Topic: [Awesome Miner] - Powerful Windows GUI to manage and monitor up to 200000 miners - page 278. (Read 703547 times)

Hello,

Thank you for your reply but that doesn't seem to be the cause. One is set to 2 minutes and the other to 3 minutes.
The 1 minute difference doesn't explain the extreme difference shown.

Also, it seems the issue only happens when you have the same pool with the same wallet/username even though the other fields are different.
I suspect that AM is not referring to each pool with its unique ID but rather seems to be identifying each pool/coin using only the wallet/user or a combination of pool url + wallet.

Hi, What's your settings in Options -> Statistics -> Update interval (minimum 2 Minutes) ?
as well as Options -> Profit Switching -> Switching interval (minimum 30 Seconds) ?

the info displayed in Coins tab is updated as per schedule set in the former (subject to API availability from providers)

the info displayed in View Details is updated as per schedule set in the latter (subject to Profit Profile Settings -> Enabled Algo, Enabled Software, Enabled Pools)

So they will almost always be out of sync in one way or another, and degree of difference is expected amplified if the interval difference of the 2 is several (tens of) minutes apart, especially if you have lots of coins with low block times and high difficulty adjust frequency enabled. (you are likely seeing lots more of mining the wrong coin scenario especially with algos such as C11, Lyra2v2, Neoscrypt, Tribus, HMQ1725, Skein ... etc. as the profitability tend to spike due to high difficulty adjustment frequency in coins that utilise these algos, many adjust difficulty every block)

Hello Patrike,

I sent you an email regarding this issue but never got a reply.
AM seems to be having issues within the profit switching calculator.
The information shown in the "Coins" tab doesn't match what is shown when you right click on the miner and you click on  "Details".
This seems to be causing AM to choose the wrong coin to mine (unless the actual issue is the "Coins" tab showing wrong info).

ah, ok, sorry thought you're trying to manage overclocking with AM and Afterburner, I'm not an AMD person btw, only have 2 RX470 vs tens of NV cards (though, my AB can set different clocks between the 2, I did let AB manage customized Fan control curves and not let AM do anything regarding GPU controls). Apart from modding BIOS timings, I left the cards with their default memclock of 1650 when booting up, and only set clock to 2000 via AB startup profile locally (not remote), gets 29MH/s each on ETH (both with Hynix VRAM) so I guess not too bad. Point being though, maybe try other OC tools like OverDriveNT, I suspect something else is causing the sync clocks rather than AM in your scenario.
[/quote]
Uhm, yes, it might helped to use overdrive, but i need to find solution within current setup, AM+Afterburner is nice combo which suits well..its just something has to be fixed within the settings, i mean..the point of AM was to manage all in one and thats what i aim for.
Thanks, hopefully this will get answered by support

[ANN] T-Rex 0.6.6 Nvidia GPU miner

New version released.

0.6.6
Speed improvements: bcd - 5%
Cosmetic change: the miner will not show duplicate hashorder messages for x16r/s, bitcore, and they are no longer labeled as WARN
To disable API altogether set --api-bind-http / --api-bind-telnet to 0.
HTTP API: new control command "shutdown" to quit the miner (see help file for details)


Downloads
v0.6.6

Linux:
CUDA 9.1: https://mega.nz/#!XMlS0CYL!ruvZvVnoRC3gTArZyzzYF-t6B-XH0UnDCN6hV227Ya4
CUDA 9.2: https://mega.nz/#!eF1Q1SJC!0hv7ZhJaOCOQQ3zPmN5AnVCEJdND7rzqkRTm-MlVca0

Windows:
CUDA 9.1/9.2:
https://mega.nz/#!TEsUSIiC!4zpoxI58Qc90bcKsFHM0RhAvA7ERsNBB5HlNYsUa9xI

ah, ok, sorry thought you're trying to manage overclocking with AM and Afterburner, I'm not an AMD person btw, only have 2 RX470 vs tens of NV cards (though, my AB can set different clocks between the 2, I did let AB manage customized Fan control curves and not let AM do anything regarding GPU controls). Apart from modding BIOS timings, I left the cards with their default memclock of 1650 when booting up, and only set clock to 2000 via AB startup profile locally (not remote), gets 29MH/s each on ETH (both with Hynix VRAM) so I guess not too bad. Point being though, maybe try other OC tools like OverDriveNT, I suspect something else is causing the sync clocks rather than AM in your scenario.

T-Rex 0.6.6 (mostly for those interested in bcd)
Speed improvements: bcd - 5%
Cosmetic change: the miner will not show duplicate hashorder messages for x16r/s, bitcore, and they are no longer labeled as WARN
To disable API altogether set --api-bind-http / --api-bind-telnet to 0.
HTTP API: new control command "shutdown" to quit the miner (see help file for details)

Linux
CUDA 9.1:
https://mega.nz/#!XMlS0CYL!ruvZvVnoRC3gTArZyzzYF-t6B-XH0UnDCN6hV227Ya4

CUDA 9.2:
https://mega.nz/#!eF1Q1SJC!0hv7ZhJaOCOQQ3zPmN5AnVCEJdND7rzqkRTm-MlVca0

Windows (x64)
CUDA 9.1/9.2:
https://mega.nz/#!TEsUSIiC!4zpoxI58Qc90bcKsFHM0RhAvA7ERsNBB5HlNYsUa9xI

FTP mirror:
http://ftp://radio.r41.ru/t-rex-0.6.6-linux-cuda9.1.tar.gz
http://ftp://radio.r41.ru/t-rex-0.6.6-linux-cuda9.2.tar.gz
http://ftp://radio.r41.ru/t-rex-0.6.6-win-cuda9.1.zip

Hi, i just thought i made myself clear enough, but this may help a bit more, please see:
1. I am not clocking up the system with AM. Clocks are set via BIOS directly. Only FANs are set and that is related to "Miner template" not GPU.
2. I already run through various videos, also official documentation, and did some re-installation of system to start from blank. None of this helped.
3. Here are screens just so you have the full picture:
https://imgur.com/7SU9Cpq
You can see there that there is only one clocking profile with FANs only set - that is working correctly.
Also you can see that the memory clocks are available, means they can be read.
One GPU is missing, because it is disconnected now, but it would show the same memory clock - HOWEVER GPU Z default clock(out of bios) is set to 1950Mhz!! Only somehow it is re-clocked to 2100 with others.
As stated, on SINGLE GPU TEST RIG this issue is not happening.

Hope this helps. Thanks for the suggestion anyways!  

https://support.awesomeminer.com/support/solutions/articles/35000086015-gpu-overclocking

Check the section on "Clocking Profile Groups", this allows you to set clocks in a mixed GPU system and apply clocking per GPU / group of GPUs

Hello all,

I am coming to you with a question in regards to GPU clocking. I just noticed that when we use Afterburner our GPU will set all on the same settings, even though the "synchronize all" is turned off. Our rigs are each specifically tuned with own BIOS (AMD part) and has different clocking. But since we use AM and especially MSI Afterburner it use the "memory clock" of the first GPU regardless. How can this be turned off? Also there is this issue that even the "memory clock" is not shown in the GPU Settings of AM. So even if i wanted to set that differently for each GPU i am not able to.

Clocks are working if i try it on testrig where only one card is able to work, it gets right clocks and has no issue.

Thank you very much, hopefully someone is able to respond. In parallel i will try info@...

Have a nice weekend!

I have a suggestion

Taking into consideration of cryptomining newcomers, IMHO, terminology change (which ever used) won't really be effective. Ppl gets confused one way or another. I do prefer it states "Wallet Address" though as it is both literally and technically accurate.

Though, for Yiimp based pools at least, it might seem logical to implement something like the Yiimp backend, where you have a database that keeps track of address and its associated coin type. It'll be a bit of work, but can be used as a building base (selection list) for the options dialogue / configuration interface wherever applicable.

So maybe something in the form of


Now, one example of implementation, is now you just need user to populate this DB like they do with lots of user defined attributes that's already available in AM (softwares, algo, pools ... etc)

Then start experimenting only with Online Services (if working well and demand is rising, I'm sure it can be extended to custom pools .. .etc), and only ones that support AutoExchange to user chosen currency (right now, Zerg, Zpool), use a dropdown list for these 2 pools in the Options->Profit Switching section like how you choose for stratum locations for nicehash, zpool, but this one looks up the YiimpAE_PayoutDB table and let user choose. Once chosen, it should set the relevant online service entries to c=ABC (unless overridden in profit profiles).

This has an advantage that user can even choose to say, have multiple wallet addresses even for a single cryptocurrency but for one period, they like to be paid out to say, Coinbase/Exchange, but it can also have other entries to wallet running on local node/mobile wallet/hardware wallet...blah blah blah. And to change, instead of having to copy address stored elsewhere and perhaps making mistakes (due to many address preferences / number of machines / custom overridden profile settings), it can all be done via the AM GUI with a few mouse clicks (possibility with Bulk Profit Profile Edit option say).
 

You can already add whatever Yiimp compatible pool that you want.
Just follow support documentation...   :
https://support.awesomeminer.com/support/solutions/articles/35000086026-add-custom-yiimp-compatible-pools

Is this ever going to be fixed?

http://stats.finnsk3.com/AwesomeMinerFail.png

My datalogging and projections have all been ruined for the last 3 weeks.

Hi Patrike

That would be great to include in AM this pool:   icemining.ca it includes BCD and other algo not available in other pools.  It's YIIMP compatible. Thanks in advance!

Thanks.

https://support.awesomeminer.com/support/solutions/articles/35000085972

basically, if using NeoRouter (no guide on AM site), just download the free binary for your OS (Windows I presume), during installation, you get to choose Server or Client installation, only install Server on ONE machine (perhaps the host where AM is running), all the remaining are Client installations. Register a custom domain on NeoRouter's site and log the server and clients into that newly created domain so they are on the same Private Network and manage it through there.

Any questions you might have, should be answered in NeoRouter's own documents / FAQ.

http://www.neorouter.com/documents

Some guide?

Personally, I use NeoRouter (www.neorouter.com) which is one of the 2 recommended on AM's support page, it's been good so far, but best avoid Windows 10 as some Win10 builds seem to stuff it up and you have to manually resolve the issue every now and then (restart the service with Network Connections window open), I just transition to Win7 instead, best is if I want to mine ETH on 3GB cards, I still can do that on Win7 and not having to run Linux. (Win 10 reserves memory making 3GB cards unusable for ETH, ETC at their current DAG size)

I have couple rigs but in different locations, how could I do to make awesome remote to recognize all of them?

Patrike,

Can you temporarily disable rules when running benchmarks? I appear to still have the stop rule runs every 10 seconds problem, and it took me a minute to realize that was killing my benchmarks in 5.5.1. I figure even of the stop rule running too frequently gets fixed that could still affect long running benchmark sessions.