[ANN][GRS] Groestlcoin | 1st to activate Segwit | Building Lightning Network - page 233.

srcxxx

sr. member

Activity: 266

Merit: 250

Quote from: ripzombie on April 14, 2014, 01:50:26 PM

r9 290 GPU , According to information on the page 9.8 MH/s ..
I have sapphire R9 290 GPU .. But 7.8 Mh/s Huh

Please Can you give r9 290 BAT file settings Huh

My Bat File

sgminer.exe -k groestlcoin -o stratum+tcp://erebor.dwarfpool.com:3345 -u FgYka3E3S9jXU7Q3iZtxtT43j4RdnnyrQP -p x -d 0,1,2 -I 18

You need to set gpu engine clock to like 1200. See my R9 290 config a few pages back in a post regarding BAMT setup: https://bitcointalksearch.org/topic/m.6030892

MinerSVK

member

Activity: 86

Merit: 10

Something wrong with cpu-pool.net , any worker - sgminer can't connect Huh

child_harold

hero member

Activity: 812

Merit: 1000

Attention DwarfPool MIiners

Please begin re-distributing your hashrate IMMEDIATELY
DwarfPool has way too much hashrate atm

So can some of you please move your mining to (in this order of priority):

http://grs.cryptohunger.com/ (fees go to coin promo)
https://www2.coinmine.pl/grs/
https://grs.suprnova.cc/
https://cpu-pool.net/grs/

all are good pools run by experienced Ops

It is important for GRS to have an evenly distributed hashrate

cheers

FiR3

full member

Activity: 192

Merit: 100

Quote from: srcxxx on April 14, 2014, 02:40:19 PM

Please join the new pool at http://grs.cryptohunger.com/

Vardiff starting from 4.

Multi core server

DDOS protected

0.8% proportional fee

Pool fees go towards GroestlCoin promotion

Just made this a status on #Groestlcoin (IRC).

child_harold

hero member

Activity: 812

Merit: 1000

Quote from: cbuchner1 on April 14, 2014, 02:43:09 PM

Quote from: child_harold on April 14, 2014, 01:36:48 PM

1. Are the hardware issues experienced with the random access from many threads something like "parallelisztion resistance"? Or in other words would that make this algo offer low(er) ASIC advantage compared to other algos in use? Would producing ASIC hardware therefor be more difficult and result in a lower / reduced ASIC advantage compared with SHA-256?

2. So NVIDIA will run 3x current speed with your "different approach"!?

3. Can you imagine an efficient ARM miner for this algo (a BOUNTY has been offered)

4) Any other info bout what makes this algo different (quirks, surprises etc) from others used in crypto mining would be enlightening Wink

thank you Christian for your generous contribution to this conversation

1) the implementer of the SPH library chose a table based approach because it works fast on CPUs. This doesn't work as fast on GPUs though. On GPUs explicit computations are usually faster than table lookups. It's just this particular approach taken that is somewhat resistant to parallelization.

2) yes.

3) maybe on 64 bit ARM platforms with 4 and more cores, using wide 128 bit SIMD operations (NEON?). I am not very familiar with ARM, so I can only speculate. In fact, our alternative approach would also apply here (and to Intel SSE2/AVX2)

4) it's very similar to AES, but somewhat upsized. The algorithm uses interesting maths based on Galois Fields GF(2^8).

Absolutely fascinating and thanks so much for your considered response.

just as a follow up to 1)… Do Bitcoin ASIC's use a table-based approach or specific explicit computations? I guess it's the latter. And can you speculate on how a Groestl ASIC might vary from a SHA-256 ASIC and the possibility that Groestl might offer a reduced ASIC advantage compared to SHA-256?

So…

ARM mining has potential (speculation appreciated) Lets get that ARM BOUNTY up!!

and…

your 3x NVIDIA performance news (that's insane!) is HUGE for Groestl!
When can we expect to see this?

So many many thanks
No more questions after this I promise

cryptonit

legendary

Activity: 3052

Merit: 1053

bit.diamonds | uNiq.diamonds

Quote from: cbuchner1 on April 14, 2014, 04:09:54 PM

Quote from: cryptonit on April 14, 2014, 03:01:06 PM

so if i understand cbuchner right its only a matter of time better performing miners on all platforms will be avaiable

it took us longer than a week and lots of debugging to come up with the new implementation. It took some serious brain skillz too. Wink

And no, we don't intend to open source this.

as i said only a matter of time

keep ur few weeks advantage i dont care for that

i care for longterm mining algo of diamondcoin and if a optimized amd miner would lose all power saving compared to scrypt

cbuchner1

hero member

Activity: 756

Merit: 502

Quote from: cryptonit on April 14, 2014, 03:01:06 PM

so if i understand cbuchner right its only a matter of time better performing miners on all platforms will be avaiable

it took us longer than a week and lots of debugging to come up with the new implementation. It took some serious brain skillz too. Wink

And no, we don't intend to open source this.

cryptonit

legendary

Activity: 3052

Merit: 1053

bit.diamonds | uNiq.diamonds

so if i understand cbuchner right its only a matter of time better performing miners on all platforms will be avaiable

will that reduce power savings compared to scrypt to zero or even optmized mining software will use less power?

we at diamond will for our coin feature relaunch choose a non scrypt algo

and groestl could be the new one

we want to avoid injection vulnerabilities of multialgos like x11 and twe
but we also want powersaving for our miner

cbuchner1

hero member

Activity: 756

Merit: 502

Quote from: child_harold on April 14, 2014, 01:36:48 PM

1. Are the hardware issues experienced with the random access from many threads something like "parallelisztion resistance"? Or in other words would that make this algo offer low(er) ASIC advantage compared to other algos in use? Would producing ASIC hardware therefor be more difficult and result in a lower / reduced ASIC advantage compared with SHA-256?

2. So NVIDIA will run 3x current speed with your "different approach"!?

3. Can you imagine an efficient ARM miner for this algo (a BOUNTY has been offered)

4) Any other info bout what makes this algo different (quirks, surprises etc) from others used in crypto mining would be enlightening Wink

thank you Christian for your generous contribution to this conversation

1) the implementer of the SPH library chose a table based approach because it works fast on CPUs. This doesn't work as fast on GPUs though. On GPUs explicit computations are usually faster than table lookups. It's just this particular approach taken that is somewhat resistant to parallelization.

2) yes.

3) maybe on 64 bit ARM platforms with 4 and more cores, using wide 128 bit SIMD operations (NEON?). I am not very familiar with ARM, so I can only speculate. In fact, our alternative approach would also apply here (and to Intel SSE2/AVX2)

4) it's very similar to AES, but somewhat upsized. The algorithm uses interesting maths based on Galois Fields GF(2^8).

srcxxx

sr. member

Activity: 266

Merit: 250

Please join the new pool at http://grs.cryptohunger.com/

Vardiff 1-1024 (good for CPU)

Multi core server

DDOS protected

0.8% proportional fee

Pool fees go towards GroestlCoin promotion

child_harold

hero member

Activity: 812

Merit: 1000

Attention DwarfPool MIiners

Please begin re-distributing your hashrate IMMEDIATELY
DwarfPool has way too much hashrate atm

So can some of you please move your mining to (in this order of priority):

https://www2.coinmine.pl/grs/
https://grs.suprnova.cc/
https://cpu-pool.net/grs/

all are good pools run by experienced Ops

It is important for GRS to have an evenly distributed hashrate

cheers

child_harold

hero member

Activity: 812

Merit: 1000

Quote from: ripzombie on April 14, 2014, 01:50:26 PM

r9 290 GPU , According to information on the page 9.8 MH/s ..
I have sapphire R9 290 GPU .. But 7.8 Mh/s Huh

Please Can you give r9 290 BAT file settings Huh

My Bat File

sgminer.exe -k groestlcoin -o stratum+tcp://erebor.dwarfpool.com:3345 -u FgYka3E3S9jXU7Q3iZtxtT43j4RdnnyrQP -p x -d 0,1,2 -I 18

for starters try upping the intensity to 20 or even 21

ripzombie

full member

Activity: 589

Merit: 100

r9 290 GPU , According to information on the page 9.8 MH/s ..
I have sapphire R9 290 GPU .. But 7.8 Mh/s Huh

Please Can you give r9 290 BAT file settings Huh

My Bat File

sgminer.exe -k groestlcoin -o stratum+tcp://erebor.dwarfpool.com:3345 -u FgYka3E3S9jXU7Q3iZtxtT43j4RdnnyrQP -p x -d 0,1,2 -I 18

child_harold

hero member

Activity: 812

Merit: 1000

Quote from: cbuchner1 on April 14, 2014, 12:25:04 PM

Quote from: child_harold on April 13, 2014, 04:51:04 PM

While you're here could you speak to the Groestl (Grøstl) algo's performance on NVIDIA compared with other algos
thanks,

Dorian (child_harold)

Groestl is quite slow because most implementations use a lot of random lookups in so-called T-Tables. Regardless whether you manage to get cached read access to these T-tables, or whether you put them into shared mem - the hardware still has issues with the random access from many threads. The same problem is seen on AMD devices, I suppose.

There are ways to work around these performance issues, and we have implemented one. I won't be more specific. Our different approach uses more power, but it is also up to 3 times as fast.

Christian

Thanks for the reply Christian:

3 questions please

1. Are the hardware issues experienced with the random access from many threads something like "parallelisztion resistance"? Or in other words would that make this algo offer low(er) ASIC advantage compared to other algos in use? Would producing ASIC hardware therefor be more difficult and result in a lower / reduced ASIC advantage compared with SHA-256?

2. So NVIDIA will run 3x current speed with your "different approach"!?

3. Can you imagine an efficient ARM miner for this algo (a BOUNTY has been offered)

Any other info bout what makes this algo different (quirks, surprises etc) from others used in crypto mining would be enlightening Wink

thank you Christian for your generous contribution to this conversation

cbuchner1

hero member

Activity: 756

Merit: 502

Quote from: child_harold on April 13, 2014, 04:51:04 PM

While you're here could you speak to the Groestl (Grøstl) algo's performance on NVIDIA compared with other algos
thanks,

Dorian (child_harold)

Groestl is quite slow because most implementations use a lot of random lookups in so-called T-Tables. Regardless whether you manage to get cached read access to these T-tables, or whether you put them into shared mem - the hardware still has issues with the random access from many threads. The same problem is seen on AMD devices, I suppose.

There are ways to work around these performance issues, and we have implemented one. I won't be more specific. Our different approach uses more power, but it is also up to 3 times as fast.

Christian

child_harold

hero member

Activity: 812

Merit: 1000

Quote from: feeleep on April 14, 2014, 08:03:32 AM

Just opened new pool (proportional, 1% fee) - for 100 first miners fee will be 0%

https://www2.coinmine.pl/grs/
Primary Node: mine1.coinmine.pl:3000

Currently looking for few miners to proof its working properly (and finding blocks).

feeleep

thanks feeleep - mining there now

Could a few of you please move your hashrate over to coinmine please?

help feeleep with his testing and spread the hashrate.

We now have 4 good pools

mikk2k2

full member

Activity: 126

Merit: 100

Think we'd be able to put together a multipool for GRS?

pallas

legendary

Activity: 2716

Merit: 1094

Black Belt Developer

Quote from: feeleep on April 14, 2014, 08:03:32 AM

Just opened new pool (proportional, 1% fee) - for 100 first miners fee will be 0%

https://www2.coinmine.pl/grs/
Primary Node: mine1.coinmine.pl:3000

Currently looking for few miners to proof its working properly (and finding blocks).

feeleep

I'm mining on it right now. Reported hashrate is too high and vardiff doesn't seem to kick in.

feeleep

legendary

Activity: 1197

Merit: 1000

Just opened new pool (proportional, 1% fee) - for 100 first miners fee will be 0%

https://www2.coinmine.pl/grs/
Primary Node: mine1.coinmine.pl:3000

Currently looking for few miners to proof its working properly (and finding blocks).

feeleep

child_harold

hero member

Activity: 812

Merit: 1000

3rd place taken, new BOUNTY for 4th!
Link to competition: https://bitcointalksearch.org/topic/groestlcoin-how-do-you-pronounce-groestlcoin-5000-grs-in-bounties-560587

This one comes from one of a Russian member of the community. It's got a fish, drum and bass and Groestlcoin and its pretty fucking funny - what else do you want??

Retweet, link etc etc if u likey Cheesy

~~3rd Video Submitted: 500 GRS BOUNTY~~ - lyubomyr

https://twitter.com/streamuse1/status/455663135111008256
https://www.youtube.com/watch?v=3QnhFeWeTAo

Topic: [ANN][GRS] Groestlcoin | 1st to activate Segwit | Building Lightning Network - page 233. (Read 558212 times)