eg. all 9xx nvidia gpus start to tlb trash at 2GB despite having 4 GB+.
Are the 10xx gpus limited in the same way?
Can anyone run Genoils testprogram to verfiy that?
(would also be neat to run it on a RX 480)
Topic: ethminer-0.9.41-genoil-1.1 - page 30. (Read 397363 times)
Anybody tested these gpus if or when the TLB starts to trash and bottleneck the gpu??
eg. all 9xx nvidia gpus start to tlb trash at 2GB despite having 4 GB+.
Are the 10xx gpus limited in the same way?
Can anyone run Genoils testprogram to verfiy that?
(would also be neat to run it on a RX 480)
eg. all 9xx nvidia gpus start to tlb trash at 2GB despite having 4 GB+.
Are the 10xx gpus limited in the same way?
Can anyone run Genoils testprogram to verfiy that?
(would also be neat to run it on a RX 480)
Anyone know oldest supported driver for AMD tahiti 79xx cards?
14.7 are low hash, I presume only 15.12+
any hack to add some dll from 15.12 to genoil miner to fix hash speed?
14.7 are low hash, I presume only 15.12+
any hack to add some dll from 15.12 to genoil miner to fix hash speed?
On 280x I get over 25% more hashrate for Exp than for ETH,
on 370 likewise,
but on 380 and 380x (4gb) hashrate for Exp is equal with Eth hashrate..
whether the cause is cards architecture or something else?
on 370 likewise,
but on 380 and 380x (4gb) hashrate for Exp is equal with Eth hashrate..
whether the cause is cards architecture or something else?
I think it's due to how it accesses memory.
280x and 380 are different generations of GCN.
On 280x I get over 25% more hashrate for Exp than for ETH,
on 370 likewise,
but on 380 and 380x (4gb) hashrate for Exp is equal with Eth hashrate..
whether the cause is cards architecture or something else?
on 370 likewise,
but on 380 and 380x (4gb) hashrate for Exp is equal with Eth hashrate..
whether the cause is cards architecture or something else?
Be sure to set all variables:
GPU_FORCE_64BIT_PTR 0
GPU_MAX_HEAP_SIZE 100
GPU_USE_SYNC_OBJECTS 1
GPU_MAX_ALLOC_PERCENT 100
GPU_SINGLE_ALLOC_PERCENT 100
Ether dag size is now ~ 60% of 3GB. So 7950 shoeld be run with GPU_SINGLE_ALLOC_PERCENT 100. 2Gb cards need this parameter since Dag was ~ 1.4-1,5GB
GPU_FORCE_64BIT_PTR 0
GPU_MAX_HEAP_SIZE 100
GPU_USE_SYNC_OBJECTS 1
GPU_MAX_ALLOC_PERCENT 100
GPU_SINGLE_ALLOC_PERCENT 100
Ether dag size is now ~ 60% of 3GB. So 7950 shoeld be run with GPU_SINGLE_ALLOC_PERCENT 100. 2Gb cards need this parameter since Dag was ~ 1.4-1,5GB
When using these with setx they are saved in reg, no need to apply them every time?
Be sure to set all variables:
GPU_FORCE_64BIT_PTR 0
GPU_MAX_HEAP_SIZE 100
GPU_USE_SYNC_OBJECTS 1
GPU_MAX_ALLOC_PERCENT 100
GPU_SINGLE_ALLOC_PERCENT 100
Ether dag size is now ~ 60% of 3GB. So 7950 shoeld be run with GPU_SINGLE_ALLOC_PERCENT 100. 2Gb cards need this parameter since Dag was ~ 1.4-1,5GB
GPU_FORCE_64BIT_PTR 0
GPU_MAX_HEAP_SIZE 100
GPU_USE_SYNC_OBJECTS 1
GPU_MAX_ALLOC_PERCENT 100
GPU_SINGLE_ALLOC_PERCENT 100
Ether dag size is now ~ 60% of 3GB. So 7950 shoeld be run with GPU_SINGLE_ALLOC_PERCENT 100. 2Gb cards need this parameter since Dag was ~ 1.4-1,5GB
What is correct -SP for suprnova? Its not listed with --help.
Suprnova is closing their ethereum pool, so go mine elsewhereI wanted to mine Exp on suprnova, they claim to have stratum port for Exp.
what would be the cause for the miner to freeze after generating EXP dag??
I tried Claymores and thought it was miner´s fault. But after trying yours, it happens the same.
Some rigs do EXP fine, but others hangs after dag creation. (same OS/mem, same cards/drivers). 280x do fine.. 7950s seems to be more affected (although some 7950s work fine)
Any help or tip will be appreciated. Thanks!
I tried Claymores and thought it was miner´s fault. But after trying yours, it happens the same.
Some rigs do EXP fine, but others hangs after dag creation. (same OS/mem, same cards/drivers). 280x do fine.. 7950s seems to be more affected (although some 7950s work fine)
Any help or tip will be appreciated. Thanks!
I have some issue with dag creation on 7950 using claymore's miner on Exp, haven't tested with Genoil yet will look into that and report back.
Edit: Seems i have same issue with genoil dag creation on the 7950's 2/3 of the 7950's fail to create the dag error 4 something...
what would be the cause for the miner to freeze after generating EXP dag??
I tried Claymores and thought it was miner´s fault. But after trying yours, it happens the same.
Some rigs do EXP fine, but others hangs after dag creation. (same OS/mem, same cards/drivers). 280x do fine.. 7950s seems to be more affected (although some 7950s work fine)
Any help or tip will be appreciated. Thanks!
I tried Claymores and thought it was miner´s fault. But after trying yours, it happens the same.
Some rigs do EXP fine, but others hangs after dag creation. (same OS/mem, same cards/drivers). 280x do fine.. 7950s seems to be more affected (although some 7950s work fine)
Any help or tip will be appreciated. Thanks!
It's good to be without a CUDA device for a while. noticed that on Tahiti and Tonga, the VGPR usage exploded (112) somewhere along the line. That would also explaing the lower performance of my kernel against Clay (and Wolf). Guess I'll have to fix that, too.
What is correct -SP for suprnova? Its not listed with --help.
Suprnova is closing their ethereum pool, so go mine elsewhere 
so are the new pascal working good on win 7? or it will be done with 1.1.8 release?
What is correct -SP for suprnova? Its not listed with --help.
can't find any stratum support on their website. does it support stratum? otherwise just use the -F flasg as dispalyed in their help.
btw ran some more tests between 1.1.7 and upcoming 1.1.8. For some reason the speed improvement has vanished. It's weird because when I was building them, I saw the improvement, but when I take the old 1.1.7 and set it up against 1.1.8, they perform equally. And when I disable the improvements in 1.1.8, I get less performance than 1.1.7 while the code should be the same. I guess playing compiler is not always the best thing to do
Will release it soon then. Big boost in DAG creation on OpenCL for sure. From 7 down to 4 seconds on 7950.
What is correct -SP for suprnova? Its not listed with --help.
Can someone share how to use latest 1.1 branch with Dwarfpool?
Adding "how to with examples" to the first post (yes, Claymore style) will be very helpful.
Adding "how to with examples" to the first post (yes, Claymore style) will be very helpful.
-S eth-eu.dwarfpool.com:8008 -SP 1 -O 0xeb9310b185455f863f526dab3d245809f6854b4d.workername -SE [email protected]
I don't know what it is with dwarfpool but for some strange reason I get a lot of rejects with them.
Can someone share how to use latest 1.1 branch with Dwarfpool?
Adding "how to with examples" to the first post (yes, Claymore style) will be very helpful.
Adding "how to with examples" to the first post (yes, Claymore style) will be very helpful.
Forgot to implement a +2% speed increase (7950) that @Wolf0 hinted me at a while back. Will likely make it to 1.1.8
only for 7950 increase ? I liked your miner most, but
another miner is still about 1mhs/card better than yours
Went from 15.1 to 15.7 on ye olde 7950, so on a 390 it could be that MH that you were looking for. Don't know, we'll see once it is released. If it is released
Forgot to implement a +2% speed increase (7950) that @Wolf0 hinted me at a while back. Will likely make it to 1.1.8
only for 7950 increase ? I liked your miner most, but
another miner is still about 1mhs/card better than yours
getting this error with a 1070
Using device: GeForce GTX 1070 (Compute 6.1)
Cuda error in func 'set_constants' at line 143 : invalid device symbol.
-U -S coinotron
Using device: GeForce GTX 1070 (Compute 6.1)
Cuda error in func 'set_constants' at line 143 : invalid device symbol.
-U -S coinotron
Amph you have gtx1070 now ? .
no another user had this issue and i posted for him here, but i'm planning to buy one soon
Ok thanks for the info
. Is it possible to use this miner to mine eth clones eg expanse on suprnova?
Yes. Ethereum uses the Dagger algo, those all use Dagger as well.
Having trouble connecting to suprnova expanse, probably i have missed something simple in batfile any help appreciated.
ethminer -U -S exp.suprnova.cc:4000 -o mercuryminer.1:x
i 23:32:11|main Connecting to stratum server exp.suprnova.cc:4000
X 23:32:13|stratum Could not connect to stratum server exp.suprnova.cc : 400
0 , No connection could be made because the target machine actively refused it
ethminer -U -S exp.suprnova.cc:4000 -u
Still have same error using above format.
Jump to: