary
Activity: 1260
Merit: 1000
Far fetched?  Really?  I think not, we are actually being conservative in our power announcements in case there is any unexpected difficulties.  Can you tell me how you arrived at your flawed conclusions?  Why are we talking about BFL again, though?  Lets talk about bASIC since it's a bASIC thread.

The only reason I am here is because Tom is lying about BFLs power.  I am responding to his lies, and since he seems to know our power situation better than we do, it must mean he knows HIS power situation and is not sharing it with his customers.  I would like to know why that is.

I thought you didn't have a prototype Josh? If that's the case how can you be so certain that BFL's power consumption numbers are so very conservative? 1w/Gh is it? Hasn't BFL been way off previously?

When did Tom lie about BFL's power numbers? If you don't have a working final product and you don't have a prototype, then aren't BFL's power consumption numbers by definition an estimate? Also, he said competitorS, and Avalon's estimates are much higher and far less narrowly defined than BFL's. Given BFL's track record on this issue, I too am inclined to doubt the validity of BFL's power consumption claims. Time will reveal all things, but your question here has been asked and answered.

Where?  Right here:

Sure I can post a "simulated" or "estimated" power usage number - but its not going to be accurate. You really think BFL is going to give 1Gh/s per watt? keep dreaming.

These energy efficiency numbers from my competitors are simply estimates and are not going to be accurate. I give you my personal guarantee that our products will have energy efficiency that is competitive to any other ASIC product on the market.

I also want to know what is "competitive."  Is 2x the power consumption considered "competitive?" (Personally, I think 10 - 20% is competitive, not +100%, but I want to know what Tom's definition is.)

Again, I don't want to make this about anything but bASIC.  Stop talking about BFL stuff (besides, who says we don't have a prototype?  I've certainly not said that.).

member
Activity: 85
Merit: 10
Soo Inaba/Josh.... Still waiting on your so called refund???
Bee nice if there was 3 months interest on that but I'm sure there isn't
sr. member
Activity: 434
Merit: 250
Far fetched?  Really?  I think not, we are actually being conservative in our power announcements in case there is any unexpected difficulties.  Can you tell me how you arrived at your flawed conclusions?  Why are we talking about BFL again, though?  Lets talk about bASIC since it's a bASIC thread.

The only reason I am here is because Tom is lying about BFLs power.  I am responding to his lies, and since he seems to know our power situation better than we do, it must mean he knows HIS power situation and is not sharing it with his customers.  I would like to know why that is.

I thought you didn't have a prototype Josh? If that's the case how can you be so certain that BFL's power consumption numbers are so very conservative? 1w/Gh is it? Hasn't BFL been way off previously?

When did Tom lie about BFL's power numbers? If you don't have a working final product and you don't have a prototype, then aren't BFL's power consumption numbers by definition an estimate? Also, he said competitorS, and Avalon's estimates are much higher and far less narrowly defined than BFL's. Given BFL's track record on this issue, I too am inclined to doubt the validity of BFL's power consumption claims. Time will reveal all things, but your question here has been asked and answered.
member
Activity: 85
Merit: 10
Inaba Please refund the following order...  Thanks...
All no this was not for sale... I would rather get the refund and buy the basic. Sorry..!!
Keep trolling Lol



Dear Bradley,

I have applied your previous payment for 2 BitForce Singles to a new order for one BitForce Single SC and have sent you a PayPal invoice for the difference ($91). Upon receipt of this payment your order upgrade will be confirmed and your order date will become 6/23/2012 which is the earliest date we began accepting pre-orders for the SC products.

Kind regards,

John
BF Labs, Inc. Customer Service

On Sat, Jul 7, 2012 at 7:59 AM, PlazzMan <[email protected]> wrote:

    Hello Customer Service,

      On Mon. June 18, 2012 a had ordered 2 BitForce SHA256 Singles as you will see from the receipt posted below.
      I would like to convert it into preorder for 1 BitForce Single ‘SC’  40 GH/s  -  $1,299
      I would have no problem paying the difference in usd which I believe is $101 difference.
      Will be awaiting an Invoice from BFL Labs.

    Thank You, Brad Goodnight
    [email protected]




    -----Original Message-----
    From: Butterfly Labs Inc <[email protected]>
    To: plazzman05 <[email protected]>
    Sent: Fri, Jun 22, 2012 12:49 pm
    Subject: Re: New submission from Order Form - BitForce SHA256 Single

    Hi Bradley, thank you for your order.

    Your PayPal payment has been received in full and your 2 BitForce SHA256 Singles are secure. Your date of purchase is 06/18/2012. I will send you tracking when your order ships.

    Thanks again for your interest in our product.

    Kind regards,
    Jody D.
    Customer Service
    BF Labs, Inc.


    On Mon, Jun 18, 2012 at 10:51 PM, Bradley Goodnight <[email protected]> wrote:

        Preferred payment method
legendary
Activity: 1260
Merit: 1000
blah blah blah I am a troll and try my hardest to make my shit company's rep even worse by trolling competitors...

Tom has already stated he will provide an estimate when he gets the boards. Since he wants to give a realistic number not a far-fetched (1 watt/gh) or broad (2-10 watt/gh) estimate. Yet here you are, bitching about transparency when you represent the company that is the LEAST transparent of all.

Why are you so concerned about this Josh? Why do you feel the need to trying to force your competitor into giving numbers that aren't based on reality? Have you received a hint at what the actual performance of your (BFL) devices are and feel threatened? Unlike BFL, Tom seems to prefer to give realistic specs that are reliable (see MMQ vs BFL single past history) and can be trusted.



Far fetched?  Really?  I think not, we are actually being conservative in our power announcements in case there is any unexpected difficulties.  Can you tell me how you arrived at your flawed conclusions?  Why are we talking about BFL again, though?  Lets talk about bASIC since it's a bASIC thread.

The only reason I am here is because Tom is lying about BFLs power.  I am responding to his lies, and since he seems to know our power situation better than we do, it must mean he knows HIS power situation and is not sharing it with his customers.  I would like to know why that is.

member
Activity: 112
Merit: 10
blah blah blah I am a troll and try my hardest to make my shit company's rep even worse by trolling competitors...

Tom has already stated he will provide an estimate when he gets the boards. Since he wants to give a realistic number not a far-fetched (1 watt/gh) or broad (2-10 watt/gh) estimate. Yet here you are, bitching about transparency when you represent the company that is the LEAST transparent of all.

Why are you so concerned about this Josh? Why do you feel the need to trying to force your competitor into giving numbers that aren't based on reality? Have you received a hint at what the actual performance of your (BFL) devices are and feel threatened? Unlike BFL, Tom seems to prefer to give realistic specs that are reliable (see MMQ vs BFL single past history) and can be trusted.

legendary
Activity: 1764
Merit: 1002
Thanks AmDD and Plazzman for the info - much appreciated.

After 6 more hours of running it's got down to 1.3% HW and 837MH/s
It's 9:50am here and will be interesting to see how it handles the day getting warm here
- though it is in my basement garage, the temp down there does change, but more slowly, so the day peek is always much lower than outside

It's set to change less the longer it runs i.e. early faster changing at startup (minutes) but slower changes up or down as time proceeds (up to an hour/hours) - that's controlled by a set of constants, but I may make them --modminer-options (something like --icarus-options) so people who have boards that perform better (or worse) or are in more (or less) stable environments can possibly change some of them (within certain limits) at runtime

kano, to what extent will you be supporting the bASIC and how well do you foresee cgminer running on Ubuntu 12.04?
legendary
Activity: 1260
Merit: 1000
Quote
Still waiting for that link to the discussion.....

Then get busy and make a post on the BFL forums and I'm sure you'll get an answer.

Quote
Why do you need deny the problems you've had with your pool? Even I sent you a mail regarding my 7.2GH/s miner showing up as 5.5 GH/s on your pool, just to get ignored.

I'm not denying there were some problems, I am denying this handy little fib from you:

Quote
Mistakes all over the place, and very defensive reactions. One time I posted a strange block time artifact, he replied that it was fixed and how could I be so stupid as to post that.

I have never responded to anyone on the EMC (or BFL) forums with "How could you be so stupid as to post that" or anything like that in relation to problems with the pool.  I have no idea what you mean about "defensive reactions" either. Like I said, post a link to the threads where it happened or you're just making it up.

Quote
I really think/hope you've got better things to do. So why don't you just do us all a favor and stop hijacking your employer's competitor's thread? Thanks.

I'm not hijacking it, I have no wish to discuss EMC or BFL in this thread.  I want to know about the power consumption of the bASIC device.  Stop bringing up things not related to the bASIC device and the thread won't be hijacked. 

Again, for those that are having trouble with this:  I want to know a specific (even if it's just a ballpark) metric of the bASIC, of which I am a pre-order customer.  I have exactly zero desire to discuss anything else in this thread.  I simply want an answer to "How much power will the bASIC consume, roughly?"  also "What is considered "competitive" in terms of power consumption?  If it consumes 2w/GH, (2x the competition) is that still considered "competitive?"
legendary
Activity: 1890
Merit: 1003
Thanks AmDD and Plazzman for the info - much appreciated.

After 6 more hours of running it's got down to 1.3% HW and 837MH/s
It's 9:50am here and will be interesting to see how it handles the day getting warm here
- though it is in my basement garage, the temp down there does change, but more slowly, so the day peek is always much lower than outside

Cool way to passively insulate from the ambient heat of the day.

I am starting to think BFL_Josh (Inaba) is avoiding the questions of ambient operating temperature for the BFL Single. Which tells me his hardware may run very hot and may need an Air Conditioner to keep it's temps down.

@ Inaba

Still waiting for that link to the discussion.....
legendary
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
Thanks AmDD and Plazzman for the info - much appreciated.

After 6 more hours of running it's got down to 1.3% HW and 837MH/s
It's 9:50am here and will be interesting to see how it handles the day getting warm here
- though it is in my basement garage, the temp down there does change, but more slowly, so the day peek is always much lower than outside

It's set to change less the longer it runs i.e. early faster changing at startup (minutes) but slower changes up or down as time proceeds (up to an hour/hours) - that's controlled by a set of constants, but I may make them --modminer-options (something like --icarus-options) so people who have boards that perform better (or worse) or are in more (or less) stable environments can possibly change some of them (within certain limits) at runtime
full member
Activity: 215
Merit: 101
Any one want to buy a BFL SC single pre order.Huh  Early order...!!! Trade you for a basic....  Smiley Oh and some coins...  Grin
PM'd
member
Activity: 85
Merit: 10
Meanwhile on the actual thread subject ... can anyone post their numbers for a recent MMQ?

I'm running the v0.4-ljr-alpha firmware (without TML) and the bitstream from back whenever (from cgminer)

I've got cgminer working OK with my MMQ but I wonder if I've been rather optimistic in adjusting the clocks?

I'll put up a pull request of the first set of changes once I'm sure it's not too optimistic and going to cause anyone's MMQ to get damaged easily.

Anyway my numbers on a 2hr 20min run are (1 diff shares):
Temps are 36 to 42
Av 838 MH/s
A:1569  R:4  HW:30 -> so 1.9% HW errors
(and yes those numbers are all correct if anyone was wondering)

Edit: here's a screen shot 20 minutes later


Kano

  I get the same result as AmDD with Ubuntu 12.05 on a 32 bit laptop and BFGminer.
So I would say you are In the right ballpark.

--Cheers 
legendary
Activity: 1176
Merit: 1001

Please provide evidence or it didn't happen.  A link to the relevant posts would suffice, thanks!  Cool story, bro.


Inaba/Josh, I'm really beginning to regret I paid you back those 138BTC what your pool's bug paid me. I thought you were a nice guy. I'm beginning to think I was mistaken.

Why do you need deny the problems you've had with your pool? Even I sent you a mail regarding my 7.2GH/s miner showing up as 5.5 GH/s on your pool, just to get ignored.

I really think/hope you've got better things to do. So why don't you just do us all a favor and stop hijacking your employer's competitor's thread? Thanks.
Best post i've read in a while.
hero member
Activity: 560
Merit: 500

Please provide evidence or it didn't happen.  A link to the relevant posts would suffice, thanks!  Cool story, bro.


Inaba/Josh, I'm really beginning to regret I paid you back those 138BTC what your pool's bug paid me. I thought you were a nice guy. I'm beginning to think I was mistaken.

Why do you need deny the problems you've had with your pool? Even I sent you a mail regarding my 7.2GH/s miner showing up as 5.5 GH/s on your pool, just to get ignored.

I really think/hope you've got better things to do. So why don't you just do us all a favor and stop hijacking your employer's competitor's thread? Thanks.
legendary
Activity: 1260
Merit: 1000
I wonder if Inaba the Douche-Nozzle spent less time trolling, and more time on his pool, if it would go for more than a day without issues.

A few months ago I switched to EMC to try it out. After running there for a week or so I notice my payout dropped by 10-15% vs. other pools. The luck seemed to be about average but there were constant connection and other issues. Inaba kept posting that it was this or that issue and no worries it's fixed, but then a day later it would crap out again. This was around when he introduced the diff server which seemed to knock out the us1,2,3 servers.

It was clear Inaba has no idea how to run a production server as he was treating production like a test environment. Mistakes all over the place, and very defensive reactions. One time I posted a strange block time artifact, he replied that it was fixed and how could I be so stupid as to post that. But his original reply that it was fixed happened BEFORE the time stamp on the problem I brought up, so obviously it wasn't fixed. Childish.

The thing is you can't do this with ASICs, you have to be 100% fully tested and 100% confident before tape-out. I used to be in the field and have completed many designs, mistakes can not be allowed. If EMC is any indication of BFL's quality control, I'd be worried if I was a pre-order customer there...

Please provide evidence or it didn't happen.  A link to the relevant posts would suffice, thanks!  Cool story, bro.

sr. member
Activity: 272
Merit: 250
Cryptopreneur
I wonder if Inaba the Douche-Nozzle spent less time trolling, and more time on his pool, if it would go for more than a day without issues.

A few months ago I switched to EMC to try it out. After running there for a week or so I notice my payout dropped by 10-15% vs. other pools. The luck seemed to be about average but there were constant connection and other issues. Inaba kept posting that it was this or that issue and no worries it's fixed, but then a day later it would crap out again. This was around when he introduced the diff server which seemed to knock out the us1,2,3 servers.

It was clear Inaba has no idea how to run a production server as he was treating production like a test environment. Mistakes all over the place, and very defensive reactions. One time I posted a strange block time artifact, he replied that it was fixed and how could I be so stupid as to post that. But his original reply that it was fixed happened BEFORE the time stamp on the problem I brought up, so obviously it wasn't fixed. Childish.

The thing is you can't do this with ASICs, you have to be 100% fully tested and 100% confident before tape-out. I used to be in the field and have completed many designs, mistakes can not be allowed. If EMC is any indication of BFL's quality control, I'd be worried if I was a pre-order customer there...

So which pools do you suggest?
legendary
Activity: 1027
Merit: 1005
2 MMQ's running BFGMiner  2.8.1 on Windows 7.

legendary
Activity: 4592
Merit: 1851
Linux since 1997 RedHat 4
Meanwhile on the actual thread subject ... can anyone post their numbers for a recent MMQ?

I'm running the v0.4-ljr-alpha firmware (without TML) and the bitstream from back whenever (from cgminer)

I've got cgminer working OK with my MMQ but I wonder if I've been rather optimistic in adjusting the clocks?

I'll put up a pull request of the first set of changes once I'm sure it's not too optimistic and going to cause anyone's MMQ to get damaged easily.

Anyway my numbers on a 2hr 20min run are (1 diff shares):
Temps are 36 to 42
Av 838 MH/s
A:1569  R:4  HW:30 -> so 1.9% HW errors
(and yes those numbers are all correct if anyone was wondering)

Edit: here's a screen shot 20 minutes later
hero member
Activity: 697
Merit: 500
This is just pathetic having a BFL shill try to stir up shit in this thread. I don't care if Inaba has a bASIC on pre-order, he posts still reek of trolling. I had considered splitting up my order between the two companies to protect myself from late delivery but I think instead I'll go bASIC and bitcoins.

Why did BFL hire Inaba? If they needed someone to spread FUD wouldn't it have been cheaper to just throw a few bitcoins at forum trolls to get the job done?
Pages:
Jump to: