Star65
Member
Offline
Activity: 109
Merit: 13
|
|
April 18, 2015, 07:25:57 PM |
|
Hhmm, on Quark it looks like Nvidia win the race. 280x - 4.4Mh/s, 750ti - 6.5Mh/s.
|
|
|
|
davidrentaobition
Newbie
Offline
Activity: 16
Merit: 0
|
|
April 19, 2015, 11:33:17 AM |
|
Hhmm, on Quark it looks like Nvidia win the race. 280x - 4.4Mh/s, 750ti - 6.5Mh/s. everyone forgot quark, please checking nicehash or westhash lists, one rig got 1G quark' hash rates!! BIG MACHINE!right?
|
|
|
|
z0n0
Legendary
Offline
Activity: 1274
Merit: 1006
|
|
April 20, 2015, 08:08:42 AM |
|
How much does a ATI 290 (noX) with the newest SGminer and .bin files make on: X11 ~ 8000 kh/s X13 ~ ? ? ? kh/s X15 ~ ? ? ? kh/s Quark ~ ? ? ? kh/s Thank you I think badman74 said he got 6.9MH/s on Quark with a 290. EDIT: Never mind, was Qubit. on quark it looks like i get 4.755mh/s on 290 (non x) at 1050/1500, worksize 256, threads 1, xI 64 Well, it still sucks, but it was a major pain in the ass to do it even approaching well on 15.3 drivers. 6.55MH/s out of 280X. Don't wanna work on it anymore for now. Full results for 270X, 280X, 290X, and 7950, plus clocks and other info (NSFW): https://ottrbutt.com/miner/quarkwolf-04182015.pngI was going to leave it alone, but I got a clever idea on a new approach to the problem. The branching in Quark make it a lot more interesting than Qubit, it made me think on it for a minute. So, I worked for most of the night on the major modifications required for the new approach, then fine-tuned the algos specifically for this Quark implementation somewhat - there's some more to do, a few more tricks up my sleeve I want to try out, plus some major code cleaning; I just wanted to get it working for now. 280X hashrate is 8.97MH/s, while temps stay in the mid 60C range on 60% fan. Other cards and more detailed results (NSFW): https://ottrbutt.com/miner/quarkwolf-04192015.pngAnd I reached 11.39MH/s on 280X with my third pass working on it, temps slightly higher - 69C on 60% fan now (NSFW): https://ottrbutt.com/miner/quarkwolf-04202015.png
|
|
|
|
winmkx
Newbie
Offline
Activity: 86
Merit: 0
|
|
April 20, 2015, 08:16:30 AM |
|
Hi!
Sorry, I'm new here.
I want to create my own coin.
What is the best GPU POW algorithm? Energy efficiency and fair distribution wise.
I know that there are huge farms on X11 algos, I'd like to avoid that.
Thanks in advance.
|
|
|
|
winmkx
Newbie
Offline
Activity: 86
Merit: 0
|
|
April 20, 2015, 09:37:07 AM |
|
Anything that's profitable, there will be huge farms on - you can make a PoW or pick one, it hardly makes much of a difference.
Thanks for your reply. But some algorithms, like Fresh or Groestl, are less popular. Yet more energy efficient. Also no multipools on them. Perhaps some other like these exist? Quark? I think that using less popular algo will improve the evenness of distribution.
|
|
|
|
pallas
Legendary
Offline
Activity: 2716
Merit: 1094
Black Belt Developer
|
|
April 20, 2015, 09:52:23 AM |
|
Anything that's profitable, there will be huge farms on - you can make a PoW or pick one, it hardly makes much of a difference.
Thanks for your reply. But some algorithms, like Fresh or Groestl, are less popular. Yet more energy efficient. Also no multipools on them. Perhaps some other like these exist? Quark? I think that using less popular algo will improve the evenness of distribution. I think it's the opposite: less popular = more chance of someone optimizing it badly and getting all the coins.
|
|
|
|
smokiepot
|
|
April 20, 2015, 12:09:18 PM |
|
how you do it ? wich settings you use ? stock kernel or you write your own one ?
|
|
|
|
Neosaan
|
|
April 20, 2015, 01:46:45 PM |
|
is it possible to hope that the data rate will be available to the common people? or are they only for the gods of coding?
|
|
|
|
z0n0
Legendary
Offline
Activity: 1274
Merit: 1006
|
|
April 20, 2015, 01:49:47 PM |
|
is it possible to hope that the data rate will be available to the common people? or are they only for the gods of coding?
Nothing's free in this world anymore my friend... ...but as they say: hope dies last. So...
|
|
|
|
Neosaan
|
|
April 20, 2015, 01:59:18 PM |
|
I understand that the work always has a cost. The issue price and when?))
|
|
|
|
Thirtybird
|
|
April 20, 2015, 07:06:24 PM |
|
is it possible to hope that the data rate will be available to the common people? or are they only for the gods of coding?
If you wait until he's squeezed all the profit out of it, he might then share some BIN's with you!
|
|
|
|
BitmoreCoin
|
|
April 21, 2015, 03:41:56 PM |
|
is it possible to hope that the data rate will be available to the common people? or are they only for the gods of coding?
If you wait until he's squeezed all the profit out of it, he might then share some BIN's with you! He has bills to pay.
|
|
|
|
lulu2003
|
|
April 22, 2015, 10:43:28 AM Last edit: April 22, 2015, 11:49:04 AM by lulu2003 |
|
are the sgminer v5 Win64 binaries other than the wolf's ones? expecting more perf from the 64bit openCL libs. edit: my error, new to GPU mining any idea with this when running with quarkcoin switch? [13:42:18] Error -11: Building Program (clBuildProgram) [13:42:18] fcl build 1 succeeded. fcl build 2 succeeded. Error: parse error. Error: parse error.
[13:42:18] Failed to init GPU thread 0, disabling device 0 [13:42:18] Restarting the GPU from the menu will not fix this. [13:42:18] Re-check your configuration and try restarting. Press enter to continue:
[13:42:26] thread_prepare failed for thread 0
C:\Users\FB\Downloads\sgminer_v5.1_2015-03-09-win32\sgminer_v5.1_2015-03-09>
|
|
|
|
jjiimm_64
Legendary
Offline
Activity: 1876
Merit: 1000
|
|
April 22, 2015, 12:41:52 PM |
|
Any chance someone with a working 290X binary for Neoscrypt could share it here?
Good work Wolf0, the best I'd seen previously was maybe 120 kh. You obviously have a talent at this.
not *.bin but *.cl i think, with this diff.rate of 12.1 optimization of neo-algo is usless for current coins...may be it will help to mine/sell new coins he will only sell, it in its priority only money ok. lets state the bounty for neo-kernel-mod-improvement may be 0.5btc? Forget it - I'm going to release the fix for 14.9. I just need to water down my own block mix - including chacha and salsa (lol, water down my salsa) - and the way it's called. The original dev fucked up those so bad they won't work on 14.x. Blake2S is so stupid, I'm amazed the dev can walk and talk at the same time, but hey, it does work on 14.9. does this mean sgminer will work with chacha? aka scrypt-jane? if not, are there any plans to add chacha to sgminer ?
|
1jimbitm6hAKTjKX4qurCNQubbnk2YsFw
|
|
|
Atomicat
Legendary
Offline
Activity: 952
Merit: 1002
|
|
April 23, 2015, 07:39:08 AM Last edit: April 23, 2015, 05:08:19 PM by Atomicat |
|
i getting 5.8mhs on qubit 1100/1500 what for a config u use for the 5.8 ?? I am getting 6.72mh/s on a 290x also 6.90mh/s on 290 Nice config. Try compiling with 14.9 drivers. My R9-290 gets 7.3 @ 1150/1375. Edit: Also a bit better with worksize 256.
|
|
|
|
Oscilson
|
|
April 23, 2015, 11:30:19 AM |
|
Just ran a couple of quick tests, the first on SGMiner 5, right from Github. The source I used was a bit old, because I forgot to pull before I did the test, so for reference, I used commit f27f8dd544a107523435363e7b26bfc294719542, pushed on Tue Jan 13 11:00:57 2015 +0100. I highly doubt anything in Qubit has changed since then, however. Anyways, the rig I tested on, Freya, currently has a 270X, 280X, 290X, and 7950. The cards are OC'd - card types, clocks, and memory types are in the notepad. Screenshot of stock SG5 on qubit (NSFW): https://ottrbutt.com/miner/qubitstock-03162015.pngI also spent maybe 2 - 3 hours or less on Qubit a while ago one afternoon. As such, the results are approaching barely acceptable in my opinion, but I did it just for the hell of it. Maybe I'll finish up with all the more obvious hashrate increases to be had at some point, then clean and polish the code, but not now. Nevertheless, here it is: Screenshot of modified SG5 with completely rewritten qubit kernel (NSFW): https://ottrbutt.com/miner/qubitwolf-03162015.pngWolf, do you have plan to release the optimized miner so that AMD cards can compete with nVidia cards?
|
|
|
|
Eastwind
|
|
April 23, 2015, 05:26:12 PM |
|
Not in SGMiner. When my own, fully custom-written one is in a usable state for end users, then it may be in there.
Are you going to release your miner publicly or privately paid for?
|
|
|
|
Oscilson
|
|
April 24, 2015, 08:31:36 AM |
|
Wolf0, when are you going to release it? How much is the fee?
|
|
|
|
dmatthewstewart
|
|
April 25, 2015, 11:28:31 PM |
|
Wow, that makes life really easy. Using my own pool but was going to set up a coinking profile since PiMP has that as a default listed pool. But for some reason though I dont think the MSI 290x Lightnings are used much for mining. I dont see any pre-configured settings for them. Coinking.io and others never seem to have my card listed. Of course, I was thinking about just getting a handful of 280x's or maybe even 7950's. Ive seen some 280x's pretty cheap and, tell me if Im wrong, but they seem to be the easiest and most miner friendly right now.
|
|
|
|
winmkx
Newbie
Offline
Activity: 86
Merit: 0
|
|
April 26, 2015, 10:06:48 PM |
|
Has anybody done per-algorithm* comparison of CPU and GPU megahashes? And/or the wattage on GPU.
Like for example: BMW 100MHS on R9 280X and 20MHS on Core i7 8 threads.
Which of the SPH algorithms are the best for CPU?
* By algoritm here I mean the primitive algos, like BMW, skein, luffa, etc. Not the combined ones like Quark, X11, X15, etc
|
|
|
|
|