Pages:
Author

Topic: [ANN] 1GH.COM - HVC / MAX / MMC / PTS / VIA Anonymous Pool ★ - page 5. (Read 75199 times)

full member
Activity: 154
Merit: 100
It is something wrong with the heavy pool ? Angry 4 hours and just an orphan block  Cry.

HVC is dead. Get out while you still can.

That was a week ago :p, I am not mining it at the moment but I don't think it is dead yet, just stale for a while :p.

Anyway thanks for the advice Smiley
legendary
Activity: 1022
Merit: 1010
It is something wrong with the heavy pool ? Angry 4 hours and just an orphan block  Cry.

HVC is dead. Get out while you still can.
hero member
Activity: 609
Merit: 500
DMD,XZC
hero member
Activity: 505
Merit: 500
sr. member
Activity: 266
Merit: 250
Just started mining pts here.
sr. member
Activity: 266
Merit: 250
Just started mining pts here.
sr. member
Activity: 364
Merit: 250
seems fucked today
full member
Activity: 154
Merit: 100
It is something wrong with the heavy pool ? Angry 4 hours and just an orphan block  Cry.
sr. member
Activity: 392
Merit: 250
New to HeavyCoin.... long time Scrypt/ScryptN/Momemtum miner though....

Big THANKS to 1GH for the effort in the cgminer-3.7.3-heavy and updating it so fast.
I decided to switch from ScryptN last night to give Heavy a shot. It's a VERY different
world tuning cgminer-heavy than other PoWs. Things I'm personally seeing :

1) memory clock doesn't seem to matter... at least that I can see as I'm mining and the cards report back 150Mhz
2) gpu clock is *very* touchy in relation to intensity. As for xintensity, i find that even more touchy.
3) extremely low power consumption while mining. impressive and yet I don't see the CPU doing much work either.

I'm running Ubuntu 13.04 and I use the GIT repository and compile my own binary for mining.
I'm using MSI R9 290 Gaming cards and although I *THINK* I've find a nice sweet hash rate,
I'd like to ask the community what other R9 290's have been getting. I'm currently getting 24Mh/s
on each of my cards at gpu 977Mhz, mem 150Mhz, I=12, T=1. Using W=64 as per suggested by 1GH.
I all feels to surreal if I'm running near peak for my rig since I'm using to tweaking higher clocks, memory,
and high I/XI. Many thanks in advance for advice and comparisons!!
here are some reference speeds for gpu's mining heavycoin:
http://heavycoin.wikia.com/wiki/Hardware_Comparison
newbie
Activity: 1
Merit: 0
New to HeavyCoin.... long time Scrypt/ScryptN/Momemtum miner though....

Big THANKS to 1GH for the effort in the cgminer-3.7.3-heavy and updating it so fast.
I decided to switch from ScryptN last night to give Heavy a shot. It's a VERY different
world tuning cgminer-heavy than other PoWs. Things I'm personally seeing :

1) memory clock doesn't seem to matter... at least that I can see as I'm mining and the cards report back 150Mhz
2) gpu clock is *very* touchy in relation to intensity. As for xintensity, i find that even more touchy.
3) extremely low power consumption while mining. impressive and yet I don't see the CPU doing much work either.

I'm running Ubuntu 13.04 and I use the GIT repository and compile my own binary for mining.
I'm using MSI R9 290 Gaming cards and although I *THINK* I've find a nice sweet hash rate,
I'd like to ask the community what other R9 290's have been getting. I'm currently getting 24Mh/s
on each of my cards at gpu 977Mhz, mem 150Mhz, I=12, T=1. Using W=64 as per suggested by 1GH.
I all feels to surreal if I'm running near peak for my rig since I'm using to tweaking higher clocks, memory,
and high I/XI. Many thanks in advance for advice and comparisons!!
hero member
Activity: 518
Merit: 500
Will there be a HVC miner optimization for non-GCN cards, ie. HD 5850, 5870, 6950, 6970, etc? They run slow, barely 4 to 4.5 mhs.

Any plan to optimize the PTS miner?
For HVC, probably not, unless AMD fixes their compiler for VLIW, which is unlikely.

For PTS, yes, but it is quite low on priority list.

Looks like AMD isn't focusing on improving VLIW any time soon. Mantle and the open API is their focus with the GCN technology.

It would be cool to see a new PTS miner.

What cards are suggested to use the "VIP" password? Would it be worth to use it on a single 7970? I also have a 7850 too and would that work ok with that?
sr. member
Activity: 462
Merit: 250
Will there be a HVC miner optimization for non-GCN cards, ie. HD 5850, 5870, 6950, 6970, etc? They run slow, barely 4 to 4.5 mhs.

Any plan to optimize the PTS miner?
For HVC, probably not, unless AMD fixes their compiler for VLIW, which is unlikely.

For PTS, yes, but it is quite low on priority list.
hero member
Activity: 518
Merit: 500
Will there be a HVC miner optimization for non-GCN cards, ie. HD 5850, 5870, 6950, 6970, etc? They run slow, barely 4 to 4.5 mhs.

Any plan to optimize the PTS miner?
member
Activity: 64
Merit: 10
Quote
GPU 0  57,0C 2420RPM | 20,13M/20,1Mh/s | A:126 R:0 HW:0 U: 1,02/m I:0
GPU 1  52,0C 1995RPM | 20,13M/20,1Mh/s | A:122 R:0 HW:0 U: 0,99/m I:0

with:

Quote
"api-allow" : "W:127.0.0.1,W:192.168.0.247",
"api-listen" : true,
"api-mcast-port" : "4028",
"api-port" : "4028",
"gpu-engine" : "1100",
"gpu-memclock" : "900",
"gpu-memdiff" : "0",
"gpu-powertune" : "20",
"gpu-threads" : "1",
"gpu-vddc" : "0.000",
"no-submit-stale" : true,
"temp-target" : "72",
"temp-overheat" : "85",
"temp-cutoff" : "90",
"temp-hysteresis" : "3",
"xintensity" : "7",
"worksize" : "64"

here
sr. member
Activity: 462
Merit: 250
Anybody getting more than 18.36 Mh/s out of a 280X? I've got two Asus Direct CUII TOPs, currently running these settings:

-g 1
--xintensity 105
-w 256
--gpu-powertune -20
--gpu-engine 1150
--gpu-memclock 1600
I'd suggest '-w 64' (register pressure is high with this kernel)

Hi, until now I follow other peoples advice and put random numbers and testing the best results, what is the reason to set -w 64, what does it mean "register pressure is high with this kernel"?
A thorough explanation would require quite a lot of introduction, let me just say that the opencl kernel uses too many registers and that makes -w 64 setting optimal for AMD (at least on GCN and with modern enough Catalyst, 13.11 and newer).

Ok, this doesn't explain much :p. Anyway I don't wan't to make you to waste much time, but I am quite interested in learn how this thing works, seems to be quite profitable Wink . Do you have any good opencl book to recommend or should I just try to follow the docs ?

Thanks!
AMD OpenCL reference, especially its chapters about optimization and hardware reference appendix is surprisingly good (given the quality of compiler itself and its support). Other than that, maybe just google and AMD support forums may be of value.
full member
Activity: 154
Merit: 100
Anybody getting more than 18.36 Mh/s out of a 280X? I've got two Asus Direct CUII TOPs, currently running these settings:

-g 1
--xintensity 105
-w 256
--gpu-powertune -20
--gpu-engine 1150
--gpu-memclock 1600
I'd suggest '-w 64' (register pressure is high with this kernel)

Hi, until now I follow other peoples advice and put random numbers and testing the best results, what is the reason to set -w 64, what does it mean "register pressure is high with this kernel"?
A thorough explanation would require quite a lot of introduction, let me just say that the opencl kernel uses too many registers and that makes -w 64 setting optimal for AMD (at least on GCN and with modern enough Catalyst, 13.11 and newer).

Ok, this doesn't explain much :p. Anyway I don't wan't to make you to waste much time, but I am quite interested in learn how this thing works, seems to be quite profitable Wink . Do you have any good opencl book to recommend or should I just try to follow the docs ?

Thanks!
sr. member
Activity: 462
Merit: 250
Anybody getting more than 18.36 Mh/s out of a 280X? I've got two Asus Direct CUII TOPs, currently running these settings:

-g 1
--xintensity 105
-w 256
--gpu-powertune -20
--gpu-engine 1150
--gpu-memclock 1600
I'd suggest '-w 64' (register pressure is high with this kernel)

Hi, until now I follow other peoples advice and put random numbers and testing the best results, what is the reason to set -w 64, what does it mean "register pressure is high with this kernel"?
A thorough explanation would require quite a lot of introduction, let me just say that the opencl kernel uses too many registers and that makes -w 64 setting optimal for AMD (at least on GCN and with modern enough Catalyst, 13.11 and newer).
full member
Activity: 154
Merit: 100
Anybody getting more than 18.36 Mh/s out of a 280X? I've got two Asus Direct CUII TOPs, currently running these settings:

-g 1
--xintensity 105
-w 256
--gpu-powertune -20
--gpu-engine 1150
--gpu-memclock 1600
I'd suggest '-w 64' (register pressure is high with this kernel)

Hi, until now I follow other peoples advice and put random numbers and testing the best results, what is the reason to set -w 64, what does it mean "register pressure is high with this kernel"?
full member
Activity: 164
Merit: 100
Should I be using the -p VIP feature?  I have a 7970 and 7770.  Mining HeavyCoin.
Probably yes, especially if you run them together.

Yes I do, in the same rig.  Will change it once I get home.  Thanks for the fast response.
sr. member
Activity: 462
Merit: 250
Should I be using the -p VIP feature?  I have a 7970 and 7770.  Mining HeavyCoin.
Probably yes, especially if you run them together.
Pages:
Jump to: