What speed are your getting CPU mining TENEBRIX?

[tacotime]

So, on a 2600K with 8 threads I should get what !?

~3.8 kh/s per core

[tacotime]

AMD users:  Try turning up the CPU-NB (CPU NorthBridge) speed, that's your L3 cache+memory controller.  It may help.

I get about a 2% gain from having my NB at 2600mhz

[Bobnova]

Every little bit help!

Intel users will likely gain more from cranking Uncore than AMD with cpu-nb, just because the AMD doesn't have to use as much of it's L3 cache.

[bulanula]

So, on a 2600K with 8 threads I should get what !?

~3.8 kh/s per core

If I get my 2600K to do around 30 khash/s I would feel EPIC !

[Vitalik Buterin]

[Lolcust]

What miner are you using, and what box config ?

[Vitalik Buterin]

What miner are you using, and what box config ?

Miner is the one here: https://github.com/Lolcust/Tenebrix-miner

My box is just a simple macbook with 2 CPUs and no modifications (except the fact that it's running Xubuntu).

[Lolcust]

What miner are you using, and what box config ?

Miner is the one here: https://github.com/Lolcust/Tenebrix-miner

My box is just a simple macbook with 2 CPUs and no modifications (except the fact that it's running Xubuntu).

what minerd startup options do you use ?

[Spacy]

what minerd startup options do you use ?

Maybe 4way? ;-)

[Vitalik Buterin]

what minerd startup options do you use ?

minerd --userpass 1:1 --url http://127.0.0.1:8697

[bulanula]

LOL no wonder you are getting all of the blocks mate with that kind of speed 

So, it seems that Apple devices really are MAGICAL huh !?

[Adamlm]

For SandyBridge CPUs (corei7 with four digit model numbers).

git the package.
./configure
(screw the flags)
gedit Makefile
find "CFLAGS = "
Change that line to: CFLAGS = -march=native -O3 -Wall -msse2 -msse3 -msse4.1 -msse4.2 -msse4 -mavx
Brought my performance from 2.6kh/s per core (running one thread per core) to 3.66-3.7kh/s per core.

Tested with 6 threads, +2 kh/s in total.
HT works now!
Just needed some hardcore flags, that's all.

I wonder if someone could compile this package for Windows i5 & i7 users ?

[bulanula]

For SandyBridge CPUs (corei7 with four digit model numbers).

git the package.
./configure
(screw the flags)
gedit Makefile
find "CFLAGS = "
Change that line to: CFLAGS = -march=native -O3 -Wall -msse2 -msse3 -msse4.1 -msse4.2 -msse4 -mavx
Brought my performance from 2.6kh/s per core (running one thread per core) to 3.66-3.7kh/s per core.

Tested with 6 threads, +2 kh/s in total.
HT works now!
Just needed some hardcore flags, that's all.

I wonder if someone could compile this package for Windows i5 & i7 users ?

I'm waiting for this as well.

[tacotime]

I wonder if someone could compile this package for Windows i5 & i7 users ?

If you want to do a port, pick up a copy of cygwin or Visual Studio and do it yourself.  It shouldn't be impossible, it's only a command line program.

[GenTarkin]

Mining w/ an i3 550 @ 4ghz using virtualbox and archbang x64 in windows 32bit host my result is about 3.8kh/s / core w/ HT. Without HT its around 2.5kh/sec / core.

[panerai]

I have two dually xeon poweredge servers I'm going to power up today to run this new CPU miner. I just hope BTC-E comes back online for trading. I just deposited BTC to get some TBX and the damn site goes down lol

[mrx]

Is it possible to get 1mh/s using scrypt? OR the miner on Mac is just calculated the hashrate mistakenly?

[MaGNeT]

Is it possible to get 1mh/s using scrypt? OR the miner on Mac is just calculated the hashrate mistakenly?

No, just error in compiling.
It doesn't really earn anything... But it looks cool 

[DeathAndTaxes]

Damn it !!! How can I optimize my crap 2600K !?

You folks with AMD Phenom II at 3.6 GHz getting 16 khash/s on only 5 cores
Me with Intel 2600K at 4.2 GHz getting 9 khash/s on all 8 threads

Fuck Intel !

The 2600K doesn't have 8 real cores.  It merely tricks the OS into thinking there are 2 cores for each real core to keep the CPU more efficiently loaded. Internally there are 4 actual cores.  


HT isn't a performance doubler because the CPU only has 4 physical cores not 8.  Some applications perform slightly faster w/ HT.  Some perform faster w/ HT off.  All out database servers have HT turned off for example because we see higher throughput w/ HT turned off.  The more the pipeline can be kept full the less advantage HT provides (eventually reducing performance when pipeline is continually full).   

Your comparison should be based on 4 cores not 8. The 2600K has 4 cores and will perform as 4 core processor when performance per thread is maxed out.

[Bobnova]

That's smoothie the fudmonster.  Ignore him.