Z9 performance and general discussion


SSH is the only way to turn off correctly your miner.
Cutting eletricity to shutdown any system isn’t good (I do it all the time :P)


Hum, weird.
Did you try to change the root password by using the Antminer GUI » System » Administration ?
Then you should have SSH access with user «root» and password «as you defined»

In addition, maybe a dumb question but, did you try your SSH access using port 22 ?


what command do you enter to shut it down?


Some says it doesn’t hash upon a given frequency, make it tricky with crazy numbers, like 731 or similar…how I wish I had mine do try it :slight_smile:


I found a link to various archive files, I asked the person or anyone whos used it to verify checksum with sha-256 but have not heard back yet, will post again if I do
This came off of the Zen z-9 discussion threadon discord



contains 18 different presets

I don’t know what the archives should normally contain


So did I just get lucky with my z9 mini and get one without the new firmware? I got on Wednesday so it’s batch 2. My firmware version is Sat May 26 20:42:30 and I have the drop down menu for overclocking. I can overclock to 700M and get 15.4ksol average. Any higher than that and one of the hashing boards turns off


Some of us must have gotten batch 1 units as part of the batch 2 shipments. I purchased mine the day it hit $850 so the firmware change must not have been implemented yet.


All of mine have the updated firmware, but got them all to OC so far.


With the Chrome “fix” or by uploading the old Batch 1 FW?

// You can still download the old FW from Bitmains support section.

/// Download: https://file11.bitmain.com/shop-product/firmware/Antminer-Z9-Mini-201805262047-500M.tar.gz


It seems that those who have gotten batch 2 units with the “old” firmware from batch 1 can’t OC much higher than 650-700 (most people seem to report a limit of 675, myself included). The batch 1 units manages OC at 750 across the board. So I don’t think that batch 2 units with the old firmware perform any better. It’s probably a result of chip binning, and maybe some firmware limitation installed (the last point is pure speculation).


I can oc my batch 2 mini up to 700M. Any higher than that and one of the chains shuts down after 30-60 min. Anybody with similar results? I’m going to have to Google “chip binning” haha


It’s a function of the PLL dividers. And yes, if you give it an invalid frequency it goes to like… 250MHz


Ok…sorry, I did everything without my Z9 on hands. It is ridiculous but I’m still waiting for them…and I’m in Shenzhen!
So all the things I said are pure speculations:grin:


I’ve just done a test @700Mhz, one board crashed. I keep 650.


Chrome fix. If anyone is having problems with that I can help.
For it to work the frequency box must be blank and it has to write to Bitmain-freq. Balance will always go back to 500 and that doesn’t matter. If you reselect Balance it will go back to 500. So if you edit the value and you are still at 500, next edit both Bitmain-freq and the Balance option value to your desired frequency and leave the frequency selector blank. That should work without flashing firmware or downloading cfg files which could be dangerous and more likely to void the warranty.


My guess is that they reduced the voltage. The chips should be the same from what I have read since they have the same frequency. I think it is likely the Z9 won’t OC so they are removing OC’ing on the Z9 mini to sell more Z9’s.


I agree on this and it seems as the most plausible reason.


Yes, it a commercial move. It makes no point to design and manufacture a new asic.


The cfg, .conf file, cannot do no harm. I think I’m happy the browser isn’t able to make permanent change to the HW.


By the way, I’m also waiting for a Z9. As soon as I receive my HW I can male some test. Now…just speculations.