[XPM] Primecoin Built-in Miner Sieve Performance Issue

[n4ru]

Here is a replacement for release5, again using the official codebase.

https://www.dropbox.com/s/76eij0zcjx091am/release11_sandybridge_i5.zip

I will pull down the old links now.

Any chance at a Linux compile?

[1714794019]

1714794019

[1714794019]

1714794019

[1714794019]

1714794019

[Zalfrin]

Here is a replacement for release5, again using the official codebase.

https://www.dropbox.com/s/76eij0zcjx091am/release11_sandybridge_i5.zip

I will pull down the old links now.

Any chance at a Linux compile?

If you're running linux you really ought to be able to compile it yourself. https://github.com/primecoin/primecoin

[wibtc]

Found more orphans than blocks using release 8, prime per second is much faster through.
Very difficult to found a block now.

[BitcoinFX]

Here is a replacement for release8 which uses the official codebase. I will have a replacement for release5 momentarily as well. Those were the most popular, if there are others people want rebuilt let me know.

https://www.dropbox.com/s/lxm220qqztpgps5/release12_x86-64.zip

Thanks!

Seems good to go. Cheers.

[Zalfrin]

Please report on which build you are experiencing intermittent synchronization problem.

I got two rig trying to run this build:
https://github.com/Zalfrin/primecoin.git

One is fine, the other stuck at block no.2. This about 5 hours ago.

So I switch back to original git, which is perfect.

Yeah, don't use my github right now until I've figured out what is causing the sync issues.

[Sunny King]

I don't have any agenda towards you or your coins.

I have an odd mixture of appreciation for your work, and being appalled by the entire presentation and follow through.  

It's clear you've done this far more casually than most people could. When I read the original announcements, I was enthused. I read that at ppcoin forums the day you posted it becuase I was looking for more information about primecoin and it's work. I waited eagerly for the release, and I've mined the coin on several boxes along with my other activities.  I mean thanks for helping me pay my bills.

Just sort of tossing an unfinished product out there and practically walking away from it is pretty shocking. The sort of "figure out yourself" attitude was surprising, for someone pushing out a change like this.  To come out and bitch at the people trying to sort it out, after all that, is really just the cherry on top.

Looks more and more shady.  Like the market ploy is more important, to you, than the innovation you've chosen to push it.  Of course it's good to have a strategy.  Whether it's a good strategy or not is another issue, and who it's good for is a third.

So, I'm fascinated with it and sharing my thoughts to receive feedback and other inputs. There are a great many man-hours being spent refining your work.  If you're not going to help, you could at least avoid complaining about the direction of the efforts.

Now everybody stop fucking with your CPUs (and the wallet) and get on that GPU miner so sunny can have his threat to Litecoin!

Or am I somehow missing the whole point?

Let's just say, you have no idea how critical time to market now is. I could wait a couple more months so to have a more 'perfect' thing for you, but then what's the impact to the overall competitive landscape? You can feel cozy in your conspiracy theory about me, whatever you like. But at least get the facts straight. The most improvement in the stock miner was from me yesterday actually. My post earlier was to warn the contributors of the danger messing with core protocol code without going through development process. Ask any competent crypto developer about whether what I said was true. I did not mean to diminish their enthusiasm to contribute and help, on the contrary I appreciate their participation.

[hiddenvalleyranch]

Anyone have an i5 haswell optimized windows binary? I'll shoot you some XPM.

[coinerd]

Let's just say, you have no idea how critical time to market now is. I could wait a couple more months so to have a more 'perfect' thing for you, but then what's the impact to the overall competitive landscape? You can feel cozy in your conspiracy theory about me, whatever you like. But at least get the facts straight. The most improvement in the stock miner was from me yesterday actually. My post earlier was to warn the contributors of the danger messing with core protocol code without going through development process. Ask any competent crypto developer about whether what I said was true. I did not mean to diminish their enthusiasm to contribute and help, on the contrary I appreciate their participation.

You're clearly an incredible genious, I doubt that anyone was going to come out with a more impressive POW than this in the near future, but I certainly don't move in your circles.

I don't have an agenda I was a little shocked at that thought. Mining 100+ primecoins nearly doubled my "crypto-worth", I can't afford an agenda yet.

Anyways whatever and It's become clear to me that I'm gnawing at this in a silly, forum-drama, over invested fashion. It doesn't change my desire to analyze your various and unclear motives here in what seems from my largely removed perspective to be, well, a perplexing strategy.

And I have no doubt the guy that seriously forks miner's efforts towards primecoin in a forum-distributed build is going to be the subject of a new level of witch hunt.

[n4ru]

Here is a replacement for release5, again using the official codebase.

https://www.dropbox.com/s/76eij0zcjx091am/release11_sandybridge_i5.zip

I will pull down the old links now.

Any chance at a Linux compile?

If you're running linux you really ought to be able to compile it yourself. https://github.com/primecoin/primecoin

Good point. Where can I find your revision/sources?

