[Guide] Dogie's Comprehensive Manufacturer Trustworthiness Guide [1st Feb 2016]

[jimmothy]

BTCGarden does have quality issues, but they also deny it (hence the unethical issues).

Source? I haven't seen any serious complaints. Also why 4/10 for size? They have probably manufactured several PH by now. Probably tied with rockminer or lightningasic.

Although I have to agree with notlambchop that size is not a good variable. Not only does size not indicate how reputable a company is, it's nearly impossible to accurately judge.

In bitmain's case its excluded because its never, ever impacted the customer deliveries as they only sell in batches, and deliver within their specified dates.

Same with asicminer and bitfury. At least with those two they didn't sell used HW.

They've also never ever specified the hardware is new, its hardware with a warranty. They aren't doing anything wrong there.

That's absurd. If you buy something and it doesn't specifically say "used" then it is assumed to be new.

We both know that used electronics are worth less than new ones and there is no excuse for bitmain to secretly sell heavily used hardware. And a 90 day warranty doesn't make it any more excusable.

[1713438276]

1713438276

[1713438276]

1713438276

[aerobatic]

Dogie -

i do appreciate you doing your ratings system, and its explained quite well...  

your ratings methodology is very good.  but the way you've applied it seems 'hit and miss' and open to interpretation and emotional bias.

here's a few anomalies for you to consider...


bitfury - mostly makes chips, not systems.   Yes, they make systems for their own private mine but they have hardly sold any of those to the public.  probably 90%+ of the systems that have been 'sold' containing bitfury chips were made by unconnected third parties utilising bitfury's chips.  yet you have them classified as a manufacturer of systems and have given them the benefit of not pre-selling, whereas the third parties who make the systems using bitfury chips, often do pre-orders right up until the chips arrive in stock.  i myself bought bitburner boards from cryptx via pre-order., months in advance of their delivery. Most people bought bitfury boards from someone else and not direct from bitfury.  now, of course, these companies are selling from stock because they've got plenty of chips, but in advance of the chips arriving in volume they were taking pre-orders.   And what about your ethics score?  You've got BitFury marked as an 'F' for a private mine, yet you've got KnCMiner scored as Ethics FF, which means 'HUGE private Mine' .  Thats a purely emotional response, and isn't borne by the facts.    Do you really think KnCMiner's private mine is bigger than BitFury's private mine?

cointerra - doesn't seem well classified according to your rating methodologies.  for instance, you scored them 1/10 for 'uses pre-orders' whereas they've been selling from stock ever since their systems arrived (months ago).  you can buy one from stock today, or any time in the last few months.  yet again, thats an emotional bias and isn't factual.   they did do pre-orders (like most companies out there) before their stock arrived.  You also rated them 1/10 for 'on time'.  Why, if you're scoring out of 10, do you only have 3 options, Yes, No and Mixed.  Why not have a linear (or log) score out of 10 for how 'on time' each company is?   This particular company delivered their systems in batches according to month and the customers orders were delivered in that specific month.   They shipped thousands of systems, and the vast majority were delivered in the month they were due (or a week or two late)  the earlier batches were a few weeks late but not 'months late'.  apart from the 1st batch (december), which was delayed a whole month and the customers who bought that batch were given compensation in the form of double the number systems.   there's no way this company could be judged 1/10 for 'on time' if you're trying to compare relative timeliness to the others.  Many of the asic companies on your list are ones that have never shipped (hashfast, blackarrow, bfl, vmc etc ) are are those infinitely late.   Your ratings are scored out of 10, so you could easily differentiate those that shipped a little late versus those that shipped very very very late (or, Never) using the entire scoring range.

Size of companies.  how do you rate that?  it seems like pure fiction.   You have AsicMiner, BitMain and BitFury rated as 'Huge' (i agree with the latter two but i think asicminer is probably not so huge any longer).  You have Spondoolies and RockMiner as 'Large', but how do you even classify those?  RockMiner as small and i think even Spondoolies would admit they're not 'large'.  How are you going to measure it?  Is this just pure guesswork?   Why not based it on a measurable or determinable statistic like Revenue, or Petahashes shipped?

Bitmine - You have them rated as Large...  How was that measured?   you ALSO have them listed as having made their own chip.  Did they really make their own chip or do they just buy it from innosilicon like so many chinese companies?    Innosilicon says they made the chip and that bitmine is 'a reseller', like everyone else.

