Search content
Sort by

Showing 20 of 42 results by nitrox1
Post
Topic
Board Hardware
Re: ANTMINER S3 Discussion and Support Thread.
by
nitrox1
on 24/12/2014, 18:13:50 UTC
cheap shit is cheap

and shit

and overpriced

and hazardous

glad i refused to partake of bitmains repetitious bait & switch game albeit a bit late

I have to disagree. Out of ten S3 that I bought, one worked slightly below ad rates (~418 Gh), one worked at par (441 GH), one at ~465GH and the rest (seven) at 478GH.
I only applied updates until August, all machines were working nonstop with nary any interruption.
I am a very satisfied S3 customer.

I've got three, on the open market, so probably the worst of the bunch.
1 x 420 2 x 470, October firmware was great, even latest latest firmware, has caused me to no longer be able to run 3 on my one PSU, and one to never get above 430GH/s. Draws more voltage, by 150W over the three.
Post
Topic
Board Hardware
Re: ANTMINER S3 Discussion and Support Thread.
by
nitrox1
on 24/12/2014, 09:08:19 UTC
I have a batch of S3+ I bought, all are running on 1300W power supplies, At first they worked, but now suddenly I can only get 228Gh/s out of them ( I see many post with the same issue), even when i reboot or try 1013 image. I am not too happy about this, what can I do? I guess I have beg you now to use a beta firmware? what to do to fix this. I just found out I missed days of mining because of these miners which are not stable.  

Can I send these piece of crap defective S3+'s back and get a refund for S5? No, because you charged me out my butt for shipping so I can't return them to you...

Every other tab i get some HTML error message.

XML Parsing Error: not well-formed Location: http://192.168.1.118/cgi-bin/luci/xxx/admin/status/miner/ Line Number 348, Column 34:   Modify voltage and Save&Apply, then need to Power off and Restart -------------------------------------------------------------^

Can anyone tell me what to do with these half baked pieces of junk that dont work with bad code, and I tried all the images, still same crap. And dont give me any crap about it's my power supply, I am using EVGA G2 1300W. And it's not just one miner.




How many S3+ on your power supply? you say batch of? you say not just one? what's the watts at the wall?
Post
Topic
Board Hardware
Re: ANTMINER S3 Discussion and Support Thread.
by
nitrox1
on 23/12/2014, 19:48:44 UTC
I have 4 x S3 antminers . I had the 10/24 firmware on them seems like MY GHS was dropping from 450 to 340+ but never going back up to 450 or even close any ideas. When I reboot the GHS jumps back up for about 10-12HRs then back down again. All set at s3+ default 225. I changed the voltage to 0770. I just installed Doges firmware on one of them to try it. Any one has any ideas it would be greatly appreciated. Thanks in advance.
I know I am quoting myself but it looks like 0720 is normal voltage for 10/24 firmware running 225 Freq and 0725 for Dogies 12/19 Firmware at 225 Freq looks right. I will have to let them run for 24hrs and see what happens. Might have to go with the 12/19 firmware if everything keeps running okay because it looks like the 12/19 is running better then the 10/24 . Ill keep everyone posted. Thanks  

My question would be, if the wattage is not changing at the wall, is it doing anything at all???

edit: referencing the voltage.. changing the freq definitely changes the watts at the wall

I just had to edit my post looks like 0720 is right for 225 freq but have to monitor for 24hrs make sure theres no unexplained drops

Seems low to me, but I'm going to try it, myself
Post
Topic
Board Hardware
Re: ANTMINER S3 Discussion and Support Thread.
by
nitrox1
on 23/12/2014, 17:41:51 UTC
pekatete
Any ideas on this one. I know you are extremely knowledgeable.I have read and used a couple of your post. Just trying to get stable at 225.  

I have 4 x S3 antminers . I had the 10/24 firmware on them seems like MY GHS was dropping from 450 to 340+ but never going back up to 450 or even close any ideas. When I reboot the GHS jumps back up for about 10-12HRs then back down again. All set at s3+ default 225. I changed the voltage to 0770. I just installed Doges firmware on one of them to try it. Any one has any ideas it would be greatly appreciated. Thanks in advance.

Thanks

Just my piece, but tuning in voltage, for frequency really helped me. Try at 225 Hz 0760 voltage, and try them again, or anywhere from 0.75v-0.76v in 0.005v increments. You can go out of these ranges but one will be the sweet spot for that frequency.

Seeing you seem to be OK powering these up, I concur with tuning in the voltages ...

pekatete
I will give the voltage tweak a try . Usually takes about 4 hrs to downgrade. Thanks

I seem to find the sweet spot after 15min, then seems to run stable. I'm getting as much as 50GH/s for 0.005v, but there are also sweet spots for each frequency too, which maybe the same voltage for the previous frequency.
Post
Topic
Board Hardware
Re: ANTMINER S3 Discussion and Support Thread.
by
nitrox1
on 23/12/2014, 17:00:19 UTC
Has anyone actually verified that changing the voltage lowers the  watts at the wall?
Yes it has been checked, and NO it does not lower watts at the wall. EDIT though in theory it should......

pekatete
Any ideas on this one. I know you are extremely knowledgeable.I have read and used a couple of your post. Just trying to get stable at 225. 

I have 4 x S3 antminers . I had the 10/24 firmware on them seems like MY GHS was dropping from 450 to 340+ but never going back up to 450 or even close any ideas. When I reboot the GHS jumps back up for about 10-12HRs then back down again. All set at s3+ default 225. I changed the voltage to 0770. I just installed Doges firmware on one of them to try it. Any one has any ideas it would be greatly appreciated. Thanks in advance.


Thanks

Just my piece, but tuning in voltage, for frequency really helped me. Try at 225 Hz 0760 voltage, and try them again, or anywhere from 0.75v-0.76v in 0.005v increments. You can go out of these ranges but one will be the sweet spot for that frequency.
Post
Topic
Board Hardware
Re: ANTMINER S3 Discussion and Support Thread.
by
nitrox1
on 23/12/2014, 13:03:56 UTC
If you have 3 s3's I would use the EVGA 1600 what g2 it has more PCI-express connectors which you will need to run them. on my S3's Each had its own EVGA 1300 watt PSU but I need one of them for me new pc build so you know what happened then.





 

That's good for your efficiency. However the more power I supply to mine, makes no difference, they max out. So the draw from each max's out, so I'm able to run a very good quality PSU at a lesser wattage, though I could do with a little more ;-) Always need more power
Post
Topic
Board Hardware
Re: ANTMINER S3 Discussion and Support Thread.
by
nitrox1
on 23/12/2014, 11:26:18 UTC
http://s4.postimg.org/umfgjln31/Second_day.jpg


This is day 2 of Dogie's Firware So far so Good Even though I had to reboot then 30 times to get the write power draw from the wall. I use A power meter at the wall and every time I reboot them I get a different power rating at startup once I get the rating I am looking for I just leave them alone I don't restart or reset  JUST LEAVE IT BE!!!!


This happened with any version of the firmware But Dogie's is the best one so far as you can see no speed drops yet it stable I am not even going to touch them!!!!!

Its something with the startup surge current EFFECTING THE CLOCK CIRCUIT the Initial surge POWER from the miner is roughly between 1300 watt to 1500 watt then settles down to 790 Watt for two S3's!!!

I think this causes the Clock circuit not to oscillate improperly. meaning it locks at the wrong frequency causing massive problems with them meaning hash rate drops!!! when the initial startup surge current stays low 900 watt it stays very stable works like it should!!! the reason why I Say the wrong Frequency is this. As you go higher in hash rate the frequency goes up the more power draw it draws from the Wall in this case. Now if the base Clock does not lock in at the expected frequency it was designed for nothing will be stable it will run erratically on you example.


Power draw after miner settles after reboot with my situation.

First boot 650 watts from wall unit will drop hash rate "high start out surge current" - unstable miner speed drops

Reboot or cold start again 750 watts after settling from the wall "high start out surge current" -   unstable speed drops again

Reboot or cold start again 790 watts after settling but this time the initial startup surge current was only 900 watts opposed to 1200 to 1500 watts "Miner runs stable as indicated in my PIC ABOVE!!!

sometimes I have to reboot 30 times in a row to get the settled power draw that I am looking for. I am a disabled Engineer I have to get My old lab test equipment out of my attic to test this theory but I cant I down have the schematic diagram for the S3 "in other words circuit blueprints" and I will never get them because BITMAIN and ALL COMPAINES keep them under lock and key they don't want another company or anybody for that matter building there hardware or making it better!!!

I think there Using A VCO circuit and Not A PLL circuit and don't have schematics I don't know what kind of circuit it is!!!



790W for 2 S3's at 250Freq and 1-1.2Ths/s ? What's the volts. I'm at 1050W on 3 S3's for Approx 1.2-1.3 Th/s
Post
Topic
Board Hardware
Re: ANTMINER S3 Discussion and Support Thread.
by
nitrox1
on 23/12/2014, 00:59:35 UTC
http://s10.postimg.org/z9ha3hty1/Neew_Miner_Status.jpg

How is that for stable!!!! This is the beta Firmware that was posted here. Stable for me. Grin Grin Grin Grin Grin


Well as we all know not all chips are created equal.
But have got it sorted out now, but took a while.
I believe new firmware might use a little more juice.
But I'm tuned now:
 0.45Th/s 231.25Hz 0.76v
0.467Th/s 231.25Mz 0.755v
0.465Th/s 237.5Mz 0.765v

Which gives me a very nice 0.76W/Ghs (A little bit higher for my PSU), so I am much happier, and does seem stable. But it took a lots of Frequency, Voltage tuning, and voltage made a real difference +/-0.05v.
Post
Topic
Board Hardware
Re: ANTMINER S3 Discussion and Support Thread.
by
nitrox1
on 22/12/2014, 17:52:05 UTC
You probably did a power cycle on the rig! Do a warm reboot via SSH and that should clear your avg figures ....

Well I've SSH'd in and rebooted. I'll give it an hour. Otherwise back to October firmware, it was working really well for me. Does the December firmware cross platforms with the S3+?

Your avg numbers should be good now ...... Its something I noticed a while back whenever I did a power cycle, but when the system cleanly reboots the numbers are OK. That is a remnant of the initial Oct firmware bugs, and to be honest I'd rather they did not try and fix it as we've seen what they are capable of in terms of fixing minor bugs in their firmware!


Yeah, no solution for me, I got averages of 309, 307, 284Gh/s. My pools has me at  at 0.9-1.2Th/s over three S3's, set to S3+ default frequency 225.
I've gone up to next frequency 231.25 and lifted voltages to 0760 and reboot through GUI after, and I'll give it another try. I was getting 450Gh/s +/- average on all three on Octobers firmware at 231.25 Freq.
Don't know if this matters but I do connect over WiFi.
Post
Topic
Board Hardware
Re: ANTMINER S3 Discussion and Support Thread.
by
nitrox1
on 22/12/2014, 14:05:43 UTC
I'm sorry to say, It looks like I'm getting the same instability, that came from the latest firmware. Hash rates are dropping, and fluctuating, I'm down to 800Gh/s, even on one round it was recorded at 500 GH/s, so either network issue as mentioned above, or the hashing part of the firmware isn't correct.
Can someone advise as to whether it is worth putting in something into voltage section like : "0.8"
I have a solid 1200W running two S3's, which used to run three S3's at 450GH/s each for months until this stupid firmware.

Nothing to do with hashing changed on this version, so its likely something on your end. Else, the feedback looks pretty good and I'll open it up to a wider release.

Has anyone else been giving the @dogie firmware a try for the last coupe of days?Huh
How's your hash rate?
I really think I have to go back to October firmware, and I'm down 200-300Gh/s over three S3's. Exactly the same setup, in the same place, with the same hardware, with the same internet connection (Once your setup you don't really move things). I have older S3's, but I don't know batch.

