Pages:
Author

Topic: ANTMINER S3+ Discussion and Support Thread - page 68. (Read 710164 times)

legendary
Activity: 1666
Merit: 1185
dogiecoin.com
I have installed the latest firmware 01.09.2015, really do not be problems, but there are always errors HW Huh

Can you post a screenshot of your status please?
It is well documented in these threads that starting with the Dec firmware, all updates result in increased HW errors  and / or occurance of X's on chips (if not both) to a large extent. That you have not picked up on this and keep advising people to update is a truly astounding feat of stupidity. I'd err on the side of advising to only update firmware if there is some catastrophic event that's occured.

That's not true. Just ignore him.
it is true, i tested it.. the 2015 firmware has way more errors then the 12/19 one..
it went from .0008% to .03% in just 24 hours running, then i got 1 x which i never had before.

i would use the 12/19 FW.. it seems stable.

perhaps its from the turning off the auto voltage?

That 0.0292% increase is so insignificant that its the equivalent of going from 460GH to 459.9GH. There are other useful improvements over the 12/19 included. As to pekatete saying 12/19 has hardware errors is insane as there was no change. And as bitsolutions says, fixing that reset button functionality is imperative and recommended for all.

tldr if you are worried about 0.1GH, use 12/19, if not use the latest.
legendary
Activity: 1736
Merit: 1006
I have installed the latest firmware 01.09.2015, really do not be problems, but there are always errors HW Huh

Can you post a screenshot of your status please?
It is well documented in these threads that starting with the Dec firmware, all updates result in increased HW errors  and / or occurance of X's on chips (if not both) to a large extent. That you have not picked up on this and keep advising people to update is a truly astounding feat of stupidity. I'd err on the side of advising to only update firmware if there is some catastrophic event that's occured.

That's not true. Just ignore him.
it is true, i tested it.. the 2015 firmware has way more errors then the 12/19 one..
it went from .0008% to .03% in just 24 hours running, then i got 1 x which i never had before.

i would use the 12/19 FW.. it seems stable.

perhaps its from the turning off the auto voltage?
sr. member
Activity: 261
Merit: 257
Stick your head in the sand with your own rigs and don't advise people to do what others (including myself) have observed and reported in these threads. By your own admission you only have seen one make of S3, but you keep assuming all S3 variants are the same in your advice. I do not expect you to alter your opinion, not even with empirical evidence contrary to your advice, but that is more a reflection on you than the authenticity of the facts. I've come to expect waffle responses from you to a large extent, but do not hold your breath 'cause if i happen to stumble on it, I'll put the record straight.

So here goes, if your S3 (and this is more true for S3's as opposed to S3+'s) is running OK, there is no performance reason to update your firmware. The Oct firmware that does not respond to the reset button is inherently irrecoverable should disaster occur, but runs a lot better than subsequent firmware releases, aka Nov & Dec 2014 and Jan 2015 firmwares.

The reset button flaw of the Oct firmware is critical, but I have yet to hear of anyone having a bricked control board without having fidgeted with firmware updates, so I'd err on not updating at the expense of increased HW % and /or X's and / or o's on chips.
People have bricked the Oct 24'th firmware a lot by messing up their network settings. I haven't seen any performance issues with the Jan firmware although I generally only look at total hashrate stats.
hero member
Activity: 518
Merit: 500
I have installed the latest firmware 01.09.2015, really do not be problems, but there are always errors HW Huh

Can you post a screenshot of your status please?
It is well documented in these threads that starting with the Dec firmware, all updates result in increased HW errors  and / or occurance of X's on chips (if not both) to a large extent. That you have not picked up on this and keep advising people to update is a truly astounding feat of stupidity. I'd err on the side of advising to only update firmware if there is some catastrophic event that's occured.

That's not true. Just ignore him.
Stick your head in the sand with your own rigs and don't advise people to do what others (including myself) have observed and reported in these threads. By your own admission you only have seen one make of S3, but you keep assuming all S3 variants are the same in your advice. I do not expect you to alter your opinion, not even with empirical evidence contrary to your advice, but that is more a reflection on you than the authenticity of the facts. I've come to expect waffle responses from you to a large extent, but do not hold your breath 'cause if i happen to stumble on it, I'll put the record straight.

So here goes, if your S3 (and this is more true for S3's as opposed to S3+'s) is running OK, there is no performance reason to update your firmware. The Oct firmware that does not respond to the reset button is inherently irrecoverable should disaster occur, but runs a lot better than subsequent firmware releases, aka Nov & Dec 2014 and Jan 2015 firmwares.

The reset button flaw of the Oct firmware is critical, but I have yet to hear of anyone having a bricked control board without having fidgeted with firmware updates, so I'd err on not updating at the expense of increased HW % and /or X's and / or o's on chips.
legendary
Activity: 1666
Merit: 1185
dogiecoin.com
I have installed the latest firmware 01.09.2015, really do not be problems, but there are always errors HW Huh

Can you post a screenshot of your status please?
It is well documented in these threads that starting with the Dec firmware, all updates result in increased HW errors  and / or occurance of X's on chips (if not both) to a large extent. That you have not picked up on this and keep advising people to update is a truly astounding feat of stupidity. I'd err on the side of advising to only update firmware if there is some catastrophic event that's occured.

That's not true. Just ignore him.
hero member
Activity: 518
Merit: 500
I have installed the latest firmware 01.09.2015, really do not be problems, but there are always errors HW Huh

Can you post a screenshot of your status please?
It is well documented in these threads that starting with the Dec firmware, all updates result in increased HW errors  and / or occurance of X's on chips (if not both) to a large extent. That you have not picked up on this and keep advising people to update is a truly astounding feat of stupidity. I'd err on the side of advising to only update firmware if there is some catastrophic event that's occured.
hero member
Activity: 518
Merit: 500
Going further than others: Don't bother, you're more likely to end up with 4 fried PSUs and potentially a damaged S3. Even if you could underclock sufficiently you'd still be cross loading 2 PSUs. And even if that wasn't a problem (which it is), you're running PSUs at 24/7 load which they're likely not designed to do. tldr, don't.
Powering each blade with a different PSU is NOT an issue - that has been sufficiently answered.
The rest of your assertions @dogie, are your usual waffle .... thin on facts, un-necessarily verbose!

I was suggesting 2 PSUs per board, because you're going to struggle to do ANYTHING with 60W per PCB. If you've got something to contribute, keep it civil please.
And what was new in your suggestion that was not expressly covered prior to your follow-up waffle?
2 PSU's per board or 3 60W PSU's per 6 pin PCI-E plug is, in the context, the same thing ... and that was adequately pointed out; Thus aside from being verbose in your going further comment, you brought nothing new to the table, well, aside from your verbosity ..... (read waffle!).
Now if you really want to know what civility entails, then you'd be glad to know that pretence is counter to the universal norms of civility ... aka, if you (and here I mean dogie) have nothing to say, do not say it!
legendary
Activity: 1666
Merit: 1185
dogiecoin.com
I have installed the latest firmware 01.09.2015, really do not be problems, but there are always errors HW Huh

Can you post a screenshot of your status please?
legendary
Activity: 1666
Merit: 1185
dogiecoin.com
Going further than others: Don't bother, you're more likely to end up with 4 fried PSUs and potentially a damaged S3. Even if you could underclock sufficiently you'd still be cross loading 2 PSUs. And even if that wasn't a problem (which it is), you're running PSUs at 24/7 load which they're likely not designed to do. tldr, don't.
Powering each blade with a different PSU is NOT an issue - that has been sufficiently answered.
The rest of your assertions @dogie, are your usual waffle .... thin on facts, un-necessarily verbose!

I was suggesting 2 PSUs per board, because you're going to struggle to do ANYTHING with 60W per PCB. If you've got something to contribute, keep it civil please.
hero member
Activity: 798
Merit: 1000
I have installed the latest firmware 01.09.2015, really do not be problems, but there are always errors HW Huh
hero member
Activity: 518
Merit: 500
Going further than others: Don't bother, you're more likely to end up with 4 fried PSUs and potentially a damaged S3. Even if you could underclock sufficiently you'd still be cross loading 2 PSUs. And even if that wasn't a problem (which it is), you're running PSUs at 24/7 load which they're likely not designed to do. tldr, don't.
Powering each blade with a different PSU is NOT an issue - that has been sufficiently answered.
The rest of your assertions @dogie, are your usual waffle .... thin on facts, un-necessarily verbose!
legendary
Activity: 1666
Merit: 1185
dogiecoin.com
Is there any way to tell which S3's have the variable voltage?

my one and only  S3 is a batch 8 everything else i have is S3+


I would have thought the PCBs look physically different in the DC/DC area, but I've personally only seen one variant. I do know that there are many, many, many variants though.
full member
Activity: 174
Merit: 100
Is there any way to tell which S3's have the variable voltage?

my one and only  S3 is a batch 8 everything else i have is S3+
sr. member
Activity: 261
Merit: 257
There seems to be a new firmware dated January 2015 on the BITMAIN website. Anybody tried that firmware?
Yes, I've tested this on many miners, it works fine and doesn't have the filesystem issue the Oct 24'th firmware has.
legendary
Activity: 1666
Merit: 1185
dogiecoin.com
I'm having issues with the wifi. I have the router right next to the miner. It gets 60% signal. But sometimes it drops the signal when mining and start beeping.

And sometimes it sees the networks but fails to connect. Any idea why this is happening? is it because there is no Antenna despite the 60% signal strength?

Can I use the antenna from an old laptop? The connection looks the same.

There seems to be a new firmware dated January 2015 on the BITMAIN website. Anybody tried that firmware?

The s3's antennae connection is quite small, are you sure its the same as your laptops? Can't hurt to give it a try. Lay the antennae so it comes out the little notch the fan's cable sits in, so you can plug in to the controller board but still keep the antennae external.

01/19 disables voltage control on S3s as not all S3s have variable voltage control. Not everyone wants that though, so users currently using 12/19 and 01/19.
legendary
Activity: 1666
Merit: 1185
dogiecoin.com
Yes the seller finally got back to my with the password and I am greatful because the webui and ssh BOTH didn't work without the password. The units would of been garbage.

And yes I am on Oct 24th firmware. Any benefits to upgrading in terms of power consumption, or hash speed?

As aarons6 said, 12/19 onwards fixes the reset button's functionality. While it won't give you greater hashing power, it is an important bug fix which will give you another layer of protection from bricked units. The last thing you want is an inaccessible unit without the reset button working. 12/19 download link can be found below if you want that one, although there is a newer version on Bitmaintech.com.

Can I get some beta testers for a new S3 firmware please? Download link is here, it *should* be good to go but as you know....

Rolls back to the last good patch in Oct (1024), but fixes voltage adjustment and the important reset button. Hopefully.... let me know.

MD5: 7DF1B12D28B25E10AAC37E2EAFC7397B
legendary
Activity: 1666
Merit: 1185
dogiecoin.com
Short question regarding alternative power supply:

I've been changing lights at my home to LEDs. I recovered 6 x AC->DC adapters (12v 60W).
Would it work if I connected each of them to one pair of PCI-e connector pins.
It gives 2 PCI-e connectors, 180W each - at least in theory Smiley

I know it would be close to MAX W needed but I intend to under-clock it anyway... so it takes +/- 200W.
In theory that sounds plausible, however, bitmain have advised against mixing / sharing power supplies, and if I get your suggestion correctly, that would indeed be a case of mixing power supplies UNLESS you intend to run each blade on 60W only (12v 5A).
Is this even possible? What is the minimum W per blade? What hash power can be expected then? Forum search is disabled so...

Going further than others: Don't bother, you're more likely to end up with 4 fried PSUs and potentially a damaged S3. Even if you could underclock sufficiently you'd still be cross loading 2 PSUs. And even if that wasn't a problem (which it is), you're running PSUs at 24/7 load which they're likely not designed to do. tldr, don't.
legendary
Activity: 3808
Merit: 1723
There seems to be a new firmware dated January 2015 on the BITMAIN website. Anybody tried that firmware?
legendary
Activity: 1736
Merit: 1006
How do you expect to ssh in with root/root if the password was changed? If root doesn't work for the webui pass it won't work for ssh either(barring some very unusual configuration). You should upgrade the firmware as soon as you are able to login as you are most likely on the very dangerous oct 24th firmware. If you can't get the password from the seller you will have to use a TTL cable to reset/reflash it manually.

Because SSH and webui logins are entirely separate, changing the webui password has no impact on the SSH password. The entire point is that he can't log into the webui and so must use SSH access, else he will likely have to send the controller in.

The Oct 24 firmware is also NOT dangerous, its stable as hell and would be great if it didn't have broken reset button functionality.

Yes the seller finally got back to my with the password and I am greatful because the webui and ssh BOTH didn't work without the password. The units would of been garbage.

And yes I am on Oct 24th firmware. Any benefits to upgrading in terms of power consumption, or hash speed?

no probably not.. but just to make the reset button work i updated my s3s to the 12/19 firmware..

it seems to be just as good as the 10/24 one.
legendary
Activity: 3808
Merit: 1723
I'm having issues with the wifi. I have the router right next to the miner. It gets 60% signal. But sometimes it drops the signal when mining and start beeping.

And sometimes it sees the networks but fails to connect. Any idea why this is happening? is it because there is no Antenna despite the 60% signal strength?

Can I use the antenna from an old laptop? The connection looks the same.
Pages:
Jump to: