 Show Posts
|
|
Pages: [1]
|
I think you will be able to have two chains, one for the 721's and one for the 741's if they wont all work on the same chain. I do have 3 separate chains on my pi now, all with 721's. I'll have the answer on Wed. jfb With the newest software, based on the code, you should be able to run A4, A6, A721 and A741 with the same raspberry pi.
on the same daisy chain at the same time? pi >>>>>>> 721 >>>>>>>721 >>>>>>> 741 would be nice to see them say this. I would be annoyed further to find I can not do the chain above. Frankly I guess I am just in a crappy mood today so rather then post I will sign off here. |
|
|
|
Those of you that have updated s9 to bmv2. Can you say whether or not hw errors are down? My 550 batch shows 10's of thousands of errors. I ran an s4 for 6 months straight with less than thousand errors. These miners really are a piece of work  BR I updated 3 S9's--a B2, B5 and B9. They are now (after 48 hrs) showing .0039%, .0006%, and .009%. Comparing number of HW errors is pretty meaningless because it depends on how long the miner has been hashing--to say nothing about how the number of errors in a given time depends on the hash rate--so comparison of number of errors between an S9 and an S4 ought to be adjusted to reflect the different hash rates, even when run for equal times. The error percentages I am seeing are quite good I think and comparable to those I got with S3's, S5's, S7's and SP20's. There was a slight improvement with the upgrade to v2.0 but what I notice most is less fan speed hunting, more consistent chip temps (around 90 degrees F) which are well within spec for the chips. Yes, these miners will get failed boards--one of my S9 B2 boards was just sent for repair (the failure occurred before the upgrade). That said, I really like my A721's as well. JFB |
|
|
|
I just upgraded my S9 from bmminer 1.0 to cgminer 4.9.2
My error rate is MUCH lower. One downside is now I can not see chip temperatures.
How did you do this? Do you know if the fan speed is being regulated by chip temps? If not, you may have created a setup for board failure. JFB |
|
|
|
I have my 2nd A7 online now and I've got to say, I really like the controller interface. One of my Controllers still has not arrived yet but I really like having the 2 A7's in daisy-chain also. I like the way the CGMiner Status shows each A7 separately in the Avalon Devices Status at the bottom. And I like that the two together are humming along at a 13.5TH/s clip here on the Kano pool!  Can you elaborate on whether you expect to have improved performance with each A7 on a separate controller? Obviously it would provide some relief if a controller would go out, but other than that would it improve performance? I have two A7's on ckpool with one controller and am getting well over 12th in 18 degree C ambient room after about 18h. jfb I don't think you're going to see any difference in hash with a dedicated controller, but the one difference that might be significant is that, if you have a specific machine that needs a reboot, with one daisy-chained controller you have to reboot all the machines on that controller, and not just one. At least I've not been able to figure out how to do it otherwise with a couple of A6s on one controller. OTOH, this A7 has not needed a boot...and the A6s are stable as well. I moved all mine to separate controllers just on GP. I have never used A6s but for the A7s there is a separate Reboot button for each A7 on the CGMiner Status page at the very bottom...   Aha...you're right...I only have one A7, so did not see multiple "Reboot" buttons. That would solve that "issue," methinks. Different from the A6 protocol. EDIT...and another "attaboy!" for the coder(s)...nice improvement. Thank you all for your input. I,m going to stick with one controller for now. jfb |
|
|
|
I have my 2nd A7 online now and I've got to say, I really like the controller interface. One of my Controllers still has not arrived yet but I really like having the 2 A7's in daisy-chain also. I like the way the CGMiner Status shows each A7 separately in the Avalon Devices Status at the bottom. And I like that the two together are humming along at a 13.5TH/s clip here on the Kano pool! Can you elaborate on whether you expect to have improved performance with each A7 on a separate controller? Obviously it would provide some relief if a controller would go out, but other than that would it improve performance? I have two A7's on ckpool with one controller and am getting well over 12th in 18 degree C ambient room after about 18h. jfb |
|
|
|
|
Hello, Here are my issues and how they were resolved. My bank does do international wire transfers for a $30 dollar fee (in a business account). My first two attempts failed because the bank code had extra X's at the end--took a while to figure that out, but eventually did. Then it failed twice because I did not have an adequate street address for the company. Canaan promptly provided this when I asked. Success--bank said it transferred $2018, but Canaan only received $2008--an additional fee was charged somewhere--my guess it was sent through an intermediate bank (Citi NY?) and Canaan got the fee charge. Fortunately Canaan waived the fee. Then I got a letter from Fedex about 9AM about the shipment--not enough info to classify, what are they? Where were they made? I replied the A721's are specialized computers and the control unit is just a computer like the Rasberry Pi and they were all made in China. I also sent links to the A721 and controller pages from the site. In addition, I said these are similar in function to the Bitmaintech Antminer s9. Released from customs later in the day I got the customs email. They are to arrive Monday the 5th, which is 5 days after the wire transfer went through. I still don't know how they were classified and how much duty I will get billed for, or how much of a fee Fedex will charge for all this. My order was for two A721 and one controller.
jb
|
|
|
|
|
