NSGminer v0.9.4: The Fastest NeoScrypt GPU Miner

[ghostlander]

Hello
Do you have a crypto currency Orbitcoin developing or forgotten about it?

The coin works. Installed a couple of seed nodes recently. Further development depends on other things to be completed first.

[ghostlander]

Uploaded NeoScrypt CUDAminer v1.0.1. It supports Kepler (GK110) in addition to Maxwells and Pascals. The original Titan, Titan Black, Titan Z, GTX 780 Ti and GTX 780. Keplers were never energy efficient, so the hash rate is about 400KH/s for 200W TDP. Well, it works at least.

[CryptoCoin101]

Uploaded NeoScrypt CUDAminer v1.0.1. It supports Kepler (GK110) in addition to Maxwells and Pascals. The original Titan, Titan Black, Titan Z, GTX 780 Ti and GTX 780. Keplers were never energy efficient, so the hash rate is about 400KH/s for 200W TDP. Well, it works at least.

It shows that NVIDIA Titan GPU is not for mining. Miners should maybe avoid it.

[phieudu2601]

Uploaded NeoScrypt CUDAminer v1.0.1. It supports Kepler (GK110) in addition to Maxwells and Pascals. The original Titan, Titan Black, Titan Z, GTX 780 Ti and GTX 780. Keplers were never energy efficient, so the hash rate is about 400KH/s for 200W TDP. Well, it works at least.

Thank you for updating CUDAminer. I will try for my RX470 to mine this coin (https://masternodes.online/currencies/NYX/). Hope it works well.

[ghostlander]

Uploaded NeoScrypt CUDAminer v1.0.1. It supports Kepler (GK110) in addition to Maxwells and Pascals. The original Titan, Titan Black, Titan Z, GTX 780 Ti and GTX 780. Keplers were never energy efficient, so the hash rate is about 400KH/s for 200W TDP. Well, it works at least.

It shows that NVIDIA Titan GPU is not for mining. Miners should maybe avoid it.

There are many ones in the Titan series. They are not bad, just expensive and their superior DP FP performance is of no value here. There isn't much difference between a Titan and a top level GTX of the same architecture in mining.

[ChekaZ]

Titan V is performing pretty well, giving nearly 3mh/s per card.

[ekavieka]

hi, I am using the nvidia gtx 1050 for the miner.
but when i run the .bat file . it just go blank before it force close.
any idea why? thanks so much

[ghostlander]

hi, I am using the nvidia gtx 1050 for the miner.
but when i run the .bat file . it just go blank before it force close.
any idea why? thanks so much

Run in text only mode with -T to locate the issue.

[ChekaZ]

hi, I am using the nvidia gtx 1050 for the miner.
but when i run the .bat file . it just go blank before it force close.
any idea why? thanks so much

add STOP ad the end in a new line of your .bat file, that will help to identify the problem.

[ku4eto]

Compiled it on a Ubuntu 16.04, using CUDA 9.0.
When running, getting awfully a lot of:

GPU0 t0: nonce *somehash* fails CPU verification!

Depending on the --mode used, i get a variable amount. --mode 1 -> doesnt even submit a single share, nvidia-smi shows 0% GPU utilization.
--mode 2 -> works, but does ~600kh/s.
--mode 3 -> works, and does 1mh/s.

Using Drivers 390, and Pascal GPU.

[ghostlander]

Compiled it on a Ubuntu 16.04, using CUDA 9.0.
When running, getting awfully a lot of:

GPU0 t0: nonce *somehash* fails CPU verification!

Depending on the --mode used, i get a variable amount. --mode 1 -> doesnt even submit a single share, nvidia-smi shows 0% GPU utilization.
--mode 2 -> works, but does ~600kh/s.
--mode 3 -> works, and does 1mh/s.

Using Drivers 390, and Pascal GPU.

There are many Pascal family GPUs. Every successful nonce reported by a GPU is verified by a CPU before being submitted to a daemon or pool. If the verification fails, there is a problem. Could be wrong settings, too much overclocking, inappropriate drivers, etc.

[ku4eto]

Compiled it on a Ubuntu 16.04, using CUDA 9.0.
When running, getting awfully a lot of:

GPU0 t0: nonce *somehash* fails CPU verification!

Depending on the --mode used, i get a variable amount. --mode 1 -> doesnt even submit a single share, nvidia-smi shows 0% GPU utilization.
--mode 2 -> works, but does ~600kh/s.
--mode 3 -> works, and does 1mh/s.

Using Drivers 390, and Pascal GPU.

There are many Pascal family GPUs. Every successful nonce reported by a GPU is verified by a CPU before being submitted to a daemon or pool. If the verification fails, there is a problem. Could be wrong settings, too much overclocking, inappropriate drivers, etc.

Seems like the issue is for some reason too low intensity. Anything below 13 and it starts failing. Also, anything higher than 16, just spams the GPU model.

[ghostlander]

Compiled it on a Ubuntu 16.04, using CUDA 9.0.
When running, getting awfully a lot of:

GPU0 t0: nonce *somehash* fails CPU verification!

Depending on the --mode used, i get a variable amount. --mode 1 -> doesnt even submit a single share, nvidia-smi shows 0% GPU utilization.
--mode 2 -> works, but does ~600kh/s.
--mode 3 -> works, and does 1mh/s.

Using Drivers 390, and Pascal GPU.

There are many Pascal family GPUs. Every successful nonce reported by a GPU is verified by a CPU before being submitted to a daemon or pool. If the verification fails, there is a problem. Could be wrong settings, too much overclocking, inappropriate drivers, etc.

Seems like the issue is for some reason too low intensity. Anything below 13 and it starts failing. Also, anything higher than 16, just spams the GPU model.

If either the miner is 32-bit or the GPU has less than 3Gb of memory, intensities higher than 16 are out of option.

[Cosmictino]

600 khs ?? sry.. but it looks for me like a rly bad joke   with NSGminer 0.9.4 my esault for each vega64 is ~80-100khs depends on powerlimit.  im running my rigs on Claymore's NeoScrypt AMD GPU Miner v1.2 with powlim -25,  and each vega64 is running on 1.7Mhs. together 1 rig with 6 cards  ~10.2 Mhs

[Borisov1974]

Really, as I did not try, I could not get a hashtag above 3.4 mh/s. The Claymore gives me 6.1 mh/s on the rig with 6 cards RX580 8Gb

[ZKxFTF]

Code:

OCL 0:                |   5.0   5.0   0.0 KH/s | A:0 R:0 HW:0 U:0.0/m

[21:46:17] Pool 1 priority changed from 0 to 1
[21:46:17] Probing for an alive pool
[21:46:17] The network difficulty has been set to 1319726
[21:46:17] Stratum from pool 0 detected new block
[21:46:17] Stratum from pool 0 requested work restart
[21:46:34] Stratum from pool 0 requested work restart
[21:46:47] Pool 0 is hiding block contents from us
[21:46:47] The network difficulty has been set to 958337
[21:46:47] Stratum from pool 0 detected new block
[21:46:55] Stratum from pool 0 requested work restart
[21:46:58] Stratum from pool 0 requested work restart
[21:47:01] Stratum from pool 0 requested work restart

please tell me, what am I doing wrong?


start.bat

Code:

nsgminer --neoscrypt -o stratum + tcp: // p2p-spb.xyz:2944 -u Your_GBX_PayoutAddress -p x

[ZKxFTF]

600 khs ?? sry.. but it looks for me like a rly bad joke   with NSGminer 0.9.4 my esault for each vega64 is ~80-100khs depends on powerlimit.  im running my rigs on Claymore's NeoScrypt AMD GPU Miner v1.2 with powlim -25,  and each vega64 is running on 1.7Mhs. together 1 rig with 6 cards  ~10.2 Mhs

You mean that Claimore is better?

[k0stas]

The speed for graphics cards will not be very important either, the first neoscrypt asic miner is out on the bittech site, if it's not fake.
https://bittech.cn.com/#hproduct4

[ghostlander]

The speed for graphics cards will not be very important either, the first neoscrypt asic miner is out on the bittech site, if it's not fake.
https://bittech.cn.com/#hproduct4

The miner can be customised to support ASICs as well. NeoScrypt may also evolve in the future.

[sorji077]

@ghostlander can u share the instructions file for windows build：

Native WIN32 build instructions: see windows-build.txt