Unofficial Spondoolies SP20 thread

[HolgerDansk]

My next rig has to be Spondoolies.

Used? Or are you seriously looking to be a B2B customer and their newer offerings?

I was thinking used product as well as new.

However, after looking at the SP-T announcement today, that sure adds a big wrinkle to BTC things.

[philipma1957]

My next rig has to be Spondoolies.

Used? Or are you seriously looking to be a B2B customer and their newer offerings?

I was thinking used product as well as new.

However, after looking at the SP-T announcement today, that sure adds a big wrinkle to BTC things.

where to look for the sp-t announcement?  link please

[Biodom]

My next rig has to be Spondoolies.

Used? Or are you seriously looking to be a B2B customer and their newer offerings?

I was thinking used product as well as new.

However, after looking at the SP-T announcement today, that sure adds a big wrinkle to BTC things.

where to look for the sp-t announcement?  link please

here
http://www.spondoolies-tech.com/blogs/news

basically a merger with an OTC stock company

[DevonMiner]

We probably already knew it .. but the end of home mining with new Spondoolies gear has now been officially confirmed by something I didn't see coming.

Can't blame them, business is business ...

[notlist3d]

We probably already knew it .. but the end of home mining with new Spondoolies gear has now been officially confirmed by something I didn't see coming.

Can't blame them, business is business ...

I keep holding out hope.  I hope they are designing them for the big data center's like SP35's etc.  And they still allow home miners to buy.

They can still sell a good amount to home miners.  Granted it's a long term sale, not selling a entire data center in a day.

[Digitalmocking]

Hi.

I want to buy 4 of these. Where can i get them in U.S.?

Thanks,

used hardware is your only chance.

SP won't be making any consumer grade hardware in the future it looks like.

[notlist3d]

Hi.

I want to buy 4 of these. Where can i get them in U.S.?

Thanks,

used hardware is your only chance.

SP won't be making any consumer grade hardware in the future it looks like.

Ebay and this forums hardware section normally have some.   Sadly no new ones, but I have had some a while no issues. 

They seem to last and preform good even after month's of running so I would not be to afraid of used market.  I have under clocked mine though I like the electric efficiency better then top speed.

[Digitalmocking]

Hi.

I want to buy 4 of these. Where can i get them in U.S.?

Thanks,

used hardware is your only chance.

SP won't be making any consumer grade hardware in the future it looks like.

Ebay and this forums hardware section normally have some.   Sadly no new ones, but I have had some a while no issues. 

They seem to last and preform good even after month's of running so I would not be to afraid of used market.  I have under clocked mine though I like the electric efficiency better then top speed.

Honestly looking at the prices I'm considering just selling my 8 and replacing them with s5's. :p

[Biodom]

Good luck to Spond; I believe that they will need it.

A thought occured to me: everybody (Spond, Bitfury, KnC) want to call themselves transaction validators, however, I surmise (without marketing research behind this statement) that thousands of miners are the driving force of transactions because they do transact with bitcoin that they mine and use it for means other than mining.
If "small" miners disappear, only to be replaced by four large companies (Bitfury, KNC, Bitmain and Spond), then whose transactions these four will be processing?
Between each other?
They will basically diminish the interest of those who are most interested now and replace it with WHAT? Marketing?
If at the end of the day bitcoin is just another debit/credit card and these four are processors, why would I need that? I have a bunch of plastic already.

[padrino]

Good luck to Spond; I believe that they will need it.

A thought occured to me: everybody (Spond, Bitfury, KnC) want to call themselves transaction validators, however, I surmise (without marketing research behind this statement) that thousands of miners are the driving force of transactions because they do transact with bitcoin that they mine and use it for other means.
If "small" miners disappear, only to be replaced by four large companies (Bitfury, KNC, Bitmain and Spond), then whose transactions these four will be processing?
Between each other?
They will basically diminish the interest of those who are most interested now and replace it with WHAT? Marketing?
If at the end of the day bitcoin is just another debit/credit card and these four are processors, why would I need that? I have a bunch of plastic already.

While centralization is not a good thing, it's inevitable and those small miners pushing and pushing, and paying premiums for the next big thing drove themselves out of the market...

If you and I use Bitcoin and neither of us mine we need miners to validate those transactions, how much control a single entity has on a given percentage of the network is the concern, if it's a handful of companies it does raise the risk if there is a hint of collusion between those companies..

The above tells me you want a free market but not the inevitable consequences of that free market..

[Biodom]

Good luck to Spond; I believe that they will need it.

A thought occured to me: everybody (Spond, Bitfury, KnC) want to call themselves transaction validators, however, I surmise (without marketing research behind this statement) that thousands of miners are the driving force of transactions because they do transact with bitcoin that they mine and use it for other means.
If "small" miners disappear, only to be replaced by four large companies (Bitfury, KNC, Bitmain and Spond), then whose transactions these four will be processing?
Between each other?
They will basically diminish the interest of those who are most interested now and replace it with WHAT? Marketing?
If at the end of the day bitcoin is just another debit/credit card and these four are processors, why would I need that? I have a bunch of plastic already.

While centralization is not a good thing, it's inevitable and those small miners pushing and pushing, and paying premiums for the next big thing drove themselves out of the market...

If you and I use Bitcoin and neither of us mine we need miners to validate those transactions, how much control a single entity has on a given percentage of the network is the concern, if it's a handful of companies it does raise the risk if there is a hint of collusion between those companies..

The above tells me you want a free market but not the inevitable consequences of that free market..

well, once there are ONLY four companies, it's not a free market anymore, more like oligopoly.
Regarding it being inevitable, I am not aware of such economic law. Personally, I would very much prefer a more distributed system.

My main point was not about centralization per se, but about a supposition that once small miners are driven out, 4 big ones will have little transactions to process, unless they do it mostly between each other, which would be artificial, of course.

[padrino]

Good luck to Spond; I believe that they will need it.

A thought occured to me: everybody (Spond, Bitfury, KnC) want to call themselves transaction validators, however, I surmise (without marketing research behind this statement) that thousands of miners are the driving force of transactions because they do transact with bitcoin that they mine and use it for other means.
If "small" miners disappear, only to be replaced by four large companies (Bitfury, KNC, Bitmain and Spond), then whose transactions these four will be processing?
Between each other?
They will basically diminish the interest of those who are most interested now and replace it with WHAT? Marketing?
If at the end of the day bitcoin is just another debit/credit card and these four are processors, why would I need that? I have a bunch of plastic already.

While centralization is not a good thing, it's inevitable and those small miners pushing and pushing, and paying premiums for the next big thing drove themselves out of the market...

If you and I use Bitcoin and neither of us mine we need miners to validate those transactions, how much control a single entity has on a given percentage of the network is the concern, if it's a handful of companies it does raise the risk if there is a hint of collusion between those companies..

The above tells me you want a free market but not the inevitable consequences of that free market..

well, once there are ONLY four companies, it's not a free market anymore, more like oligopoly.
Regarding it being inevitable, I am not aware of such economic law. Personally, I would very much prefer a more distributed system.

My main point was not about centralization per se, but about a supposition that once small miners are driven out, 4 big ones will have little transactions to process, unless they do it mostly between each other, which would be artificial, of course.

I think maybe I misunderstood your point, and I think it's sourced from a misunderstanding of transaction processing..

Every Bitcoin transaction is processed, those 4 companies don't process their own, but everyone using Bitcoin, you are implying once small miners are driven out the volume of transactions go down. I would submit that small miners are already a small subset of transactions, I know many more people using Bitcoin that have never mined, than people that mine and use Bitcoin..

[alh]

How much time do you think SP20 could do a good work? 2017? or less?

because 4 machines + 4 Power supply (EVGA 1300) cost 2800 usd more or less. and it will give you 400 per month running at 1.4 th each one (5.6 th). so you will get your money back in 8 months, suppose you have cheap energy. Am I right?

Just using $.05/Kwh takes about $130 of that $400 in terms of raw electricity consumption for the mining hardware. That's a VERY favorable electrical rate. Gives you an operating profit of $270 per month. That assumes that there are no upward difficulty adjustments during the 10 months for payoff. At the end, you'll still have residual value in the power supplies, probably much less in the SP20's.

Yes it can work. You just won't get "rich" doing it (my definition of "rich").

[philipma1957]

How much time do you think SP20 could do a good work? 2017? or less?

because 4 machines + 4 Power supply (EVGA 1300) cost 2800 usd more or less. and it will give you 400 per month running at 1.4 th each one (5.6 th). so you will get your money back in 8 months, suppose you have cheap energy. Am I right?

No you are wrong as you do not need 4 evga 1300 to run them.

Under clock them to 1100gh each and you can run 2 of them on 1 evga 1300 lowers your buy in price  by 300-400 usd

you use .52 watts a gh at the wall


if you push them like a mofo  to 1500-1600gh you need a evga  for each one and you spend 1200 watts to get say 1600gh


so 2 evgas using 2300watts  runs 4 doing 4400gh

or 4 evga  using  4800watts  runs 4 doing 6400gh


the first case is almost always the better way to run sp20's..


 in the second case

the power efficiency  is .75 watts a gh and you purchased 2 extra evga's .  The blow your gear  up or burn your gear up  risk is much higher.

and you start out 300 to 4oo usd in the hole due to 2 extra psus.

One last thing the extra 2500 watts spent causes close to 10,000 btu 's of extra heat.  So cooling is harder to do.

[philipma1957]

How much time do you think SP20 could do a good work? 2017? or less?

because 4 machines + 4 Power supply (EVGA 1300) cost 2800 usd more or less. and it will give you 400 per month running at 1.4 th each one (5.6 th). so you will get your money back in 8 months, suppose you have cheap energy. Am I right?

No you are wrong as you do not need 4 evga 1300 to run them.

Under clock them to 1100gh each and you can run 2 of them on 1 evga 1300 lowers your buy in price  by 300-400 usd

you use .52 watts a gh at the wall


if you push them like a mofo  to 1500-1600gh you need a evga  for each one and you spend 1200 watts to get say 1600gh


so 2 evgas using 2300watts  runs 4 doing 4400gh

or 4 evga  using  4800watts  runs 4 doing 6400gh


the first case is almost always the better way to run sp20's..


 in the second case

the power efficiency  is .75 watts a gh and you purchased 2 extra evga's .  The blow your gear  up or burn your gear up  risk is much higher.

and you start out 300 to 400 usd in the hole due to 2 extra psus.

One last thing the extra 2500 watts spent causes close to 10,000 btu 's of extra heat.  So cooling is harder to do.

Thanks for your answer.

If in the second case you put an air condition near that raid, what would you think the result could be?

 An insane waste of power in terms of watts.  

if you have a 4 pack of units maxed   pulling 4800 watts vs a pack of units cruising pulling 2300 watts.

I only mentioned the extra 2500 watts = 10000 btus of heat  you also have close to 10000 btus from the first watts.

so a better way to write it is:

 20000 btus

vs 10000 btus

you want to vent the heat out  of the room with fans not ac the heat away.

here is a picture of one of the cheapest ways to cool a room with 6 sp20's

https://bitcointalk.org/index.php?topic=872014.msg9741334#msg9741334



Here are some more photos from link above .  movers blanket lowers sound a lot. note open garage door







third shot  white blanket is fire proof I need a few pieces of sheet metal to feel really safe





fourth

 clamp stops  gear from falling    stack is    (s-3) sp20 sp20 sp20 sp20 (s-3)



fifth

 some hot air tries to come back into garage those fans blow it out



sixth
fire proof material protects a board also fan brings in cooler air------------------------------------ This is a 4 month winter setup



seventh   sound is reduced a lot all fans are 50 or 60  this garage is attached to my home and  basement den for Home Theater.  I can watch tv and movies more important my wife thinks it is not crazy loud

[notlist3d]

How much time do you think SP20 could do a good work? 2017? or less?

because 4 machines + 4 Power supply (EVGA 1300) cost 2800 usd more or less. and it will give you 400 per month running at 1.4 th each one (5.6 th). so you will get your money back in 8 months, suppose you have cheap energy. Am I right?

No you are wrong as you do not need 4 evga 1300 to run them.

Under clock them to 1100gh each and you can run 2 of them on 1 evga 1300 lowers your buy in price  by 300-400 usd

you use .52 watts a gh at the wall


if you push them like a mofo  to 1500-1600gh you need a evga  for each one and you spend 1200 watts to get say 1600gh


so 2 evgas using 2300watts  runs 4 doing 4400gh

or 4 evga  using  4800watts  runs 4 doing 6400gh


the first case is almost always the better way to run sp20's..


 in the second case

the power efficiency  is .75 watts a gh and you purchased 2 extra evga's .  The blow your gear  up or burn your gear up  risk is much higher.

and you start out 300 to 4oo usd in the hole due to 2 extra psus.

One last thing the extra 2500 watts spent causes close to 10,000 btu 's of extra heat.  So cooling is harder to do.

Thanks for your answer.

If in the second case you put an air condition near that raid, what would you think the result could be?

Just like philipma1957 said the worst thing you could do is use AC to cool down your miners.  You will be paying way to much to get them cool.

It's best done with regular fan's and an exhaust.  It's a little tricky to get setup right, but ac is not the way to go.

[alh]

One small item of note. While I completely agree that 4 underclocked SP20's are the way to go, Phil's "open door" picture might be a shade misleading. At the time that picture was taken, I'll bet the air temp outside was 5C or less. It's not obvious to me exactly how to best cool a miner when the incoming air is 35C. I get the exhaust plan and all that, but you can't get low temps when you start with 35C air.

[notlist3d]

One small item of note. While I completely agree that 4 underclocked SP20's are the way to go, Phil's "open door" picture might be a shade misleading. At the time that picture was taken, I'll bet the air temp outside was 5C or less. It's not obvious to me exactly how to best cool a miner when the incoming air is 35C. I get the exhaust plan and all that, but you can't get low temps when you start with 35C air.

No it obviously is a winter photo.  During the summer that would not be a option.  In most houses you would have AC escaping outside.

Winter is much easier mining then summer.  Much more enjoyable.

[philipma1957]

One small item of note. While I completely agree that 4 underclocked SP20's are the way to go, Phil's "open door" picture might be a shade misleading. At the time that picture was taken, I'll bet the air temp outside was 5C or less. It's not obvious to me exactly how to best cool a miner when the incoming air is 35C. I get the exhaust plan and all that, but you can't get low temps when you start with 35C air.

My temps ranged from 0 f to 70 f

or -18c to 22 c

At 70 f = 22 c  this was still good for cooling.

I have mined in summers and the attached garage  with the door open does not work well.

But and here is a big but : In  my house the first floor is kind of a partial walk out basement.  So the garage side does not have a man-made hill of dirt but the other side does. So central air 1 zone system means a really cold 1st floor.

So I shut the back door of the garage and I can run up to 2-2.5k  of gear. Then the basement/first floor ends up comfortable matching the top floor of the home temp wise.

[alh]

So it would be 4 sp20 + 2 evga 1300w.

it would made 1100 ghz * 4 = 334 usd per month more or less.

4 sp20 + 2 evga 1300 = 2500usd more or less. I could see the ROI in 8 months more or less.


What do you think  about this calculation?

Other thing, with this raid, how much KW/h do you think it could consume by hour?

Thanks

Think about it for a minute. Two 1300W power supplies are 2600W. That's 2.6 KW. Run it for an hour, and you have used 2.6KWh.

In your ROI calculations, have you taken into account any possible difficulty adjustments that will likely happen over the next 8 months? While the last 60 days have been relatively "quiet" in terms of difficulty changes, that hasn't always been the case. If you aren't familiar with difficulty adjustments, then read up on them. Every 2016 blocks, the Bitcoins "network" adjusts the difficulty to try and maintain a steady pace of about 14 days to produce those 2016 blocks.If you look back at 2014 for example you can easily find an 8 month period where the difficult adjustment in total was 250%. That means that during your 8th month you are only effectively hashing at 40% of what your were hashing during month 1.

The bitcoinwisdom.com website has the list of pash difficulty adjustments for at least a year. Some were pretty brutal in 2014 (i.e. over 25% in one jump).

It's important that you have a complete picture before you plunk down $2500 on your "money printing" hardware. You MIGHT break even in 8 months. Don't be surprised if you don't.