Look, I was giving you a suggestion, but you said I'm full of shit. So that's it.
If you are saying that EVP on machines that are AES-NI is slower than Wolf's, it's all fine, I really don't care. On my AMD FX machine EVP is faster than Wolfs. Period.
I already wasted too much time in something that it's not even my business, I thought I was giving you a good advise, my bad.
I apologize if my claim was wrong. (on my machine it's true, but maybe on yours isn't, I don't care anymore)
Like I said, have a nice night, it's time for to go to bed.
There's no need to pout. You're making a claim that the AES_NI optimized hodlminer-wolf is slower than the
original on AMD CPUs with AES_NI. That is a serious claim and no one else has reported this so you need proof.
This goes beyond being childish, you have made a serious charge against a respected dev whio was paid to
develop the AES_NI enhanced miner.
You've already gone too far, you need to follow through with proof or make a full retraction with an apology to Wolf.
You have to calm down, I haven't make any charge against nobody, what are you talking about? listen to your self!
All I did was telling you that on my machine, EVP was using AES-NI and it is faster than wolfs code. Period. There is no 'charges' against nobody, please!!
And like I said, that only applied to the source code on the hodlcoin, I even told you the file where it was used: hodl.cpp
My suggestion was that you could use EVP alone (whitout wolfs code) because it uses AES-NI automatically. Where is the accusation??
You are making a stupid and simple suggestion a matter way, way out of proportion. You doesn't even understand that I was trying to suggest something to improve your code.
So, end of the story. I'm not gonna continue with this stupid conversation, it when from a simple suggestion to "charges against a respected dev'. Go figure!
You don't realize the sum of your claims. You claim the original hodlminer is faster than Wolf's AES_NI optimized version.
I'm not saying it isn't true in your case but you need to show proof to convince me. I can't replicate your results and
they in fact prove the opposite. You're on AMD and I'm on Intel, there're may be something there that should be pursued but you
haven't given me anything I can work with. You're coming accross as a troll. Prove me wrong.
The original hodlminer was not promoted to use AES_NI and Wolf was contracted to implement one. His version
produced higher hashrates and no one has complained. Either no one is mining hodl with AES enabled AMD CPUs,
they aren't paying attention, or you're wrong. There is some foundation to your claim because Wolf is optimized
for Intel, its performance on AMD may not be optimum. You can provide the answer if you stop throwing a hissy
fit and stick to the issue. It's up to you.
Edit: In your last 5 posts you haven't provided one piece of new information. Instead you're acting like a spoiled child
whining because I didn't give you praise. Grow up.
). Someone know how to install on CentOS 6.5? Specific parameters on configure command?
