Maybe because they're sick of having over-reaching corporations peddling their slave-labor wares at ridiculous markups. Which is obviously not the case when you buy an iPhone.

I would rather build my own because I like building stuff.

Okay. PM me again when I have announced stock of either of those things.

Right, and I'm saying that's entirely expected because the Pi's ports are current-limited and the whole thing's being sourced from a weak-sauce micro jack. Page one, post 10:

Yep, your hub sucks.

Stock is 100MHz. Stock voltage should be good for 150-175MHz but note they pull 4W at 100MHz so more like 6-7W at higher speeds. If your hub can't handle more than 1A to a port you'll have trouble. If you're plugging the stick directly into a Pi's port it definitely can't handle it.

Maybe your hub sucks.

And like I told you when you emailed, you need to look up how to assign Minera to use a custom cgminer. I'm pretty sure that process has been described three or four times in this thread.

Not yet. My firmware guy's been AWOL since sometime in November, so I'm just gonna do it myself. That's pretty much the last step, since all the hardware was finalized and ready for assembly before Thanksgiving. But we've shipped over 3100 2Pacs since Thanksgiving and that ate up a lot of time, seeing as that would have been about a 10-week job not too long ago.

But now I've got enough competent help and they're all trained at everything so we can still keep up production with me only putting in 3-4 days per week on it, which means 1-2 days per week of R&D time. Step one, finish the Terminus pod. Step two, newer and better things.

Naw, looks like the guy I was responding to deleted his post.

Anyways it wasn't anger so much as incredulity.

Seriously? I literally just said I'm out of stock, have been for a week and will continue to be for several weeks.

Anyways the first post also says I don't ship to Europe.

All y'all still asking for these need to take note of two things - the part in the first post which says SUPPLY IS LIMITED and the recent post by me saying they're out of stock for several weeks.

I would guess that's a regulator module replacing the onboard circuit which is responsible for providing node-level IO voltages to the top several sets of chips.

A fan?

I would bet it's not much different from previous generations. We used the BM1385's datasheet's comm specs to write a BM1384 driver which previously had been adapted from BM1382 code. Seems likely they wouldn't start over from scratch when four generations have already proven it works.

I might have torn into an S9 by now if they hadn't stayed balls expensive for the last year.

That's certainly interesting.

Every time I hear "pre-loved" with respect to machinery, I immediately think of a fetishist doing lewd stuff with it and, well, ew. Stickiness tends to reduce resale value.

You are using a cgminer without the gekko driver. Read the first post.

Not really sure what you mean by that. There's no information to share that hasn't already been given.

That's quite thoroughly weird. Both hashrate and WU correspond perfectly to 200MHz operation.

Gonna have to take that up with VH. Only thing I can think of is, the chips have a clock speed they default to after reset. I don't know if it's 200MHz but it's in that neighborhood. Maybe the string got reset, or the freq-set command got lost somewhere, and cgminer claims it's running 100MHz because that's the speed it thought it told the stick to run at, but the chips are still operating at the higher default speed. I don't know if that makes any sense, since the work timeouts are probably based on clock rate so you'd think it'd be hashing on the same work twice before getting updated in which case it'd be returning a whole lot of duplicate nonces. As far as I know, cgminer either prints out a dup-nonce warning or, on older versions, chalks it up as a HW.

Anyways. Weird. I'd guess it's a software thing. VH wrote the driver so he could give a better idea of what's going on.