I have been active in Bitcoin mining section for about 2 years, seeing someone like -ck who is the core developer of Cgminer which 90% of mining gears use today with below-average merit score of about 300 merit, and seeing great developers such as but not limited to thierry4wd, taserz and braiins with "member" rank or close, gives a clear sign that the mining boards lack merit.
If you check the support child board alone, you will easily spot a dozen of posts that deserve merit and got none, mining is the "backbone" of Bitcoin, and the mining board is highly undervalued (I could be
a little biased for obvious reasons
). Bitcoin mining boards combined got
1.79% of the total forum merit, while the subsection of altcoins announcements alone got
6.07% , not to mention about other boards that add just a little value to the forum and bitcoin and have a much higher rate than Bitcoin Mining.
If my application is not approved (which is understandable), the least that should be done IMO is increasing the
small merit allocation for the (two ?) merit sources who are active in the mining section.
Below are 10 posts that either received no merit, or that I think deserve more merit.
I have removed some parts because the content was too much, for further evaluation please visit the links and check the whole post.
ONEThis miner is a clone version of Cheetah a1 miner and it seems their old website aisenminer.top is suspended. I can't even find the firmware on Google for this miner.
Anyway, can you try some steps below and maybe it can solve your issue.
- Find the reset button near the SD card slot then press the button for 15 seconds then release. Test again after.
- If it doesn't work try to disassemble(I don't know exactly the miner what it looks like) and look for ribbon cables connected from the control board to hashboard. Remove them all then reattach them and also the power supply on 3 hashboard after that reattach the ribbon cable and power supply to all hashboard and control board. After that test it again.
- If the issue still persists try to check the hashboard one by one remove the ribbon cable and power supply to two hashboard then test it again. Check the kernel logs for changes and if it's running fine without the other two. If it's running fine then do the same on the other two hashboard to test which one is broken.
- Since you have many working miners why not try to take the hashboard from working miner and use it to test it on the broken miner. So that we can test if the issue is on the hashboard/program or if the 3 hashboard is broken. Also, check the power supply and maybe the power is not giving enough power that is why it's not running properly.
Take note: If these miners are still under warranty much better don't disassemble these broken units instead contact the seller if where you bought this and tell them about your issue.
TWOEdit: I've removed some of the irrelevant banter.
My KDB pool code reports
ANY work generated that is ever empty, of the work generated once every 30 seconds over the years.
Over the life of my pool, that would amount to less than you can count on one hand.
This, of course, is the same work that is available to EVERY pool at the same time, not just my pool.
Generating work:
Most pools, I imagine work the same as here - we do a call to bitcoind to get a template that already includes all the transactions we will mine.
Some pools may manipulate the transactions in their pool code - but we don't.
It is possible for us to mine an empty block under very rare circumstances out of the pool's control.
But that has never happened here for the 2429 blocks we have found.
Those circumstances are: a pool generates work that contains all the known transactions on the network.
That pool (here or elsewhere) then finds a block. During that short time (30s or less) no other transactions are seen on the network.
Thus after submitting the block found, the next call to get a work template from bitcoind will have no new transactions.
I log whenever the work is empty - I think it's happened less times than you can count on one hand in 6 years
Thus during those very few rare events someone would have to have found a block to produce an empty block out of their control.
So yeah maybe less than half a dozen possible empty blocks in 6 years for all pools on the network - the rest are them doing it on purpose.
I make my bitcoin debug.log report the transaction information for each work it generates, and I can also see in the bitcoin debug.log what is available in the first work after any block change (VERY relevant example further below)
----
"Mandatory" empty blocks have to do with the time between the last work generation sent to the miner that found the block, before finding a block, and the next work generation once your pool found a block.
The transactions that appear on the network during that time (or unused by the block, still in the mempool) decide it ...
It doesn't matter how big or how small my pool is, my pool is part of the bitcoin network and sees the same transactions as every other well connected pool on the network - though mine is most likely the most well connected of all pools
When a block change occurs, my pool will ALWAYS see every transaction that was in the block, that's how bitcoin works.
It will also see what transactions were left over and available to use mining for the next block, in the new work it will send out to the pool's miners.
This is the time gap/number of available transactions that decides if pools MUST generate an empty block with the new work they send to their miners.
This number is the "maybe" less than half a dozen in the past many years.
What the other pools do, to purposefully produce empty blocks, is to not use a template with transactions generated by bitcoind, but simply generate work based on the hash of the block just found on the network, and nothing more - they cannot include transactions in the work when they do this, since they don't already know which network transactions have already been included in the block they don't want to get the transaction details of.
i.e. they aren't verifying the full block before producing a new block work template for their miners.
This was the cause of the failed/incorrect fork by Bitmain and F2Pool in 2015 - they lost all their 5 or 6 blocks on that bad fork since they built new blocks, on a fork of the network based on an invalid block, produced by another pool that ignored the network rule changes.
You could probably force most of the current large pools onto an invalid fork by mining a block with an invalid transaction in it
---
It takes less than 200ms from receiving a block in bitcoind, until it can produce a full new block template with transactions.
I have a message in debug.log when the block arrives with a timestamp to microseconds, another when it has been processed, and another when the pool has been sent the new work.
The
largest block in the last 8 hours was
631950 it was
1783877 bytes.
bitcoind debug.log shows (as I mentioned)
2020-05-27 19:07:53.985131 ProcessNewBlock
2020-05-27 19:07:54.025189 Pre-allocating up to position 0x3000000 in blk02094.dat
2020-05-27 19:07:54.107985 UpdateTip: new best=000000000000000000067af76e3e524beabb9557f71413d8db9b88760e445d3b height=631950 version=0x20002000 log2_work=91.982404 tx=533594598 date='2020-05-27 19:07:33' progress=1.000000 cache=36.9MiB(236180txo) warning='75 of last 100 blocks have unexpected version'
2020-05-27 19:07:54.108131 Block 000000000000000000067af76e3e524beabb9557f71413d8db9b88760e445d3b provided by 107.191.117.193:8333
2020-05-27 19:07:54.158265 GetBlockTemplate called
2020-05-27 19:07:54.160724 CNB
2020-05-27 19:07:54.174358 CreateNewBlock(): block weight: 3964486 txs: 122 of 10917 fees: 0.04177550 sigops 20166
Also of interest in this case, it had to extend some disk space for the block.
Total time from when it arrived until the pool got new work 19:07:54.174358 - 19:07:53.985131 = 189227 microseconds
189 millisecondsAs for empty blocks, the one I see in the last 8 hours is 631926 by
ViaBTCI'll include the block before as well for more information:
2020-05-27 16:36:48.892330 ProcessNewBlock
2020-05-27 16:36:48.913863 Leaving block file 2093: CBlockFileInfo(blocks=99, size=133620045, heights=631826...631924, time=2020-05-26...2020-05-27)
2020-05-27 16:36:48.914006 Pre-allocating up to position 0x1000000 in blk02094.dat
2020-05-27 16:36:48.950898 Pre-allocating up to position 0x100000 in rev02094.dat
2020-05-27 16:36:49.011947 UpdateTip: new best=00000000000000000010dab51e5208c538fce5634104fbd059da24140911efe7 height=631925 version=0x27ffe000 log2_work=91.981925 tx=533547943 date='2020-05-27 16:36:37' progress=1.000000 cache=14.5MiB(52565txo) warning='72 of last 100 blocks have unexpected version'
2020-05-27 16:36:49.012075 Block 00000000000000000010dab51e5208c538fce5634104fbd059da24140911efe7 provided by 107.191.117.193:8333
2020-05-27 16:36:49.071051 GetBlockTemplate called
2020-05-27 16:36:49.071136 CNB
2020-05-27 16:36:49.088969 CreateNewBlock(): block weight: 3964991 txs: 1933 of 17715 fees: 0.18324281 sigops 13534
2020-05-27 16:37:08.482568 ProcessNewBlock
2020-05-27 16:37:08.490710 UpdateTip: new best=0000000000000000000f1b87afb1b95a5e681736ea387b60a8bd150b1ec8bb30 height=631926 version=0x3fff0000 log2_work=91.981944 tx=533547944 date='2020-05-27 17:06:23' progress=1.000012 cache=14.5MiB(52772txo) warning='73 of last 100 blocks have unexpected version'
2020-05-27 16:37:08.568545 CreateNewBlock(): block weight: 3964339 txs: 1903 of 17821 fees: 0.21419789 sigops 13493
In this case firstly, you can see the block before, it was 20 seconds before it.
Secondly, in this case the work kano.is generated included 1933 transactions and fees: 0.18324281 BTC
However as can be seen from the next block on the block chain, it was only 315 bytes ... instead of our "block weight: 3964991"
(You'll have to check the 315 bytes number on any block explorer to see it's an 'empty' block)
i.e. our work that miners were working on was a full block, but
ViaBTC miners were working on an empty block when they found it.
It took 196639 microseconds to process i.e.
197 milliseconds... and of course it
didn't take
197 milliseconds longer than ViaBTC - they didn't process it in 0ms
You can run a bitcoind yourself and you can see this information - though you'll have to patch it and recompile it to display the extra information
I've been around in bitcoin since 2011 working on code related to it.
Working on mining code, pool code, and even simple patches to bitcoin itself.
---
Now as to why not to mine on these "empty block" pools?
Well as is clear in all discussions about empty blocks, those pools don't fully process the block before sending out new work.
Thus they can fork bitcoin and cause all sorts of problems with exchanges and transactions, if the empty block they generate is based on an invalid block e.g. with an invalid transaction in it.
This occurred before in 2015 by Bitmain and F2Pool where they forked the bitcoin network and kept ahead of the correct fork for 5-6 blocks with a large % of the bitcoin network, until certain people contacted them and told them they were mining on an invalid fork (I contacted Bitmain ...)
THREE
easy to see taller psu and skinny controller case on the t17e the unit on the left.
this is the t17e
this is also the t17e
here they are mining
I will post numbers again.
but t17e seem to be in the 53-55th area.
the s17pro on low is about 43th
I have 2 measurements on power for the t17e 3356 watts and 3120 watts
the 3356 is on a killawatt style meter this reads high
the 3120 is on a pdu the pdu reads low
so 3120/54th = 57.77 watts a th
and 3356/54th = 62.15 watts a th
57-63 watts a th range for a t17e
53-55th hash range
note a one speed unit.
the controller for the s17 and s17 pro is this one.
https://shop.bitmain.com/product/detail?pid=000201908151010180899X61guRp068Cthe psu for the s17 and s17 pro is this one.
https://shop.bitmain.com/product/detail?pid=0002019072316001724716dkNtX50679the label on the t17e psu reads apw9+
the label on the s17 pro psu reads apw9
FOUR
(Review/Guide) AvalonMiner 821 11.0 Th/s, 1200W Bitcoin (SHA-256) ASIC miner
Click for a bigger version. Picture copyright (c) Canaan Creative.
