ASICminer vs Antminer results question

[MCTV]

I have three block erupters and five ant miners.  The ant miners together show 9.9Gh/s the block erupters together show 1.01 Gh/s.  No problem there. But the 3 much slower block erupters are only 1000 blocks away from keeping up with the ant miners, when both started at the same time.  Showing 6026 blocks for the ant miners and 5100 blocks for the block erupters. 

Can anyone explain what I'm seeing? Are the block erupters more efficient at mining? Here are the bgfminer windows:

http://i62.tinypic.com/35jzh3k.png

Thanks!
MCTV

[jamesc760]

It's the number of shares, not blocks, (A)ccepted. And the shares accepted are higher-level for the ant miners vs lower-level shares for the BE's.

Even though the rate of Accepted shares seems low (6000 vs 5000) on the Ant miners, you get more btc because they are higher difficulty shares.

I guess you can say that this is a case of Quality vs Quantity.

Which would you rather get, $8 or $1 for each share found? Ant miners are giving you $8 for each Accepted share vs $1 for BE. I'm making up the $ figures here, but the concept is accurate.

[Rannasha]

Many pools will adjust the difficulty of shares to the hashing power you have available. The more hashing power, the higher the difficulty of each share. Of course, this is taken into account when computing your payouts.

In your case, the Antminers are computing shares that are 8 times more difficult than what the Block Erupters are doing. So if you account for that, you could say that the Antminers have done the equivalent of ~48000 easy shares versus the ~5000 of the BE's. This corresponds very well to the ratio of the hashrate of both setups.

[hurricandave]

You have the AMU's set at Diff=8 with 0 hardware errors and the ICA's set at Diff=1 with dozen's of errors. I don't think that the numbers can be reliably compared one vs. the other until you get the error ratio and difficulty to match.