Author

Topic: Ω (Read 4248 times)

newbie
Activity: 8
Merit: 104
Ω
January 15, 2014, 12:25:56 PM
#19
I had the same problem with yvg1900's miner. It is a fantastic and very well optimized miner, however there is a caveat:

It uses 1GB per thread, so if your setup has less GB than threads, you need to reduce # of threads.

Here is what I did with one server that has 24 threads and only 8GB of RAM:

nthreads = 3 av=3 (that turns off hyper threading) - and run this twice. But as a start I just ran one process with 6 threads using avg.

Then use a regular miner that only uses 1GB of RAM with the other 18 threads.

so 6GB + 1GB + whatever is left for the code/system fits in 8GB of RAM and those servers produce about 7HPM, not spectacular but they only did 4HPM before that. In the pools however this reports as 10-11HPM so not sure which one is the correct.

For the ones that report as 31HPM in the server and about 40-45HPM in the pool, they are 32 cores with 64GB of RAM. Yes - I found a fantastic deal for those servers, but I can not tell you where I got them from since that might affect my ability to get more.

Cheers! Hope this is helpful.


Ok, I understand the problem. This is a 2cpu/2GB vps, but of course the OS needs some ram as well. I edited the config file as below, but then I get the message "Error: Parameter [nthreads] not supported for ProtoShares". Sorry if I'm being really noobish here...

Code:
threads = 0
mining-params = mmc:av=3&aesni=on&nthreads=3
mine = getwork://[email protected]:80:8880:8881:8882:8883/mmc
mine = getwork://[email protected]:8080:8081:8082:8083/mmc
#proxy = socks4a://127.0.0.1:9150
compact-stats = 1
print-timestamps = 0

Hmm... there is threads=0 at the top, don't add it as &nthreads=3 to mining params, that's not how it works. And also you can't have 3 threads on 2GB.

Use yam -c yam-mmc.cfg. protoshares is PTS, MemoryCoin is MMC.
newbie
Activity: 8
Merit: 104
January 14, 2014, 03:12:13 PM
#17
Soon to release: The Evil-Knievel-Killer-Miner!
It generates valid blocks without any neccessity of AES or SHA256.

I already had seen 10,000 HPM on my 5 year old notebook.
All I want is to earn some money first for my new Ferrari Italia, then I will make it open source  Wink

(no sarcasm whatsoever)

Ok that's fine locally - but I don't think you understand how it works - the pool and the network will not accept your work.

Because it is not just about the nonce, it has to also use the AES to find the correct offset for the block - and does that 50 times.

Other coins only check the nonce for 00s but this one also traverses the 1GB RAM and calculated 64KB xor 50 times and the address is chosen based on AES, so not sure you can avoid it.
sr. member
Activity: 420
Merit: 250
January 15, 2014, 05:52:33 PM
#15
I had the same problem with yvg1900's miner. It is a fantastic and very well optimized miner, however there is a caveat:

It uses 1GB per thread, so if your setup has less GB than threads, you need to reduce # of threads.

Here is what I did with one server that has 24 threads and only 8GB of RAM:

nthreads = 3 av=3 (that turns off hyper threading) - and run this twice. But as a start I just ran one process with 6 threads using avg.

Then use a regular miner that only uses 1GB of RAM with the other 18 threads.

so 6GB + 1GB + whatever is left for the code/system fits in 8GB of RAM and those servers produce about 7HPM, not spectacular but they only did 4HPM before that. In the pools however this reports as 10-11HPM so not sure which one is the correct.

For the ones that report as 31HPM in the server and about 40-45HPM in the pool, they are 32 cores with 64GB of RAM. Yes - I found a fantastic deal for those servers, but I can not tell you where I got them from since that might affect my ability to get more.

Cheers! Hope this is helpful.


