sp_ (OP)
Legendary
Offline
Activity: 2926
Merit: 1087
Team Black developer
|
 |
October 07, 2017, 06:16:46 PM |
|
Probably someone else would wrote the xevan code,
xevan has been harvested for months with a descent profit. The profit is now gone. So the big farms that invested in the development of a private xevan miner are loosing alot of money right now. The profit used to be 3x per card. Feels like a robbery indeed... Should we blame krlx? He just took the bounty that was there for him to be taken, and he doesn't share the profit with the real developers. Low moral, but not a criminal. The big farms just lost $5 a card per day and they have more than 100 cards. The only way to keep the profit is to optimize the code. Again.
|
|
|
|
scryptr
Legendary
Offline
Activity: 1798
Merit: 1028
|
 |
October 07, 2017, 06:30:55 PM |
|
The Xevan miner was just released. Another alexei rippoff with some added new padding bytes. The miner was profitable for 6 hours. and when the windows binary was released, there was no profit left. So the opensorce miner is now worth nothing. Krlx got his bounty of 0.5btc, and with your logic, he should share the bounty with the real developers of the ccminer right? 99% of the code is written by me and the other devs, he just changed a few lines of code to claim the bounty. 2-3 days of work. 0.5btc. Nice inside job..
YOU MAY BE RIGHT ABOUT PROFITABILTY-- I mined for months with CPUminer-OPT, and quietly stacked BitSend away. Occasionally, I would mine with the SGminer for Xevan, on 4 x RX 470 cards. That would calculate at about 5+ MH/s, total. BitSend would peak every so often, and my BitSend would sell at 2-3 x the normal market price. My 1080ti is now doing almost as much hash as my RX 470 4-card rig, about 4.9MH/s at "-i 23". Most of the hash on YIIMP is now from KRNLX's code, and I think the market will suffer. If the "mine and dump" strategy is employed by the majority of the miners, it may be a long time before another price peak. My 1080ti does fine with KRNLX's code, but my 960 cards kept crashing. If I set the intensity for those cards to "-i 16", they were stable for ~10 minutes, but still crashed. I hope that KRNLX will refine his work. --scryptr
|
|
|
|
BitTeo
|
 |
October 07, 2017, 06:36:19 PM |
|
Gigabyte 1070 g1 gaming (micron memory) Bitcore sp-mod #3  Bitcore Sp-mod #2 is crashing on the same clocks and intensity and I can max do 20.8 on the same card with all the settings I have tried. I work at pc while mines can not put intensity at 25, but +750 memory is not too much ? .
|
|
|
|
abudfv2008
|
 |
October 07, 2017, 06:53:14 PM |
|
Probably someone else would wrote the xevan code,
xevan has been harvested for months with a descent profit. The profit is now gone. So the big farms that invested in the development of a private xevan miner are loosing alot of money right now. The profit used to be 3x per card. Feels like a robbery indeed... Should we blame krlx? He just took the bounty that was there for him to be taken, and he doesn't share the profit with the real developers. Low moral, but not a criminal. The big farms just lost $5 a card per day and they have more than 100 cards. The only way to keep the profit is to optimize the code. Again. So as I said before. The "dirty" work is done. Now sp_ can make some tricks and release an optimised version. But I think optimised version already exist - someone was mining with 7 rigs with avg speed of 58Mh/s - it is more than 10*1080ti per rig with krnlx version.
|
|
|
|
skdsw
|
 |
October 07, 2017, 07:26:49 PM |
|
SP where is my bitcore SP 3? I am tired to ask you to send it ) You already confirmed my mail What else I need to do to get sp3? ))))
|
|
|
|
Amph
Legendary
Offline
Activity: 3276
Merit: 1072
|
 |
October 07, 2017, 07:35:17 PM |
|
Gigabyte 1070 g1 gaming (micron memory) Bitcore sp-mod #3  Bitcore Sp-mod #2 is crashing on the same clocks and intensity and I can max do 20.8 on the same card with all the settings I have tried. on lower tdp the diff is minimal between 2 and 3, not worth it, i tried and i get +2mh only on all 6 card but the fee kill it...
|
|
|
|
fr4nkthetank
Legendary
Offline
Activity: 2294
Merit: 1182
Now the money is free, and so the people will be
|
 |
October 08, 2017, 12:19:23 AM |
|
"So the big farms that invested in the development of a private xevan miner are loosing alot of money right now. The profit used to be 3x per card. Feels like a robbery indeed..."
screw the big farms, if they want to invest money to dev a miner, they take a risk. morals ? morals ? wow. just wow. so its all libertarianism and capitalism, but once you get a private miner people go full commie I swear to god...
|
|
|
|
hanskan
|
 |
October 08, 2017, 01:39:28 AM |
|
Gigabyte 1070 g1 gaming (micron memory) Bitcore sp-mod #3  Bitcore Sp-mod #2 is crashing on the same clocks and intensity and I can max do 20.8 on the same card with all the settings I have tried. any chance you could post hasrates for 1060 6gb, 1070 and 1080 ti with 70% tdp ?
|
|
|
|
pallas
Legendary
Offline
Activity: 2716
Merit: 1094
Black Belt Developer
|
 |
October 08, 2017, 03:38:11 AM |
|
The Xevan miner was just released. Another alexei rippoff with some added new padding bytes. The miner was profitable for 6 hours. and when the windows binary was released, there was no profit left. So the opensorce miner is now worth nothing. Krlx got his bounty of 0.5btc, and with your logic, he should share the bounty with the real developers of the ccminer right? 99% of the code is written by me and the other devs, he just changed a few lines of code to claim the bounty. 2-3 days of work. 0.5btc. Nice inside job..
You clearly didn't look at the code for more than a couple seconds. Or maybe not at all.
|
|
|
|
GeeTee-X
Newbie
Offline
Activity: 38
Merit: 0
|
 |
October 08, 2017, 10:07:09 AM |
|
It looks as if mod#3 does not read the intensity value from the config file. When I use the exact same config file on mod#2 then it does work. Can you fix this?
|
|
|
|
du44
|
 |
October 08, 2017, 10:55:37 AM |
|
I have made a xevan miner now. How fast is the opensource and on wich launchconfig?
6.3 mh 1080 ti oc'ed with cuda 9.0 x32
|
hello
|
|
|
sp_ (OP)
Legendary
Offline
Activity: 2926
Merit: 1087
Team Black developer
|
 |
October 08, 2017, 11:37:08 AM |
|
but what about the 1070. 1060 6g and 1060 6gb? Here are some preresults. xevan sp-mod 1 (100% tdp) 1. gtx 1070 g1 gaming (3850 KHASH) 2. gtx 1060 windforce 6gb (2550 KHASH) 2. gtx 1060 gainward 3gb (2420 KHASH) 
|
|
|
|
rednoW
Legendary
Offline
Activity: 1510
Merit: 1003
|
 |
October 08, 2017, 12:09:58 PM |
|
but what about the 1070. 1060 6g and 1060 6gb?
Here are some preresults. xevan sp-mod 1 (100% tdp)
1. gtx 1070 g1 gaming (3850 KHASH) 2. gtx 1060 windforce 6gb (2550 KHASH) 2. gtx 1060 gainward 3gb (2420 KHASH)
public miner does 4mhs @ 109W (60%tdp) gtx1080 Throw you overpriced overclocked gtx1070 to the toilet ))
|
|
|
|
sp_ (OP)
Legendary
Offline
Activity: 2926
Merit: 1087
Team Black developer
|
 |
October 08, 2017, 12:21:37 PM |
|
public miner does 4mhs @ 109W (60%tdp) gtx1080
So you are saying that the higher model, more expensive card is only 3.8% faster than the gtx 1070 running the sp-mod?
|
|
|
|
rednoW
Legendary
Offline
Activity: 1510
Merit: 1003
|
 |
October 08, 2017, 12:26:01 PM |
|
public miner does 4mhs @ 109W (60%tdp) gtx1080
So you are saying that the higher model, more expenisve card is only 3.8% faster than on the gtx 1070 with the sp-mod? It is 50% more efficient with only 10% price gap. One can spend 0.05btc better to buy gtx1080 or rx vega56 instead of buying gtx1070 and paying you. And will get much more profit with lower power consumption ))
|
|
|
|
sp_ (OP)
Legendary
Offline
Activity: 2926
Merit: 1087
Team Black developer
|
 |
October 08, 2017, 12:53:18 PM |
|
If you read the topic of the other thread, krlnx's version is doing 3.3MHASH on the gtx 1070(NOT profitable). Mine is doing 3850 (+16.66%) (profitable). Thats why I ask the members of the forum to test and verify my assumptions.
|
|
|
|
rednoW
Legendary
Offline
Activity: 1510
Merit: 1003
|
 |
October 08, 2017, 12:56:36 PM |
|
If you read the topic of the other thread, krlnx's version is doing 3.3MHASH on the gtx 1070. Mine is doing 3850 (+16.66%). Thats why I ask the members of the forum to test and verify my assumptions.
Ok, I made a test with my palit jetstream micron memory, 100 %tdp. My own compile cuda 7.5 x64 3.73mhs palgin compile cuda 9 x86 3.86mhs So you can go and work further. You gain nothing.
|
|
|
|
bakednblazed
Newbie
Offline
Activity: 10
Merit: 0
|
 |
October 08, 2017, 12:56:53 PM |
|
If you read the topic of the other thread, krlnx's version is doing 3.3MHASH on the gtx 1070(NOT profitable). Mine is doing 3850 (+16.66%) (profitable). Thats why I ask the members of the forum to test and verify my assumptions.
yeah, but all you're doing is relying on cuda9 to make your 'existing' *cough* code quicker. no skill there.
|
|
|
|
sp_ (OP)
Legendary
Offline
Activity: 2926
Merit: 1087
Team Black developer
|
 |
October 08, 2017, 01:00:28 PM |
|
I use cuda 7.5.
|
|
|
|
sp_ (OP)
Legendary
Offline
Activity: 2926
Merit: 1087
Team Black developer
|
 |
October 08, 2017, 01:01:19 PM |
|
Ok, I made a test with my palit jetstream micron memory, 100 %tdp. My own compile cuda 7.5 x64 3.73mhs palgin compile cuda 9 x86 3.86mhs
On what clocks and config?
|
|
|
|
|