[PASC] PascalCoin: Induplicatable NFT

[adaseb]

R7-370's crashing without errors.

Looks like I can get 4 GPU's working per rig. After than the miner just crashes on the 5 and 6th ones.

My 370s didn't crash but my 7850 crashed.

Best cards were the 7970/280x, my R9 290 wasn't that much faster and no matter what I changed, I or C, it still hashed the same.

[mitache365]

Website? Slack? Something?

[aigeezer]

Thanks, Vorksholk!

My R9290X (Hawaii) crashed quickly for all intensities from 23 down to 10. It seems to be stable at intensity 4, running cool at over 80% GPU usage.

I'll start working intensity back upwards again Real Soon Now to find a sweet spot.

Edit: Update - intensity 9 looks stable, whereas 10 was not in an earlier test. I'll do more careful checking as time permits, but for now the sweet spot for this rig seems to be i9.

Edit: Update - I now think I was getting artifacts from another miner that had not completely shut down and which caused the crashes at higher intensities. I'm retesting everything.

[Eyedol-X]

Ok im out. Started mining when diff was 20GH/s and had 40% of the entire network hashrate and now at 100GH/s, not worth it without an exchange at least.

I'm out as well, pending an exchange add and a pool.

I succeeded in what I wanted to do though, learn how it all worked and earn a few thousand coins in the process. Hopefully this coin has some actual value some day and it will make it worth it.

I may dedicate 1 GPU on this for the time being but I'm still deciding.

Great job on the miner Vorksholk!

2x R9 390s mining for about 8 hours now without an issue here, not sure why everyone is getting crashes but the system I used is on Windows 7 and I'm using 16.7.3

[Aetsen]

I can only assume someone is attacking us again with huge hashing power. 4X increase in less than 3 hours. I put 6,000 MHs on mining and haven't gotten a block in 3 hours...

[Vorksholk]

Quick note for anyone wondering about my config that I tested this miner on:
Windows 7 x64 Ultimate
Latest AMD Driver (Crimson 16.10)
GTX 1080 and RX 480 installed (1080 attached to display, RX 480 not connected to display, platform 1)

RADEON 7970 are MURDERING this coin at 800MH/S

what intensity?

Everything is default.


Doesn't work on RX470/RX480

Does it give any kind of an error, or just crash?

I can get it to mine with 1 GPU, but the other 3 don't run. Trying to set up another user account now and syncing the wallet to test if it will work for targeting the 2nd GPU. Opening multiple miners with different device targets didn't work on an account together. The first card runs at 100%, but the others are at 0% doing nothing. Also noticed when I find a nonce it finds and submits the same nonce 6 or 7 times in a row.

If you're seeing duplicated nonces, try decreasing intensity and/or cyclesize.

So the benchmark only works on 1 GPU, cant test on any other GPUs.

Good catch--unfortunately probably can't fix for the next few days, since I'm traveling. If anyone else has Visual Studio or similar set up to compile AMD OpenCL host code (aka AMD app sdk installed, IDE configured), they could follow the instructions in the readme to make the benchmark version. Additionally, anyone can open the .cl file and change the single F in the targetY = 0xF0000000 line, which will make shares go 16x faster. Then mine with the normal miner, and divide the output number by 16 to get actual hashrate (but shares will be fast enough to work for benchmarking purposes, just not as nice).

This coin is very good with Tahiti based GPUs like 7950 / 7970 / 280X

Bad for 290 / 470 / 480 / Any Pitcairns

That's really weird--I developed and tested the miner only on an RX 480 system. Maybe those GPUs had better instruction sets for mining, or the compilers from OpenCL to device kernels were better optimized for mining on older devices where SHA-256 mining was more relevant and AMD developers wanted to get more of an edge for their cards to increase sales?

@Vorksholk
Hey mate , My 480s are crashing all the time , Ive tried everything ( underclock , undervolt , etc) , bur did not try change the driver.
Which driver you are using ?

Crimson 16.10, does it give any kind of error or just crash? Did you make sure your device and platform arguments were correct?

Thanks, Vorksholk!

My R9290X (Hawaii) crashed quickly for all intensities from 23 down to 10. It seems to be stable at intensity 4, running cool at over 80% GPU usage.

I'll start working intensity back upwards again Real Soon Now to find a sweet spot.

Edit: Update - intensity 9 looks stable, whereas 10 was not in an earlier test. I'll do more careful checking as time permits, but for now the sweet spot for this rig seems to be i9.

Huh, what hash rate are you getting at intensity 9 on that card?

help me please

i don't understand that characters..read the readme file of the miner.

install the latest java run time environment so you can run "PascalProxyv2.jar" and input the right syntax (it is in the readme.txt file)

I did everything according to the instructions but I did not start out miner

Yeah, the miner name you set in the Pascal Wallet needs to be 8 characters, it's not the name of the miner .exe file itself.