Ok, I understand the problem. This is a 2cpu/2GB vps, but of course the OS needs some ram as well. I edited the config file as below, but then I get the message "Error: Parameter [nthreads] not supported for ProtoShares". Sorry if I'm being really noobish here...

Code:
threads = 0
mining-params = mmc:av=3&aesni=on&nthreads=3
mine = getwork://[email protected]:80:8880:8881:8882:8883/mmc
mine = getwork://[email protected]:8080:8081:8082:8083/mmc
#proxy = socks4a://127.0.0.1:9150
compact-stats = 1
print-timestamps = 0

Hmm... there is threads=0 at the top, don't add it as &nthreads=3 to mining params, that's not how it works. And also you can't have 3 threads on 2GB.

Use yam -c yam-mmc.cfg. protoshares is PTS, MemoryCoin is MMC.


Ok, got it working now. I deleted the nthreads line, and put "1" at the threads line. Apparently 1 thread is the max at a 2gb ram droplet.
sr. member
Activity: 420
Merit: 250
January 15, 2014, 08:27:49 AM
#14
I had the same problem with yvg1900's miner. It is a fantastic and very well optimized miner, however there is a caveat:

It uses 1GB per thread, so if your setup has less GB than threads, you need to reduce # of threads.

Here is what I did with one server that has 24 threads and only 8GB of RAM:

nthreads = 3 av=3 (that turns off hyper threading) - and run this twice. But as a start I just ran one process with 6 threads using avg.

Then use a regular miner that only uses 1GB of RAM with the other 18 threads.

so 6GB + 1GB + whatever is left for the code/system fits in 8GB of RAM and those servers produce about 7HPM, not spectacular but they only did 4HPM before that. In the pools however this reports as 10-11HPM so not sure which one is the correct.

For the ones that report as 31HPM in the server and about 40-45HPM in the pool, they are 32 cores with 64GB of RAM. Yes - I found a fantastic deal for those servers, but I can not tell you where I got them from since that might affect my ability to get more.

Cheers! Hope this is helpful.


Ok, I understand the problem. This is a 2cpu/2GB vps, but of course the OS needs some ram as well. I edited the config file as below, but then I get the message "Error: Parameter [nthreads] not supported for ProtoShares". Sorry if I'm being really noobish here...

Code:
threads = 0
mining-params = mmc:av=3&aesni=on&nthreads=3
mine = getwork://[email protected]:80:8880:8881:8882:8883/mmc
mine = getwork://[email protected]:8080:8081:8082:8083/mmc
#proxy = socks4a://127.0.0.1:9150
compact-stats = 1
print-timestamps = 0
legendary
Activity: 1260
Merit: 1168
January 14, 2014, 02:48:19 PM
#13
This message was too old and has been purged
full member
Activity: 196
Merit: 100
January 14, 2014, 01:15:02 PM
#12
Makes sense.  I have 1GB of unfragmented memory free for each of my cores so I guess 14HPM for 10 threads isn't too bad.

I can see 30-40 using 32 cores lol, but not for the price you mentioned.  Must have been a rare find.

I've been mining for about 2.5 hours and made 1 measily MMC heh.

http://mmcpool.com/en/user?wallet=MJwtdFaewbLTaCsgzAJsk54VDwo4abZ9tx
full member
Activity: 196
Merit: 100
January 14, 2014, 11:47:24 AM
#11
I get about 14HPM with my desktop sandy bridge CPU so I'd be interested to know what kind of servers are being used to get 30-40HPM each
sr. member
Activity: 420
Merit: 250
January 14, 2014, 11:24:49 AM
#10
What version of the miner to use for an AMD FX-8320 Vishera?
I'm guessing yam-yvg1900-M7j-win64-bdver2.zip  bd stand for bulldozer right?

Correct.


I have tried it with two pools now, doesn't change anything. Could it be something with digital ocean? I don't have a problem with them connecting to any other pools (for other coins).

