From: "Luke-Jr" <luke@dashjr.org>
To: Jeff Garzik <jgarzik@exmulti.com>
Date: Fri, 25 May 2012 00:57:38 +0000
User-Agent: KMail/1.13.7 (Linux/3.2.12-gentoonestfix-intelwr; KDE/4.8.1; x86_64;
	; )
References: <CA+8xBpdBe4yR6xkCODL6JQ41Gyx9eWcGGGvcQVt7DCmaEnAhbg@mail.gmail.com>
	<201205250045.24540.luke@dashjr.org>
	<CA+8xBpfOh-61z_7e1jzz7ZYV4eiCCi=ruQbKBuQp1juuSdYdbQ@mail.gmail.com>
In-Reply-To: <CA+8xBpfOh-61z_7e1jzz7ZYV4eiCCi=ruQbKBuQp1juuSdYdbQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: Text/Plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Message-Id: <201205250057.39749.luke@dashjr.org>
Cc: bitcoin-development@lists.sourceforge.net
Subject: Re: [Bitcoin-development] Punishing empty blocks?
Precedence: list

On Friday, May 25, 2012 12:51:09 AM Jeff Garzik wrote:
> On Thu, May 24, 2012 at 8:45 PM, Luke-Jr <luke@dashjr.org> wrote:
> > On Thursday, May 24, 2012 4:33:12 PM Jeff Garzik wrote:
> >> Comments?  It wouldn't be a problem if these no-TX blocks were not
> >> already getting frequent (1 in 20).
> > 
> > FWIW, based on statistics for Eligius's past 100 blocks, it seems 10% (1
> > in 10) of 1-txn blocks is not actually unreasonable. This also means
> > these 1-txn mined blocks are not necessarily harming Bitcoin
> > intentionally. Anyone care to figure out the math for how fast miners
> > need to finish processing transactions to reduce the number of 1txn
> > blocks?
> 
> Look at the time since last block, and correlate with the number of
> non-spam TX's in the memory pool at the time.  It is obvious which
> ones are quick blocks (<60 seconds since last block, no big deal) and
> which ones are the lazy miners (> 120 seconds since last block).

Block times are not accurate enough for that.