In order of size of company, i suspect a more realistic list would be:   BitFury is 1st, KnCMiner 2nd, Bitmain 3rd, Cointerra 4th, etc..    this size rating probably works regardless of whether you measure by petahash or revenue.  I doubt asicminer is even relevant at this time.  They're no longer Huge.  They were HUGE when the network was less than 1 Petahash - now they're a minnow - and got left behind when the others designed better chips.

[jimmothy]

In order of size of company, i suspect a more realistic list would be:   BitFury is 1st, KnCMiner 2nd, Bitmain 3rd, Cointerra 4th, etc..    this size rating probably works regardless of whether you measure by petahash or revenue.  I doubt asicminer is even relevant at this time.  They're no longer Huge.  They were HUGE when the network was less than 1 Petahash - now they're a minnow - and got left behind when the others designed better chips.

Are you forgetting that asicminer had 60PH worth of wafers to sell as of ~2 months ago? That's probably more than bitfury has ever produced.

Anyways it's pointless to speculate because nobody knows the exact (or even approximate) size of every company.

Maybe use age instead of size?

[BitCoin Operated Boy]

HashFast should have gone with ranking to 0 because they have failed to deliver 95% of equipment and what they delivered was with 6 months delay. Also they are crrently going through bankruptcy court hearings so they literally ceased to exist.

Black Arrow communication ranking shoudl go to zero. They have completely stopped any customer service 2 months ago. They have never answered any phone calls and they have been ignoring any customer's e-mails for last 2 months. They also, every day, on their forum thread they keep cenosring-removing any posts related to seeking any legal action against them.

[aerobatic]

In order of size of company, i suspect a more realistic list would be:   BitFury is 1st, KnCMiner 2nd, Bitmain 3rd, Cointerra 4th, etc..    this size rating probably works regardless of whether you measure by petahash or revenue.  I doubt asicminer is even relevant at this time.  They're no longer Huge.  They were HUGE when the network was less than 1 Petahash - now they're a minnow - and got left behind when the others designed better chips.

Are you forgetting that asicminer had 60PH worth of wafers to sell as of ~2 months ago? That's probably more than bitfury has ever produced.

Anyways it's pointless to speculate because nobody knows the exact (or even approximate) size of every company.

we have no evidence asicminer ordered 60 PH of chips.  thats a huge amount, and they don't seem to be in evidence anywhere.  And anyway, where'd they get the cash to buy such a large order?  that would cost tens of millions of dollars.   Also, their chip isn't the world beater they thought it was going to be and ended up pretty uncompetitive... if they had any sense they'd not order too many as it'd be unprofitable to run it this late in the game.  when they were originally planning to make such big orders it was going to be the best chip by a huge margin and much lower power consumption than it ended up.  Its a common problem that has befallen almost every asic company - of the chips not performing quite as good as they expected (yes, even bitfury whose chips were rated at 5 GH and ended up at 2.5 GH... same with Cointerra shipping a 1.6 TH instead of a 2 TH system, initially).

[bitsalame]

In order of size of company, i suspect a more realistic list would be:   BitFury is 1st, KnCMiner 2nd, Bitmain 3rd, Cointerra 4th, etc..    this size rating probably works regardless of whether you measure by petahash or revenue.  I doubt asicminer is even relevant at this time.  They're no longer Huge.  They were HUGE when the network was less than 1 Petahash - now they're a minnow - and got left behind when the others designed better chips.

Are you forgetting that asicminer had 60PH worth of wafers to sell as of ~2 months ago? That's probably more than bitfury has ever produced.

Anyways it's pointless to speculate because nobody knows the exact (or even approximate) size of every company.

we have no evidence asicminer ordered 60 PH of chips.  thats a huge amount, and they don't seem to be in evidence anywhere.  And anyway, where'd they get the cash to buy such a large order?  that would cost tens of millions of dollars.   Also, their chip isn't the world beater they thought it was going to be and ended up pretty uncompetitive... if they had any sense they'd not order too many as it'd be unprofitable to run it this late in the game.  when they were originally planning to make such big orders it was going to be the best chip by a huge margin and much lower power consumption than it ended up.  Its a common problem that has befallen almost every asic company - of the chips not performing quite as good as they expected (yes, even bitfury whose chips were rated at 5 GH and ended up at 2.5 GH... same with Cointerra shipping a 1.6 TH instead of a 2 TH system, initially).

Why do you ridicule yourself by claiming there are no proof, when you clearly haven't done your homework?

[FrictionlessCoin]

AMT ethics?  5/10 ... utterly ridiculous... should be 0/10.

They have no proof that they delivered a working model to anyone.

They sent a bounced check to their original manufacturer.

They shipped broken equipment to their users.

They never accepted RMA from anyone.

They don't even have a legit office!

[aerobatic]

Why do you ridicule yourself by claiming there are no proof, when you clearly haven't done your homework?

care to link to some proof?   there's 140 PH on the network earlier today, and asicminer asics cant power any more than 10 PH of it (and more likely < 5)... seems unlikely to be 60 PH or anywhere close.  Or, are you saying its sometime later this year?

60 PH, at today's wholesale price of $0.75-$1/GH (for complete systems, not asics.. round figures)... would be $60 M..  do you think there's customers out there buying $60m of bitcoin mining hardware, at today's bitcoin price?  especially systems that arent power competitive and could be obsolete by xmas compared to the newer generation that are better?

[dogie]

@dogie
Hi, I see what you mean now.     See if this sort of thing would be easier to update.  Instead of having separate tables for each company, it's just one long table, with the top row made up of spacer gifs, which you can adjust to whatever you want with the "width" tag.

May or may not be easier for you to update later.

Will delete this post as soon as you reply.

Yeah gifs are the alternative solution, although I've seen the image proxy sometimes bunk out when you try and put 100+ images in one post and then shit goes down. I've got your code saved thanks.

I wired this thing up to an excel spreadsheet which generates me the code automatically, so its easy to update atm.

[dogie]

do you think there's customers out there buying $60m of bitcoin mining hardware, at today's bitcoin price? 

Absolutely, yes, yes, yes. That's only 100,000 bitcoins.

[dogie]

In order of size of company, i suspect a more realistic list would be:   BitFury is 1st, KnCMiner 2nd, Bitmain 3rd, Cointerra 4th, etc..    this size rating probably works regardless of whether you measure by petahash or revenue.  I doubt asicminer is even relevant at this time.  They're no longer Huge.  They were HUGE when the network was less than 1 Petahash - now they're a minnow - and got left behind when the others designed better chips.

Are you forgetting that asicminer had 60PH worth of wafers to sell as of ~2 months ago? That's probably more than bitfury has ever produced.

Anyways it's pointless to speculate because nobody knows the exact (or even approximate) size of every company.

Maybe use age instead of size?

Tried that, companies can sit semi dormant and build reputation for free, again circumventing the system.

[dogie]

This new guide looks really good.  I love the composite scores.  that makes it really easy for people.

Black Arrow

Communication 4/10
Ethics 5/10

They are about though, even if through negative interactions they are interacting with the community. One of their management PMed me within a few hours debating one of the ratings. The ethics has to go into a specific category. There maybe more types of infractions added in the future, like live lawsuits but for now I can't put them down for any of them other than generic.

[dogie]

HashFast should have gone with ranking to 0 because they have failed to deliver 95% of equipment and what they delivered was with 6 months delay. Also they are crrently going through bankruptcy court hearings so they literally ceased to exist.

Black Arrow communication ranking shoudl go to zero. They have completely stopped any customer service 2 months ago. They have never answered any phone calls and they have been ignoring any customer's e-mails for last 2 months. They also, every day, on their forum thread they keep cenosring-removing any posts related to seeking any legal action against them.

I can't just mash a company's score to 0. They're listed as bankrupt and no one can even buy anything from them. They will slowly drop off into inactive (which currently isnt in this thread). They're currently joint 3rd last.

Black arrow are around, see above post. I can't punish them for using the moderated forum functionality, there are plenty of non moderated blackarrow threads you can post details about, including their CEO's address.

[dogie]

AMT ethics?  5/10 ... utterly ridiculous... should be 0/10.

They have no proof that they delivered a working model to anyone.

They sent a bounced check to their original manufacturer.

They shipped broken equipment to their users.

They never accepted RMA from anyone.

They don't even have a legit office!

