Is the 300x performance boost attributable to the threading model vs process model?
Was the code for the threading model written by hand or was it translated from the WIP threading model the human PG team is busy with as part of the 2028 roadmap?
I didn't say we should ignore splits. To reiterate: most third-party data providers include splits by restating historical prices. This only works if you want to analyse performance at individual stock level by ignoring holdings. But if you want to do performance analysis for your own portfolio of stocks, relying on restated prices is incorrect.
Eg.
If you bought say 10 stocks at the old price, you cannot take the restated historical price with your 10 stocks and you cannot take the units held of 10 stocks with the new prices. You have to account for this via transactions to increase your holdings (for splits) by leaving the prices historically the same.
Valuations are straightforward in a world without corporate actions like splits, consols, dividends etc.
Most (all?) data providers rerate historical prices after corporate actions that affect prices and this is not really the correct way to do continuous valuations using the restated prices.
ORMs and NoSQL were a direct result of the desire to have an object-oriented data worldview in a programming landscape dominated by OO languages. They were not introduced because programmers didn’t want to learn SQL (implying incompetence etc).