Bitmain's Released Antminer S9, World's First 16nm Miner Ready to Order

[ComputerGenie]

I haven't been able to find an answer to this, but after reading about ASIC boost, it appears that when mining on Antpool, a S9 should use ~20% less power?  Has this been verified by anyone?  Or is this only for miners hashing for Bitmain themselves.

It seems you have a misunderstanding of what you read, or you read too much speculation.
The process of finding a single share is reduce, and thus the speed of finding that share. While it does, in theory, "use ~20% less power", that is per share. An ASIC produces a share and goes on to the next one. It's not an overall reduction in consumption of power, just a reduction in the power required to produce a single share (thus yielding more overall shares per amount of energy consumed).

As fanatic26 said, no one has implemented it, yet, on the main network; however, if they did and you joined their pool, you would still consume the same power per 24 hour period that you currently do (it would just be a more efficient usage of your power).

It is astounding the lack of understanding people have over this whole asicboost thing...

Much like everything else in the crypto space.

[1715018131]

1715018131

[1715018131]

1715018131

[1715018131]

1715018131

[1715018131]

1715018131

[NotFuzzyWarm]

I haven't been able to find an answer to this, but after reading about ASIC boost, it appears that when mining on Antpool, a S9 should use ~20% less power?  Has this been verified by anyone?  Or is this only for miners hashing for Bitmain themselves.

Sigh...
With all that has been said here how did you ever come up with that conclusion
A) Bitmain has time and again stated that EVEN FOR THEIR OWN USE IN CHINA it has NOT been used on the main BTC net. They have only experimented with AB on testnet. Many very knowledgeable folks here have given their confirmation that as far as can be seen it is not being used in any pool on the main network.

B) Because of 'A' NONE of their miners get a "boost" if ran on AntPool.

[RoadStress]

No word about www.antbleed.com here yet? 

[ComputerGenie]

No word about www.antbleed.com here yet?  

Because adults know that minerlink stuff was added to firmware way back in 2015, abandoned, and that there's not much removed from the successive S7/S9 fw, just added to it.

[RoadStress]

No word about www.antbleed.com here yet? 

Because adults know that minerlink stuff was added to firmware way back in 2015 and that there's not much removed from the S9 fw, just added to it.

So everyone bought the miners knowing this issue?

[ComputerGenie]

No word about www.antbleed.com here yet?  

Because adults know that minerlink stuff was added to firmware way back in 2015 and that there's not much removed from the S9 fw, just added to it.

So everyone bought the miners knowing this issue?

It's not an "issue". At the time that it was initially done, it was to be a service available to let you remotely monitor and control all of your rigs from 1 central location (and was part of the selling point of S7s for some people). Multiple desktop apps were created that allowed you to do the same thing, so Bitmain abandoned the idea but never removed the code. Years later, new ill-informed donkeys arrive on the crypto scene, discover the remnant code, and claim there's something nefarious about it.  

FFS, the S7s still even have a tab for it (even though it was never publicly activated).

[HagssFIN]

BITMAIN: What is your official answer on this "Antbleed" -case?

[RoadStress]

It's not an "issue". At the time that it was initially done, it was to be a service available to let you remotely monitor and control all of your rigs from 1 central location (and was part of the selling point of S7s for some people). Multiple desktop apps were created that allowed you to do the same thing, so Bitmain abandoned the idea but never removed the code. Years later, new ill-informed donkeys arrive on the crypto scene, discover the remnant code, and claim there's something nefarious about it.  

FFS, the S7s still even have a tab for it (even though it was never publicly activated).

So is it possible or not to remotely shut down a BITMAIN miner due to their sloppiness and un-professional attitude? I'm not sure that I understand what are you stating.

[NotFuzzyWarm]

BITMAIN: What is your official answer on this "Antbleed" -case?

Please see my replies on this:
https://bitcointalk.org/index.php?topic=1888573.msg18762495#msg18762495
and
https://bitcointalk.org/index.php?topic=1888573.msg18762648#msg18762648
I will verify with my s7's tomorrow.

[ComputerGenie]

So is it possible or not to remotely shut down a BITMAIN miner due to their sloppiness and un-professional attitude? I'm not sure that I understand what are you stating.

What I'm saying is that, much like all software that doesn't get a complete overhaul recode, old code still remains. For the delusional conspiracy-theorists that believe in this drivel of Bitmain shutting down rigs without your consent, and you're a responsible computer component owner, then you're not still using "root" as your password and no one else can access your equipment and you needn't worry anyway.

Do you buy your tinfoil hats or do you make them yourself?  

[HagssFIN]

Yeah it used to be a service, but why hasn't it been modified in recent firmwares?
This is a security breach in my opinion because it can be used in a wrong way.
I commonly worried about device internet security safety, mostly with "internet of things" devices but now with a good reason with bitcoin miners as well.
It is something that should be looked into 100% or we will be fucked up.
We should pay more attention in general at device security.

[NotFuzzyWarm]

Yeah it used to be a service, but why hasn't it been modified in recent firmwares?
This is a security breach in my opinion because it can be used in a wrong way.
I commonly worried about device internet security safety, mostly with "internet of things" devices but now with a good reason with bitcoin miners as well.
It is something that should be looked into 100% or we will be fucked up.
We should pay more attention in general at device security.

Well as I posted in the other thread, yes it should be removed as part of due-diligence. And as with most embedded networked devices seems the designers just-don't- give-a-shit.

http://www.electronicdesign.com/embedded-revolution/beware-brickerbot-iot-killer is one Robin Hood's shot at IoT 'security'

[HagssFIN]

Yeah it used to be a service, but why hasn't it been modified in recent firmwares?
This is a security breach in my opinion because it can be used in a wrong way.
I commonly worried about device internet security safety, mostly with "internet of things" devices but now with a good reason with bitcoin miners as well.
It is something that should be looked into 100% or we will be fucked up.
We should pay more attention in general at device security.

Well as I posted in the other thread, yes it should be removed as part of due-diligence. And as with most embedded networked devices seems the designers just-don't- give-a-shit.

http://www.electronicdesign.com/embedded-revolution/beware-brickerbot-iot-killer is one Robin Hood's shot at IoT 'security'

Yeah, we are definitely at the same page.
I'm just interested if Bitmain would have any comment to this or will they be quiet as usual.

[ComputerGenie]

...yes it should be removed as part of due-diligence...

No one in the software industry ever does that. As a parallel example, there are unused telemetry remnants from Vista in Windows 10.

[NotFuzzyWarm]

...yes it should be removed as part of due-diligence...

No one in the software industry ever does that. As a parallel example, there are unused telemetry remnants from Vista in Windows 10.

And probably still even some old OS/2 dll's a well. In the early 90's before their tiff with IBM, MS wrote some of the code under contract and kept their rights to it. I know the Win OS's as recent as XP has DLL's tagged as OS/2, not sure it the Vista/Win7 re-code finally took them out.

[ComputerGenie]

Yes, Vista was a recode (for the most part, anyway).

[HagssFIN]

Bitmain gave an official comment on the "Antbleed" case.

Antminer Firmware Update - April 2017
https://blog.bitmain.com/en/antminer-firmware-update-april-2017/

Code:

There is a new controversy regarding a feature in the firmware of Bitmain’s Antminer series of Bitcoin miners which allegedly claims that Bitmain can remotely shut down Antminers. This is feature was designed and coded by the same team that is responsible for the firmware of Antminers. As the firmware has always been open-source, the feature was never intended to be malicious.

We need to clarify the intention of having this feature. We planned to add this feature to the code to empower customers to control their miners which often times can be hosted outside their premises. This was after more than one incident of miners being stolen from a mining farm or being hijacked by the operator of the mining farm:

In 2014, around 1,000 Antminers were withheld from the owner by a hosting service provider in Shenyang, China.
In 2015, around 2,000 units of Antminers were withheld from the owner by a hosting service provider in Georgia.
In 2017, Bitmain’s own miners were withheld and sold without its consent in Canada.
This feature was intended to allow the owners of Antminer to remotely shut down their miners that may have been stolen or hijacked by their hosting service provider, and to also provide law enforcement agencies with more tracking information in such cases. We never intended to use this feature on any Antminer without authorization from its owner. This is similar to the remote erase or shutdown feature provided by most famous smartphone manufacturers.

However, this feature was never completed. We started to develop this feature since Antminer S7 and wanted to finish its development on the Antminer S9. We hoped to make it a useful feature that we could advertise to our customers. But, due to some technical problems, we were unable to finish the development of this feature and shut down the testing server in December 2016.  It is a bug to leave the code there before the feature is fully complete and acknowledged to the users. This bug has now been pointed out in context of Bitcoin’s scaling roadmap debate and has caused considerable misunderstandings within the Bitcoin community. We apologize for this.

Following are the models affected by this bug:

Antminer S9
Antminer R4
Antminer T9
Antminer L3
Antminer L3+
Since MITM and DNS hijacks can exploit this bug we would like to apologize to our customers for a degraded level of security caused by this. We should also note that the existing stratum protocol is vulnerable to MITM and DNS hijacks, and that the mining industry needs to work together on design of the next generation mining protocol.

Moving on, we have released the new updated source-code on GitHub and new firmware on our website which removes this bug. We recommend all Antminer owners to upgrade their firmware to the updated firmware from the list below and strongly advise to not download any firmware provided by a third-party which can lead to unexpected outcomes in function and can also be potential phishing attacks.

Source code on GitHub

Source-code for Antminer S9, T9 and R4: https://github.com/bitmaintech/bmminer-mix
Source-code for Antminer L3 and L3+: https://github.com/bitmaintech/setup-A8 which includes https://github.com/bitmaintech/ltc_frimware inside.

Alternatively, we advise owners of mining farms that run Antminers to set the DNS record on their routers of auth.minerlink.com to 127.0.0.1 in the /etc/hosts file by the following method:

1. Use ssh to login to the miner
2. Run command:   echo “127.0.0.1     auth.minerlink.com” >> /etc/hosts
3. Run command:   sync
 

The controversy around this code has brought our attention to improve the design in order to address vulnerabilities that were pointed out by the community recently. For this, we thank the community for pointing out bugs and contributing to our open-source codebase. We would also like to take this opportunity to express our solidified belief in the open source community and express our renewed commitment to improving the quality and testing methods of our code.

We will continue the development of this feature to provide a technical protection for mining rig owners to host their miners in remote locations. We will add a switch to this feature, and this switch will be closed by default. Customers will be able to set up and configure their own authentication server. Before we finish this development, such a code will not be in the firmware we release from now on.

[ComputerGenie]

Any takers on laying bets as to the next Bitmain conspiracy that the tinfoil hat brigade will come up with? 

[HagssFIN]

These are reasonable concerns. I can not make sense why you always take Bitmain's side mr. Genie, no matter what case.

If nobody would have talked anything about this, NOTHING WOULD HAVE CHANGED!

I can bet my ass on that.

[ComputerGenie]

These are reasonable concerns. I can not make sense why you always take Bitmain's side mr. Genie, no matter what case...

I don't "always take Bitmain's side". I do, however, always approach the topic from the direction of both an informed consumer and a practical adult looking at a business grade piece of equipment.  

There's an inherent irony when viewing the complaints of the new firmware that alienates 1% of users and the complaints of the firmware that was designed to server 1% of users. That's the part where I "side with Bitmain"; it seems that people want to cry about something (simultaneously crying "you should cater to the 1% like me" and "you shouldn't cater to the 1% that might want remote control").

...If nobody would have talked anything about this, NOTHING WOULD HAVE CHANGED!

I can bet my ass on that.

And I'm guessing you advise all Windows users that Microsoft can control their computers at will, too? After all, phone-home telemetry is phone-home telemetry.