As with the post a few up, I can't punish them for an ethics infraction which doesn't exist yet in the guide, and we don't actually have proof for. In the future they may be hurt with the addition of a 'legal action' infraction. They have/did have working models, I've seen them.

[dogie]

bitfury - mostly makes chips, not systems.   Yes, they make systems for their own private mine but they have hardly sold any of those to the public.  probably 90%+ of the systems that have been 'sold' containing bitfury chips were made by unconnected third parties utilising bitfury's chips.  yet you have them classified as a manufacturer of systems and have given them the benefit of not pre-selling, whereas the third parties who make the systems using bitfury chips, often do pre-orders right up until the chips arrive in stock.  i myself bought bitburner boards from cryptx via pre-order., months in advance of their delivery. Most people bought bitfury boards from someone else and not direct from bitfury.  now, of course, these companies are selling from stock because they've got plenty of chips, but in advance of the chips arriving in volume they were taking pre-orders.   And what about your ethics score?  You've got BitFury marked as an 'F' for a private mine, yet you've got KnCMiner scored as Ethics FF, which means 'HUGE private Mine' .  Thats a purely emotional response, and isn't borne by the facts.    Do you really think KnCMiner's private mine is bigger than BitFury's private mine?

I've not classified them as anything other than a company who makes chips - which is a huge positive. It means they've raised significant capital and have invested significant money in getting it right. They're much less likely to disappear overnight then someone who has just bought a small number of chips and integrated them. Bitfury's mine is not directly competing with their customers because they're two separate things entirely - they raised $10M for a farm. They don't have preorders waiting. KNC on the otherhand do, they raised preorder funds several times from the community, promised not to compete against their customers, then used all that cash (once orders went) to fund a super mine. After that they stopped caring about the average consumer because they no longer had to. That has been evidenced by the recent failed Neptune launch and now the even more preorders being offered. That's why their mine punishes them (a whole extra 2 points may I add...)

I'm not sure what you mean by an emotional response, I've not had dealing or orders with either company, I have no financial investment or incentive either way. I couldn't really care less what they do.

cointerra - doesn't seem well classified according to your rating methodologies.  for instance, you scored them 1/10 for 'uses pre-orders' whereas they've been selling from stock ever since their systems arrived (months ago).  you can buy one from stock today, or any time in the last few months.  yet again, thats an emotional bias and isn't factual.   they did do pre-orders (like most companies out there) before their stock arrived.  You also rated them 1/10 for 'on time'.  Why, if you're scoring out of 10, do you only have 3 options, Yes, No and Mixed.  Why not have a linear (or log) score out of 10 for how 'on time' each company is?   This particular company delivered their systems in batches according to month and the customers orders were delivered in that specific month.   They shipped thousands of systems, and the vast majority were delivered in the month they were due (or a week or two late)  the earlier batches were a few weeks late but not 'months late'.  apart from the 1st batch (december), which was delayed a whole month and the customers who bought that batch were given compensation in the form of double the number systems.   there's no way this company could be judged 1/10 for 'on time' if you're trying to compare relative timeliness to the others.  Many of the asic companies on your list are ones that have never shipped (hashfast, blackarrow, bfl, vmc etc ) are are those infinitely late.   Your ratings are scored out of 10, so you could easily differentiate those that shipped a little late versus those that shipped very very very late (or, Never) using the entire scoring range.

"they did do pre-orders (like most companies out there) before their stock arrived". You answered it yourself. Again I have no investment or incentive in any of this, so its not really fair for you to call it an emotion reponse because you don't agree with my opinion.

You can see how much people (like you) are arguing about 1 and 2 points all over. To have 10 different scaling numbers within 10 categories for each company is insane and unusable. I'm not sure if you've worked with this style of rating system before, but the aim is to have as few categories as you can get away with. The more categories, the less useful and useful a rating system is.

I think cointerra was significantly later than 'a few weeks', you can link me to something if you want.

Size of companies.  how do you rate that?  it seems like pure fiction.   You have AsicMiner, BitMain and BitFury rated as 'Huge' (i agree with the latter two but i think asicminer is probably not so huge any longer).  You have Spondoolies and RockMiner as 'Large', but how do you even classify those?  RockMiner as small and i think even Spondoolies would admit they're not 'large'.  How are you going to measure it?  Is this just pure guesswork?   Why not based it on a measurable or determinable statistic like Revenue, or Petahashes shipped?

Asicminer is huge, huge huge huge. Every unit a multitude of companies sells has 100s of their chips in, its an astronomical scale. You're not in Asia where most of their business gets done but they are crazy.

Bitmine - You have them rated as Large...  How was that measured?   you ALSO have them listed as having made their own chip.  Did they really make their own chip or do they just buy it from innosilicon like so many chinese companies?    Innosilicon says they made the chip and that bitmine is 'a reseller', like everyone else.

They're large because of their chips, and while I dispute they're 'just a reseller', it still wouldnt matter. They're the ones with the partnership and they're the ones powering a significant portion of consumer devices on the market today.

In order of size of company, i suspect a more realistic list would be:   BitFury is 1st, KnCMiner 2nd, Bitmain 3rd, Cointerra 4th, etc..    this size rating probably works regardless of whether you measure by petahash or revenue.  I doubt asicminer is even relevant at this time.  They're no longer Huge.  They were HUGE when the network was less than 1 Petahash - now they're a minnow - and got left behind when the others designed better chips.

Your 'inside information' is either wrong or old.

[aerobatic]

Your 'inside information' is either wrong or old.

can you produce or link to any data that shows the relative size of asicminer?

what about just looking at list of mining pools (blockchain.info or organ of corti).  we can see where kncminer is.. we can even see where bitfury might be.  perhaps we can see some of the others.   we also know what bitmain says they are.  if we add up the network percentages that bitfury (40%), bitmain (20%), kncminer (10%?) and a few others (15%?) say they are (and hopefully can prove it someday)... then there's not a lot of unknowns left.  asicminer is a public company.  did they disclose just how big they are?

[bitsalame]

Your 'inside information' is either wrong or old.

can you produce or link to any data that shows the relative size of asicminer?

what about just looking at list of mining pools (blockchain.info or organ of corti).  we can see where kncminer is.. we can even see where bitfury might be.  perhaps we can see some of the others.   we also know what bitmain says they are.  if we add up the network percentages that bitfury (40%), bitmain (20%), kncminer (10%?) and a few others (15%?) say they are (and hopefully can prove it someday)... then there's not a lot of unknowns left.  asicminer is a public company.  did they disclose just how big they are?

Let me invite you to do some basic estimation of AM's worth.
Total dividends distributed so far are 0.5911 BTC per share.
Bitfountain, AM parent company, own 236,038 shares.
You do the math.

[k9quaint]

Your 'inside information' is either wrong or old.

can you produce or link to any data that shows the relative size of asicminer?

what about just looking at list of mining pools (blockchain.info or organ of corti).  we can see where kncminer is.. we can even see where bitfury might be.  perhaps we can see some of the others.   we also know what bitmain says they are.  if we add up the network percentages that bitfury (40%), bitmain (20%), kncminer (10%?) and a few others (15%?) say they are (and hopefully can prove it someday)... then there's not a lot of unknowns left.  asicminer is a public company.  did they disclose just how big they are?

Let me invite you to do some basic estimation of AM's worth.
Total dividends distributed so far are 0.5911 BTC per share.
Bitfountain, AM parent company, own 236,038 shares.
You do the math.

That statement can be true if ASICMiner currently holds 0% of the network and all the BTC was earned in the past.
ASICMiner hash rate can only be inferred from dividends. They might be operating at a temporary loss in order to re-invest past proceeds into new mining hardware. Or they might be under reporting their earnings in order to skim some BTC off the top. Or some combination of the two.

Also, other farms might be reporting as someone else in order to hide their size (e.g. if they were mining with customer hardware instead of shipping it).

