Quote from: Peter R on August 07, 2014, 12:20:18 AM

Quote from: impulse on August 07, 2014, 12:01:31 AM

Miners still get to decide which transactions that they want to include...

And some miners do stupid things. "BCPool" created 17,976 single-satoshi dust outputs from that "1Enjoy" address by mining a bunch of spam transactions in block #309657 and again in block #309740. Assuming each output is 50 bytes, that's 17,976 x 50 B = ~1 MB of garbage that will possibly forever sit in each node's RAM, as ouputs are not prunable if they never get spent (and these ones cost more to spend than they are worth!).

I'm not saying the limit shouldn't be raised, I'm just trying to present some empirical data that illustrates some things that have gone wrong.

Code:

BLOCK #309740: 10,486 DUST OUTPUTS
ADDRESS SENT FROM TX HASH N_OUTPUTS
===================================================================================================================
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z b2b6112e73cbe0937a1b60c9abfc4c2ca6e26b2612c90ec598b1cd28787a553e 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z 5e5d17901378b8835861d8cc0df2b5f9bf3c86759c3821f96470cd852344d799 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z b2bf7525618ed5402024d5bff6f477a2093965d3f11ececbc2b6a9b5bbe1f5d5 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z 3df8f4a7e06486bb3b0700935a16fc6d4fad397dcb6c88b9c4b6a7453301aa54 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z ae05ddfc8aa206b213c71ec0e96723dcd4ff03a71ca76b469d225d1c60d9e262 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z f75b455b4df94bcdcd54985b4aeea9948d2a1dca0f63c65b85ad2d941050c5cf 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z ee61b911610f66f539832699fbbf4ab3955c8bd5ad0cfa570ff500dedcde5bf8 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z a22186cbcf82cac66da17183209f20f76088934c931bec131fe71ad2f250e8f8 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z 0469286403543b7d794fe52661135a80142c47ac541793c703fd637a4e9dc0bf 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z a71dc9abd8bae377b661b0c288222acf2ab62fc7bf96b0bab2dd5354fbc91643 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z f437d94c1cfe818c44f4505f3e6506839113ec747d815a867e953a4164276047 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z eb95d902eb841ce23c9c1466e010cbb05d43d1c248376440f9dd56ee1bc8b3e5 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z b62416ceb3453d6dcba95213e4e222915438a4b70eba37e09099110b20be0d52 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z f3aba1c064e9c9a1d49db3bce42c07c184a0b329366994000fa49f0bc8f3ba16 749

BLOCK #309657: 7,490 DUST OUTPUTS
ADDRESS SENT FROM TX HASH N_OUTPUTS
===================================================================================================================
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z 2e7fdb43e2a0fe17cfe2d283a2da4d375204700aef4bf2c31056125f4383b121 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z 3d1105d7c2cd36705b2163fa2d0f74973377467aefea72156a30444f330acf71 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z 5c4e56943f3128629ebecb41f5916bd7bc1dceaccc37f29e59c385464f2bf558 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z 536adfb745ffaa8c1fb698414475f5ca7d82846eadd912cfe0a76045d78c3198 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z 45c7544374e88de47d9be2e3e100f4aa10aafd9469c4dba8d55a870f83c312ab 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z e7ae2378a44a940a8c872a63e43fa67c5a023df4df3f9ca8011611422fc2861c 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z d864e3cc07dc79f221bd91210b274e136159ccbb5c9eafad906d9d08c5558ce5 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z 3dedbc086d5cfc81bd3614a1b3fab1bb4d73d8982d9552e811f84f0118a1fc12 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z 7337d7ac353a9283c65a1ab99d28f97997784a0fb26d0dc0b59f2fa2556c4460 749
1Enjoy1C4bYBr3tN4sMKxvvJDqG8NkdR4Z 634a7fdcba3fe91da07f44e6b9bee1c659cbfca27076bd8b61984d062c071651 749

Sure, I'm not actually in favor of no limits either, but I am in favor of increased block sizes.