Topic: BiblePay - New Coin Launch - Official Thread - page 60. (Read 119833 times)

I have a dual processor setup with two Intel Xeon CPU E5-2620s.. and I can confirm that is horrible. My regular desktops are performing better than the Xeon's.

I think u just answered your own question ^^"

Or if your still confused.. maybe check out http://biblepay.org

Maybe- Im checking that too.  I certainly see a similar net result, but, I dont see any errors in measurements in the code, so Ill keep checking.  Either way, the xeon multiprocessor is performing horribly.

No- we dont get anything, and have to "create" the project.  We only get something after sactuaries go live if the IT expenses are approved.

The idea of that coin is something new for me, but very exciting. Is bible coin related to any religion? I'm interested about impact on that currency whoch could it bring as a consequence. But hopefully wont be damaging.The fact that you count charity activity is very constructive, and for sure will bring much good for coin and needy. If that coin will help people, I hope we will be able to see more that projects, i approve it and your hard work.

We only fight for our right;)
 isn't devs get a small amount of coins from each block?

BTW. Hash2 has rised, but vpu usage now is jumping. Before was stable.
P.S. Thank You Bible_pay

Not a bad idea. On other pools, I always set my fee's to 2-5%.

Validated also. 4700-ish to 6100-ish.

You and your Oliwer mate are such slave drivers... i wouldnt want to work for u =D =D

How about , pay him? And he "might" actually focus on what "you" want. =D

but still isnt full power like on linux = ROBERT can you check continuously? there is still place for improvement

 

I can confirm that the hps2/shares is higher on my midrange single processor windows (7/8.1/10) machines after update. First guess is that my hash speed has risen by about 30%. Thank you for your strong commitment to this project!

biblepay: WIN WALLET HASH is FIXED ===== thaaanks +25%

Is there a real problem or is it a reporting problem.

My two nearly identical machines (the Linux has 2GB more RAM and runs on USB vs my Win7) show dramatically different hashps on getmininginfo (1.0.3.7) but pretty much do within about 10-15% actual shares/ H2pS.  I'm going to spin up another Win7 that will be exactly the same setup (same router even) as the Linux box (minus it running from HDD vs USB) and see how they compare.

linux 2core= 15594.65   30858.4 

the heavenly heavens

So it looks like we made a little progress on the windows side today- the newest version seems to improve the hash speed on windows single processor machines (and Probably the Ryzen), but it does not improve the multiprocessor machines (like the xeon) yet, feel free to download v1.0.3.8:
www.biblepay.org

Ill keep working on the xeon hash speed next.

I pretty much do the same on my GPU mining rig, I tested using a HDD and there was no noticeable impact on that, so followed the conclusion that it would likely be the same on this.  I've got a few extra HDD roaming around but I'm not seeing much difference in my debug log versus my windows box.  So again, I'm guessing the difference is normal, plus when I am currently looking at the leader board, I'm (at this moment) 15th on the Linux box in H2pS and to be grabbing much higher would actually seem incorrect as I know Tiras knows what they're doing and is pulling about the same.

Running Linux off a USB stick might actually be slowing you down.  I did for a while and noticed odd delays, which turned out to be I/O delays reading and writing from/to the USB stick.  In the end it fell over (although part of that may have been the age of the USB stick and heat causing contact issues on the connectors).

A spare hard drive or an SSD would almost certainly be better, or working out how some distros load themselves into a ramdisk before booting, so the USB stick is not required while it is running.

add new strong cpu linux machine

thanks for all  tiras,virus,UK guy

First off, I have a misnamed worker.  My W3503 is what the box started but I upgraded the chip to a X5650, at the time the pool was so flakely I was scared to change the worker name to reflect it's actual chip and then I forgot when the pool got more or less stable.  So my W3503 is a X5650 six core, 12 thread chip.  I also think the hashps in Windows is a bit off on the display side.  That w3503 worker (X5650 chip) runs Win7, shows 9KhS but routinely pulls 20KhS or better in H2pS (usually above 25K).  It's been running long enough to discount luck.  So in part, I think some of the issue is internal reporting is off, but your actual work rate is higher.

My actual Linux X5650 is xeon_linux01 and it shows 44KhS but only pulls 28K H2pS on average.  Sure, that's a decent margin, but I'm running a very stripped down version of Xubuntu off a USB, so I would expect a somewhat higher actual rate.  Windows is pretty bloated in that regard and it can only be a drag on performance.  So on average, my Linux box is about 15% faster.  That's certainly within the realm of reason.

The Dev is overloaded.  We basically have a one-man coin (he has some support from a few others but for the most part the buck stops with him).  Right now, the priorities should be wallet stability which is very dependent on pool stability.  And pool stability is very dependent on anti-DDOS.  So these are rightly the priorities.

In the end, you can install or upgrade the windows version in probably 3 minutes flat.  If you want to spend 2 hours to gain 15%, follow the Linux Installation instructions, hope you don't miss a step and go for it.  Until the wallet is stable though, you're looking at an hour or better every time there is a wallet update to upgrade your Linux.  But really, the difference is not nearly as significant in the real world as the hashps report shows.  So it's shouldn't be a very high priority against the weight of those other items.

it happened to me, i have a few machines and all point to one worker and hps still count as 1 machine