[wibtc]

Here is a replacement for release5, again using the official codebase.

https://www.dropbox.com/s/76eij0zcjx091am/release11_sandybridge_i5.zip

I will pull down the old links now.

Any chance at a Linux compile?

If you're running linux you really ought to be able to compile it yourself. https://github.com/primecoin/primecoin

I compiled it myself. But primecoind is very big, 50MB. Don't know why. It works, but no block found so far.

[Zalfrin]

Here is a replacement for release5, again using the official codebase.

https://www.dropbox.com/s/76eij0zcjx091am/release11_sandybridge_i5.zip

I will pull down the old links now.

Any chance at a Linux compile?

If you're running linux you really ought to be able to compile it yourself. https://github.com/primecoin/primecoin

I compiled it myself. But primecoind is very big, 50MB. Don't know why. It works, but no block found so far.

It has a bunch of debug data in it still. Run "strip primecoind" to decrease the size.

[Zalfrin]

Here is a replacement for release5, again using the official codebase.

https://www.dropbox.com/s/76eij0zcjx091am/release11_sandybridge_i5.zip

I will pull down the old links now.

Any chance at a Linux compile?

If you're running linux you really ought to be able to compile it yourself. https://github.com/primecoin/primecoin

Good point. Where can I find your revision/sources?

The code I was building with for my old releases is here: https://github.com/Zalfrin/primecoin

But, it's STRONGLY recommended to use the code on the official github until the kinks on my fork have been worked out.

[akspa]

Here is a replacement for release5, again using the official codebase.

https://www.dropbox.com/s/76eij0zcjx091am/release11_sandybridge_i5.zip

I will pull down the old links now.

Any chance at a Linux compile?

If you're running linux you really ought to be able to compile it yourself. https://github.com/primecoin/primecoin

I compiled it myself. But primecoind is very big, 50MB. Don't know why. It works, but no block found so far.

run 'strip primecoind'

Otherwise, you're running the binary along with its debug and line numbers. Strip will literally strip this data out and give you a binary that's roughly 8-10MB.

edit: looks like I'm late to the strip party. In any case, I learned yesterday how to compile my own builds of the client, all thanks to this currency

how is this release for sandybridge i7?

I believe the SB i7 is just marketing for the most part - the cpuflags are identical with the i5 SB, except the i7 has more cache and higher speeds. I'm running an i7 2600k with the SB binary, and currently using the 4 physical cores with a speed of ~500pps. When I ramp it up to use all 8 logical cores, I'm averaging just over 1000pps. Maybe there's some tweaking to be done for us i7 users, maybe not.

\/ \/ \/

[Tgspring]

Here is a replacement for release5, again using the official codebase.

https://www.dropbox.com/s/76eij0zcjx091am/release11_sandybridge_i5.zip

I will pull down the old links now.

how is this release for sandybridge i7?

[Zalfrin]

Here is a replacement for release5, again using the official codebase.

https://www.dropbox.com/s/76eij0zcjx091am/release11_sandybridge_i5.zip

I will pull down the old links now.

how is this release for sandybridge i7?

It will work fine for that, if that's what you're asking... if you're asking how it is for a sandybridge i5, that's just because I ran with -march=native, and I have a sandybridge i5 processor in this PC.

[Zalfrin]

Here is one more build, it's probably extraneous with the two I've already provided, but if anyone wants to try it out they can. Again built from official codebase. This one is optimized for the core i7-avx.

https://www.dropbox.com/s/b3a5d3k6xt0ezvx/release13_corei7-avx.zip

[Tgspring]

Here is a replacement for release5, again using the official codebase.

https://www.dropbox.com/s/76eij0zcjx091am/release11_sandybridge_i5.zip

I will pull down the old links now.

how is this release for sandybridge i7?

It will work fine for that, if that's what you're asking... if you're asking how it is for a sandybridge i5, that's just because I ran with -march=native, and I have a sandybridge i5 processor in this PC.

thanks,i'm using i7.

[mrmanager]

Here is a replacement for release8 which uses the official codebase. I will have a replacement for release5 momentarily as well. Those were the most popular, if there are others people want rebuilt let me know.

https://www.dropbox.com/s/lxm220qqztpgps5/release12_x86-64.zip

Thanks for the update.

[akspa]

Here is one more build, it's probably extraneous with the two I've already provided, but if anyone wants to try it out they can. Again built from official codebase. This one is optimized for the core i7-avx.

https://www.dropbox.com/s/b3a5d3k6xt0ezvx/release13_corei7-avx.zip

Is this for the SandyBridge avx, or Haswell's avx2?

[dbasql]

Here is a replacement for release8 which uses the official codebase. I will have a replacement for release5 momentarily as well. Those were the most popular, if there are others people want rebuilt let me know.

https://www.dropbox.com/s/lxm220qqztpgps5/release12_x86-64.zip

Thanks for the update.

This ^^^^ Thanks