[aerobatic]

Actually.. this could be easier than i originally thought...  asicminer is public, so we can figure out exactly how big they are (or were, last quarter).

here's their most recently published accounts for the quarter feb-may 2014.

assets :-
https://docs.google.com/spreadsheets/d/12wJMe3A6Ie8ris86m2OJ_1vucurdcVtM3fDkNCjKIRI/edit?usp=sharing

cashflows :-
https://docs.google.com/spreadsheets/d/1phCAhMR-9_AaRXKICfN_M7gw4rI1hsOtRD_6PElmLOM/edit?usp=sharing

if we look at how much electricity they bought in the above spreadsheet, you can see its a total of 500,000 Yuan (across three data centers)... which is $80K US Dollars (for a quarter).  Divide by 3 and thats $27K/month.   I know how much electricity costs in the USA.. but alas china and HK arent cheap places to buy electricity so its probably going to be at least the same price (and suspect its more expensive. )

http://en.wikipedia.org/wiki/Electricity_pricing

lets use some safe assumptions.  that it costs about 8 cents per KwH (its probably more).  And lets assume that theyre already using their latest chips (and already achieving 1 W/GH from their new BM200 asics, whereas in reality, this period from feb-may 2014, they were using their older and less efficient silicon.

Assuming they buy electricity reasonably well.. and have the newer better asics already deployed... they have a mine thats a max of 0.47 Petahash (assuming zero cost for cooling - lets assume some cost for cooling and we're on say 0.4 Petahash), based on the electricity theyre using.   Compare that to knCMiner's mine thats 4.5 PH, and theyre about 1/10th the size of KnCMiner's mine.  Approx.   And thats giving them the benefit of the doubt that they were already using their newer more efficient silicon.

You sent a post detailing the total dividends to date.  is that relevant?   those dividends go to investors, not the company.  Thats not cash the company can spend, unless the investors choose to re-invest it.

the other calculation you could do is look at revenues.   yet again, asicminers' revenue is public, and the info is in the above two spreadsheets.

or lets also look at market cap on the havelock site.. which quotes AM1 shares trading at 0.165 BTC/Share with a total market cap of 4673 BTC.  ( $2.8m in usd equivalents.)   Maybe you can help me understand the relationship between havelock AM1 shares, and the total market cap of AM ?    is it 1:1 or is there some fraction of AM thats listed on havelock and that this doesnt represent the full market cap (i assume the latter) ?

[bitsalame]

Actually.. this could be easier than i originally thought...  asicminer is public, so we can figure out exactly how big they are (or were, last quarter).

here's their most recently published accounts for the quarter feb-may 2014.

assets :-
https://docs.google.com/spreadsheets/d/12wJMe3A6Ie8ris86m2OJ_1vucurdcVtM3fDkNCjKIRI/edit?usp=sharing

cashflows :-
https://docs.google.com/spreadsheets/d/1phCAhMR-9_AaRXKICfN_M7gw4rI1hsOtRD_6PElmLOM/edit?usp=sharing

if we look at how much electricity they bought in the above spreadsheet, you can see its a total of 500,000 Yuan (across three data centers)... which is $80K US Dollars (for a quarter).  Divide by 3 and thats $27K/month.   I know how much electricity costs in the USA.. but alas china and HK arent cheap places to buy electricity so its probably going to be at least the same price (and suspect its more expensive. )

http://en.wikipedia.org/wiki/Electricity_pricing

lets use some safe assumptions.  that it costs about 8 cents per KwH (its probably more).  And lets assume that theyre already using their latest chips (and already achieving 1 W/GH from their new BM200 asics, whereas in reality, this period from feb-may 2014, they were using their older and less efficient silicon.

Assuming they buy electricity reasonably well.. and have the newer better asics already deployed... they have a mine thats a max of 0.37 Petahash, based on the electricity theyre using.   Compare that to knCMiner's mine thats 3 PH, and theyre about 1/10th the size of KnCMiner's mine.  Approx.   And thats giving them the benefit of the doubt that they were already using their newer more efficient silicon.

You sent a post detailing the total dividends to date.  is that relevant?   those dividends go to investors, not the company.  Thats not cash the company can spend, unless the investors choose to re-invest it.

the other calculation you could do is look at revenues.   yet again, asicminers' revenue is public, and the info is in the above two spreadsheets.

or lets also look at market cap on the havelock site.. which quotes AM1 shares trading at 0.165 BTC/Share with a total market cap of 4673 BTC.  ( $2.8m in usd equivalents.)   Maybe you can help me understand the relationship between havelock AM1 shares, and the total market cap of AM ?    is it 1:1 or is there some fraction of AM thats listed on havelock and that this doesnt represent the full market cap (i assume the latter) ?

Omg, I just pasted four freaking lines and you still can't interpret that information.
The company is majority stockholder.
I am outta here, I can't deal with this crap anymore.