damn just started my 1 7970 rig that i lost for awhile cause i change cpu from a k8 to k10 found a block within 5mins
When I rebooted my PC twice yesterday, I found a block in 12 minute after the second reboot. Solo mining makes people become superstitious lol
|
|
|
I can post settings for any card...lol
Just use a proper miner like yacminer 3.5 and off you go.
Question: what's that bunch of code ghostlander posted? :/
If you have something for the 290x (or just the basic setup with yacminer to work on that scrypt:64) Is there a benchmark mode ? I just mine to test... Use the regular scrypt flags, and select n factor 6. Something like this: yacminer --scrypt --nfmin 6 --nfmax 6 -R 11264 -g 1 --buffer-size 3600 --lookup-gap 1 --worksize 256 --gpu-engine 1000 --gpu-memclock 1375 --temp-cutoff 95 --gpu-fan 100 --auto-gpu --temp-target 85 --temp-overheat 90 You can also try -g 2 and see how it goes...
|
|
|
Maybe somebody can code a program that gives you the values and automaticly compensates for N factor changes? or would this be a giant coding job?
automatically gives you what values? like changes your configuration/settings based on n factor? yeah for example i say i have a r7950, and it provides my with my settings every time the N factor changes Could you put your config? i mean as an ultracoin service yes that could be done. might be better to generate conf files on the website - enter your config, display the proper settings. easy to set up and maintain, doesn't have to be distributed. I think it'll work for multi gpu setups as well...just have a 64bit OS and enough ram, say 8gb-16gb. I actually know how to tweak for each N factor thanks to Thirtybird, the yacminer developer. His readme cover this perfectly fine.
|
|
|
open sourcing it would be a problem. Last time we did this (heavycoin) the AMD guys caught up within a few days...
if it gives us an edge then screw open source, just give us the exe's have to find a way to do this without running afoul of the GPL Hmmm, primegpu sold licenses for x amount of coins. That's ok in my book, will keep the amd miners out for a while.
|
|
|
I can post settings for any card...lol
Just use a proper miner like yacminer 3.5 and off you go.
Question: what's that bunch of code ghostlander posted? :/
|
|
|
I'd reccomend against investing in cpu mining hardware at this point in time...gpu miners are out for xpm, and soon will be open sourced and massively deployed.
So, buy some xpm if you want cause price will soon increase as we get rid of the botnets at last!
|
|
|
I'll experiment with 13.9 and the latest drivers to compare performance and post results. 14.3 is running fine for me, not slow at all. I'll download it then...it will allow me to use mantle while at it :p
|
|
|
Sorry but that's a bunch of crap.
I can mine all n factors just fine on gpus that are out NOW. (Correct, you can get 2.5Kh/s on Yaccoin with 1 single gpu out of all made; What are you going to mine with in 55 days?) Even more so, higher n factors start to normalize hashing power between devices of different price ranges so in the end it's better for descentralization, takes the power away from the huge scrypt farm operators and back into the little guy with a cpu and a bunch of cheaper gpus. (at least till the big guys start to figure out they need different hardware for optimal results) (Huge farms leave if there's no profitability, I would not tout that as a plus. For the "little guy" he will need to buy new hardware too, GPU, CPU and / or RAM; why would the little guy with low funds put in more money when they can just mine 100's of others without?).
Your statements are not correct, and I can prove them wrong really easily, at least on the hardware side. (this post has no proof, so yes please provide refutation to my points) Yac has been around for a long time now, and it has battled the insta-mine burden for a while, lots of negative press...but it's still a pretty solid coin, and excellent for small time miners. (agree)
"-have no long-term network security" -> this just got invalidated. (how might I ask) Anyone can solo mine a coin with higher n factors, at least way easier than solo mining a scrypt coin or something like vertcoin which attracted all the farm dudes of the world (they are all lazy and have tons of outdated hardware at their disposal). (again, the fact that you can solo mine is NOT something you should tout as it's Bad. This meas poor network hash rate and thus poor network security.)
If you apply any of your "validation" to BTC or LTC then their values could not be where there are. Even at the NY conference there was discussion about BTC network security and if it can be attacked still. The answer was yes. That is a security issue and will always be there. LTC is at the point where it must have ASICS, so that it too can move forward in pricing and security. Any crypto that does not SUPPORT increasing hash rate cannot survive; N factor or no. A coin can have no real market cap without a corresponding amount of network security. Any coin that will be around in 10 years will keep this as a core principle. LTC will survive and they understand this; which is why they are NOT changing N factor. All of this is enough and we are not even talking about the many security shortcomings inherit with higher Nfactor People will need to learn how to tweak the miner for their gpus, most people are lazy, and super high end gpus are not as fast as they are with lower n factors, whereas lower end hardware fares a little better. What better than that for descentralization of the hash rate? EDIT: best settings for a 7950 using yacminer 3.5: yacminer.exe -o stratum+tcp://whatever:3333 -u ivanlabrie.1 -p x --scrypt-chacha --nfmin 4 --nfmax 30 --starttime 1388361600 --no-submit-stale --expiry 10 --scan-time 1 --queue 0 --gpu-engine 0-1100 --gpu-memclock 1500 --gpu-fan 0-100 --temp-target 75 --temp-cutoff 100 --temp-overheat 90 --auto-gpu --auto-fan --rawintensity 7168 --buffer-size 2560 -g 1 -w 256 --lookup-gap 2 https://github.com/Thirtybird/YACMiner/releases
|
|
|
Excellent work, meeting and update!! Very nice to see all the plans moving forward. Also it was nice to put a face to the bum; as it were. Sounds good bumface.
When the next nfactor hits, Can we expect cpu-mining to be a bit more realistic?
expect an N factor delay withing some time (extending the gpu era) -1 UTC being the only scrypt-chacha coin on its own schedule makes me sad. I was looking forward to seeing how your momentum did when people had to figure out how to mine at N=13 and 14... Excellent news, except for this part: "-N factor delay (at least 2 more years of GPU mining)"
WHAT!?
What exactly does that mean? I can gpu mine coins well into n factor 15...come on!
The nfactor changes was going to fast to early We need to be able to take ppl in july that gpu mine And the way UTC was going that would'nt happen. So bumface is rolling back to an earlier nfactor or just increasing the time before the next changes happen That sounds pretty bad to me...rolling back n factor ugh. Whatever, I hope it helps increase popularity but that kills it for me. I think it'll introduce a new generation of dumpers that won't give a dang for the coin, whereas higher n factors would keep dedicated miners onboard. The change was made to not follow the same fate as Yacoin. There is absolutely no reason to increase N to such high levels so fast. I guess thirtybird gave one " I was looking forward to seeing how your momentum did when people had to figure out how to mine at N=13 and 14" not sure I would want to design a coin "just to see if you could mine it". Higher N factors hurt network security more than a concern of miners dumping a few coins. Also, keeping lower N's will allow a larger base instead of just " keep dedicated miners onboard." UTC does not want to follow the detrimental path Yacoin is on and stuck with. Verification of this can be seen here: http://www.cryptocoincharts.info/period-charts.php?period=alltime&resolution=day&pair=yac-btc&market=cryptsyYou guys seem to want UTC to: -Drive away miners -Perpetually loose value rather than increase -Run N factors that are resistant to devices that won't be produced for years if not a decade. (n 14-15) -have no long-term network security If we wanted these aspects there is always Yac.......... Sorry but that's a bunch of crap. I can mine all n factors just fine on gpus that are out NOW. Even more so, higher n factors start to normalize hashing power between devices of different price ranges so in the end it's better for descentralization, takes the power away from the huge scrypt farm operators and back into the little guy with a cpu and a bunch of cheaper gpus. (at least till the big guys start to figure out they need different hardware for optimal results). Your statements are not correct, and I can prove them wrong really easily, at least on the hardware side. Yac has been around for a long time now, and it has battled the insta-mine burden for a while, lots of negative press...but it's still a pretty solid coin, and excellent for small time miners. "-have no long-term network security" -> this just got invalidated. Anyone can solo mine a coin with higher n factors, at least way easier than solo mining a scrypt coin or something like vertcoin which attracted all the farm dudes of the world (they are all lazy and have tons of outdated hardware at their disposal).
|
|
|
Excellent news, except for this part: "-N factor delay (at least 2 more years of GPU mining)"
WHAT!?
What exactly does that mean? I can gpu mine coins well into n factor 15...come on!
The nfactor changes was going to fast to early We need to be able to take ppl in july that gpu mine And the way UTC was going that would'nt happen. So bumface is rolling back to an earlier nfactor or just increasing the time before the next changes happen That sounds pretty bad to me...rolling back n factor ugh. Whatever, I hope it helps increase popularity but that kills it for me. I think it'll introduce a new generation of dumpers that won't give a dang for the coin, whereas higher n factors would keep dedicated miners onboard.
|
|
|
I'll experiment with 13.9 and the latest drivers to compare performance and post results.
|
|
|
Excellent news, except for this part: "-N factor delay (at least 2 more years of GPU mining)"
WHAT!?
What exactly does that mean? I can gpu mine coins well into n factor 15...come on!
|
|
|
No prob!
I gravitate to skein most of the time when mining myr too...groestl feels like I'm wasting power mining on a 280x lol
|
|
|
Well, I meant for MY rig and cards.
I don't think I can go higher on the core and maintain good thermals (would need higher voltage and I'm using a case).
|
|
|
I think I can beat this bad luck of mine...hvc clocks eh. So, core over core/mem ratios a la scrypt.
EDIT: Win 8 automatically syncs my time periodically.
all win does it can still get out of sync i never use time.windows.com always nist Yup, nist.gov here as well... 1075mhz core seems to be the sweetspot. I underclocked mem to 1250mhz and it resulted in a slight speed bump. From 2.5 to 2.55cpd with two cards.
|
|
|
I think I can beat this bad luck of mine...hvc clocks eh. So, core over core/mem ratios a la scrypt.
EDIT: Win 8 automatically syncs my time periodically.
|
|
|
I suck...or you need more than 2 cards for optimal results.
Question: what gpu clocks are you guys using? Stock, super stable oc?
This is a wild guess but i think you have always had bad luck in this game because you use the desktop you mine on which i think is dramatically reducing your chances That and the oc I was using might not be 100% stable for this...have no way to tell. I'll underclock to stock clocks me thinks, live with the 'lower' cpd. The dedicated rigs I have use linux (pimp distro ) so yeah, needed to use my desktop for this. EDIT: testing ideal clocks for scrypt, I guess it would show what my max stable core oc is...not sure if the ideal scrypt setup would be the same for xpm though.
|
|
|
the z87x-oc is one of my go to boards
|
|
|
I suck...or you need more than 2 cards for optimal results.
Question: what gpu clocks are you guys using? Stock, super stable oc?
|
|
|
|