[1200 TH] EMC: 0 Fee DGM. Anonymous PPS. US & EU servers. No Registration!

[rgod14]

Throwing 1GH your way. This pool looks VERY promising imo. 

[1713525807]

1713525807

[1713525807]

1713525807

[1713525807]

1713525807

[innervisi0nn]

am i seeing this right...i see a paypal option?

going to throw about 1.8 ghash/s towards this..only thing is blocks are taking long to solve (doh!)

[Patches]

Small glitch:

It seems your API is broken, at least for me. I get a similar JSON response no matter whether I use 'api.php?key=<yourapikey>' or 'api.php?key=<yourapikey>&poolstats=yes' etc. which seems to be a userstats report and the data returned does not appear to be for my account (the hash rate and number of workers are wrong).

Otherwise my workers seem to be mining just fine.

Edit:

Actually, I just checked and the API is reporting the same information as the webpage summary box which is incorrect. Even though I have only two workers registered, it is reporting 7 active miners and current speed 3X what is shown on the "My Workers" page. Still, I can't get the API to show anything different with '&poolstats=yes', '&userstats=yes' or '&blockstats=yes' they all just report something like this:

Code:

{"hashrate":"1.34 GH\/s","active_workers":"6","round_shares":"360252","round_duration":"1d 00:01:04","avg_shares_block":"0"}

Another edit:

Ah I see, that is actually the total poolstats. Where have all the miners gone? In that case, the problem is just that the API won't return the userstats or the blockstats for me.

[organofcorti]

1.5 hours on pacrim server last night with full 6990: 705 shares
2 hours 5 minutes on pacrim server tonight, half 6990 (the other is on ozco.in atm): 458 shares

These are a little bit on the low side,  at other pools I usually get around 850 to 950 shares/hour for full gpu, around 450/hour for half gpu.

I noticed tonight that although I go no invalids I did get 9 long poll exceptions after long polls, and the hash rate reported by the miner (poclbm) was a bit on the low side. I'm on deepbit with the same gpu right now, and am getting my usual hashrate.

I'll have another go tomorrow night.

Cheers

[rgod14]

Argh, woke up to connection problems

[Inaba]

Seems the US getwork server locked up last night with no explanation.  I'm looking into it and will be implementing something to restart it if it freezes like that in the future.  Sorry about that.

Organofcorti: I will look into that issue and get that resolved hopefully today.

Patches: I forgot to update the api information page.  The method is actually &action=poolstats, workerstats or userstats.

I've updated the api_info page with the appropriate information.

innervisi0nn: Yep, that's right, we have a direct Paypal payout option.  The blocks have been unkind to us the last couple blocks.

Everything should be working at the moment.  I'm going to make changes to the UK and PACRIM servers to speed them up today, I just need to rewrite some code and test it.

[The Koolio]

what are peoples connections and idles like? i might send 8 g hash your way if they are good. Im in uk

[Inaba]

I will be rolling out the change on the PACRIM server in the next two hours or so.  If you PACRIM people would report back if it's improved your speed, I would appreciate it.  If it works, I will roll it out to the EU server as well at that time.  I'll make another post when the update is complete... it's taking awhile to load the data a half world away!

[Inaba]

I made some changes on EU and PACRIM to speed things up, how's the speed going?

I have one more step that should increase response dramatically but it's taking a RREEAAALLYYY long time to load the data, so that's still in the pipeline.

[organofcorti]

I made some changes on EU and PACRIM to speed things up, how's the speed going?

I have one more step that should increase response dramatically but it's taking a RREEAAALLYYY long time to load the data, so that's still in the pipeline.

I'll let you know in about 10 hours (damn my work IT for not allowing SSH out!)

[organofcorti]

gpu1: 3.25 hours on pacrim, poclbm reporting ~ 375 Mhps: 732 shares (expected hashrate=268 Mhps)
gpu2: 3.25 hours on other pool, poclbm reporting ~ 365 Mhps: 999 shares (expected hashrate=366 Mhps)

The hashrate your site reported was similar to the expected hashrate based on shares/hour. There were only three long poll exceptions.

I also noticed that the gpu on eclipsecm would periodically drop in hashrate and then recover.

Maybe it's just bad luck on my end? It's been pretty similar each night though. Any ideas, Inaba?

[Inaba]

What was the time frame that you ran the miner?  I made some adjustments to some things, but I don't think it affected PACRIM. 

I finally got all the data loaded, took a lot longer than I expected, but I will bring that data online later tonight and that should speed things up or if not, it will tell me exactly where the latency bottleneck is. 

[organofcorti]

I ran the miners both from 20:05 to 23:35 AEST (10:05 to 13:35 GMT). It's weird that the miner seems to be hashing alright from the poclbm report, but just not producing shares. I'll have another go tonight on pacrim *and* us to compare them. It might just have been a string of bad luck.

[rgod14]

Solved 4th block an hour ago with 25 GH/s. Come join the fun! Seriously, I've been in almost all the pools and this is the one I see having the most promise. Plus the dude who runs it seems to be a nice, smart guy.

[kakobrekla]

Yeah, but no pay?

[Inaba]

Thanks rgod14!

Kakobrekla: No pay?  What do you mean?

Organofcorti: I think the problem is lying with database latency issues.  The US server is having some problems as well and I think I've isolated it to the latency.  I need to rewrite much of the work server share handling and then I will probably look at changing out the DB server to something more robust.  Well, actually the server itself is pretty robust, it's network latency that is causing the issue I think.

[rgod14]

Yeah, but no pay?

Gotta wait 120 blocks for confirmation.

[Inaba]

You can look at the current block stats to see how many blocks have been confirmed.

Eventually, I will make the option to avoid the 120 block delay, but in the beginning here, I can't afford the losses if we have an invalid block.  Eventually though if the pool is large enough I will do away with it.

[mmortal03]

I am seeing a bug when changing the automatic payout threshold.  After I changed it to 0.10, it changed itself to 0, and now won't change at all.

[Inaba]

Sorry about that!  I was treating that as an integer instead of a float, so anything less than 1 was getting set to 0.

It's fixed now, though.  You should be able to change it to a fraction of a btc, but no less than .02.

I apologize to the PACRIM people; I've been rewriting a huge chunk of the worker servers share and statistics code to speed things up dramatically and in the process I think I've identified what's been slowing down PACRIM.  It's late here, though and I don't want to push the new code to the servers unless I can watch it for awhile in case it crashes.  I plan on pushing it out tomorrow along with the original changes to speed things up... between the two it should make PACRIM just as fast as the US server hopefully.

I've already made some changes on the backend though and that might solve a good portion of the speed issues.