Quote from: PCMiner on April 30, 2014, 09:25:10 PM

Quote from: pooler on April 30, 2014, 07:06:08 PM

Quote from: PCMiner on April 30, 2014, 06:40:34 PM

I just did a full re-install of my OS and noticed a new version of CPU miner was out (Before I was on 2.3, first one with stratum built in). Anyway, I've installed the new version and while it works, with the same hash rates as before, I'm noticing I'm getting a ton of "accepted xx/xx (%) (booooo) compared to the older version. Has something changed in newer versions? If I stick my old boot hard drive into the machine the 2.3 version of CPU miner on it, it works on the same pools with nearly 100% accepted rate. With this new version I'm getting accepted rates around 60-70%.

Weird. It would be nice if you could try all versions from 2.3 to 2.3.3 (they are all available at SourceForge) and see which of them seem not to work. Also, please post (or send me via pm) the exact command-line options you are using so that I can test if there's a problem with the server you are connecting to.

Originally I was on 2.3 as I mentioned in the post, and now I am on 2.3.2 and it's working fine. When I rm minerd from the usr/bin/minerd and replace it with the 2.3.3 version I get much lower.... well, it still says accepted but says "boo" and the percentage drops.

2.3.2 seems to work fine with near 100% accepted over several hours.

This was tested over the last 2 days on both give me coins and clevermining.

Launch strings are:
minerd --url stratum+tcp://us.clevermining.com:3333 --userpass :x -t 24
and
minerd --url stratum+tcp://ltc.give-me-coins.com:3334 --userpass :x -t 24

I tried less threads to see if that helped and it made no difference. This is a 2012 MacPro upgraded with dual 6 core 3.46 Xeon 5690's. 32GB RAM. Fresh install of OSX 10.9.2. The same result happened when tested on my older 2008 3.2Ghz 8 core MacPro. with 10.9.3 (Appleseed Beta release).

After borrowing a friend's MacBook and testing, I can confirm this issue.

It is not an issue specific to version 2.3.3, and actually doesn't even seem to be a bug in the code at all, but for some reason the OS X binary for version 2.3.3 produces 1/3 wrong hashes for algorithm scrypt. Binaries for all other platforms are not affected. The problem appears to be caused by a bug in the particular compiler used to produce the binary (Apple's version of Clang for OS X Lion), and is only triggered at -O2 or higher optimization levels. The issue does not appear if using gcc on OS X, nor does it appear with the latest version of Clang on Linux.

I have now updated the OS X package for version 2.3.3 with a binary compiled at -O1. It would be nice if you could confirm that it works.