Pages:
Author

Topic: Custom Innosilicon A2 Terminator image - Anx Edition - page 5. (Read 22709 times)

sr. member
Activity: 345
Merit: 250
Hi,

I have a query that I would like to check before I do anything. On my stats list of the A2 it has the ID as 0 to 4. Am I correct in assuming it should go up to 5 and one of the boards is either disconnected somehow or dead? Or is there something else I have missed.
legendary
Activity: 1596
Merit: 1000
My A2s all have ~19/17W fans (12V/1.6A or even 12V/1.4A). May I ask what kind of fans you are using now?

Honestly, I put some slower fans in there since the stock Nidec BETAV TA450DC seemed to be pretty overkill.  I replaced the 3 front fans with Yate Loon D12SH-12 fans (~2200rpm) and added two 140mm fans to the back (Yate Loon D14SM-12).  My temps seem to be pretty close to what they were with the stock fans (maybe 2-3 degree difference at their hottest point), but the A2 is a lot quieter and uses about 50 watts less at the wall.

I'm sure if you had a bunch of them in a data center it wouldn't really matter and you'd want the extra airflow with the stock fans, but in a residential setting with just a few units these replacement fans seem to be pretty suitable.

Yeah, if I were running them in my house instead of my garage I might care about the noise. My A2s are at a disadvantage, though. My 60MH only has 2 fans pushing air through in the back, and the 88mh only has 3 fans in back pushing air through.
member
Activity: 63
Merit: 10
My A2s all have ~19/17W fans (12V/1.6A or even 12V/1.4A). May I ask what kind of fans you are using now?

Honestly, I put some slower fans in there since the stock Nidec BETAV TA450DC seemed to be pretty overkill.  I replaced the 3 front fans with Yate Loon D12SH-12 fans (~2200rpm) and added two 140mm fans to the back (Yate Loon D14SM-12).  My temps seem to be pretty close to what they were with the stock fans (maybe 2-3 degree difference at their hottest point), but the A2 is a lot quieter and uses about 50 watts less at the wall.

I'm sure if you had a bunch of them in a data center it wouldn't really matter and you'd want the extra airflow with the stock fans, but in a residential setting with just a few units these replacement fans seem to be pretty suitable.
legendary
Activity: 1596
Merit: 1000
What is the highest clock speed available with the ANX firmware? Can someone share some screenshots of it?

The highest clock rate on mine is stock (1200) - I don't support any overclocking, only underclocking, as I don't have the source to cgminer so can't modify it in the same way Emdje does on his.  So for right now you've got a choice between functionality and working on all (my latest A2's didn't work with his firmware) terminators on mine, and overclocking on most terminators on his...


I see, thanks. Being able to run at 1300MHz is the most important factor for me, unfortunately. I'll keep an eye on your thread with my fingers crossed!
hero member
Activity: 687
Merit: 511
What is the highest clock speed available with the ANX firmware? Can someone share some screenshots of it?

The highest clock rate on mine is stock (1200) - I don't support any overclocking, only underclocking, as I don't have the source to cgminer so can't modify it in the same way Emdje does on his.  So for right now you've got a choice between functionality and working on all (my latest A2's didn't work with his firmware) terminators on mine, and overclocking on most terminators on his...
legendary
Activity: 1596
Merit: 1000
Mark,

Your firmware doesn't support overclocking, correct? Any chance you'd consider adding it into your next version? I want to try your firmware, but I run my A2s at 1300MHz since the W/MH scale very well and my power is cheap enough.
What kind of performance/power usage are you seeing at 1300MHz?  I'm seeing about 9.58W per MH on one of my units, and about 9.14W per MH on another unit.  I don't have stock fans in my A2s, so my numbers may be a little different than others (the stock A2 fans are nearly 25 watts a piece, so I put in some more efficient fans).

My A2s all have ~19/17W fans (12V/1.6A or even 12V/1.4A). May I ask what kind of fans you are using now?


My 60mh was one of the first built, and had 2 really sub-par 25mm fans. I replaced those with Delta 38mm fans, which are 1.6A each going from memory. The 90mh unit came with 3 high quality 38mm 1.4a nidec fans, and those are still running strong after a year and a half or so.

What is the highest clock speed available with the ANX firmware? Can someone share some screenshots of it?
legendary
Activity: 1612
Merit: 1608
精神分析的爸
Mark,

Your firmware doesn't support overclocking, correct? Any chance you'd consider adding it into your next version? I want to try your firmware, but I run my A2s at 1300MHz since the W/MH scale very well and my power is cheap enough.
What kind of performance/power usage are you seeing at 1300MHz?  I'm seeing about 9.58W per MH on one of my units, and about 9.14W per MH on another unit.  I don't have stock fans in my A2s, so my numbers may be a little different than others (the stock A2 fans are nearly 25 watts a piece, so I put in some more efficient fans).

My A2s all have ~19/17W fans (12V/1.6A or even 12V/1.4A). May I ask what kind of fans you are using now?
legendary
Activity: 1596
Merit: 1000
Mark,

Your firmware doesn't support overclocking, correct? Any chance you'd consider adding it into your next version? I want to try your firmware, but I run my A2s at 1300MHz since the W/MH scale very well and my power is cheap enough.
What kind of performance/power usage are you seeing at 1300MHz?  I'm seeing about 9.58W per MH on one of my units, and about 9.14W per MH on another unit.  I don't have stock fans in my A2s, so my numbers may be a little different than others (the stock A2 fans are nearly 25 watts a piece, so I put in some more efficient fans).

I've got a "60MH" and a "90MH" from their initial release way back when they cost $10000 a piece in April of 2014. At stock, they actually ran closer to 52MH/s and 86MH/s. At 1300MHz I get about 60MH/s @ 600w and 90MH/s at 900w at the wall running on 240v with Corsair AX860i PSUs in each. About 10w/MH all in.
member
Activity: 63
Merit: 10
Mark,

Your firmware doesn't support overclocking, correct? Any chance you'd consider adding it into your next version? I want to try your firmware, but I run my A2s at 1300MHz since the W/MH scale very well and my power is cheap enough.
What kind of performance/power usage are you seeing at 1300MHz?  I'm seeing about 9.58W per MH on one of my units, and about 9.14W per MH on another unit.  I don't have stock fans in my A2s, so my numbers may be a little different than others (the stock A2 fans are nearly 25 watts a piece, so I put in some more efficient fans).
legendary
Activity: 1596
Merit: 1000
Mark,

Your firmware doesn't support overclocking, correct? Any chance you'd consider adding it into your next version? I want to try your firmware, but I run my A2s at 1300MHz since the W/MH scale very well and my power is cheap enough.
hero member
Activity: 687
Merit: 511
Interesting - are the lights solid or flashing?  When they are connected and working properly they should be solid green (when you boot it starts as flashing green until I think CGMiner connects to them).  I have had blades 'die' on me, and they continue to flash while all the others are solid.  If I remember correctly, I think the blades still show up in the live stats, but they just never hash.

My experience with zoomhash was similar to madmartyk, if you run into issues they're usually very good about sorting them out - one of the advantages of buying from them.
legendary
Activity: 2702
Merit: 1030
Yes I am a pirate, 300 years too late!
I really don't know what to tell you.  I will say the support at ZoomHash is excellent.  When I ordered my unit I ordered a 110, but received an 88.  After doing everything they wanted it was the opinion that I in deed get an 88.  Well they made me whole.  There was price difference and they took care of it.
newbie
Activity: 4
Merit: 0
Thanks for the suggestions.

I've tried all 3 images and all have the same result.

I know it is connecting to the internet as I can see an additional worker on Litecoinpool.org when it is running - sadly the worker just sits idle.

I'm reluctant to open the box until I talk with Zoomhash (there is a seal on the case), and if an RMA is warranted, the extra hassles aren't worth it.

The boards doo light up green, things look healthy, but who knows what might have shaken loose inside during shipping?

The ANX firmware certainly gives the most info, all the blades report healthy, the status shows the pool as "alive"  --- it seems so close, but as of yet, no cigar.

Thanks for the feedback.

Robert
legendary
Activity: 1612
Merit: 1608
精神分析的爸
Not really related to the ANX firmware, but make sure your miner has a default gateway set and can reach the internet.
HTH
legendary
Activity: 2702
Merit: 1030
Yes I am a pirate, 300 years too late!
Have you downloaded and try using the ANX Edition firmware?  I got my 88 from ZoomHash and that what was on it.  Are all the boards lighting up?  Have you opened up the box and made sure all the cables are seated?
newbie
Activity: 4
Merit: 0
I'm back to the stock image and seeing the following stats.

https://dl.dropboxusercontent.com/u/3191371/stats1.jpg

My config is https://dl.dropboxusercontent.com/u/3191371/config.jpg

Another capture is https://dl.dropboxusercontent.com/u/3191371/summary2.jpg

Ghash isn't showing my worker.

As said in my previous post, I'm willing to pay someone if they can help me get this new miner working.   

I'm starting to suspect that I have a bad box, but I don't know what might be wrong? Huh

Thanks in advance to anyone who might be able to help.

