You mean bullish?

I assumed completely "averaging" it out.
Single events are too hard to estimate, no one can guess those. Hence, assuming success, ie. all the problems can be worked through.
Every startup has issues, sometimes very critical ones, even Paypal had their share of those.

If governments try to interfere with Bitcoin, it has already seen success, and people will resist because interfering with Bitcoin would interfere with other rights too.



Which would push the price up even faster.
Thing is, more there is adoption that much more the price HAS to increase. There won't be sudden increase of BTC around, so only way for it to work is for price to increase. Which will cause some people dumping their BTC, but in overall people will keep on hoarding.
If you notice, the first thing i did was to adjust for hoarding. Conservatively, but still.

The 20$+ prices last year was a bubble very obviously, but i think we are now at somewhere around the "real value" of BTC.
So yes, next year this time, price could be in 40$+ range, but unlikely, the value growth is likely to be more "exponential" of nature, the first few years value climbing up slowly, starting to get more speed next summer, and ~5½ years from now on, as the amount of new BTC coming around is getting lower, and adoption rate increases over time, reaching huge velocities when critical mass is reached, therefore huge increase in value very short period of time, followed by a short crash due to people cashing out, a moment of conservative investment into BTC followed by steady value growth (Assuming adoption does not keep increasing) until finally stabilizing.

**EDIT: The numbers are there, you don't need to be a rocket scientist to understand what happens to the value of 18million BTC if there is 408million users wanting piece of the cake.

just realized i forgot to adjust for inflation.

I started this by comparing to Paypal, and by finding some statistics of Paypal.
Please note that all of this is based on the assumption Bitcoin will have very conservative success - but success over the long term nevertheless.

350mil+ Paypal users, from 2009 data we can estimate that that 135million of them are active users.
Active account average balance was 45$ in 2009, we shall use that without correction for expected growth.
This means 6 075million USD in just active account balances on Paypal in 2011.

So by 2022, if Bitcoin achieves just 5% of the current mass of Paypal, which could be very conservative, especially if Bitcoin is made easier for Joe Averages and some kind of Killer application becomes reality, ie. eBay + Craigslist type of events. Technological advances will make a huge influence as well to make this a reality, think about how things were in 2002.

If i calculated right there will be already ~19.03mil BTC in the wild by 2022.
That seems too much, but i based it on the halving of mined return per 210,000 blocks or 4 years.
If it's too much, it only makes this even more conservative.

If we assume 5% get lost forever by this time, that leaves us with 18.08mil BTC.

If even 5% of Paypal's mass today is reached by 2022 this means these will represent 303.75million USD, or 16.80USD/BTC. Seems far too conservative? Because it is.

Due to nature of Bitcoin, people will be inclined to hoard and keep them, in hopes of future gains. Bitcoin is also a value store, unlike with Paypal you are mad to keep any significant balance there for any period of time.

So we can assume safely people will keep atleast 4 times more, ie. 180$ or more in their balance.
This already leaves us with 67.20USD/BTC. That looks a lot more realistic - still conservative tho

But it remains extremely conservative STILL.  It does not account for new people getting connected, the growth of eCommerce, and that many wants to use Bitcoin also as a offline transaction medium, which AFAIK is happening to some degree already? ie. cash based exchanges, actual coins & bills etc. Which is impossible with Paypal.
We will buy more and more stuff online, and for many parts of the world internet is just beginning to see widespread adoption in some parts of the world. There still remains billions to be connected to internet. With the raise of better and better mobile devices, these people will get connected.

Just from eCommerce growth we can assume at least 50% growth by 2022, which is probably off by order of magnitude.
Add to that it's quite likely we will probably see nearly double the internet users by 2022, let's say 80% for a total of 58% of current world population using internet.

eCommerce growth will bring us to 100.80USD/BTC
Internet user growth will bring us to 159.26USD/BTC.

Unlike credit card companies, and almost any method of transactioning out there has fees of at least % for merchants. Only cash, and in most parts of the world bank wire transfers are free, neither of which is easy to automate for some portions.
This will mean a lot of Merchants will choose to support BTC when value slightly stabilizes, easy to automate, no transaction fees, likely to just increase in value - what's there not to like?
Assume this adds an conservative 10% to the % using BTC.

Russia, United Arab Emirates, Asia, are pretty much blocked off Paypal - In real world this would easily triple the amount of user base. We will choose an very conservative 50% for this figure too. This too would of course increase adoption within Merchants but we are not adjusting for that separately.

Merchant adoption increase: 175.19USD/BTC
Russia, UAE, Asia: 262.79USD/BTC

Add that to other potentially disrupting technologies: Think pre-paid Bitcoin "Credit cards" to use @ grocery stores, this "CC" would store the keys required for the grocery store to make the transaction from your wallet. Remember that with real cards merchants are trusted like this too. And this isn't even that hard to do.For short term someone could offer regular prepaid Mastercards which are prepaid using BTC

To recap the percentages of increase from 16.80$/BTC Are:
Hoarding/Investing into bitcoins: 400%
eCommerce Growth: 50%
Internet userbase Growth: 80%
Merchant Adoption: 10%
Russia, UAE, Asia Access: 50%
IMHO, these should be calculated in exponential fashion as in described above.

The 16.80$/BTC came from:
5% of Paypals mass 2011
135mil active users with 45$ average balance representing 6 075million USD.
18.08Million BTC in circulation
(135mil * 5%) * 45$ = 303.75Million USD
303.75 / 18.08 = 16.80$/BTC

Using 25%:
1 518.75million USD
84$/BTC.
Growth factors:
Hoarding: 336$/BTC
eCommerce: 504$/BTC
Userbase: 907.2$/BTC
Merchants: 997.92$/BTC
Russia, UAE, Asia: 1496.88$/BTC

My instinct is telling me however that we are going to see somewhere in vicinity 1000$/BTC.

If we use the 45$ average balance figure, and 10% of internet users are actively using BTC in 2022:
408million users.
18 362million USD.
1015$/BTC.

This holds better with my instincts

I did do some heavy assumptions here, but please do point out if i had blatant mistakes. Just remember on this i assumed mainstream success if the value seems too high to you.

If you did like my speculation, please donate something: 1983sUSuHXAWSMnYCX1ZJ2RFZBPPQwNVYD


References:
http://wiki.answers.com/Q/How_many_members_does_PayPal_have
http://statspotting.com/2011/09/paypal-statistics-one-fifth-of-global-e-commerce-conducted-via-paypal/
http://www.quora.com/What-is-the-average-users-PayPal-balance
http://techcrunch.com/2011/01/03/j-p-morgan-global-e-commerce-revenue-to-grow-by-19-percent-in-2011-to-680b/
http://www.internetworldstats.com/stats.htm

Details please

If someone does an cost-effective and easy solution for hard power cycling, i would not hesitate to spend several hundred euros immediately for just testing gear

So if someone finishes that Arduino design and makes it easy to order complete units i would buy several immediately.
Even if i have to solder the 230V (110V) power cables onto the relay board wouldn't really matter.

Another thing i'm looking for is something which can switch from power source to another, say network power goes out, it changes to inverter operated from a bunch of car batteries kept charged by a cheap 40€ trickle charger.

Arduino with relay board and ethernet shield works out to something like 275$ for around 40 ports. 160$ of which is relay boards, 8 of them.
Tho requires a bit of work to get running, but ought to be rather easy after that.
Not a perfect solution, but half way there.

I've done quite a bit research on this for servers, namely how to build a cluster of cheap servers on the cheap.

One option is to use Arduino with a Relay board and Ethernet Shield.
That will give you maximum of around 40 relays, depending upon model of Arduino used etc.
Parts can be sourced from: http://arduino-direct.com and eBay
Some information: http://arduino-info.wikispaces.com/ArduinoPower#4-8

As for finished product i ended up choosing IP Power 9258T and Avocent Direct PDU for initial setups for it's low buy-in price.

For KVM i ended up choosing Belkin Omniview for larger setups and Aten CS1716i for smaller setups.
For big setups Belkin Omniview works out somewhere around 30-40$ per node, depending upon do you get your gear from cheap or expensive place.
Aten CS1716i is a few $ more expensive per node only, but the Belkin also provides saner setup by the means of using CAT5 cabling

BTC fine, with somekind of escrow?

you didn't answer my earlier question. And that machine can't quite reach 2Ghash. More like 1.5-1.8

The rig comes with everything, just put power on, and start miners?

Now you are trying to just annoy me.
"If you was a competent linux user you would know this and edit your configuration files accordingly.". I do admit i don't use it on desktop - because it sucks as desktop if you need to get work done, but that is quite an insult. I probably have 100x the experience you do (I do happen to manage hundreds of servers).

Systems in question which i've tried:

ASUS P5LD2SE Mobo
Celeron 2.66Ghz CPU
1Gb DDR2 667
PowerColor HD5830
Kingston USB Stick for boot

Another System:
ABIT IN9 mobo (their flagship before they went bust)
Core2Duo X6800 (yes, extreme series a bit overkill for this)
4x1Gb DDR2 667
2xClub3D HD5870
Kingston USB Stick for boot


One did work, but i can't recall the specs of the machine anymore. Very similar to those tho.

Catalyst etc. complains lack of drivers.

aticonfig init makes new xorg.conf, but then X won't start up.

Persistence does not seem work, used unetbooting to create the sticks, and specified ~1Gb for persistence. Hasn't mattered tho so far that much i've been arsed to look into it.

This isn't a linuxcoin issue it's just how the drivers work.
It is VERY much Linuxcoin issue, if newer Linuxcoin does not work but older does. Simple as that, doesn't require a genius to understand.

why can't you just VNC into the machines ?.
So linuxcoin has VNC setup out of the box? Tbh, i did not bother to take a look if it were, i did not see NX or VNC references in any obvious places.

now am I meant to make linuxcoin better if everyone just posted "it don't work" ?
I did reference earlier discussion about this very same problem, on the previous page someone else was having similar problems.

What I'm not interested in is people posting rants on this thread moaning about broken machines and not posting anything useful
Those machines work perfectly, ty for asking. It's linuxcoin which does not.

I reiterate a bad workman always blames his tools.
So it's my fault that LinuxCoin has bugs? What does it say about developer who always blames users "it can't be a bug! i created it so it's bug free, it's your fault for not putting effort work around the 'features'!"

I did accept the SDK -.- I tried from terminal, and from system, and as a regular user etc.
and keymap has nothing to do with being broken. Being broken has everything to do with the drivers not working anymore. There are also other minor bugs and glitches which are annoyances.

Also a bad workman is someone who does not recognize when tools are broken, or does not recognize when a better one is required.

Plugging monitor after booting -> when you boot up say 5 machines from cold, are YOU going to wait for one to boot up individually, or power up all at the sametime?
Or if you meant "why would you need monitor after booting up", maybe starting the miner and verifying clocks. Didn't intend to just idle those machines.

aticonfig can see the adapters, but not phoenix, or amdoverdrivectrl etc. complains lack of ATI drivers. i used aticonfig initialize, but that is to no avail -> restarting X causes only a black screen, won't come back.

I would call that broken, when out of box advertised features do not work, not even after sensible amount of research & trying, whereas a older tool works as expected

Btw, Ad hominem is an argumentation mistake, quite an serious one.

i have the same adapter not found errors with 0.21b, the older a version worked fine.

Works only on 1 machine, couple others does not work.

Also on top of that, if i do not have monitor plugged in while booting, when i plug it in it complains "Video mode not supported", meaning it's running resolution beyond the tiny old Dell LCD i'm using.

So i'm mostly running the older version as the newer version does not work at all for me.
Trying the above mentioned -> http://developer.amd.com/sdks/amdappsdk/assets/ati_stream_sdk_installation_notes.pdf section 2.2

EDIT: Typing all kinds of special characters on US keymap wasn't my idea of fun, it's just easier to use old version or roll my own in the next upcoming weeks than fool around with something obviously broken.

From 338.9 to 352Mhash/s on my 6950 (only @860) ~3.86% gain

Will donate a bit as well if this is stable

Nope, not using the latest, but atleast the 2 first queries are mine, and my added indexes are in place.

Haven't yet checked what the diffs are.

I'm running about 1.1Ghash against this, and i noticed severe drop in earnings too, due to the more frequent phoenix crashes.

Will need to put aside 2 identical rigs to properly test is there difference.
My rejected rate is also "through the roof", at about 7% via this proxy! But again, need the testing setup of 2 identicals to verify.

Need to go buy more 6950s or 6770s to setup 2 identical rigs

Atleast i do, Yes, 66.5 dBa

In all seriousness tho:
http://www.gcaudio.com/resources/howtos/loudness.html
That is about as much as normal conversation.

Currently at my workstation the noise level is probably plenty above 66.5.

PSU is both a transformer and AC/DC switching unit. First you need the transformer to get to proper voltage, then you put in a transistor bridge, few diodes etc. and you got yourself DC. I also would recommend a regulator and few small capacitors and low Ohm resistors, and a heatsink for the regulator(s). The combination of regs, capacitors and resistors should filter everything out

Working with high amperage low voltage is safer than high voltage low amperage IMHO, because the electricity does not "jump" as easily. High voltage is needed for being able to go through highly resistive materials (see Tesla coils for example. That's also why engine spark wiring insulation is such an important factor). My hobby used to be RC cars, and i had 7.2V 400A peak battery packs... I was more worried about getting them pierced while driving (and resulting explosion & fire) than getting a shock.

Here is how you can build it:
http://engknowledge.com/power_supply_design.aspx

I would design it to be quite parallel, meaning multiple transformers, multiple bridges, multiple capacitors etc etc.
So that any single component has smaller impact if it fails on operation, or is otherwise flaky. I would design regulators, capacitors etc. well beyond required spec to reduce risk of failure.

Also, doing it from this low level you could add some basic electronics to handle backup power, see for example: http://www.dreamgreenhouse.com/designs/12v/index.php (links to their UPS project),  http://tech.icrontic.com/articles/super_ups , http://1wt.eu/articles/alix-ups/

Last one is the most intresting one infact

Google is your friend, but i wouldn't suggest trying this with no past experience on the required scale....

I can sell you some at 0.3BTC(3.6€)/ea. for quantities over 10. For less 0.38BTC(4.56€) each. I got 20 in stock, more coming in several weeks. Includes shipping.

Can accept also Paypal, Moneybookers, SEPA wiretransfer.

that looks insane, more pics please!

currently testing the flexible mining proxy and it reports almost 7% rejected! :O

Haven't checked direct to pool rate.