Can you post your .bat string?  Have you tried the other ports for mmc pool? (80 / 8880 / 8881 / 8882 / 8883)

It is a linux (Ubuntu 13.10 x64) machine, so no bat file. I start with "./yam --config yam.cfg". My yam.cfg file looks like this:

Code:
threads = 0
mining-params = mmc:av=1&aesni=on
mine = getwork://[email protected]:80:8880:8881:8882:8883/mmc
mine = getwork://[email protected]:8080:8081:8082:8083/mmc
#proxy = socks4a://127.0.0.1:9150
compact-stats = 1
print-timestamps = 0
legendary
Activity: 854
Merit: 1000
January 14, 2014, 11:17:41 AM
#9
What version of the miner to use for an AMD FX-8320 Vishera?
I'm guessing yam-yvg1900-M7j-win64-bdver2.zip  bd stand for bulldozer right?

Correct.


I have tried it with two pools now, doesn't change anything. Could it be something with digital ocean? I don't have a problem with them connecting to any other pools (for other coins).

Can you post your .bat string?  Have you tried the other ports for mmc pool? (80 / 8880 / 8881 / 8882 / 8883)
sr. member
Activity: 420
Merit: 250
January 14, 2014, 11:12:59 AM
#8
Hmm, since I already have a digitalocean account, I decided to try MMC mining. I am using a 2CPU/2GB node, but keep getting this:

Code:
Target OK
Checking params [mmc:av=1&aesni=on]...
Params OK
Checking MMC Stage 1 optimizations compatibility...
Checking MMC Stage 2 optimizations compatibility...
OK: MMC optimizations are compatible
MemoryCoin: Memory usage 1024M per thread, Algorithm Variation 1
Using 2 mining threads
  Will mine 2 rounds for miner developers to support development of the next version
  Follow @yvg1900 on Twitter to get information on new version availability on time
New MemoryCoin Block nTime 1389713466
MMC Agg. SPM: ?, HPM: ?; Rnds C/I: 0/0, Don. C/I: 0/0; Cfg/Thr SPM: ?/?, Cfg/Thr HPM: ?/? 0 rnds AV=1, ART=?
  work.mmcpool.com: On-line, Shares Submitted 0, Accepted 0
New MemoryCoin Block nTime 1389713471
New MemoryCoin Block nTime 1389713475
MMC Agg. SPM: ?, HPM: ?; Rnds C/I: 0/0, Don. C/I: 0/0; Cfg/Thr SPM: ?/?, Cfg/Thr HPM: ?/? 0 rnds AV=1, ART=?
  work.mmcpool.com: On-line, Shares Submitted 0, Accepted 0
Killed

Anyone who knows what to do? I am using the "yam-yvg1900-M7j-linux64-generic" download.



In the readme provided with the miner:

Code:
6. Known issues

If your miner shows "PTS Agg. CPM: ?; Rnds C/I: 0/0, Don. C/I: 0/0; Cfg/Thr CPM: ?/? 0" over an extensive period of time, consider restarting your miner.
In most cases this caused by problems with network connection or temporary pool server unavailability.


I have tried it with two pools now, doesn't change anything. Could it be something with digital ocean? I don't have a problem with them connecting to any other pools (for other coins).
legendary
Activity: 854
Merit: 1000
January 14, 2014, 10:53:28 AM
#7
Hmm, since I already have a digitalocean account, I decided to try MMC mining. I am using a 2CPU/2GB node, but keep getting this:

