Unless you're running two sticks and that's the combined total, you're either wrong about the frequency (22GH means 200MHz) or something's seriously hosed.

I recommend you jump over to the 2Pac Support thread and get VH's cgminer instead. It supports the Compac and is a heck of a lot better than Novak's original driver.

As of right now, I'm out of stock on seconds. It'll be a couple weeks before I have more.

If you've paid in the last couple days, your order will ship tomorrow.

Kill JavaScript and a lot of garbage goes away.

Even though the purchasing request instructions are given in big bold letters in the very first post, I'm still getting a lot of messages from people who ask how to purchase stuff. Often they're not even very clear about what they want.

Normally I'd ignore them outright because, while I like money, I really like not dealing with idiots. However, since this is for the good of someone I care about, I'll suspend that policy and give a response. However, since it's still my policy to be a cranky jerk to anyone who asks me to do his homework for him, the best you'll get is a dry response indicating you should acknowledge the clear instructions given in big bold letters in the very first post of the sales thread. After that you're on your own.

Did you look into any of the things people were talking about three pages or so ago?

To be quite honest, I'm surprised they hadn't done it already. I guess they would have if they'd had another batch ready sooner.

Known issue. Use a batch file instead.

Aye, but African swallows are non-migratory.

I have no idea what you're looking at. On older prototypes of the 2Pac and the original Compac the EN line was snaked between 1K and 0.1uF to the 5V island, but on all the 2Pac production versions it's internally routed to below the 2R2. Again, nothing buck-side interacts with 5V directly. PGood is not connected to anything, never has been on any version of any stick I've yet made, production or prototype. I'm at the shop now and just verified that against the layout files.

If you drill out that via under the 5V pin, the only things left touching 5V are the high-side FET and input caps, through the visible surface trace and an identical inner-layer trace interconnected with nine vias.

I generate footprints manually in a text editor; it's not too bad.

The PGOOD is NC. Like I said, nothing on the buck ties to any 5V line that doesn't come through the 2R2.

BM1384 datasheet is, as far as I know, still accessible on Bitmain's website.

Actually the original Compac was one of the first boards I laid out myself. It came after a couple BM1384 breakout boards, but before that Novak did most of the layouts. The 2Pac's buck circuit is almost identical to the original Compac's.

Only the buck is downstream of the 2.2ohm. It's sensitive to ripple, to the point if Vin drops below about 4.2V (I think) the buck will drop out. This causes a nice power cycling during high overclock if your hub's power lines aren't too good. The LDOs and LED aren't susceptible (the worst case is, the top node's LDO will have issues if the 5V line goes below about 2.8V) and I didn't want the extra power draw inside the ripple buffer because, you know, I^2R losses and limited capacitor storage.

If you drill out the via under the 5V pin, you shouldn't need to cut any copper planes. If you're looking to isolate the USB jack's 5V line, just drill out that via, hold the pin off the board and run a jumper from it to the uphill side of the 2R2.

Pulling the 2.2ohm should isolate the buck chip from all 5V. Even the tied-high enable line pulls from below that. I put that RC on there to help isolate the buck from ripple due to input impedance when the load's cranked up.

You still have to worry about the node-level 1.8V LDOs and the LED which are tied directly to 5V.

Other than that, the high side FET's the only thing remaining.

Which, by the way, I believe the entire power section has 2 interconnected layers of power plane and 2 interconnected layers of ground plane. Didn't want trace impedance to be the reason someone's overclock attempt failed. It's not a booby trap so much as an overbuild.

I don't have one in front of me, but I think if you look just above the 5V pin on the USB jack there's a via right there. Or maybe it's under the pin. Not sure offhand. Anyway that connects the 5V line to the meandering trace powering the buck (via 2R2), the LED circuit and LDOs. Drilling that out cleanly will probably disconnect everything, and then you can inject 5V back into the line at any of the taps mentioned above. Lifting that pin will then isolate the plane going to the FET, which you can use with any voltage you want (provided your input caps don't asplode).

But please note your inductor is only rated for 17A.

I hadn't talked to him in a while so I emailed on Monday, haven't heard back yet. Earlier on he swore up and down that I'd have access to chips so I really hope he follows through on that, even though it's taking me forever to be ready for 'em.

I don't expect to be doing too much with BM1384 after January or February. Starting this week I'll have R&D time again so better stuff will be back in active development. I've already got half a layout done for the next step in BF16 testing, which should be fun, and another thing or two in the works.

More stable perhaps, but less efficient. The bulk of power use in a CMOS IC is the charge required to flip the transistors on and off. The current use is proportional to frequency for obvious reasons - more flips per second means more electrons moving back and forth per second. In order to flip fast enough, a higher voltage might be required in order to have more "pressure" pushing those electrons into and out of the gates fast enough to keep up, so voltage tends to increase as frequency increases. This is where your efficiency comes into play. For a given voltage, the power use is directly proportional to frequency, but as frequency increases the voltage has to as well - so power use increases from both voltage and current increases, which means more overall power per unit work.

Bitmain's original S5 announcement post (https://bitcointalk.org/index.php?topic=902305.0) has a chart for this.

Voltage has nothing to do with the workload, just the hardware's stability. In general, the higher the voltage at a given frequency, the more stable it'll be (until heat or junction breakdowns become an issue).

The amount of work done is proportional to the operating frequency. I see about WU 1.53/MHz with a 2-chip stick, so 225MHz should run about 344 - plus or minus, because USB traffic and pool latency and whatever.

Stock sticks run about 1.26V, which for BM1384 puts it chip-level roughly 0.25J/GH

Seconds run between 1.3V and 1.4V, 0.26 to 0.34J/GH chip-level.

According to Bitmain's BM1384 efficiency charts, that's what can be expected. It's worse efficiency, but still reasonable. If it won't go stable at 1.4V or below, I won't sell it. Strip the chips and start over.

Also, everyone who's messaged in the last week and I haven't gotten back to you yet - we just finished the "Christmas rush" assembly last night about 7:15PM. Three months ago we pushed past 400 units per week for the first time. We shipped almost 2100 units in the last three weeks to get stock to resellers to ship before Christmas. Today I finally have some breathing room so I'll catch up with everyone.

Seconds orders to date nearly exhausted my seconds supply. They only appear when a stick fails to perform during stock testing, something I actually try to avoid; only about 5% of sticks fail and become seconds candidates, so now I have about 100 standing by to be tested as seconds. I expect about 70% of them to pass seconds testing. Gonna start on that today.

Seconds testing happens about once a month, when I have a free day available and enough candidates stacked up to make it worthwhile. Testing for these is actually more extensive than for stock sticks because I always start out the voltage in the lowest range and adjust upward after every failure. Some sticks get run through four times before they pass. Takes a full day to get a batch of 100 out, where with stock sticks we can test 400-500 in a good day.

That's the basis for the "limited supply" statement in the first post.

Maybe one of these days I'll get the firmware finished up. The guy I handed the job off to still hasn't finished, and 2Pac manufacture leading up to Christmas has been seriously kicking my butt. But I've got a good crew and things'll slow down soon, so I'm hoping to open up at least one day a week specifically for R&D. Finally get some stuff done. In addition to the BM1384 Terminus pod, I've got two CNC builds, some other machinery, Bitfury design and a new stick idea piling up. Starting to run out of back burners.

But we'll get caught up here soon.

Yes, the same driver already has baseline support for the Terminus pod because VH is super thorough. I think with the right USB adapter his code could run an entire S5.

I was figuring on offering just the miner, and also a full kit with the miner, USB cable and power brick. Probably could make a preconfigured Pi an add-on option as well. I don't know what the USB load is but one Pi should be able to handle several pods so it wouldn't have to be a one-to-one thing.