[ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX] - page 600.

Severaz

newbie

Activity: 48

Merit: 0

Quote from: cbuchner1 on April 09, 2014, 09:39:50 AM

Quote from: Severaz on April 09, 2014, 09:15:42 AM

adding more types of alogo's dosent accomplish anything in the long run

well we also support single-algo coins using Blake, Keccak (cudaminer), and FugueCoin, GroestlCoin (ccminer).

A bit of diversification is certainly a good idea, however the flood of un-original clone coins ("crap coins") is real problem. It makes it much harder for truly original concepts to succeed.

ya i was just saying that in the long run more algo's dosent help for asic resistance

the ltc baffles me when they first made the coin they were the asic answer now they embrace because its too hard.. the man got to them Cheesy

if they changed it they would loose faith because of the people that already had asics or have ordered them the price would drop.. if they didnt the price may stay the same or increase short term..

cbuchner1

hero member

Activity: 756

Merit: 502

Quote from: Severaz on April 09, 2014, 09:15:42 AM

adding more types of alogo's dosent accomplish anything in the long run

well we also support single-algo coins using the Blake and Keccak algorithms (cudaminer), and FugueCoin, GroestlCoin (ccminer).

A bit of diversification is certainly a good idea, however the flood of un-original clone coins ("crap coins") is real problem. It makes it much harder for truly original concepts to succeed.

Severaz

newbie

Activity: 48

Merit: 0

Quote from: cbuchner1 on April 09, 2014, 09:30:14 AM

Quote from: Severaz on April 09, 2014, 09:15:42 AM

a coin like vtc developers wont be scared to hard fork or whatever is needed to resist them and thats all thats really

really? ltc developers deemed it too risky to attempt (and they didn't see enough of a justification). A failed hardfork can completely ruin a coin.

yep thats the way they decided.. thats why i dont like ltc anymore no balls .. its a bitcoin clone not a alternative

Bearclaw

newbie

Activity: 52

Merit: 0

I tried the binaries posted by djm34 and it looks like I am getting a little better scrypt performance with the new version:

GTX 780Ti (~20-40 khash) improvement, it might be me but I think I'm seeing a little more CPU usage as well, but it's a little too early in the morning for me

https://i.imgur.com/x6fdFVi.png

However scrypt-jane appears to be broken:

https://i.imgur.com/aJloKrJ.png

cbuchner1

hero member

Activity: 756

Merit: 502

Quote from: Severaz on April 09, 2014, 09:15:42 AM

a coin like vtc developers wont be scared to hard fork or whatever is needed to resist them and thats all thats really

really? ltc developers deemed it too risky to attempt (and they didn't see enough of a justification). A failed hardfork can completely ruin an established coin.

Severaz

newbie

Activity: 48

Merit: 0

Quote from: cbuchner1 on April 09, 2014, 09:03:28 AM

Quote from: cdogster on April 09, 2014, 09:00:43 AM

Hey Christian,

Are there any other improvements that could be made for scrypt mining with the GTX 750 Ti's with the Maxwell chips? It seems like there is unlocked potential with these cards.

I couldn't find such potential yet. Wink

And face it: scrypt GPU mining will most likely be unprofitable in 3 months, even for the power-efficient 750 Ti's... we're looking to get alternatives working in time (Quark, X11, Heavycoin, possibly Riecoin too)

standard scrypt will be done but there will always be something to mine that will be profitable too many people that are mining looking for something.. just need to think outside the box .. something with a changing alogorithm not multiple will probably prevail as asics cant be made they will just change the algo..

a coin like vtc developers wont be scared to hard fork or whatever is needed to resist them and thats all thats really needed

adding more types of alogo's dosent accomplish anything in the long run

cbuchner1

hero member

Activity: 756

Merit: 502

Quote from: cdogster on April 09, 2014, 09:00:43 AM

Hey Christian,

Are there any other improvements that could be made for scrypt mining with the GTX 750 Ti's with the Maxwell chips? It seems like there is unlocked potential with these cards.

I couldn't find such potential yet. Wink

And face it: scrypt GPU mining will most likely be unprofitable in 3 months, even for the power-efficient 750 Ti's... we're looking to get alternatives working in time (Quark, X11, Heavycoin, possibly Riecoin too)

cbuchner1

hero member

Activity: 756

Merit: 502

Hmm, Linux compilation of cudaminer is borked.

cpu-miner.c won't build, most likely because Alexey used C++'isms... which works on Windows because I complile this module with the /TP flag in order to trick it into allowing inline delarations of variables (and other things) requiring C99 support which Visual Studio 2010 is lacking)

EDIT: it's fixed!

Christian

cdogster

newbie

Activity: 47

Merit: 0

Hey Christian,

Are there any other improvements that could be made for scrypt mining with the GTX 750 Ti's with the Maxwell chips? It seems like there is unlocked potential with these cards.

Thank you for your efforts.

ltcnim

legendary

Activity: 914

Merit: 1001

Quote from: cbuchner1 on April 09, 2014, 08:31:32 AM

Quote from: ltcnim on April 09, 2014, 08:09:32 AM

has anyone tried the 337.12 beta driver under linux? I just wanted to check it, since I thought the overclocking feature would be nice. however, my hashrate for scrypt-N went down from ~650khash/s to ~620khash/s and I was not able to figure out how to overclock the cards from CLI (running ubuntu 13.10 server).

I've inquired nVidia about the command line OC support. Let's see what they have to say...

Christian

great, thx!

cbuchner1

hero member

Activity: 756

Merit: 502

Quote from: ltcnim on April 09, 2014, 08:09:32 AM

has anyone tried the 337.12 beta driver under linux? I just wanted to check it, since I thought the overclocking feature would be nice. however, my hashrate for scrypt-N went down from ~650khash/s to ~620khash/s and I was not able to figure out how to overclock the cards from CLI (running ubuntu 13.10 server).

I've inquired nVidia about the command line OC support. Let's see what they have to say...

Christian

bigjme

sr. member

Activity: 350

Merit: 250

Quote from: restless on April 09, 2014, 08:19:15 AM

Guya, anyone tried to run cudaminer/ccminer as service?
I have PC which is set to shutdown at 18, but can safely auto-started at 19

, tho autologin is no go
ANy idea will the program function ok if started when no user logged in?

if the video drivers initiate then maybe

restless

legendary

Activity: 1151

Merit: 1001

Guys, anyone tried to run cudaminer/ccminer as service?
I have PC which is set to shutdown at 18, but can safely be auto-started at 19

, tho autologin is a no go
Any idea will the program function ok if started when no user is logged in?

ltcnim

legendary

Activity: 914

Merit: 1001

has anyone tried the 337.12 beta driver under linux? I just wanted to check it, since I thought the overclocking feature would be nice. however, my hashrate for scrypt-N went down from ~650khash/s to ~620khash/s and I was not able to figure out how to overclock the cards from CLI (running ubuntu 13.10 server).

ivanlabrie

hero member

Activity: 812

Merit: 1000

Excellent news! Cheesy

djm34

legendary

Activity: 1400

Merit: 1050

Quote from: cbuchner1 on April 09, 2014, 06:01:54 AM

Quote from: djm34 on April 09, 2014, 05:12:35 AM

Ok here are the binaries compiled on windows x86: http://www36.zippyshare.com/v/12355342/file.html
You will need the usual libraries to use it (I only included the exe and cudart dll). If you have no idea about what I am talking, please don't use it.

I'll will try to run some test a bit later.

edit: ok I can give several pool in arguments (this seems to work) however I don't know how to switch

I guess you'd have to DDOS one of the pools to find out , lol Wink

or just pull the network cable for a while, and plug it back in. Hopefully it would connect to the next pool in the list.

Christian

I don't have enough resource to ddos a pool...
Solomining on the wallet, then killing the wallet worked though

bigjme

sr. member

Activity: 350

Merit: 250

It will only switch if you get an connection error. Maybe try putting in wrong urls and see if it switches?

cbuchner1

hero member

Activity: 756

Merit: 502

Quote from: djm34 on April 09, 2014, 05:12:35 AM

Ok here are the binaries compiled on windows x86: http://www36.zippyshare.com/v/12355342/file.html
You will need the usual libraries to use it (I only included the exe and cudart dll). If you have no idea about what I am talking, please don't use it.

I'll will try to run some test a bit later.

edit: ok I can give several pool in arguments (this seems to work) however I don't know how to switch

I guess you'd have to DDOS one of the pools to find out , lol Wink

or just pull the network cable for a while, and plug it back in. Hopefully it would connect to the next pool in the list.

Christian

djm34

legendary

Activity: 1400

Merit: 1050

Ok here are the binaries compiled on windows x86: http://www36.zippyshare.com/v/12355342/file.html
You will need the usual libraries to use it (I only included the exe and cudart dll). If you have no idea about what I am talking, please don't use it.

I'll will try to run some test a bit later.

edit: ok I can give several pool in arguments (this seems to work) however I don't know how to switch
edit 2: I am certainly not doing things right...
I tried to test the failover.
So I put in first line, the wallet and in the second a pool.
Then I kill the wallet, however I just get connection refused, retry after 15 sec.
It does not seem to want to switch to the pool

edit 3: I didn't use "-r ".
okay the failover works. After killing the wallet and having retried one time (-r 1), cudaminer switched to the pool with correct user name and password

Regarding the error handling, I inadvertently tested it. I put a wrong argument in the command line, cudaminer started ignoring the wrong argument instead of closing (I personally prefer the old way, rather than starting with some wrong argument...)

cbuchner1

hero member

Activity: 756

Merit: 502

Quote from: bigjme on April 09, 2014, 03:39:52 AM

I will dig through the source code when i drag my ass out of bed and see what i can find. Will give it a shot. Any idea which file the new commands may be under?

cpu-miner.c has all the command line flags, also --help may be showing them already.

Here's the only bit of documentation I've received by EMail.

Quote

I think I’ve managed to implement proper failover support. It can switch between http, longpoll and stratum in any combinations, and currently supports 16 pools. When the last one stops responding, it can loop over the pools (when --loop-pools option is specified) or exit the app. To use it, just specify –o several times. It can handle same or different user names for pools.

Topic: [ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX] - page 600. (Read 3426922 times)