Code:
Target OK
Checking params [mmc:av=1&aesni=on]...
Params OK
Checking MMC Stage 1 optimizations compatibility...
Checking MMC Stage 2 optimizations compatibility...
OK: MMC optimizations are compatible
MemoryCoin: Memory usage 1024M per thread, Algorithm Variation 1
Using 2 mining threads
  Will mine 2 rounds for miner developers to support development of the next version
  Follow @yvg1900 on Twitter to get information on new version availability on time
New MemoryCoin Block nTime 1389713466
MMC Agg. SPM: ?, HPM: ?; Rnds C/I: 0/0, Don. C/I: 0/0; Cfg/Thr SPM: ?/?, Cfg/Thr HPM: ?/? 0 rnds AV=1, ART=?
  work.mmcpool.com: On-line, Shares Submitted 0, Accepted 0
New MemoryCoin Block nTime 1389713471
New MemoryCoin Block nTime 1389713475
MMC Agg. SPM: ?, HPM: ?; Rnds C/I: 0/0, Don. C/I: 0/0; Cfg/Thr SPM: ?/?, Cfg/Thr HPM: ?/? 0 rnds AV=1, ART=?
  work.mmcpool.com: On-line, Shares Submitted 0, Accepted 0
Killed

Anyone who knows what to do? I am using the "yam-yvg1900-M7j-linux64-generic" download.



In the readme provided with the miner:

Code:
6. Known issues

If your miner shows "PTS Agg. CPM: ?; Rnds C/I: 0/0, Don. C/I: 0/0; Cfg/Thr CPM: ?/? 0" over an extensive period of time, consider restarting your miner.
In most cases this caused by problems with network connection or temporary pool server unavailability.
full member
Activity: 224
Merit: 100
January 14, 2014, 10:42:54 AM
#6
What version of the miner to use for an AMD FX-8320 Vishera?
I'm guessing yam-yvg1900-M7j-win64-bdver2.zip  bd stand for bulldozer right?
sr. member
Activity: 420
Merit: 250
January 14, 2014, 10:34:04 AM
#5
Hmm, since I already have a digitalocean account, I decided to try MMC mining. I am using a 2CPU/2GB node, but keep getting this:

Code:
Target OK
Checking params [mmc:av=1&aesni=on]...
Params OK
Checking MMC Stage 1 optimizations compatibility...
Checking MMC Stage 2 optimizations compatibility...
OK: MMC optimizations are compatible
MemoryCoin: Memory usage 1024M per thread, Algorithm Variation 1
Using 2 mining threads
  Will mine 2 rounds for miner developers to support development of the next version
  Follow @yvg1900 on Twitter to get information on new version availability on time
New MemoryCoin Block nTime 1389713466
MMC Agg. SPM: ?, HPM: ?; Rnds C/I: 0/0, Don. C/I: 0/0; Cfg/Thr SPM: ?/?, Cfg/Thr HPM: ?/? 0 rnds AV=1, ART=?
  work.mmcpool.com: On-line, Shares Submitted 0, Accepted 0
New MemoryCoin Block nTime 1389713471
New MemoryCoin Block nTime 1389713475
MMC Agg. SPM: ?, HPM: ?; Rnds C/I: 0/0, Don. C/I: 0/0; Cfg/Thr SPM: ?/?, Cfg/Thr HPM: ?/? 0 rnds AV=1, ART=?
  work.mmcpool.com: On-line, Shares Submitted 0, Accepted 0
Killed

Anyone who knows what to do? I am using the "yam-yvg1900-M7j-linux64-generic" download.
legendary
Activity: 854
Merit: 1000
January 14, 2014, 08:25:08 AM
#4
Is it time to fire up the xeons again?  Grin

Edit: yvg1900 in the mix: https://www.dropbox.com/sh/jvp4wwek8jpohj7/RlW6hzYqTz?tm=1389705870605

awwww yeah
newbie
Activity: 54
Merit: 0
January 14, 2014, 08:16:37 AM
#3
What CPUs are you using in the servers?  And where are you getting the servers from?  Just started testing this out myself now.
hero member
Activity: 572
Merit: 500
January 14, 2014, 05:17:40 AM
#2
dammmmm, whats the cost/rental to run that many servers
newbie
Activity: 8
Merit: 104
January 14, 2014, 04:45:02 AM
#1
Ω
Jump to: