Thanks to a little research by Nick in the Coindesk comments and Burrito on IRC, it appears the location is:

Internet Cafe Worldwide Connection
25 Boswell Street
Holborn
London WC1N 3BW

According to Coindesk a new Robocoin machine went on-line in London today (London's second Bitcoin ATM and first two-way machine), and there were crowds there for the launch.

http://www.coindesk.com/robocoin-machine-heats-competition-londons-bitcoin-atms/

Anyone know exactly where this new machine is located?  And also where it was announced - presumably it must have been announced somewhere if there were crowds, but my Google-fu is failing me...

It's run by Global Bitcoin ATM but they don't seem to have any information on their website yet

http://globalbitcoinatm.com/

roy

Scan sell a bunch of pro audio gear (http://www.scan.co.uk/shop/pro-audio) and take Bitcoin.  Not sure if they will ship outside Europe, but if they have what you want it can't hurt to drop them an e-mail and ask.

The PSU is an ATX PSU and is internal (although I'm running with my own PSU, mounted externally using various extension cables).

My (600GH/s) unit is powered as follows.  24-pin motherboard connector to the small controller board which is stacked on top of the Pi (and which powers the Pi and fans).  A six pin PCI-E to each 200GH/s hashing module (three of them in my system).  A standard floppy connector to power the backplane that connects the modules.  From memory, I think that's it - but it's certainly all just connectors on a standard ATX PSU.

The supplied PSU (and indeed the one I'm using) are of a kind that has a physical hard power switch next to the mains lead.  The controller board (and via it the Pi and display) are powered by the standby power from the motherboard connector.  So when you turn on the hard power switch, the system powers up the Pi and the display, and the Desk starts to boot up on standby power (even though the PSU is really still 'off').  After the Pi boots, it then turns the ATX PSU on, and then the controller board powers up the fans and of course at this point the modules are now powered by the PSU, and the system starts up MinerOS and cgminer and starts hashing.

Hope that helps

roy

Some comments on the software:

The temperature display on the front panel shows the first temperature sensor - would be nicer if it showed the higher of the two.

The target and cutoff temperatures don't seem to be changeable so I can't test if they actually work.  There's no fan control as far as I can tell (the fans just run at full speed I think) so I guess the target temperature doesn't do anything yet.  That's no big deal.  But would be nice to know if the cutoff works.  I have a suspicion it isn't implemented yet - at least, I couldn't find any code in MinerOS either handling this or passing the temperatures through to cgminer.  It would be nice to have a working thermal cutoff in place before warmer weather arrives.

Sometimes some of the AJAX URLs stop responding.  I've had the module info disappear from the dashboard page - and also the temperatures and total speed.  The odd thing is that a reboot doesn't fix this, but the dashboard does start working again eventually.

And according to the home page the current pool speed is 0.00H/s

Grr, it would pick tonight to do this, just when I'm commissioning a new miner...  :-(

Web site showing 0GH/s on my workers, despite shares being accepted.  Anyone else seeing this?

Is it possible that the machine you used to gpg encrypt them is compromised?

Oh, sure, my first statement is speculation on my part.  (I thought you were saying that there was an official statement from Gox saying that they lost all the coins - sorry if I misunderstood.)

But the way in which the coins are being moved is very specific.  What we know is that these coins are being moved in a way that is consistent with the way the PHP code that purports to be the leaked Mt Gox code would automatically move coins.  That much is pretty much indisputable.  So the possibilities are:

1. 200,000 coins have been loaded into the Gox hot wallet, or
2. Someone with access to 200,000 coins (the thief?) is making it look like 200,000 coins have been loaded into the Gox hot wallet, or
3. The leaked code is a fake, so the movement of the coins being consistent with the way the leaked code would move them proves nothing, and this transaction is somehow unrelated to Gox (or at least it doesn't mean what it appears to mean)

I haven't looked at it, but other people have found past evidence on the blockchain that confirms that this really is how the Gox hot wallet works.  (3) sounded pretty implausible anyway, but I'm willing to discount it pretty much completely based on this evidence.

So this leaves us with (1) and (2).  There was some discussion on reddit as to which of (1) or (2) were the case, but gmaxwell claims to have found evidence via the Gox API (which is apparently still live) that strongly suggests (1) is the case.  I haven't looked at this evidence but I'm inclined to trust gmaxewell's evaluation of the evidence.

So, based on the above (and always remembering that this may be out-of-date as I write it, and certainly as you read it): what we very strongly believe, right now (although of course can't say with 100% certainty) is that someone has loaded 200,000 coins into the Gox hot wallet.

Yes, we don't know who, how, what or why.  So yes, the rest is speculation on my part.  (But if anyone has a more plausible explanation as to how or why someone has loaded 200,000 coins into the Gox hot wallet I'm all ears.)

Well, I wasn't really talking about the border experience.  But I will say this:

Sure, as a visitor I get the whole 50 quetions when visiting the US and it's a bit disconcerting.  But I find it a bit weird that Americans coming home get the same experience....

As a Brit, coming home to Britain, they just scan my passport, loop up at me and nod, and maybe say barely one word, if that.  I know our border controls are just as bad as yours at making visitors unwelcome (at least visitors from outside the EU/EEA) but as a citizen coming home we never get that kind of treatment. 

Here in the UK many people carry no ID.  In fact, in the UK, quite a lot of people don't actually possess any government issued photo-ID at all.  We seem to manage just fine here without the ID culture that the US and many other European countries seem to espouse.

"Exactly" what?  I'm saying I don't recall any of the official statements actually saying that they've lost all the coins.  If you have a reference to such an official statement I'd be interested.

Depends where you are in the EU.  Here in the UK, we don't have so much of an identify culture.  I'm always annoyed and angered when visiting the US that I'm asked to show my ID documents every time I visit a bar.  And I'm told that in the US you can be arrested for failing to carry your documents with you when driving a car, even if you are legally entitled to drive and are breaking no (other) laws.  As a Brit, I find that pretty shocking.  Here in Britain we're not used to being required to routinely carry identity documents, as one clearly is (IME) in the US.

roy

The official line, in the legal documents filing for bankruptcy, says nothing about how many coins they have left, only how many they lost.

Because that's what the hot wallet code does.  We know this because the PHP code has been leaked.  There was some discussion as to whether this was really the Gox hot wallet, or a thief (who has read the leaked code) deliberately faking it to look like the Gox wallet.  But Greg Maxwell pretty much conclusively confirmed it using the Gox API.

See the links I posted above.

No, it would appear that these are the remaining 200,000 coins that Gox has left after the theft of the 850,000 coins.  And that they are being moved from Gox cold storage to the Gox hot wallet. 

See:

http://www.reddit.com/r/Bitcoin/comments/1zt2hc/i_made_a_network_map_of_where_the_coins_went_from/
http://www.reddit.com/r/Bitcoin/comments/1zshct/4ee89f7cf824a85ad5f11d52604ffdebe9f01302bcea8ddec0/cfwnd2v?context=3

As to what Gox are doing with these coins, who knows?  But due to the bankruptcy proceedings it's unlikely that anything will happen quickly.  It could just be part of an audit to prove that these few coins are still accessible.

roy

My miner was delivered Monday 3 March.

roy

The Bitmine chips might be that efficient on paper, but E's review of his 600GH Coincraft Desk shows it taking 1.08 J/GH at the wall in normal mode.  So right now Bitmine's power efficiency, measured at the wall, is about on a par with KNC or Bitfury.

As for low power mode, E's figures show barely any improvement over normal mode if I'm reading them right (down from 1.08 to 1.06 J/GH).  So looks like no real low power mode, at least for now.  Of course, this is probably because the chips are not being undervolted, which is needed for low power mode.

I'd like to believe this will be fixed in a future firmware release, but does anyone know whether the hardware currently shipping actually supports changing the core voltage?

roy

ETA: Anyone know what efficiencies Cointerra and Hashfast are getting at the wall?

I doubt it.  You'll recall that Bitmine offered a choice of a reservation queue and a preorder queue.  I always assumed they did it that way because they expected that only very few people would want to preorder.

In the end, though, the community chose to preorders pretty much their entire production for several months, and the reservation queue became essentially pointless.