I can only assume someone is attacking us again with huge hashing power. 4X increase in less than 3 hours. I put 6,000 MHs on mining and haven't gotten a block in 3 hours...

Attacking... or just mining a ton for their own benefit. Probably just someone with a large AMD farm who wants some Pascal

------------------------------------------------------------------------------
For anyone interested, the current difficulty (2Cxxxxxx) implies a network hashrate of ~58 GH/s, and the current blockrate at this difficulty implies a hashrate of ~100 GH/s. That's about 512x the speed of a few weeks ago.

[go6ooo1212]

6 * 480 rig tested on win 8.1 x64.
There is no other display connected ,AMD platform 0 only.
Tested with 3 different drivers ( 16.9.2; 16.10.2; 16.7.4).
Tested with many values of Intensity 4,8,10,13,15,18,20,21,22,23,24,25
Results are the same - miner crashes by the time it gets a nonce.
When you increase the intensity above 23 , it could not submit a nonce
Ive tried to underclock also - same crash..

Am I missing something to try ?

EDIT: The NV 1070 rig is working pretty well with Miner 1.03...

[arielbit]

@ qwep1

how about the 8 characters of miner's name?

do you mean to rename miner

i just did that..

from readme..

3. Your miner name must be exactly 8 characters long. The miner expects that the input is exactly 176 total bytes (which is achieved by using a secp256k1 key and a 10-character name)  NOTE: NOT 10 like before! 8 characters, because the last two will be used to identify each GPU!

[adaseb]

6 * 480 rig tested on win 8.1 x64.
There is no other display connected ,AMD platform 0 only.
Tested with 3 different drivers ( 16.9.2; 16.10.2; 16.7.4).
Tested with many values of Intensity 4,8,10,13,15,18,20,21,22,23,24,25
Results are the same - miner crashes by the time it gets a nonce.
When you increase the intensity above 23 , it could not submit a nonce
Ive tried to underclock also - same crash..

Am I missing something to try ?

EDIT: The NV 1070 rig is working pretty well with Miner 1.03...

Flash your RX back to stock bios. Your hacked bios memory straps are probably messing this up.

[go6ooo1212]

6 * 480 rig tested on win 8.1 x64.
There is no other display connected ,AMD platform 0 only.
Tested with 3 different drivers ( 16.9.2; 16.10.2; 16.7.4).
Tested with many values of Intensity 4,8,10,13,15,18,20,21,22,23,24,25
Results are the same - miner crashes by the time it gets a nonce.
When you increase the intensity above 23 , it could not submit a nonce
Ive tried to underclock also - same crash..

Am I missing something to try ?

EDIT: The NV 1070 rig is working pretty well with Miner 1.03...

Flash your RX back to stock bios. Your hacked bios memory straps are probably messing this up.

It is with the stock BIOS...

[adaseb]

6 * 480 rig tested on win 8.1 x64.
There is no other display connected ,AMD platform 0 only.
Tested with 3 different drivers ( 16.9.2; 16.10.2; 16.7.4).
Tested with many values of Intensity 4,8,10,13,15,18,20,21,22,23,24,25
Results are the same - miner crashes by the time it gets a nonce.
When you increase the intensity above 23 , it could not submit a nonce
Ive tried to underclock also - same crash..

Am I missing something to try ?

EDIT: The NV 1070 rig is working pretty well with Miner 1.03...

Flash your RX back to stock bios. Your hacked bios memory straps are probably messing this up.

It is with the stock BIOS...

No idea then. I had crashing issues also. It mined for a few seconds and then crashed and I just switched to another GPU. Didn't want to flash back to stock bios.

[theboccet]

Hi Vorksholk!!!
Please can you give me some help to deal with this error ? "failed to start kernel -54"
seen it when i put pause in the bat file.

config : windows 10 build 1607 / GTX 750 / core i5-4460 3.20Ghz / 8Go Ram

Thank you

[arielbit]

after my dll problem is fixed and the miner started i had a crash in two of my 280x rigs, i just reduced the core clock to 1100 and 1110...no crashes since then, running for hours now...all are stable even the 4x 390 rig and the mixed 290x, 2x rx 480 and r9 390 rig

drivers are 15.7.1 and 16.8.1

like Vorksholk i'm on windows 7 64bit ultimate.

bios are stock, settings and intensity are default.

...the only problem i had was my desktop pc with my amd fury card not mining due to "The batch file cannot be found." in the proxy, i removed the fury card and mixed it with the 280x's rig and run pretty stable there.

[Ursul0]

6 * 480 rig tested on win 8.1 x64.
There is no other display connected ,AMD platform 0 only.
Tested with 3 different drivers ( 16.9.2; 16.10.2; 16.7.4).
Tested with many values of Intensity 4,8,10,13,15,18,20,21,22,23,24,25
Results are the same - miner crashes by the time it gets a nonce.
When you increase the intensity above 23 , it could not submit a nonce
Ive tried to underclock also - same crash..

Am I missing something to try ?

EDIT: The NV 1070 rig is working pretty well with Miner 1.03...

Same here. (Ive checked 16.9.1 and 16.10.2 full install/drv only) got constant crashes on 370 and 470. The rig I'm testing has few of them - all behave the same way... succeeded to get 1 block somehow, but it crashes every few minutes.
I strongly suspect it have something to do with the system... mine is Win10, AMD, ASRock FM2A88X PRO+ (not willing at this point to play with more stable rigs, working on ROI )

EDIT: I have tried single stock 370 (extremely stable card otherwise) with no OC - it's all the same - crashes in a few moments, sometimes submits something, mo.... it also NEVER gave me a proper hashrate(like during benchmarking) sometimes there's none, sometimes it's way off -


P.S. Don't mind me f*kng with it, I've tried in 'sterile' environment as well

[lobstermd]

6 * 480 rig tested on win 8.1 x64.
There is no other display connected ,AMD platform 0 only.
Tested with 3 different drivers ( 16.9.2; 16.10.2; 16.7.4).
Tested with many values of Intensity 4,8,10,13,15,18,20,21,22,23,24,25
Results are the same - miner crashes by the time it gets a nonce.
When you increase the intensity above 23 , it could not submit a nonce
Ive tried to underclock also - same crash..

Am I missing something to try ?

EDIT: The NV 1070 rig is working pretty well with Miner 1.03...

Same here. (Ive checked 16.9.1 and 16.10.2 full install/drv only) got constant crashes on 370 and 470. The rig I'm testing has few of them - all behave the same way... succeeded to get 1 block somehow, but it crashes every few minutes.
I strongly suspect it have something to do with the system... mine is Win10, AMD, ASRock FM2A88X PRO+ (not willing at this point to play with more stable rigs, working on ROI )

I fixed my crashing by changing the properties of the clminer to windows 7 and run as admin. 290 noce so far and not a single block lol.

[adaseb]

Hi Vorksholk!!!
Please can you give me some help to deal with this error ? "failed to start kernel -54"
seen it when i put pause in the bat file.

config : windows 10 build 1607 / GTX 750 / core i5-4460 3.20Ghz / 8Go Ram

Thank you

I had that on a 6990. Must be too old to mine with.

[aigeezer]

Thanks, Vorksholk!

My R9290X (Hawaii) crashed quickly for all intensities from 23 down to 10. It seems to be stable at intensity 4, running cool at over 80% GPU usage.

I'll start working intensity back upwards again Real Soon Now to find a sweet spot.

Edit: Update - intensity 9 looks stable, whereas 10 was not in an earlier test. I'll do more careful checking as time permits, but for now the sweet spot for this rig seems to be i9.

Huh, what hash rate are you getting at intensity 9 on that card?

Thanks - it was a red herring, I now think.

I hadn't looked closely enough at the outputs (doh) and was AFK for a while. At low intensities it would display huge real numbers from time to time, but never find a nonce or build up hashrate.
I switched back to high intensity tests and found the miner no longer crashed. I now think the crashes were caused by another miner that did not shut down completely before yours started.

All is well now - I'm back to running at I23, finding nonces, with hashrate fluctuating around 700-850 MH/s so far. GPU utilization is near 100% and temperature around 80C. I'll keep an eye on that, since I burned out a few graphics cards mining back in the day.

[Ursul0]

I fixed my crashing by changing the properties of the clminer to windows 7 and run as admin. 290 noce so far and not a single block lol.

Thx, it does the trick!

[Aetsen]

I'm still mining and getting nonces. Just no blocks being found. My miners seem to have an issue finding the same nonce multiple times and submitting it over and over. It seems odd even with a larger farm that this miti person is getting most of the blocks. Only about 80,000 MHs total right now and I am mining with 6,000-7,000 MHs of the networks total. This is an image of just one rig hashing with 4 GPUs. http://imgur.com/2o9jwng

[Vorksholk]

I'm still mining and getting nonces. Just no blocks being found. My miners seem to have an issue finding the same nonce multiple times and submitting it over and over. It seems odd even with a larger farm that this miti person is getting most of the blocks. Only about 80,000 MHs total right now and I am mining with 6,000-7,000 MHs of the networks total. This is an image of just one rig hashing with 4 GPUs. http://imgur.com/2o9jwng

Try adding the argument "c60" or "c30" to each one, do you still get duplicates?

The duplicates are caused by the miner not stopping to get work before exhausting the nonce range (and then starting over). Lowering the cycle count will let it get work more often.