[ANN] Spondoolies-Tech - carrier grade, data center ready mining rigs

[jimmothy]

Bick I honestly think it's time for you to admit that estimated specs are bullshit.

The only thing they are good for is giving investors hope and intimidating the competition.

[1714141137]

1714141137

[Guy Corem]

Bick I honestly think it's time for you to admit that estimated specs are bullshit.

The only thing they are good for is giving investors hope and intimidating the competition.

That's our roadmap. It based on tech we're developing.

[Guy Corem]

True when they called about the fire. I can see in the miner stats that the fire was already 1 hour going. So they where all sleeping ;-).  and stupid me trusted them with all this equipment. That's my big mistake. But again, I can't turn back time. No one hurt and a lesson learned.

Man, you are VERY passive for losing a ton of money and profit. You must be some rich kid playing with mom's money or never earned the money yourself? Usually when you earn money and lose it, it hurts a lot more then when just given to you.

I have patient with your weekend trolling, started after I didn't agree to a deal you requested privately.

I don't have patient for your inappropriate and wicked comments here and before.

Guy

I don't think I used the message system here in months. I think you got the wrong dude, also my comments aren't even near being wicked. I've seen a lot worse.

I was referring to email exchange we had few months ago regarding SP10. I really don't care about your posts as long as they're dealing with hardware ROI, Spondoolies-Tech, etc. Keep posting them.

Your remarks towrads cowboyminer are wicked. Maybe there are worse outside this thread, but it's not an excuse.

Guy

[jimmothy]

Bick I honestly think it's time for you to admit that estimated specs are bullshit.

The only thing they are good for is giving investors hope and intimidating the competition.

That's our roadmap. It based on tech we're developing.

Are you guys already working on the ASIC design that you estimate will achieve 0.05 j/gh at $0.03/gh or is it just an extrapolation/hope?

Also is it still the plan to have $0.2/gh at 0.1 j/gh hardware by the end of the year?

[Guy Corem]

Bick I honestly think it's time for you to admit that estimated specs are bullshit.

The only thing they are good for is giving investors hope and intimidating the competition.

That's our roadmap. It based on tech we're developing.

Are you guys already working on the ASIC design that you estimate will achieve 0.05 j/gh at $0.03/gh or is it just an extrapolation/hope?

Also is it still the plan to have $0.2/gh at 0.1 j/gh hardware by the end of the year?

Yes to the first, No to the second.

[2112]

Well, we all know how good Spondoolies' projections turn out to be. With AM, Bitfury and Spondoolies all talking 0.2 J/Gh for the next generation of ASICs early next year, 0.05 J/Gh mid 2015 for the generation after that is clearly marketing bullshit.

With total absence of a reliable technical information about the chips it is hard to make predictions. About the only source available are the "preliminary data sheets". But even with those one could make educated guesses.

1) If the future chips still contain BIST circuitry/instructions that means that they still weren't able to internalize their own experience that dedicated test circuitry is nearly a complete waste of silicon resources and that the chips have to be continuously calibrated while running, not just tested at the start.

Other things that may indicate that they were finally capable of doing a true full custom/analog design:

2) BIST is still present, but no longer gives binary result PASS/FAIL, but some vector of values

3) PLL programming information will have additional parameters besides "frequency", either duty cycle (instead of the default 50%) or explicit phase shifts (for multi-phase clocking)

4) package information showing more than one "core" voltage, eg. positive Vdd and negative Vss (in addition to ground GND)

(3) and (4) would mean that they are capable of exploiting imperfections in the "C" of "CMOS".

Edit: Anyway, there is a Jewish proverb (approximately): Don't ask the tailor if the suit he's making is going to fit you.

[Bicknellski]

Bick I honestly think it's time for you to admit that estimated specs are bullshit.

The only thing they are good for is giving investors hope and intimidating the competition.

Quoted for future reference.

I have no doubt they believe they can reach those numbers.

I have not seen once in these forums or in their media releases any attempt other than to challenge the competition to keep pace with them. What you see as a ploy I see as supreme confidence in a team. Having met a number of these people don't kid yourself on that these people are a solid professional team. They work at being a team all the time and it shows.

If you feel threatened as a shareholder of AM stock ok that is fine I have no money in SPTech or AM at this point. Given the economies of scale required I doubt I will be buying any hardware unless it is for use at our school on altcoins as part of our ICT curriculum we have. Given the mess with AM currently you should be more worried about what they are promising given SPTech has been pretty much spot on with most of the design and delivery of their products.

It is time to admit they rarely if ever have failed to make their customers happy. That has nothing to do with what is possible or probable chip design wise but SPTech is very very careful to craft their message in order to not fail at the critical moment and even the time they did miss their numbers they compensated their customers proactively. As one of those customers I was happy to receive the unit even though it didn't hit the numbers it certainly mined well, retained its value right up until it was sold and I came out on top even when BTC was plummeting to sub 400. As a consumer I can't be any happier than that really. Now look at AM threads? Are you really flinging shit here and not there? Seriously.

Now we are not talking home miners anymore. We are talking larger scale outfits with massive needs. Who would you want? I think that question gets answered pretty fast after you meet Guy and his team. As well that is why you see teams like Bitmaintech or Asicminer and especially BitFury starting to take a more serious approach to building a team that can win not just survive. Right now SPTech has the better team all around.

One other glaring aspect is the fact that of the big four SPTech is the only one not playing with larger mining operations or cloud mining as part of their portfolio. That has implications as they have less exposure to the issues of large scale farms as an investment and a much more defined focus on product and design alone. I really think many are going to opt for SPTech over those who are in direct competition with them for ever decreasing btc from mining. I think they have a significant advantage over everyone else since they are only supporting the build of a customers mining operation. There is no direct competition unlike BitFury, Bitmaintech, KnC or Asicminer or Rockminer that impact their potential industrial clients bottom line when mining.

[samsonn25]

Bick I honestly think it's time for you to admit that estimated specs are bullshit.

The only thing they are good for is giving investors hope and intimidating the competition.

That's our roadmap. It based on tech we're developing.

Are you guys already working on the ASIC design that you estimate will achieve 0.05 j/gh at $0.03/gh or is it just an extrapolation/hope?

Also is it still the plan to have $0.2/gh at 0.1 j/gh hardware by the end of the year?

Yes to the first, No to the second.

Currently asics need to be priced $ 190 TH at .6 watts efficiency just to  break even. Its hard to believe to see on the market by mid next year that there will  be $30 1TH machines with the increased  efficiency.  Not to mention  difficulty  will  be  5-6 times higher  than today.

[Guy Corem]

... Its hard to believe to see on the market by mid next year that there will  be $30 1TH machines with the increased  efficiency. ...

The form factor is certainly not 1TH $30 machine ...

[zvisha]

I got notification from a client that version 2.5.7 caused him to loose ASICs on machine.
I looked at the machine and that is a possibility.
Please don't upgrade (and if you upgraded - downgrade back to 2.5.6) till I investigate

[zvisha]

Is anyone else getting errors like this on firmware 2.5.7 and 2.5.9?

Code:

LOOP[8] ON
24: disabled (Asic all engines fail BIST)
25: DC2DC/1/:[vlt1:671 vlt2:674(DCl:794 Tl:794 Ul:749) 50W  74A  38c] ASIC:[ 85c (120c) 580hz(BL: 580)   18 (E:193) F:0]
26: DC2DC/1/:[vlt1:669 vlt2:674(DCl:794 Tl:794 Ul:749) 48W  71A  38c] ASIC:[ 85c (120c) 555hz(BL: 555)   12 (E:193) F:0]

They get cleared by restarting minergate, seem to happen at random, and often come with very large (0.5 to 2 TH/s) reductions in hashrate. I started to get them recently on our machines, and they annoy me.

In other news, I did another test on the effects of fan speed at lower temperature (around 9C -- it's night here now), and they came out 4 GH/s in favor of 70. My guess is that at intake temps above 17C, fan speeds at 80 are better, and below 16C, fan speeds at 70 are better.

I am investigating.
Please downgrade to 2.5.6.

[zvisha]

Does anyone else see problem with 2.5.7?

[xandriel]

Does anyone else see problem with 2.5.7?

I had been running 2.5.7, then upgraded to 2.5.9 last night. Neither were giving any noticeable errors. I have since downgraded to 2.5.6 as a precaution.

I should add that this is for the SP31

[Biffa]

Does anyone else see problem with 2.5.7?

Not noticed a problem with mine on 2.5.7.

[Mabsark]

One other glaring aspect is the fact that of the big four SPTech is the only one not playing with larger mining operations or cloud mining as part of their portfolio. That has implications as they have less exposure to the issues of large scale farms as an investment and a much more defined focus on product and design alone. I really think many are going to opt for SPTech over those who are in direct competition with them for ever decreasing btc from mining. I think they have a significant advantage over everyone else since they are only supporting the build of a customers mining operation. There is no direct competition unlike BitFury, Bitmaintech, KnC or Asicminer or Rockminer that impact their potential industrial clients bottom line when mining.

When buying miners though you've got to add in the costs of shipping, which is a considerable amount. Look at the SP20, for example.

It costs $1190 and shipping is $115 for me. That's almost 10% of the cost. Then you need a 1300W PSU which costs about another $200, so about $1500. Given that the hashrate is 1700 Gh/s and it consumes 1200W then it costs about 0.88 $/Gh @ 0.71 J/Gh. If electricity cost 0.1 $/kWh, that's 2.88 $/day. With the network hash rate at 290,522,381 Gh/s, 1700 Gh/s would represent 0.000585% of the hashing power, therefore would receive that much of the reward giving 0.0211 BTC/day. At 344 $/BTC, that's 7.26 $/day, minus the electricity costs that's 4.38 $/day and dropping.

Compare that with AMHash which you can purchase for <= 1.25 mBTC, which at 344 $/BTC is upto 0.43 $/Gh. After fees, 1700 AMHash shares would pay out 0.0216 BTC/day and dropping.

So, 1700 AMHash shares are equivalent to running an SP20 with electricity costs of 0.1 $/kWh, only they cost less than half the price of an SP20 and there's simply no hassle. It's simply a far more efficient way of selling hash power and I'll be seriously surprised if Spondoolies' don't start offering cloud mining services too.

[zvisha]

Does anyone else see problem with 2.5.7?

i dont recall what i updated to but it was the latest non experimental version.

as of 2.5 hrs ago my sp20 is offline with no obvious cause.

i am busy with data recovery or else i would have noticed earlier.

Send port forwarding to my mail/skype I will take a look

[Bicknellski]

<snippy snip>

I am not buying miners to earn BTC or ROI and cloud mining is useless as an educational tool for my school. I am not personally buying any unprofitable cloud mining you have me confused with someone else obviously or didn't read my post all the way through. I am pretty certain you don't have all the numbers for larger shipments per kg that SPTech might have for large palette shipping but again I am not going to be buying millions of dollars of miners am I. I suspect again your current AMHash with current tech is not going to be profitable even for AM to run much longer. AMHash may want to order something in 2015 that out performs AM stuff so you can call SPTech then.

[RoadStress]

Does anyone else see problem with 2.5.7?

I had one unit that had the same problem as jtoomim, but restart got the chip back up. I am downgrading to 2.5.6 right now.

[TKeenan]

Bick I honestly think it's time for you to admit that estimated specs are bullshit.

The only thing they are good for is giving investors hope and intimidating the competition.

That's our roadmap. It based on tech we're developing.

Yeah, but this time maybe you ought to check for 'slow corners' before you take shitloads of people's money and deliver woefully short performance. 

[mavericklm]

'slow corners' go with that vocabulary in amt, ba or knc thread!!!
I see top hardware, good communication, best user support, nice ethics!!!
Da fuck do u want???