Sounds like to me you have no clue what you're talking about because he's not wrong.

Looks like the kernels have become popcorn.

Delicious.

As a continuation of my AHPPlus report using v1.3/1.3.2:

48h: Accumulated another 1.25977 mBTC, making earnings approximately identical to what NH would've given me.
72h: Added another 1080Ti and accumulated another 1.64016 mBTC, which is about 0.03 mBTC more than what NH would've given me.

With that being said, I'd say my confidence interval in this case is relatively low. My previous testing a few weeks back with AHP24h led me to profits that were nowhere close to estimates. I haven't tried the regular AHP batch file, but I've found that AHP24h profit estimates are a good approximation (once you reach high trust levels) of what you should expect to earn.

Yessir.

To update anyone who was following my long-ass report, I'm scheduled to make much more on AHP today, probably looking at 1.35-1.40 mBTC depending on luck, which is MUCH closer to default estimates than what I experienced in my other 96 hours.

In an attempt to see if I could match earnings estimates, I decided to hop onto ahashpool for 24hrs. I had spent 72hrs on AHP before, and the results were less than satisfactory. That test is documented in my older posts if anyone's interested.

Equipment:

- 3 1080Tis @ 200w
- 1 1070Ti @ 150w

I downloaded the release of NPlusMiner 1.3 and I was on my way. I selected xevan, hsr, phi, tribus, x17, Nist5, Lyra2RE2, blake2s, and neoscrypt as my algos. I also selected skein, but again, it never stopped benchmarking so I just ditched it because I'm impatient. The benchmarking process yielded me 0.000061901 BTC; this amount will be subtracted from my first 24hrs later.

- My switching gain was 4%
- I suffered about 30 minutes of downtime
- The test started (benchmarking finished) on Feb 21 9:35pm. I stopped it at 10:35pm of the following day to account for the downtime.

I waited another 30 mins for any stray earnings to register. My total earned was 0.00118892 BTC. After subtracting the earnings from the benchmarking section, I'm left with 0.001127019 BTC. Let's break this down to what each card amounted to approximately:

- 1080Ti: 0.3 mBTC*3 cards = 0.9 mBTC
- 1070Ti: 0.22 mBTC*1 card = 0.22 mBTC

I haven't dabbled with any other Yiimp-based pools except for Zergpool briefly; most of my mining is from NH. In the same time frame, I've earned about 0.34-0.37 mBTC for each 1080Ti and about 0.23-0.26 mBTC for each 1070Ti. My actual earnings are usually within very close proximity to estimates. The greatest outlier I've seen was less than 5%. This leads me to my conclusion.

The good: After reaching 100% trust level in NPlus, the "average BTC/D" reading is extremely accurate. I still enjoy the UI, there's something classic about it.

The bugs: Skein benchmark never stops for me, this has happened to me twice in isolated situations.

The switching behaviour is also very strange. The first time I boot after finishing benchmarks, my sleep time is 30 secs. X17 is the most profitable at this time. As X17 warms up during this minute, the program sees that X17 is the most profitable algo, yet, it'll restart X17 and I gotta warm up again. Talking about warming up, without changing the default intensity, I warm up starting from around 53mh all the way to 70mh. Once in a while, I'll warm up from 64mh to 70mh. Not sure what's causing the latter, but I wish I could reproduce it. Takes about 10 mins to warm this damn thing up if I have to start from 53mh.

After my first round with 30 sec sleep time, it turns to 180 secs. I didn't touch the "-interval" option at all. It's not even in my batch file. HSR is now the most profitable algo. I think to myself "alright, the program probably just thinks it's temporary, so it hasn't switched algos". I wait for 20 minutes. The switcher hasn't refreshed once ever since sleep time turned to 180 secs. The computer isn't frozen and neither is the miner. I think to myself again "ah fuck it, X17 will probably be on top soon enough anyways, I'll just leave it and let's see how long it takes". Almost 2 hours pass and I can safely conclude it's probably a glitch at this point. I manually restart. After the reboot, we're back to 180 sec sleep time; the miner refreshes and switches properly now.

The ugly: Outside of the BTC/D from reaching 100% trust, the earnings estimates are still wildly inaccurate from my experience. I'm not sure why this is, but fuck AHP is all I'm saying.

Thanks for reading.

As requested (kinda).

http://www.strawpoll.me/15122211

Just took the latest files off your repo (the makings of 1.3 I'd assume) and tested it with a 1070Ti.

Bench-marking of skein and tribus for ZergpoolPlus went on for 20-30 mins before I manually cancelled them. X17 and NIST5 working fine so far. I'm wondering if this is because there's no "-interval" command in the start.bat file (was I supposed to re-insert that?). Also profits seem stupid-high right now, let's see how accurate they are.

Also is it possible for me to just replace Palgin's "hsrminer_neoscrypt.exe" with just_a_miner's fork (rename it to the same .exe)? Or will I have to fiddle with the Powershell files too? On second thoughts I should just try it for myself.

For what it's worth, I've had a fairly good experience with HSRminer, particularly with Just_a_miner's fork. I like to restart it once every 8 hours or so and it'll often produce the hash rates I see on client-side.

I've heard good things about KlausT if you're willing to mess with intensity, but I haven't tried that myself. Stock KlausT is underwhelming IME. NH's Excavator (1.4.4 I believe?) is also excellent for Neo when it comes to 1080Tis/TitanXPs. Not sure what their licensing details are, but might be worth replacing HSRminer if it's (relatively) easy to integrate.

Just had a chance to give this a shot after the rave reviews. Here are my brief, anecdotal stats of 6x1070Tis running @ 85% TDP:

PalginHSR: 8.18kH/s
NH's newest version of Excavator: 8.1kH/s
Klaust 8.18 (NHML 1.8.1.: 7.7kH/s
J_A_M HSRfork (high intensity): 8.23kH/s

Of course, this is before any fees, so you can really see J_A_M's fork and Excavator pull through as my primary miners. I also favor HSR's UI. YMMV with hash rates (as seen throughout all of these miner threads), so please do your own testing and check pool-side rates.

@Just_a_miner: Please consider adding an ETH wallet

Thank you for the testing, always good to have another data point. However, I think 200 minutes is far too short. 12hrs would be much more appropriate, or better yet, 24hrs.

Well I guess this has become the ahashpool discussion thread to some degree.

On my end, it appears they've removed their 0.005 BTC transfer option after ~10 days, despite the fact that fees haven't increased drastically since a few days ago. Guess I'm stuck here with my pants down until I reach 0.01 BTC again (they transferred me 0.005 BTC 2 days ago). Perhaps I'll email them and see if they can make an exception.

EDIT: I've contacted them and apparently they're still doing 0.005 BTC minimum transfers every 10 days, they just don't explicitly state it on the website for whatever reason.

Didn't really see a change in performance with my single V56 (figured I'd just throw it in for testing's sake) as performance went up by 0.2% (well within margin of error). However, I really like what you've done with the wealth of features (in particular, the restart function if it drops below a hash threshold) and the UX/UI looks relatively aesthetic compared to other miners I've used.

Definitely a project to be proud of and something I'd gladly pay the 0.65% fee for.

Thanks for the insight. I was wondering if you ever just see flat lines on your profitability graph. I'm working with about 147 and 57 Mh rigs, and there are stretches of 30-60 mins where my graph completely flatlines. This never happens with me when I'm just mining Blake2s, so if this inconsistency is a regular occurrence (because I don't have enough computing power), I might just point it at B2s unless Phi is significantly more profitable.

https://i.imgur.com/nJWAbno.jpg

~11hr (~6.75h to ~16.25h on the graph) test for hsrminer on Nemosminer.

Ignore the rollercoaster at the 5h mark, I was playing around with settings. The drop at 17h is a result of a 30min internet outage, this screws up the following moving average.

As you might be able to see, the relevant moving average is about 2820-2880kh. I'd say it's closer to a 2850-2860kh average if you hover above the data points. This is server-side on AHP24h (pure Neoscrypt). Client-side, I'm averaging 2930-2950kh.

So where'd the extra 100kh go? Take into account a 1% fee and that leaves us at 2900-2920kh. From here we're still missing a bit. I'm going to assume AHP takes the 1% from your actual profits, and doesn't just cut your hash rate accordingly. The only logical explanation is my 99% efficiency, which would take us to 2871-2890kh. Still missing 10-20kh, but in the grand scheme of things, that's pretty negligible. I've seen a lot of people claim hsrminer doesn't deliver the promised client-side numbers, but my experience differs. I'm not trying to discredit any claims, but I would encourage everyone to try it and see whether it works for them.

Stats in the Powershell window are fairly accurate, may be the 1070Ti should be closer to ~142w.

Specifics:

1070Ti @ 85% TDP, 80c target, +200 core, +400 memory (may be I should clock this down too), ~58c @ 65% fan
1080Ti @ 85% TDP, 80c target, +190 core, +100 memory (+300 mem didn't really help for me in Neo), ~52c @ 85% fan

Ambient temp is ~21-22c.

All in all, I'm much more concerned about my Blake2s hashrates being quite different from what I'm getting client-side.

Yeah, pretty much. Basically someone bids for a small amount of NIST5 in your case, and NHML flings a bunch of computers to NIST5 to fill the order. Problem is, if you switched, you're competing with everyone else now. The NH system gets gamed like this by buyers fairly often. That's why I was saying limiting Nemos to just Neo/Equihash (usually) so you can enjoy the extra gaps of lowered difficulty when everyone leaves temporarily, while not having to "warm up" for Neo again.

Got an MSI Air Boost V56, does 935/935 hash @ 950 HBM through PPT mods. Draws way less than 150w (130w if hwinfo's anything to go by). XMR-STAK intensity is 1940/1940 IIRC. Ambient temperature usually 65f or lower, cards never go above 57c under those conditions.

I have 2x Asus STRIX V56s and 10x Sapphire Pulses coming in. Think I should just sell them and find some 1080/1080Tis for relatively normal prices? Pretty tired of maintenance and tweaking on the AMD side. It's fun for the first couple cards, then it just becomes burdensome. Alternatively I could just hoard the cash and wait for Volta/Ampere?

Desktop:

CPU: Intel Core i7 6700k @ 4.0GHz
RAM: 8GB DDR4 @ 2400MHz
GPU(s): 1x Nvidia 1080Ti, 7x Nvidia 1070Ti

OS: Windows 10
Browser: Google Chrome