bOS lets you control voltages/freq/fans on a board by board basis, don't think you can do that with BM's FW.

It would appear Innosilicon has released a video of the T3 in action on their blog/homepage, link below:

http://www.innosilicon.com/html/news/31.html

Items back up for sale with new price!

Is that pot something I can adjust while the PSU is powered?

I see in the description and the manual for the APW5 that there is supposed to be an ability for the user to “change output lines and fine-tune the output voltage.” Any ideas on how to adjust the output voltage? Here’s what the inside of one of my APW5’s looks like:

Well that explains why the AB firmware won’t install for zachj00, it’s only compatible with the Xilinx boards. I wasn’t sure if he could get the 2017-04-28 FW, so was checking to see if his board didn’t have the requisite 512 or 1024 mb of ram.

Zach, can you tell us exactly what your R4 overview page says for your File System Version? I’m curious if you’re on a different 2017 FW that BM no longer allows us to download.

Is it possible you're one of the few with an incompatible control board? Can you post the start up sequence from the kernel log? Curious if you have neither 512 or 1024 mb like mentioned in the FW title.

Also, check and see if your control board says XILINX on it. If not, the AB FW is incompatible with your board.

What would make me feel way better would be for someone to figure out how to power a T/S15 with an APW5 or something similar. I'm far from an EE and based on what others with more knowledge have said about the efficiency gains all these new machines receive from integrated PSU's, I doubt it would be possible without some serious jerry-rigging.

The case could be made that MicroBT has no experience with 7nm (in terms of produced miners) and may not be as successful with the efficiency advantages they eked out from 16nm, the node size they've been working with for what, 3 years? They also may not have the capital to purchase/manufacture 7nm chips at a scale which will allow for a price competitive product.

I think he was saying that he provided both a shipping and a billing address but, out of fear they would ship to his billing address, opened a ticket 20 minutes after placing his order to confirm they were shipping to his shipping address and not his billing address. Pangolin confirmed they would ship to the shipping address. Tracking number is provided by Pangolin upon shipment which showed the package was going to the billing address instead of the shipping address.

I've had this problem before with large and small retailers, especially when the shipping/billing are in the same state or even zip code (my billing address is only 3 miles from my shipping address and something like 5 miles from my office haha).

Given this is my first all-new Bitmain (BM) product, if you exclude the S7-LN because it wasn't using new chips, I'm in the strange position of hoping BM stays around so they can properly support my investment. Since getting into mining in 2016 and subsequently learning of BM's bad actions this is the first time I've felt this way haha.

Maybe I'll settle with my conscience by hoping they keep operating but on a much smaller scale?

To make it clear why I am sharing these thoughts in relation to this thread, I've noticed that someone in this thread with a T15 that shipped after mine is on firmware dated December 29th, whereas the FW on my T15 is dated December 18th, which is also the latest officially listed FW from BM. Why would they ship the next batch with newer firmware and not list that fw for download? Are the batches different in terms of hardware? If they're different, why not give them a distinct subgroup and list that different machine's FW on the support page?

Which brings me to why I worry: are things at BM so bad that the chaos in the C-suite is affecting decisions regarding FW releases and support? They've released 3 new ASICS with integrated PSU's without offering any way to purchase the PSU (APW8) on its own, let alone share the specs on the PSU so we can be informed on our machines and ensure they're being run in the proper environment.

As far as I can tell they still haven't even realized the specs of the APW8!

Panel says 208v but multimeter reading directly from the plug on the PDU says 200v.

I'm seeing 21TH @ ~1310-1330w in LPM

My T15 is still hashing away without issue. 23.3 TH/s at the pool, ~1540 watts.

Just to confirm, you’ve tried the Antminer-R4-all-201704280718-autofreq-user-Update2UBI-NF.tar.gz fw, correct?

Are you getting both blinky lights from the Ethernet ports?

One thing, which I assumed was because I swapped an Ethernet cable from an A4+ to the T15, is that when I powered up the T15 is it stole the IP address from the A4+. Given this may not be an isolated incident caused by my switch, are you using static IP’s or dhcp? If DHCP, you may want to move all the existing miners to static IP’s and see if the missing miners start showing up.

Yes, that would be correct! However, be aware that you will the lose the ability from the stock firmware to set your own fan speeds and freq when you upgrade to the UBI filesystem.

That is the ASICBoost FW from Bitmain. However, if you are not on the other listed firmware:

Antminer-R4-all-201704280718-autofreq-user-Update2UBI-NF.tar.gz

The update will not work. The update listed above changes the file system to UBI or something like that which is necessary for the AB firmware to be compatible as it uses a UBI filesystem but does not contain the necessary files to changeover the existing filesystem.

I don't trade crypto on an exchange or anything like that, but I've sold some BTC (e.g. buying miners) over the past year and will be reporting using FIFO.