My advice to you, yohan:

Size each unit so that its fully-loaded configuration is about 960W. This will allow users to put exactly 2 of these units per 120V-20A circuit, or 4 per 240V-20A circuit, or 9 per 208V-30A 3-phase circuit (while not exceeded 80% of its maximum current rating). Rationale: in datacenter environments, users pay a fixed monthly price per circuit. A circuit not used at its maximum capacity (or a config not fully loaded to not over-consume) is wasted money. BFL failed to follow this advice of mine when designing their 1250W mini rig.

We are looking to make this rack stand alone and yes it is up against Butterfly's larger products. We would like to make this run free of a host PC. This is very much a big system concept. I hope we do a better product that the competitors but time will tell on that point. It's very unlikely we will do a PCIe backplane for this although we are looking at these for our general HPC products. The bought in industrial backplanes are usually very expensive so it's unlikely we would but that in. However we can use one of our standard ones that we have designed or even a derivative of one of them. The cost is quite reasonable doing it this way.

I would forget any secondary value on any sort of mining kit. If you are banking on that your equations will be wrong. FPGA families have replacements on average every 2 years and the old family will have limited value even still as chips never mind in a system that has either to be reused or silicon recovered. GPUs are even worse for this. Try selling a 2-3 year old GPU. It might have cost £500 but in 2/3 years you can buy a brand new board of equivalent performance usually for less than £100. Second hand maybe it goes for £30. I for one would not want to buy a ex-mining GPU given the stress put on them but of course most people don't mention that on Ebay.

We are not likely to use PCIe in this way but it is a possibility for a quick picture have a look at http://www.schroff.co.uk/internet/html_e/index.html. So what we are taling about is a rack with card guides and a backplane at the back of the rack. The cards slide in and connect to the backplane.

The backplane standard could be an industry standard or just something we do to suit the purpose. The main thing is that the metal work and supporting guides are standard things. With a full height 19" rack we might be looking at fitting up to 4-8 sub racks depending on height we adopt on the sub-rack and what we have in power supplies. So we might be able to do an entire rack with 0.5-1 TH/s if I can add up correctly. Then it is just a case of adding racks. In data centres you might find hundreds of these sorts of racks.

Lets start by clarifying that it won't be a single big board. Those are actually expensive to make and there is no logic in this design that needs that approach. The architecture is more of a controller card with processing cards linked by a backplane that wires it all together.The backplane should let us have 12 working boards maybe up to 19 in one run depending on design decisions made.

Similarly PCIe is a bit of overkill for this one internally but could be used to link a rack to a PC or several levels of rack. We might do a PCIe card but that is a different project.

28nm at the moment may not be viable especially Artix that is probably 6 months away. 28nm may also be expensive initially. With what we are doing on bitstreams, and partner offerings, might mean FPGA type might be relatively irrelevant. It's more about the system cost than anything. 45nm may still be the best option today but that is one of the things that we are looking at.

Voltage that is used for whatever FPGA we use is being looked at. We might put in a VRM but that is probably more complicated than is necessary and has it's own cost. There are other ways to do this.

Historically fans have been a reliability thing and we might put in monitoring or PWM but these features do have a cost themselves both in materials and electricity. That needs to considered given the fans we are currently using on Cairnsmore1 have 100K+ Hrs lifetime (11-12+ years) and have a 6 year warranty. The floating bearings that have come in the last few years are a lot to do with this reliability and alternative approach might be a planned fan replacement maintainence schedule.

A fan tray may be the way we do cooling for this design and that would have whatever fan headers are needed. We have some other ideas here as well and more on those when we have thought them through as little better to see if they are viable.

Power wise this will take power from the backplane and there won't be a choice there. The processing card isn't a replacement for Cairnsmore1 but for the bigger rack market. Different backplanes are a possibility including a mini setup maybe with a cut down number of slots but that is for later after we get the big solution out in the wild.

Yohan

Ok I will start this one off by saying that we are looking at a range of FPGA technologies to base the product on not that any of you might think we would start putting GPUs in our products although you never know. Some of our decisions will be based on how we do this week with more fully loading the Cairnsmore1.

We are probably thinking of 19" rack as a basis for this product and we already have some backplane designs that we did previously that might be useful either directly or in an adapted form. This also fits well with power supply availability and so on. This would also allow a modular purchase of a system that would be easy to upgrade and add to as time goes on.

One of our aims will is to be a very competitive FPGA solution in the market for large scale mining.

We have an initial individual card target of 4-5 GH/s+.

We are looking at our cooling technology and we are testing a new idea this week in a different product that might get adopted into Cairnsmore2.

Timeline - we are likely to limited by FPGA lead time which is typically 6-8 weeks so August-September is likely to be the initial availability of this system.

What we would like to hear from you guys is what you would like in interfaces? USB?, Ethernet - 100/1G/10G?, Cable PCIe?

And any particular features you might think we need to include?

Yohan

Motion Raised! Opposing what the motion page reads, and enforcing what the contract states: 70% of shareholders must vote "yes" for the motion to pass.

https://glbse.com/vote/view/24

If this motion passes, capital will be raised to purchase five Quad XC6SLX150 Boards, which are featured here: https://bitcointalk.org/index.php?topic=78239.0
1400 shares will be issued at 0.6BTC each. Remaining funds from this capital will be invested in other GLBSE companies/bonds, which will ultimately increase Cognitive's earnings. This will more than double Cognitive's current hashrate. The operator will maintain the contractually granted 10% of the company and will receive 140 shares.

The default size?  Yah no. 

Unfortunately, I'm seeing the exact same problem on four different iMacs (all of the same model, OS version, etc.) regardless of whether the firewall is on/off. The only difference between it working and not is whether I run the March or May version.

Although I don't know much about the getwork protocol, eventually the March diablominer receives a response from the pool like the following (which I understand to mean that the work was accepted):

As the result of sending a non-empty "params":[] value.

Whereas the May diablominer only receives new blocks after POSTing the following

I sniffed the May diablominer with wireshark for twenty minutes and never saw a non-empty "params" value sent. The old one sends one every minute to few minutes.

I'm not sure how else to troubleshoot this. Have any of the command line options changed (--url being the only other one I use that I didn't already mention)?

I suppose I could try another pool, but the only variable here seems to be the newer version.

What's with the double spaced lines of output in this new version?  I personally don't like it.

If you were interested in say mounting in a 3U rack a bolt-on mechanical board edge could be made to make the board the right width. A wider or longer variant of Cairnsmore1 would also be possible. That's only a few minutes of design time. It's then more a manufacturing thing where it splits the volume and we buy in 2 PCBs rather than 1. However if there is a serious requirement for this and enough orders say 25+ units we could do it for a small increase in cost. Other than the PCB all the other components can stay the same and we get the volume pricing advantages of that. Worst case (small numbers) the pricing delta might be about GBP £25 / US $40 per board. If that is of serious interest to anyone it is worth contacting us. It's something we can do without a serious effect on the current batch build. The current heatsinks should work well in a rack config if suitable runner spacing is used as we considered side blow when we choose them as much as the downward standard configuration.

So which is it then BFL is great because it has got case or it sucks because the cooling is no good in that case. I for one am failing to see the point your trying to make here putting these boards in a case is going to lead to the same cooling problems as with the BFL.

This may be unrelated to typhoon's problem, but the latest version of diablominer fails to submit to and/or connect to the mining pool I've been using.

I'm running on 2011 iMacs, running OS X 10.7.4, with AMD Radeon HD 6750M (512 MB of RAM) and Apple OpenCL 1.1 (Apr 9 2012 19:41:45).

The command line arguments I'm using include:

./DiabloMiner-OSX.sh -w 64 -v 2 -na

The May 2012 version outputs the following:


The March 2012 version outputs the following:


Are there any options I can use to give more verbose messages?

On an unrelated note, I've recently noticed this model iMac go down to ~50-60 mhash/s from a previously consistent ~70 mhash/s. Could this be related to May's change(s) or the latest Lion update from Apple?

But if the BTC price were to drop to say $0.50 or rise to say $50 before the Bitcoinica positions were closed and the accounts settled with the clients  then the real fun begins.

Don't know why your even bother mentioning the BFL labs as a good idea then they use a similar design, I've seen more than a few my BFL is throttling posts on here.

While a case maybe be nice for those that want it I have no problems with an open air solution in fact I prefer that option not only does it save me money on something I don't want/need to have my boards will run cooler due to the lack of a case. Plus could be a nice business to someone like yourself that wants case get some made and if most people are thinking like you then they can buy them from you.

I agree completely, but only under the circumstance that the "better" option is available immediately (or available at a MUCH better cost/performance ratio)

If I have funds to buy say 8U filled with Cairnsmore1 right now (32 boards) and they are available in the next 30days, yet it would be 2-3 months before even the possibility of a higher density better solution, I'd rather be mining for that 2-3 months on the slightly less optimal Cairnsmore1 option, and re-evaluate when the available options change.

Also considering the smaller boards let me grow in a more fluid fashion, I can increase my hashing power in a fairly aggressive curve.

IE: 32 boards at 4x FPGA each, lets just for round math assume 800MHash even per board. So 25.6Ghash from that cluster, that lets me mine around 475BTC per month at current difficulties. That 475 can be re-invested in roughly 2 more boards (At the inflated "full" price). those 2 new boards generate about 30BTC a month, So month 2 I have 505BTC, buy 2 more boards, then I have 60 spare next month (plus the 30 from previous month) buy 2 more, now I Have 90 spare, plus the 90 from the previous 2 months, now that's 180 on top of the base 475, which buys me 3 boards. and so on. Typical exponential growth curve, allowing me to add quite a bit more hashing power in the timeframe it would take just waiting for the new boards to possibly come out.

I realize my math above may be skewed or off (it's all off the top of my head) But I've also done forecasts based on this (with real numbers in extensive spreadsheets). Looking at smaller boards versus larger boards with slightly higher performance/$ ratio, and the smaller boards won out in longterm ROI. (because they grew faster due to the fluid expansion). Especially when the larger option wouldn't be available initially for some time.

This is of course offset if the higher end board is much lower cost per MHash but I would be surprised if doing a much larger board dropped more than 10-15% off the price per chip. (just personal opinion though lol so take with a grain of salt)

I'm not sure where you're getting the 24" card thing from?

I'm not trying to be an ass, I just think we're perhaps arguing 2 different points. You seem to be arguing the "best case optimal high density mining" solution, and I'm talking about "How can I best mount the cairnsmore1 cards in a rack".

I'm planning on mounting them similar to this:
https://bitcointalk.org/index.php?topic=70611.msg844733#msg844733
(that's my current icarus setup)

But using a 4U actual rackmount case rather than a wooden 3U case.

Mounted vertically the cards will need about 2" wide, a rackmount case is about 17" wide on the outside (so at least 16" - 16.5" of usable inside room side to side) and about 14" deep for the motherboard compartment (12" - 13" for the actual motherboard plate). These cards are like 6" - 7" deep so I can stack 8 of them side by side, 2 deep in the same arrangement I have in that photo. Filling the back of the case (like I said redirect the powersupply to the front of the case in place of the 5.25" drive bays).

This gives me room for 16x cairnsmore1 cards in a 4U rackmount case, with tech available right now (not a hypothetical future product). so a total of 16 FPGAs per Unit of rack space.

I'm not talking about an "optimal end solution for best case high density bitcoin mining" that's obviously going to require some higher density special purpose hardware, like the Merrick3 adjusted for higher power density, and put into a standard PCIe motherboard. OR the Merrick1 modified for a few less chips (maybe half, like 50 chips) and higher power density. I agree completely, a 100% custom 1U case with a custom board (or boards) holding a lot of chips will result in the best density. But that product doesn't exist, and won't exist for at least a few months (and that's if someone starts working on it right now).

The cairnsmore1 will be available in a couple weeks, so I'm talking about the best way to mount THOSE boards in a rackmount enclosure.

Also to note, I'm not talking about custom cases. I'm talking about an off the shelf case, which has removable drive cages already, you just unscrew it and remove it. Problem solved. Sure you need some custom brackets/clips to mount things, not a problem. That's what a 3D printer is for

Also yeah I planned on using the HALE90 as well in my build.

Does that make a little more sense?