This was shared long ago on Twitter. Same agreement used for Azure BaaS.
Lastly, any input on this?
https://v.cash/forum/threads/basic-math-proves-digibyte-cannot-scale-debunk-it.447/#post-6509I'm not known for trolling, just truth seeking, I own digibyte for years and seek some simple answers and explanations. Regards.
"
DigiByte has a 15 second block timing. I can create a single DigiByte transaction that is 998,000 Kilobytes (@ zero cost) in size that takes a modern computer 43 seconds to validate. That said, if you run these types of performance tests against DigiByte's network slower nodes will get backed up further and further forking off onto their own chains. The only nodes on the "main chain" would be extremely fast nodes that can validate these blocks in < 15 seconds. This would split the network into 100's of disagreeing networks.
Disclaimer: This is a 100% technical post and I'd like someone to debunk it with provable technical facts. Thanks!"
First thank you for your concern and for supporting DigiByte. Now let's answer your question.
1) You cannot create a single 998,000 KB TX for free and expect any node to relay it. Every DGB client out there has this code in it:
https://github.com/digibyte/digibyte/blob/master/src/main.h#L74/** The maximum size for transactions we're willing to relay/mine */
static const unsigned int MAX_STANDARD_TX_SIZE = 100000;
In case you are wondering that is 100 KB or 0.1 MB. No way you could actually create a TX that large and get it relayed.
Bloating the network with multiple transactions gets very expensive very fast (as will be demonstrated below).
2) Each algo has a block timing of 1 min 15 seconds. Microsoft research showed on the BTC network it takes an average of 6.5 seconds for 50% of nodes to receive a new block. 95% in 40 seconds with a mean of 12 seconds. So our 1 min 15 second individual algo block time is well within this time frame.
http://www.tik.ee.ethz.ch/file/49318d3f56c1d525aabf7fda78b23fc0/P2P2013_041.pdfIn order to 51% attack DGB and fork the chain you would need to take over 93% of SHA256 and 51% of the other 4 algos at the same time. (virtually impossible) or else the difficulty adjustment (Multi-Shield) will easily prevent it.
3) With future modifications to DigiByte core things will get even faster as far as block validations are concerned (secp256k1)
Finally here is a screen shot from the live DigiByte network test after DigiSpeed (we did many similar tests internally on a test network before releasing the hard fork).
It is very difficult and expensive to send this many TX's in such a short amount of time. We pushed through almost 10,000 TX's in just over a minute and the network handled it just fine. In fact, the main worldwide DigiByte network performed better than our internal test network.
Check these blocks for yourself on DigiExplorer:
http://digiexplorer.info/block/e5ff5243c291d088e286bbe5ea449100487c6346667fcfc84f89d85971074880Now compare that to this recent snap shot of BTC blockchain with full blocks. It took over an hour just to process 5,000 TX's.
So DGB easily hits 10,000 TX's in 1 min. BTC takes over an hour for 5,000 TX's.
The proof is in the Blockchain.