@MichaelPatak DLL is good. Consistency targets with it make it pretty challenging. One great day with the market moving, you either cap yourself or really extend out the timeframe to get a payout.
@ThePrimeagen 100x is a bit much, but **early** prototyping is vastly faster. The rest, I'm not so sure. Especially in a team environment, the benefits of AI are very dubious.
I was talking about rendering. You're greatly over-simplifying it for rendering. I used to do that and the ROI was not great, especially when OS or hardware/driver changes product subtely different results. Noise filtering helps with the issue, but only partially solves it.
I moved to exactly what you said: deterministic file structures of layout coordinates and metadata in text and binary files.
@ThePrimeagen@rfleury This works great, until you try to expand this type of testing to different UI scaling, OS versions, or decide to make some changes to how something works. Then you get to regenerate all of the baselines and re-verify everything.
I've got an agent in a loop optimizing a renderer with the goal to minimize frame times (and tests to measure). It got times down from 88ms to 2ms and allocations down from ~150K to 500. Sounds good, right? Wrong. This is exactly why agent psychosis is a big fucking problem.
As an experiment, I rewrote the Ghostty core render state in Go, with access to identically laid out data structures as Ghostty and the exact same validation tests. I made a purposely naive renderer (simple, correct, but slow). 88ms per frame with 150,000 allocations (horrendous, lol)!
I then kickstarted a Ralph loop to bring the frame times down. I told it it can't modify input data structures or the public API or tests (they're correct), but it can do anything else it wants. It got to work.
It has worked for about 4 hours. I've spent around $350 on this experiment so far. The results?
88ms => 1.5ms
150K allocs => ~500 allocs
Incredible right? Nope.
My hand-written renderer I ported has frame times (same benchmark) of ~20us (0.020ms) and 0 allocations in the update path.
This is the problem with psychosis and lacking systems understanding. If you don't understand the system, you're going to accept that this is an incredible result. If you understand the system, you'll see better solutions immediately and can do roughly 75x better on throughput.
The people who blindly trust agent output are in the former camp. They're sheeple, overdrinking from a fountain of mediocrity.
Standard disclaimer: I use AI all the time. I like AI. The point I'm making is to not blindly accept results. Think. Analyze. Learn.
What's the problem? Now you need to make another $750k from your $3k trailing drawdown live account to get back those profits in the vault. /s
Don't worry, if you fail, we have a chaser for you! Really high-priced re-attempts to win back that vault money!
Strongly disagree here. You shouldn’t have to write library code and app code different.
I also think there should only be two access levels: public and internal.
Public means API I need to version and maintain.
DAMN. Former Washington Gov. Christine Gregoire, a Democrat, goes scorched earth on the current Democrats leading the state. Says they have no clue how bad their policies are for the economy.