Quote from: Amph on July 05, 2017, 08:24:59 AM

Quote from: bensam1231 on July 05, 2017, 03:15:24 AM

Quote from: Silver166 on July 05, 2017, 01:33:16 AM

Hello everyone! I've been using alexis miner for a while and it gave me 1300 k/h on my gtx 1070 for neoscrypt Cool

. Few days ago I bought two 1080 ti, and I find out that they do only 550 k/h with alexis miner. When I switch to tpruvot ccminer 2.0 I start getting 1000 k/h for 1070s and 1600 k/h for 1080ti. Is there any way to fix this abnormally low 1080 ti hash rate for alexis miner? Huh

1080 and 1080ti have problems with like half the algos for Nvidia, which is why most people don't buy them.

half...again with this, it's like only 2 algo worth mentioning if there are other they are on shitcoin not worth mining

...dagger... ...monero... The high volume markets on dagger and monero... I'm drawing a blank on those, I literally cannot see what would be worth mining with those algos. I like how you don't actually understand how mining works yet love giving advice on mining. Pretty much any memory hard algo suffers the same fate. NeoS is a memory hard algo. I guess all coins are created equal though and no one uses those algos. Weird, we're already at 3 if you're looking 'purely' at algos.

Maybe 'algos' was the wrong word since there are so many algos people don't even know and mine with anymore, lets go with half the coins worth mining, and 4/5 of the volume for gpu cryptos. And that doesn't include the actual performance you're getting on algos that aren't completely crippled running on GDDR5X.

Quote from: Silver166 on July 05, 2017, 08:26:58 AM

Quote

1080 and 1080ti have problems with like half the algos for Nvidia, which is why most people don't buy them.

I wouldn't say so, neoscrypt mining is the only problem I got, every other algos doing just fine with 1.7x-1.9x of 1070s hashrate.

You're supposed to get +85% performance. That's in a absolutely best case scenario as it's how many more cuda cores a 1080ti has over a 1070, there is no +90% unless the 1070 you're basing it off of isn't the best in the lot. Out of all the algos I've tried Lyra2v2 is probably the one that scales the best, some you'll see anywhere from +50-70% of a 1070s performance. And that's not in the memory hard algos.

Here lets go down the list.

Bitcore is 58% faster 1070 to 1080ti
x17 is 56% faster
Tribus is 65% faster
Skein itself is 65% faster
Blake14r(Decred) is 69% faster
Equihash is 51% faster
Library is 68% faster
Lyra2v2 is 69% faster

I don't really care what BS you guys are smoking and how much you hate, you don't need to believe me and instead listen to the one guy that OC'd the crap out of his card and somehow managed to hit 1k on skein for five seconds before his system hard locked and his risers started on fire. All of this is publicly available. You can look at testing on CMB, Whattomine, or Epsylons benchmark part of his site. This also coincides with my personal testing.

If there are some sort of other 'good' algos, feel free to add them to the list. That is literally everything worth mining. While a 1080ti isn't crippled, in other words only getting 30% faster performance or sometimes even negative in the case of NeoS, it's no where close to +85%.

You guys never cease to amaze me. These are the sort of people you're listening to for investing +$10k. Hell the very fact that they haven't spent any surmountable amount of time themselves finding this stuff out makes me wonder if they're absolutely retarded and don't know what they're doing or they really don't have more then a couple cards. Who the hell spends thousands of dollars on business upgrades and doesn't even take the time to benchmark stuff and look around for results that confirm what they got?

http://yiimp.ccminer.org/bench

And 'every other algo' pretty sure you're excluding some there... like dagger...

You know what, why don't you guys give me a algo that IS actually hitting +85% over a 1070 with a 1080ti. Lets talk about those.

Oh hey look Shub did some actual research, just read your post after I wrote this... Also coincides largely with my results. If you guys didn't know Epsylon keeps a large database of benchmarks on his site so it doesn't depend on the '1 guy that OC'd skein to 1k and everyone wonders why they don't get it.

Quote from: lexele on July 05, 2017, 11:55:48 AM

Well how did you compare? Because at least on skein, with same consumption, 1080ti is >100% of a 1070

Your 1070 sucks and/or you suck at OCing/cooling/power delivery/everything it takes to be a miner.