Robert
newbie
Activity: 4
Merit: 0
Hi

I just received my first A2 last night, and have been trying to get it going without much luck.  I've tried the stock Zoomhash firmware, and yours - which is pretty amazing - but I just can't seem to get the box to do any mining.

https://dl.dropboxusercontent.com/u/3191371/status1.jpg

I've tried different pools, but the most success I have has is with Litecoinpool.org - it at least shows a worker, albeit that worker is doing nothing.

If anyone could help me out, I would be happy to share some coin.

Thanks

Robert
legendary
Activity: 1612
Merit: 1608
精神分析的爸
As promised here is a howto explaining the growing and shrinking of SD card images and filesystems.

Growing is useful if you want to grow the filesystem from your image to take advantage of all space available on your SD card.
Shrinking is useful if you want to write i.e. an 8GB image to a 2GB SD card (given the content of the image is less than 2GB).

You will need a Linux computer with gparted installed. Many Live-CDs for Linux and BSDs that include gparted can be used so you can use these steps without devoting a computer to Linux permanently (though you really should Wink).

We will first deal with growing an image to your SD cards full size, since that is slightly easier to do (nvm, shrinking is not that hard either).

Growing an image on the SD card
First write the image you have downloaded from the OP to your 8GB SD card with something like

Code:
sudo dd if=~/Desktop/AnxA2-09012015.img of=/dev/sdX bs=512

Make sure that you use the correct device for sdX in the above /dev/sdX!
The below two commands help you to identify the device name of your SD card:

If you just have plugged it in, check the output of the below command (the last lines to find out what device name your SD card has)
Code:
dmesg

[1883229.800218] scsi119 : usb-storage 8-2:1.0
[1883230.804097] scsi 119:0:0:0: Direct-Access     USB2.0   CARD-READER      1.01 PQ: 0 ANSI: 2
[1883230.804824] sd 119:0:0:0: Attached scsi generic sg2 type 0
[1883230.814085] sd 119:0:0:0: [sdb] Attached SCSI removable disk
[1883230.831082] sdb: detected capacity change from 7948206080 to 0
[1883230.844116] sd 119:0:0:0: [sdb] 15523840 512-byte logical blocks: (7.94 GB/7.40 GiB)
[1883230.850080] sd 119:0:0:0: [sdb] No Caching mode page found
[1883230.850084] sd 119:0:0:0: [sdb] Assuming drive cache: write through
[1883230.866076] sd 119:0:0:0: [sdb] No Caching mode page found
[1883230.866080] sd 119:0:0:0: [sdb] Assuming drive cache: write through
[1883230.880111]  sdb: sdb1 sdb2


The other command below shows your partition tables, look for one that looks like the size of your SD card:
Code:
cat /proc/partitions

On my PC with one 80GB harddisk (sda) the output looks like:
major minor  #blocks  name

   8        0   78150744 sda
   8        1     498688 sda1
   8        2          1 sda2
   8        5   39061504 sda5
   8        6    3905536 sda6
  11        0    1048575 sr0
   8       16    7761920 sdb
   8       17      57344 sdb1
   8       18    2831360 sdb2

Obviosuly, /dev/sdb is the SD card with 8GB capacity.

Once the image is written to your SD card, run the following command (again replacing /dev/sdX with your device name):
Code:
sudo gparted /dev/sdX

A window like the following will appear:


Make sure that all partitions are unmounted by right-clicking them and selecting "unmount". You will not be able to resize the partition and filesystem while it is mounted:


Now you are ready to resize. Right-click on the partition /dev/sdX2 (sdb2 in my case) and select Resize/Move:


A new window will appear that allows you to resize:


Grab the right edge with your mouse and drag it completely to the right:


You can also enter values directly below instead of dragging with your mouse.
Once you are done, press Resize/Move.

A warning will eventually appear that you can safely ignore if you did everything right until here.

Now select Edit|Apply All Operations and give it some minutes to do its job:


Once this has finished, your image is resized to the full SD card size and you can start to use it in your miner.

Shrinking an image file to fit on a smaller card
Now this gets a bit more work as we need to do all work on the image file as we obviously can't just dd it to a smaller card and work on the card directly. Still it is very easy to do so:

First of all we need to mount the image to work on its filesystem. Linux offers the loop device system for such tasks. To create a device from the ANX image do as follows:
Code:
sudo losetup /dev/loop0 ~/Desktop/AnxA2-09012015.img

Now you can again run gparted but against the loop device:
Code:
sudo gparted /dev/loop0

As above in growing the filesystem you can now shrink the second partition to your needs. In our case I want to make the image to fit on a 2GB SD card. Since the boot partition needs some space and we do not want to stretch our luck too much, we make the partition ~1700MB:

Note:Actually one could make some maths to get the right matching number for partition size, but that's beyond the scope of this howto and in this case I don't mind wasting a few MB. In fact the image, after written to SD card can simply be grown to the full 2GB size of the target card with above growing procedure.

