Re: [bitcoin-dev] [BIP Draft] Datastream compression of Blocks and Transactions

2015-11-30 Thread Matt Corallo via bitcoin-dev
I'm really not a fan of this at all. To start with, adding a compression library that is directly accessible to the network on financial software is a really, really scary idea. If there were a massive improvement, I'd find it acceptable, but the improvement you've shown really isn't all that mu

[bitcoin-dev] [BIP Draft] Datastream compression of Blocks and Transactions

2015-11-30 Thread Peter Tschipper via bitcoin-dev
@gmaxwell Bip Editor, and the Bitcoin Dev Community, After several weeks of experimenting and testing with various compression libraries I think there is enough evidence to show that compressing blocks and transactions is not only beneficial in reducing network bandwidth but is also provides a sm

Re: [bitcoin-dev] Test Results for : Datasstream Compression of Blocks and Tx's

2015-11-30 Thread Jeff Garzik via bitcoin-dev
Thanks for providing an in-depth, data driven analysis. It is surprising that zlib provides better compression at the high end. I wonder if that is due to our specific data patterns - many zeroes - which probably puts us into the zlib dictionary fast path. On Sat, Nov 28, 2015 at 4:41 PM, Pete