My 5s average is 450Gh/s which is about right,  but my average  is 250GH/s which is not right.

You probably did a power cycle on the rig! Do a warm reboot via SSH and that should clear your avg figures ....

Well I've SSH'd in and rebooted. I'll give it an hour. Otherwise back to October firmware, it was working really well for me. Does the December firmware cross platforms with the S3+?
Post
Topic
Board Hardware
Re: ANTMINER S3 Discussion and Support Thread.
by
nitrox1
on 22/12/2014, 09:28:52 UTC
Hi
I have a Problem With my Antminer s3
in miner status page i see a 250 GH
and 16 ASIC are work
http://i57.tinypic.com/33os3et.png
PLZ HELP ME

You are missing a chain.  Check power supply.  Connections.  Reboot.  If it still is no good try swapping some cables around to isolate where the problem is.




thank you for your help , i check PSU , change PSU , change cable , reset factory ... nothing change

 

Maybe the cable from the controller board to the hashing board is not on properly. Have you opened it up yet to check internal cables?

no i didn't open the case , ok i'll check it


i opend it up and check internal cables , checked it . nothing found !
one of board's LED is on the other side is off and not work


Have a look at the not working board for dry solder points.


http://upload.wikimedia.org/wikipedia/commons/thumb/2/20/Cold_solder_joint2.jpg/220px-Cold_solder_joint2.jpg http://upload.wikimedia.org/wikipedia/commons/7/72/Gebrochene_loetstellen.jpg
Post
Topic
Board Hardware
Re: ANTMINER S3 Discussion and Support Thread.
by
nitrox1
on 22/12/2014, 09:19:51 UTC
I'm sorry to say, It looks like I'm getting the same instability, that came from the latest firmware. Hash rates are dropping, and fluctuating, I'm down to 800Gh/s, even on one round it was recorded at 500 GH/s, so either network issue as mentioned above, or the hashing part of the firmware isn't correct.
Can someone advise as to whether it is worth putting in something into voltage section like : "0.8"
I have a solid 1200W running two S3's, which used to run three S3's at 450GH/s each for months until this stupid firmware.

Nothing to do with hashing changed on this version, so its likely something on your end. Else, the feedback looks pretty good and I'll open it up to a wider release.

Has anyone else been giving the @dogie firmware a try for the last coupe of days?Huh
How's your hash rate?
I really think I have to go back to October firmware, and I'm down 200-300Gh/s over three S3's. Exactly the same setup, in the same place, with the same hardware, with the same internet connection (Once your setup you don't really move things). I have older S3's, but I don't know batch.

My 5s average is 450Gh/s which is about right,  but my average  is 250GH/s which is not right.
Post
Topic
Board Mining support
Re: bitmain s3 serial connect?
by
nitrox1
on 21/12/2014, 09:07:15 UTC
0.1BTC for Mr. nitrox1
0.05BTC for Mr. pekatete

assuming both are male  Wink

Thank you both very much!

Thank you very much, and yes I am last time I checked.
Post
Topic
Board Mining support
Re: bitmain s3 serial connect?
by
nitrox1
on 21/12/2014, 02:26:49 UTC
Well for the next person like me I think it's only fair I explain how I got here.
I flashed a firmware over WiFi and I ended up with a corrupt bootloader/kernel.
After contacting Bitmain (bitmain wanted a lot of bitcoin for a new board) and reading on Here  (bitcoin.org) it was suggested to start by getting a £3 usb to serial/TTL converter, I have the one I have pictured in previous posts.
I connected my serial connection to the 3 debug pins, and opened putty using a serial connection on the right COM port, and at a baud rate of 115200
After presuming the square debug pin was earth/ GND, I had rubbish output.
I connected my serial converter as shown in previous photos, but from right to left, square to round, round, went TX (Square), Rx (Middle Round) GND (Outer round). I found using GND made a lot of things better.
After starting a putty terminal, and then turning on the flashing red led of death S3, I got a very large output of information. But if I turned it on and pressed "p" it would holt the boot process, and allow me a connection into the S3 "Easy-Link> "
As you can see from previous posts typing help allowed me to make sure that my computer was talking to the S3

I had a stable firmware already in possession, but they can be found here https://www.bitmaintech.com/support.htm?pid=007201407180243004432lBQW28O0633 (mind you this is how I got here to begin with, check for a good firmware, cross reference bitcoin.org S3 support forum)
Unzip the firmware to get a .bin file

I'm using Windows 8.1 64-Bit
So I next had to get a TFTP sever program which I acquired from http://tftpd32.jounin.net/

In windows network center I found out the IP address of the S3 with both a LAN cable connected directly from my S3 to my computer and the serial to USB connected, which in my case happened to be 192.168.1.250.
I had to change this by going in to "network and sharing center" and the click on "ethernet", then under "internet protocol version 4 (TCP/IPv4)" select properties, and use the static IP address of 192.168.1.100 subnet 255.255.255.0

While putty is open, and the S3 is on, and the Easy-Link> prompt is still there
Type ping 192.168.1.100 and check the output says that the IP address is alive

After that start TFTPD64 or 32
Set current diectory to where your firmware is
Set the server interface to 192.168.1.100 (If you can't do this, you need to go back to network and sharing center part)
In the settings menu you need to set TFTP menu setting to "Bind TFTP to this address 192.168.1.100"
THEN TURN OFF YOUR FIREWALL EVEN WINDOWS FIREWALL, ESPECIALLY WINDOWS FIREWALL, but you don't need internet if your worried

Then fingers crossed type in to putty after the "Easy-Link>" prompt

Easy-Link>tftpboot 0x80000000 antMiner_S320xxxxxx.bin (your firmware name)

Easy-Link>erase 0x9f020000 +0x7c0000

Easy-Link>cp.b 0x80000000 0x9f020000 0x7c0000

Easy-Link>bootm 0x9f020000

You will get a lot of output from putty, and after the last command, the S3 should actually boot up, and then allow you to have web GUI access through the default IP address 192.168.1.99

I hope this helps someone as much as others have helped me. Thanks
Ty Excuse bad spelling, grammar, and typos
Post
Topic
Board Mining support
Re: bitmain s3 serial connect?
by
nitrox1
on 21/12/2014, 01:27:54 UTC

OK having problems with TFTP server part.


That is all you need to sort out now and you are home and dry!
I gave you a link to setting up a tftp server; go through it again and ensure you have (to the last letter) done all the steps, including but not limited to completely disconnectiong from the network and ONLY connecting the board to you PC / laptop ethernet. Do not try to cut corners as networking can be finicky at the very best.


I don't believe it!!!! number 5 is really alive.

It was windows firewall!!!!!! It was you mentioning, to re read your post!

Post
Topic
Board Mining support
Re: bitmain s3 serial connect?
by
nitrox1
on 21/12/2014, 00:59:24 UTC
Easy-Link>

That is the uboot prompt, so simply type (after saving the antMiner_S320141013.bin file in the tftp server root dir as directed by bitsolutions):

Easy-Link>tftpboot 0x80000000 antMiner_S320141013.bin

Easy-Link>erase 0x9f020000 +0x7c0000

Easy-Link>cp.b 0x80000000 0x9f020000 0x7c0000

Easy-Link>bootm 0x9f020000



OK having problems with TFTP server part.

I've got ping ok:

Easy-Link> ping 192.168.1.100
eth1 link down
Using eth0 device
host 192.168.1.100 is alive
Easy-Link>

I'm trying to use TFTP2.exe and TFTPd64 but not having much luck
Post
Topic
Board Mining support
Re: bitmain s3 serial connect?
by
nitrox1
on 21/12/2014, 00:19:01 UTC
can't seem to setup any other server than over .1.250, even though everything is .1.100
Post
Topic
Board Mining support
Re: bitmain s3 serial connect?
by
nitrox1
on 20/12/2014, 23:57:34 UTC
Sure fire fix assuming hardware is ok and uboot intact is uboot based recovery, interrupt the boot process and set your computers IP to a static IP of 192.168.1.100. Start a TFTP server on your computer serving antMiner_S320141013.bin and run the following commands on your miner.

Code:
tftpboot 0x80000000 antMiner_S320141013.bin
erase 0x9f020000 +0x7c0000
cp.b 0x80000000 0x9f020000 0x7c0000
bootm 0x9f020000

Just put the firmware bin file into the root directory on a tftp server running on 192.168.1.100 subnet mask 255.255.255.0 and execute the above commands from uboot to flash it.

OK, that is new.? Pardon, ex-squeeze me, what? don't understand, sorry.
I have a windows computer, and I have TFTP2.exe, can't get TFTP2 to find S3
Windows says IP is .1.250, which the S3 says is alive.
TFTP2.exe, only does something on .1.100 or .1.1 which while typing TFTPboot, is what the S3 says it's addresses are.
Sorry for my ignorance.
Post
Topic
Board Mining support
Re: bitmain s3 serial connect?
by
nitrox1
on 20/12/2014, 23:51:18 UTC
Thanks for the output, I'll browse through and see if I can spot a light at the end of the tunnell.

By the way, I meant the command altogether, i.e:

run mtd -r erase rootfs_data



yes did a long time ago:

Easy-Link> run mtd -r erase rootfs_data
## Error: "mtd" not defined
Easy-Link>

Edit: Really appreciate your help. or Elf-lp ;-)
Post
Topic
Board Mining support
Re: bitmain s3 serial connect?
by
nitrox1
on 20/12/2014, 23:30:57 UTC
Yes, I did , won't except commands, run mtd, mtd, or erase rootfs_data
How do they fail? Is it verbose and if so, could you paste here?
Also, run a printenv command and report back.

Printenv:
Easy-Link> printenv
bootargs=console=ttyS0,115200 root=31:02 rootfstype=squashfs init=/sbin/init mtdparts=ar7240-nor0:128k(u-boot),1024k(kernel),6912(rootfs),64k(config),64k(ART)
bootcmd=bootm 0x9f020000
bootdelay=1
baudrate=115200
ethaddr=0x00:0xaa:0xbb:0xcc:0xdd:0xee
ipaddr=192.168.1.1
serverip=192.168.1.100
stdin=serial
stdout=serial
stderr=serial
ethact=eth0

Environment size: 345/65532 bytes

edit:
list of tried commands:

Easy-Link> run
Usage:
run     - run commands in an environment variable

Easy-Link> run mtd
## Error: "mtd" not defined
Easy-Link> erase rootfs_data
Usage:
erase   - erase FLASH memory

Easy-Link> mtd
Unknown command 'mtd' - try 'help'
Easy-Link> run erase rootfs_data
## Error: "erase" not defined
Easy-Link> help
fake   - fake
?       - alias for 'help'
boot    - boot default, i.e., run 'bootcmd'
bootd   - boot default, i.e., run 'bootcmd'
bootm   - boot application image from memory
cp      - memory copy
erase   - erase FLASH memory
help    - print online help
httpboot- entering the backup mode.
md      - memory display
mm      - memory modify (auto-incrementing)
mtest   - simple RAM test
mw      - memory write (fill)
nm      - memory modify (constant address)
ping    - send ICMP ECHO_REQUEST to network host
printenv- print environment variables
progmac - Set ethernet MAC addresses
reset   - Perform RESET of the CPU
run     - run commands in an environment variable
setenv  - set environment variables
tftpboot- boot image via network using TFTP protocol
version - print monitor version
Easy-Link> httpboot
eth1 link down
eth1 link down
uip init...
http init...
eth1 link down
eth1 link down
eth1 link down


edit:
Any way to use "mw" command, memory write command, then "bootm" command to boot from memory, to flash through GUI?