1) General information, other items needed
2) Unpacking
3) Overview
4) Controller unit setup
5) Physical setup for the miner and controller
6) Configuration
7) Testing
8] Power consumption measurement
9) Conclusion
10) How to order Last updated in: February 25, 2018.
Steemit blogging website:https://steemit.com/review/@hagssfin/review-guide-avalonminer-821-11-0-th-s-1200w-bitcoin-sha-256-asic-minerNicely formatted version (February 25, 2018) is available for download here (Mega.nz cloud service):https://mega.nz/#!Zl8zlRSZ!8JKeDllHjtBPUW3ctF3Rap-Dau4TqAVPR6Voy-bS4B0
1) General information, other items needed (top)AvalonMiner 821 is a Bitcoin (SHA-256) ASIC miner using 16nm ASIC chip technology.
It is manufactured by
Canaan Creative (headquarters in Beijing, China).
This miner is a 8th generation AvalonMiner, and it uses a new generation 16nm Bitcoin ASIC chip called
A3210.
This new generation ASIC chip makes this miner more power efficient than the previous, 7th generation AvalonMiners.
AvalonMiner 821 does not have a built-in controller included and you need a separate controller, AvalonMiner Controller or any of the Raspberry Pi versions to run it.
A nice feature in my opinion is that up to max. 20 AvalonMiners can be chained to be controlled by a Raspberry Pi controller unit.
This saves you some money with network infrastructure cost if you have a large mining operation.
AvalonMiner 821 weights 4.7 kg and outer dimensions are 370mm (L) x 150mm (W) x 136mm (H).
The nominal hashrate is 11.0 Th/s with a nominal power consumption of 1200 Watts.
This is a review and a unofficial setup instructions guide for the AvalonMiner 821.
Technical specifications:AvalonMiner 821
Hash rate: 11.0 Th/s, -5% ~ +10%
Power consumption: 1200W, +0% ~ +20% @ 11TH/s, 25°C, 93% PSU Efficiency, 12V AC, Wall-Plug
Power efficiency: 0.109J / Gh/s Wall-Plug
DC voltage input: 12 ~ 12.6 VDC
ASIC chip type: A3210 (16nm SHA-256 ASIC chip)
ASIC chip quantity per unit (miner): 104 chips
Number of hash boards per unit: 4 hash boards
Cooling: 1x 120mm x 38mm 12VDC 2.7A fan; manufacturer & model: Delta QFR1212GHE (mounted in rear for a pull configuration).
Weight: 4.7 kg
Dimensions: 370mm (L) x 150mm (W) x 136mm (H)
- A server grade power supply with at least 8x PCI-E 6-pin connectors,
every connector with a dedicated cable, preferably at least 16AWG wire gauge.
I used a 1600W power supply called Sorcerer Power Supply Unit to power my miner.
This is an official custom power supply manufactured by Canaan Creative. - C13 power cord suitable to your country's electrical socket
- AvalonMiner Controller or Raspberry Pi.
- 5V 2.4A USB wall wart for the Controller.
- Micro USB cable between controller and the 5V 2.4A USB wall wart.
- Ethernet cable to go from the controller to your router/switch.
- Computer to set initial pool and configuration settings.
2) Unpacking (top)I received this AvalonMiner 821 as a sample unit for this review.
It doesn't have the info label printed on it, but this is a ready mass production unit for sure.
Packaging was good as always with Canaan hardware.
Sturdy design and lots of foam to protect the miner.
Using DHL Express for the delivery was nice and fast.
The packaging principle pretty much goes along same principles as it was with the previous 7th generation AvalonMiner.
(See
chapter 2 in this A741 review thread:
https://bitcointalk.org/index.php?topic=2588105).
3) Overview (top)In these following pictures you can see the AvalonMiner 821, AUC3 converter and Sorcerer Power Supply Unit.
Click the picture to show a bigger picture. Pictures taken by HagssFIN.
I had pleasant previous experience buying hardware from official distributor Cryptouniverse, so I bought one Sorcerer Power Supply Unit from them for this A821 setup.
I bought a C13 to Type F (schuko) power cord for the power supply.
Please make sure that the C13 power cord has a high amp rating.
I also had to buy a wall wart for the controller unit and a Micro USB-USB A Cable.
Please make sure that the cable is a 'fast charging cable' with high amp rating.
Same as with the A741 review I did previously, I also did a disassembly for this AvalonMiner 821 unit and I took a ton of pictures.
The disassembly and assembly for the A821 is quite easy, only one screw driver (with Phillips screw head tip) is needed.
The miner design continues the same style as with previous AvalonMiner 7 generation and it is really sturdy and effective in terms of cooling effect.
AvalonMiner 821 disassembled. Click for a bigger version. Picture taken by HagssFIN.
4) Controller unit setup (top)When it comes to the controller unit, you have two options:
1. You can use the AvalonMiner Controller.
Using it is very straight forward, because it is ready assembled and software is already installed on the Micro SD card.
AvalonMiner Controller. Click for a bigger version. Pictures taken by HagssFIN.2. You can build one yourself by using a Raspberry Pi 3 and other supplies.
So in this next part I'm gonna show you how to build a controller by myself,
using a Raspberry Pi 3 single-board computer and some other supplies.
It's really easy to source these parts needed and I used my local electronics store to find these supplies.
Here is my shopping list:
• Raspberry Pi 3
• Transparent plastic case for the Raspberry Pi 3
• 8GB Kingston Micro SD card
• 5VDC 2.4A USB wall wart
• USB-A to Micro USB-B cable
Supplies. Click for a bigger version. Pictures taken by HagssFIN.Here are the instructions for doing the controller setup from scratch.
Instructions:1. AvalonMiner Controller’s firmware supports the following types of Raspberry Pi single-board computer:
• AvalonMiner Controller (Canaan Creative)
• Raspberry Pi 1 Model B
• Raspberry Pi 1 Model B+
• Raspberry Pi 2 Model B
• Raspberry Pi 3 Model B
2.A minimum capacity of 4GB memory card is required.
There is no requirement for brands, speed rating is recommended for Class 10.
(Please be select type of memory card based upon your selected Raspberry Pi model)
• Raspberry Pi 1 Model B and Model B+ using a SD card
• Raspberry Pi 2 Model B using a Micro SD card
• Raspberry Pi 3 Model B using a Micro SD card
• AvalonMiner Controller using a Micro SD card
3.We need a SD card reader.
When you buy a Micro SD card, you want to make sure it comes with a SD card adapter.
Micro SD card with a SD card adapter. Click for a bigger version. Picture taken by HagssFIN.Put your memory card to the SD card reader.
Integrated SD card reader in my PC case front panel. Click for a bigger version. Picture taken by HagssFIN.4.Depending on your Raspberry Pi model, select and download the firmware version.
Raspberry Pi 1 Model B Firmware:https://canaan.io/downloads/software/avalon821/openwrt/latest/rpi1-modelb/openwrt-brcm2708-bcm2708-rpi-ext4-sdcard.imgRaspberry Pi 2 Model B Firmware:https://canaan.io/downloads/software/avalon821/openwrt/latest/rpi2-modelb/openwrt-brcm2708-bcm2709-rpi-2-ext4-sdcard.imgRaspberry Pi 3 Model B Firmware:https://canaan.io/downloads/software/avalon821/openwrt/latest/rpi3-modelb/openwrt-brcm2708-bcm2710-rpi-3-ext4-sdcard.img5.Install a tool for writing a image to SD card, if you don't have one installed already.
Win32 Disk Imager:
https://sourceforge.net/projects/win32diskimager/6.Open the tool software and select your SD card reader device letter.
Click for a bigger version.7.First click the folder icon.
Then locate and select your downloaded Openwrt firmware for the Raspberry Pi.
Click for a bigger version.8.Click Write, and then click Yes, to start writing your image.
Click for a bigger version.9.Wait for the process to complete.
Click for a bigger version.10.When the following message is displayed, your program was successfully written.
Click for a bigger version.Click ok and exit.
If you followed these steps successfully, your controller is initially completed.
5) Physical setup for the miner and controller (top)Connecting everything together is quite simple, if you are familiar with ASIC miner hardware.
I tried writing these instructions simple as possible by using specific step pictures as help.
1.Connect 5V power to the controller unit.
Click for a bigger version.2.Connect Ethernet cable to the controller unit.
Click for a bigger version.3.Connect the Micro USB-USB A cable and 5-pin AUC3 cable to the AUC3 I2C-USB communication converter unit.
Click for a bigger version.4.Connect that Micro USB-USB A cable other end to the controller unit. You can use any of the four USB ports available.
Click for a bigger version.5.Connect the other end of the 5-pin AUC3 cable to the miner unit.
If you have more than one miner unit, you can daisy chain max. five miners per one AUC3 converter
and you can connect max. four AUC3 converters to one controller unit.
Click for a bigger version.6.Connect eight pieces 6-pin power cables with invidual cabling to the miner.
I would strongly advice using a server grade power supply with at least 1400W power capacity.
In this setup I'm using the official custom power supply unit, Sorcerer Power Supply Unit.
It is able to handle max. 1600W power capacity with 230V mains voltage electricity here in Finland.
Click for a bigger version.7.Plug in the 5V USB wall wart to power on the controller unit.
8.Connect a C13 power cord first to the power supply unit and then to the wall socket.
Make sure that the power switch in the power supply unit is turned on.
If you like a video tutorial better, there are two good A821 setup tutorial videos made official Canaan distributors Blokforge (Blokforge official channel)
and Cryptouniverse (video released in TechMagnet channel, presenter Stefan is the CTO of Cryptouniverse OG).
Blokforge tutorial video link: https://www.youtube.com/watch?v=z8r0tbUmgCUCryptouniverse tutorial video link: https://www.youtube.com/watch?v=-wa2L97ReOw6) Configuration (top)You can follow these instructions for the controller configuration:
1.Connect the power to the controller, and make sure the controller and computer is configured to be connected to the same network.
(You can also use the controller connected directly to the computer with a network cable).
2.The default IP for the controller is 192.168.0.100. The IP address of the computer needs to be adjusted to access the controller.
3.Use your web browser to access “192.168.0.100”, and you will see the following login screen.
4.There is initially no password, so you can enter after click Login Overview screen.
5.You can click Configuration to configure your miners and see your pool information.
For example, you can set up three pools and corresponding miners with the default execution priority pool1> pool2> pool3.
6.If you need to change the network settings for the controller, click on Network to configure, click Save & Apply, and then power down your unit, and manually restart the controller.
7. After the configuration, change your computer back to the automatic IP address acquisition.
8.If the controller is properly connected to the Internet, you can access different sections in the miner web configuration page, for example the Cgminer configuration page and the miner status page. As long as a miner is correctly connected via AvalonMiner USB Converter to this controller unit, you will see them on the overview page and see relevant information on these devices.
Canaan Creative controller instructions for the A721 and A741: https://canaan.io/question/how-do-i-connect-an-avalonminer-721-and-avalonminer-controller-including-raspberry-pi-1-2-and-3/These same instructions can also be applied in the A821 controller setup process.
7) Testing (top)After I was ready with all the setups and configurations, it was time to test the miner.
Mining was started using the default settings.
The average hash rate was after close to two hours running time was 11.2 Th/s.
FIVEextra) Testing different voltage offset values and writing down miner performance resultsNote: This extra chapter is updated in February 25, 2018.
This is a second and successful attempt to test the voltage offset values.
To get the voltage offset value working, we need to put it as a Cgminer command
in the 'More Options' field found in the Cgminer configuration page.
Here is an example: "--avalon8-voltage-level-offset -2".
See more details in screenshot images below.
The original review was done with the default setting for the voltage offset, which is
0.
I will go ahead and test the miner with every available voltage offset value.
I will let the miner run for at least
one hour and then I look at the results.
Just like with the previous AvalonMiner generation, it is possible to choose a value from -2 to +1 for the voltage offset.
Instead of changing the hash board frequency setting, AvalonMiner performance can only be controlled by changing the voltage offset value.
In the Controller Dashboard, what does the Voltage Offset option do?
We added a Voltage Offset option for users because each AvalonMiner is different with each other because of the inclusion of so many high performance chips. The AvalonMiners are very sensitive with voltage, so the “Voltage Offset” option is used for tweaking the AvalonMiner’s voltage. If the miner works well, we suggest to not change the voltage.
And, the lower voltage offset, the lower power. So if the miner works well with -2, it just means the miner has better health.
Source: https://canaan.io/question/in-the-controller-dashboard-what-does-the-voltage-offset-option-do/This time I was able to see difference in performance and power consumption value, when testing each different voltage offset value.
Power consumption was moving in about 1200 - 1300W area, and the hash rate was in 11.2 - 11.4Th/s area.
Best power efficiency results were achieved with the voltage offset value
-2.
Please notice that these power consumption results are measured power at wall,
so the power supply power efficiency also has an effect in the result.
| Voltage offset value | Hash rate (Th/s) | Power consumption (W) | Power efficiency (J per Gh/s) |
| -2 | 11.23 | 1195 | 0.1064 |
| -1 | 11.21 | 1222 | 0.1090 |
| 0 | 11.36 | 1249 | 0.1099 |
| +1 | 11.39 | 1295 | 0.1137 |
Cgminer status page view and Cgminer configuration page view, when the voltage offset value is -2. Click for a bigger version.
Cgminer status page view and Cgminer configuration page view, when the voltage offset value is -1. Click for a bigger version.
Cgminer status page view and Cgminer configuration page view, when the voltage offset value is 0. Click for a bigger version.
Cgminer status page view and Cgminer configuration page view, when the voltage offset value is +1. Click for a bigger version.
Power consumption measurement, when the voltage offset value is -2. Click for a bigger version.
Power consumption measurement, when the voltage offset value is -1. Click for a bigger version.
Power consumption measurement, when the voltage offset value is 0. Click for a bigger version.
Power consumption measurement, when the voltage offset value is +1. Click for a bigger version.
SIXHi all, I took a big decision, I decided to release my custom firmware for the symbolic sum of 0.005BTC WITHOUT ANY DEVFEE as often requested by the comunity ... a single purchase, has unfolded unlimited!
I know that once released, he will browse the web, thank you to try to respect my work with a purchase or a donation there after;), 0.005btc is not much ... I'm going in the direction from the community and I did not include any hidden fees;)HI ALL !!! After reflextion, i take big decision, and KILL ALL firmware S9 !!! not paid for me ? is not simpaty for my work, not problem, so, S9 is now old machine, i take with pleasure donate 
Way for install this :
1 - you need unlocked firmware (before 2019) if not, downgrade by SD Card with T9+ firmware (see on bitmain.com)
2 - Install original firmware S9J mode low power/hashrate (containe the bmminer with full control coded, Yeah !!! error by bitmain)
3 - install my patch for take full control ! (contain bmminer not modified, if you want, open my patch and take only other files than bmminer by ssh for proof !) i included bmminer taked on S9J firmware only for more facility to flash on other miner)
4 - ENJOY !!! and thanks me lol
download link : S9 / S9I / S9J
- Original firmware for S9J (compatible with S9 / S9I / S9J) :
https://easyupload.io/b2xgv5- My patch (need flash S9J firmware before) :
https://easyupload.io/5khk9pdownload link : T9+
- Original firmware for T9+ ONLY :
https://easyupload.io/1gr2dj- My patch customised for T9+ (need flash flash T9+ firmware in the link for unlock all feature) :
https://easyupload.io/tzzj55Release :
- Compatible with S9 / S9I / S9J (auto or not) and (T9+ available by other files)
- ASICBOOST (On / Off)
- ORIGINAL Bitmain Firmware
- NO DEV FEE !!!
- OVERCLOCK / UNDERCLOCK
- Frequency control
- Voltage board control (By increment 5mv)
- FAN Fixed / Auto
- Show PCB temp
- No Problem on all pool supported AsicBoost
- Is original firmware, absolut no problem for reflash other original firmware
- Same perform with all other custom firmware, TAKE FULL CONTROL ! down to 75w/th and up to 18,5th 
Do not ask me the sources, I do not have them, for the good reason that I work with a BMMINER of origin bitmain (drawn in a firmware), then bypassed the system via the other file ... nothing 'is hidden
I am not a noobie here, have confidence in me;)
Ps : Kano, is for you !!! kiss !!! and now, all detractor, please stop broken my burns !!!
Donate :
LTC : LgTLyhXEeF3p6oshxPwcacfCXezpdNXjhc
BTC : show on my firmware
DOGE : DKTK4bXwKLHEmcXc5SG81RecnfJkQa6NK1
SEVENHi all!
I know some people have had a problem with their T9 + who lost 2 hash board, especially after flashing the firmware Dollemizer! or build own bmminer by source code
What happens, the bad firmware erases the board DSPIC32 to reprogram them, but as it does not have the HEX file, it reprograms nothing, and therefore the chain is definitely lost!
I also had the problem after trying aa BMMINER custom that had exactly the same effect! so I got hooked on the question ... and HERE IS THE SOLUTION!
pre-requisite:
- a PICKIT 3 case
- MPLAB X IPE
- Bitmain HEX original file (pic32)
After :
1 - connect to the board, pin1 in front of the PICKIT arrow
2 - Connect on MPLAB
3 - Enter in advanced mode and take "power target circuit from tool"
4 - ERASE the pic (use if you sure that the hash board is hs)
Erasing...
Erase successful
2018-09-19 22:17:36 +0200 - Programming...
5 - Put in Hex File my Hex file extracted from good board (download here
http://www.k-upload.fr/afficher-fichier-2018-09-19-103561fd6dspic33ep16g.hex.html ) or (
http://www.mediafire.com/file/mk8ivrax849ztxp/DSPIC33EP16GS202-ORIGINAL-T9%252B.hex/file )
Connecting to MPLAB PICkit 3...
Currently loaded firmware on PICkit 3
Firmware Suite Version.....01.54.00
Firmware type..............dsPIC33E/24E
Programmer to target power is enabled - VDD = 3,250000 volts.
Target device dsPIC33EP16GS202 found.
Device ID Revision = 4003
DEVSN0 = 00000000
DEVSN1 = 00000000
2018-09-19 22:16:21 +0200 - Hex file loaded successfully.
Loading code from C:\Users\xxx\Desktop\Firmware antminer D3 et L3+\pump firmware T9+\Prog PIC\DSPIC33EP16GS202.hex...
6 - "Program" and wait !
Programming...
The following memory area(s) will be programmed:
program memory: start address = 0x0, end address = 0x1dff
Programming/Verify complete
2018-09-19 22:18:13 +0200 - Programming complete
once the cards flash with PicKit, do not forget to reflash the minor with official firmware bitmain before reboot the minor ...
to do this, do it before you flash the card, or just connect the controller card (without the hashboard) and flash the official firmware, power off the miner, reconnect the hashboard and turn on again.
Essenciel is that the first start after flashing the board is done with an official firmware on the minor (with hashboard connected) otherwise the problem will come back immediately..
Now your board is reprogramed ! and working !!!
For any advice or request do not hesitate!
If you like it;)
BTC: 1FWpeHCs7rB7LgWUiU2j2GgwuZpEYtfGm2
LTC: LShLPeXRMgRg3K2DP1ZMmWXbfS3Fh4ATNQ
EDIT : FREE GIFT ! for no error / brick your miner after compile your own code for bmminer
Modifi this "define" to "undefine" on the bitmain code on "driver-btm-c5.H" (the effect : no flash pic)
241 #define ENABLE_RESTORE_PIC_APP // if enabled, will restore PIC APP when the version is not correct!!!
EIGHTAs an example for setting minimum speed of 10% and max of 30%, under More Options use
--avalon8-fan 10-30
Full list of A8 commands:
USE_AVALON8
OPT_WITH_CBARG("--avalon8-voltage-level",
set_avalon8_voltage_level, NULL, &opt_set_avalon8_voltage_level,
"Set Avalon8 default level of core voltage, range:[0, 15], step: 1"),
OPT_WITH_CBARG("--avalon8-voltage-level-offset",
set_avalon8_voltage_level_offset, NULL, &opt_set_avalon8_voltage_level_offset,
"Set Avalon8 default offset of core voltage level, range:[-2, 1], step: 1"),
OPT_WITH_CBARG("--avalon8-freq",
set_avalon8_freq, NULL, &opt_set_avalon8_freq,
"Set Avalon8 default frequency, range:[25, 1200], step: 25, example: 800"),
OPT_WITH_ARG("--avalon8-freq-sel",
set_int_0_to_7, opt_show_intval, &opt_avalon8_freq_sel,
"Set Avalon8 default frequency select, range:[0, 7], step: 1, example: 7"),
OPT_WITH_CBARG("--avalon8-fan",
set_avalon8_fan, NULL, &opt_set_avalon8_fan,
"Set Avalon8 target fan speed, range:[0, 100], step: 1, example: 0-100"),
OPT_WITH_ARG("--avalon8-temp",
set_int_0_to_100, opt_show_intval, &opt_avalon8_temp_target,
"Set Avalon8 target temperature, range:[0, 100]"),
OPT_WITH_ARG("--avalon8-polling-delay",
set_int_1_to_65535, opt_show_intval, &opt_avalon8_polling_delay,
"Set Avalon8 polling delay value (ms)"),
OPT_WITH_ARG("--avalon8-aucspeed",
opt_set_intval, opt_show_intval, &opt_avalon8_aucspeed,
"Set AUC3 IIC bus speed"),
OPT_WITH_ARG("--avalon8-aucxdelay",
opt_set_intval, opt_show_intval, &opt_avalon8_aucxdelay,
"Set AUC3 IIC xfer read delay, 4800 ~= 1ms"),
OPT_WITH_ARG("--avalon8-smart-speed",
opt_set_intval, opt_show_intval, &opt_avalon8_smart_speed,
"Set Avalon8 smart speed, range 0-1. 0 means Disable"),
OPT_WITH_ARG("--avalon8-th-pass",
set_int_0_to_65535, opt_show_intval, &opt_avalon8_th_pass,
"Set A3210 th pass value"),
OPT_WITH_ARG("--avalon8-th-fail",
set_int_0_to_65535, opt_show_intval, &opt_avalon8_th_fail,
"Set A3210 th fail value"),
OPT_WITH_ARG("--avalon8-th-init",
set_int_0_to_65535, opt_show_intval, &opt_avalon8_th_init,
"Set A3210 th init value"),
OPT_WITH_ARG("--avalon8-th-ms",
set_int_0_to_32767, opt_show_intval, &opt_avalon8_th_ms,
"Set A3210 th ms value"),
OPT_WITH_ARG("--avalon8-th-timeout",
opt_set_uintval, opt_show_uintval, &opt_avalon8_th_timeout,
"Set A3210 th timeout value"),
OPT_WITH_ARG("--avalon8-th-add",
set_int_0_to_1, opt_show_intval, &opt_avalon8_th_add,
"Set A3210 th add value"),
OPT_WITH_ARG("--avalon8-th-mssel",
set_int_0_to_1, opt_show_intval, &opt_avalon8_th_mssel,
"Set A3210 th mssel value"),
OPT_WITH_ARG("--avalon8-lv2-th-add",
set_int_0_to_1, opt_show_intval, &opt_avalon8_lv2_th_add,
"Set A3210 lv2 th add value"),
OPT_WITH_ARG("--avalon8-lv2-th-ms",
set_int_0_to_32767, opt_show_intval, &opt_avalon8_lv2_th_ms,
"Set A3210 lv2 th ms value"),
OPT_WITH_ARG("--avalon8-lv3-th-add",
set_int_0_to_1, opt_show_intval, &opt_avalon8_lv3_th_add,
"Set A3210 lv3 th add value"),
OPT_WITH_ARG("--avalon8-lv3-th-ms",
set_int_0_to_32767, opt_show_intval, &opt_avalon8_lv3_th_ms,
"Set A3210 lv3 th ms value"),
OPT_WITH_ARG("--avalon8-lv4-th-add",
set_int_0_to_1, opt_show_intval, &opt_avalon8_lv4_th_add,
"Set A3210 lv4 th add value"),
OPT_WITH_ARG("--avalon8-lv4-th-ms",
set_int_0_to_32767, opt_show_intval, &opt_avalon8_lv4_th_ms,
"Set A3210 lv4 th ms value"),
OPT_WITH_ARG("--avalon8-lv5-th-add",
set_int_0_to_1, opt_show_intval, &opt_avalon8_lv5_th_add,
"Set A3210 lv5 th add value"),
OPT_WITH_ARG("--avalon8-lv5-th-ms",
set_int_0_to_32767, opt_show_intval, &opt_avalon8_lv5_th_ms,
"Set A3210 lv5 th ms value"),
OPT_WITH_ARG("--avalon8-lv6-th-add",
set_int_0_to_1, opt_show_intval, &opt_avalon8_lv6_th_add,
"Set A3210 lv6 th add value"),
OPT_WITH_ARG("--avalon8-lv6-th-ms",
set_int_0_to_32767, opt_show_intval, &opt_avalon8_lv6_th_ms,
"Set A3210 lv6 th ms value"),
OPT_WITH_ARG("--avalon8-lv7-th-add",
set_int_0_to_1, opt_show_intval, &opt_avalon8_lv7_th_add,
"Set A3210 lv7 th add value"),
OPT_WITH_ARG("--avalon8-lv7-th-ms",
set_int_0_to_32767, opt_show_intval, &opt_avalon8_lv7_th_ms,
"Set A3210 lv7 th ms value"),
OPT_WITHOUT_ARG("--avalon8-iic-detect",
opt_set_bool, &opt_avalon8_iic_detect,
"Enable Avalon8 detect through iic controller"),
OPT_WITH_ARG("--avalon8-nonce-mask",
set_int_24_to_32, opt_show_intval, &opt_avalon8_nonce_mask,
"Set A3210 nonce mask, range 24-32."),
OPT_WITH_ARG("--avalon8-nonce-check",
set_int_0_to_1, opt_show_intval, &opt_avalon8_nonce_check,
"Set A3210 nonce check, range 0-1."),
OPT_WITH_ARG("--avalon8-roll-enable",
NINE
Thanks, -ck.
The simplest solution to the port problem that I can think of is to let the pool's stratum server listen on port 80 instead of port 443. This frees up port 443 for the web server to use for HTTPS traffic. This makes a bit more sense anyway, since port 80 is intended for unencrypted HTTP traffic, and the stratum protocol is itself unencrypted. Outgoing connections to port 80 are also usually allowed by default in most firewalls.
One downside to this approach is that visitors to the pool's website would have to manually add the
https:// prefix to the pool's web address every time they enter it into their browser's address bar, because the pool's web server can't listen on the occupied port 80 to redirect them to port 443—unless the pool's website is (a) included in the
HSTS preload list and/or (b) incorporates the EFF's
HTTPS Everywhere rulesets to allow visitors using the
HTTPS Everywhere browser extension to be automatically directed to port 443. The upside to this approach is that you do not need to configure the web server to forcibly redirect unencrypted HTTP requests to HTTPS URIs.
If you prefer to reserve port 80 for the pool's web server to use, then another simple solution would be to let the pool's stratum server listen on port 53 instead of port 443. This frees up both port 80 and port 443 for the web server to use. Port 53 is meant for unencrypted DNS traffic, and so outgoing connections to port 53 are also usually allowed by default in most firewalls. This approach requires you to configure the web server to
forcibly redirect unencrypted HTTP requests to HTTPS URIs. Registering the pool's website on the HSTS preload list and incorporating HTTPS Everywhere rulesets remain highly recommended when using this approach.
A more complicated solution that doesn't involve setting up an additional network interface with additional IP addresses on the same machine, would be to
terminate TLS connections at a HAProxy instance on a cheap VPS that in turn relays web traffic to and from the upstream web server (ideally encrypted, over a private network). Upsides to this approach include (a) allowing you to leave the pool's current setup on the dedicated server mostly untouched, and (b) freeing up the pool's dedicated server from not only having to manage TLS connections with web clients, but also from having to manage web traffic with clients at all. The downside to this approach is that it involves more effort and money to set up and maintain, because you would have to manage two servers instead of one.
And then there's setting up an additional virtual network interface with additional IPv4 and IPv6 addresses on the same machine, and binding the pool's stratum server to one virtual interface and the web server to another. This seems quite straightforward in concept, but I do not have any experience with this approach, and so I can't say much about it, nor can I recommend it with confidence. It does seem cheaper and simpler than setting up HAProxy as a reverse proxy on another VPS, though.
TENWhy all miners need to mine on a pool that pays them the tx fees.
That is a really hard question to answer.
But I see only BTC in trouble as the math is dead solid correct that BTC can have fee rate jacked up with ease.
Do I mine at Zpool yes not BTC
Do I mine at NiceHash yes not BTC
Do I mine at Miningpoolhub yes not BTC
At the moment BTC has an easy way to manipulate fees by the top Six pools
the math is there.
You can argue that mining with
https://pool.viabtc.com/pool/btc/profit/ALL/that they pay back some of the fees. all of the coins above 100% is some of the fees.
So to make sure I am not feeding the fee monster and not getting killed by variance.
I mine 2 s-9s on viabtc.--------- no variance
I mine 2 s-9s on mmpool. ------ high variance
If alot of miners did this problem goes away
and you will not get crushe with high fees or high variance.
so ½ on viabtc pps ---------- no variance
½ on kano-------------------- high variance
or ½ on viabtc pps ---------- no variance
½ on mmpool---------------- high variance
or
½ on viabtc pps --------------no variance
½ on ck's pool ----------------high variance
any pattern like the one above
would be helpful
Date Hashrate Total Profit Unit Output(TH/s) PPS Profit PPLNS Profit SOLO Profit PPS+ Rate
2017-12-26 22.844T 0.00166932 0.00015943 0.00135316 0.00031616 0 118.43%
2017-12-25 22.541T 0.00377373 0.00016741 0.00290061 0.00087312 0 124.9%
2017-12-24 22.719T 0.00425017 0.00018707 0.0029248 0.00132537 0 139.5%
2017-12-23 22.555T 0.00494303 0.00021915 0.0029121 0.00203093 0 162.95%
2017-12-22 22.510T 0.00536606 0.00023838 0.00289729 0.00246877 0 177.8%
2017-12-21 22.788T 0.00492021 0.00021591 0.00294698 0.00197323 0 160.28%
2017-12-20 22.867T 0.00415326 0.00018162 0.00294023 0.00121303 0 135.61%
2017-12-19 22.492T 0.00376132 0.00016723 0.00289554 0.00086578 0 124.7%
2017-12-18 22.656T 0.00438107 0.00019337 0.00341745 0.00096362 0 123.07%
2017-12-17 24.742T 0.0046689 0.00018870 0.00377236 0.00089654 0 118.81%
2017-12-16 27.328T 0.00523549 0.00019158 0.00415086 0.00108463 0 121.08%
2017-12-15 27.102T 0.00501499 0.00018504 0.00412332 0.00089167 0 116.76%
2017-12-14 27.754T 0.00533809 0.00019233 0.00420781 0.00113028 0 121.78%
2017-12-13 27.653T 0.00518566 0.00018752 0.00419169 0.00099397 0 118.76%
As you see above they do pay some back extra.
The math for excluding this style of mining on their pool is not as compelling.
I run 2 s-9's there see above
I run my other 2 s-9's here
http://mmpool.org/statisticsEstimated Payout
Rank...Id........DGM Estimate......PPS..................... Shares............. Rate (MHash/s)
1 a5d78fff 9.74241459 0.00000000 130,161,014,576_____ 0
2 71602a00 0.43753559 0.00000000 20,405,489,465______ 23,917,084
3 2d0ba29d 0.21326891 0.00000000 12,671,491,314______ 5,583,990
4 c932f3ed 0.23847335 0.00000000 11,985,725,219______ 0
5 5e935067 0.05308415 0.00000000 3,806,752,856_______ 13,209,683
6 433aa491 0.07248369 0.00000000 3,183,604,233_______ 1,072,527
7 0f7b9499 0.03441520 0.00000000 2,603,052,116_______ 0
8 d7ec4cdf 0.02936538 0.00000000 2,460,009,494_______ 5,852,435
9 8dae603b 0.02509840 0.00000000 1,746,057,383_______ 0