Again apply these changes by selecting "Edit|Apply All Operations" and get yourself a coffe (better two).

Once things are done from gparted we are left with an image that would fit on 2GB SD card but has ~6GB unused space at the end and actually is still an 8GB image.
So to be able to dd it to our 2GB sd card we need to truncate the image to the space currently used.
First we need to find out where out second partition we just resized now ends:
Code:
sudo fdisk -l /dev/loop0

Which will give you an output like below:
Disk AnxA2-09012015.img: 7948 MB, 7948206080 bytes
255 heads, 63 sectors/track, 966 cylinders, total 15523840 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x0002c262

             Device Boot      Start         End      Blocks   Id  System
AnxA2-09012015.img1            8192      122879       57344    c  W95 FAT32 (LBA)
AnxA2-09012015.img2          122880     3809279     1843200   83  Linux


The important number is the last block of the last partition: 3809279

From the above output we also know that the block size is 512 bytes. That leaves us with the following formula to determine where we will be truncating the image:

(3809279+1) * 512 = 1950351360

Note: the +1 is because blocks count from 0 so we adjust for that with one block added.

Now can trim our image to the size we calculated above:
Quote
truncate --size=1950351360 ~/Desktop/AnxA2-09012015.img

You should now be able to dd the above image to a 2GB SD card disk.

I hope this helps one or the other when fiddling with images. If you have feedback or corrections please let me know.
hero member
Activity: 851
Merit: 1000
Do You Even Onion Bro?
Thanks for the information  where is the link to your sd card download for the cgminer i want try your version for the a2 terminator sounds like it works better then the other one i been using

The link is at the very first post in this thread, you can get the current version as well as older version - when I release a new one, I update that first post so you only have to look one place to get it.

I will say that Emdje's firmware has more flexibility in terms of what you can do with the device - mine has more of the convenience things, that make it a bit easier to manage a bunch of them.  Either way the experience is much nicer than the stock one.  Wink

i have one question why is it when i look at the mgh the first one on the page saying 96.mgh5s the second one saying 96.mgh average and then the pool one saying76mghs is this normal or is the program that im using

There usually is a difference between what you see locally and what the pool sees - and of course what the pool sees is all that really matters.  The difference typically comes from discards and rejects, which knock your rate down on the pool side, so you'll probably want to look into that first.  The speed you're getting is pretty strange, 96MH would be low for a 110, and high for an 88... If I had to guess, and it would be a big guess, it's that you're running a heavily overclocked 88 - the higher you overclock, the more hardware errors (and in turn rejects) you get, so you can crank up you apparent to higher speed, but your actual speed is lower than normal.  You might try dropping down to 88 and see what happens - if your pool rate increases to be much closer, then that's what's happening.  Overclocking is tricky, it's all a balance of managing heat/power/errors...

Thank you for the information the speed of the miner has always made me wonder im going to try your version see what happen to  the miner  i will update to let you know how it turn out THANKS
hero member
Activity: 687
Merit: 511
Thanks for the information  where is the link to your sd card download for the cgminer i want try your version for the a2 terminator sounds like it works better then the other one i been using

The link is at the very first post in this thread, you can get the current version as well as older version - when I release a new one, I update that first post so you only have to look one place to get it.

I will say that Emdje's firmware has more flexibility in terms of what you can do with the device - mine has more of the convenience things, that make it a bit easier to manage a bunch of them.  Either way the experience is much nicer than the stock one.  Wink

i have one question why is it when i look at the mgh the first one on the page saying 96.mgh5s the second one saying 96.mgh average and then the pool one saying76mghs is this normal or is the program that im using

There usually is a difference between what you see locally and what the pool sees - and of course what the pool sees is all that really matters.  The difference typically comes from discards and rejects, which knock your rate down on the pool side, so you'll probably want to look into that first.  The speed you're getting is pretty strange, 96MH would be low for a 110, and high for an 88... If I had to guess, and it would be a big guess, it's that you're running a heavily overclocked 88 - the higher you overclock, the more hardware errors (and in turn rejects) you get, so you can crank up you apparent to higher speed, but your actual speed is lower than normal.  You might try dropping down to 88 and see what happens - if your pool rate increases to be much closer, then that's what's happening.  Overclocking is tricky, it's all a balance of managing heat/power/errors...
Pages:
Jump to: