Posts: 14
Threads: 3
Joined: Apr 2014
05-07-2014, 08:48 PM
(This post was last modified: 05-07-2014, 08:49 PM by haagens.)
Estimates is what I'm looking for. Assuming development stays at X pace for a Y set of time, what is a good estimate of when there will be noticeable improvements?
For example we know that rpcs3 started out with just homebrews up until 2014. Factoring in a stasis development time and upticks and downticks here and there, should anyone expect 2017-2018 to be the the time frame when rpcs3 gets serious (runs a handful of commercial games, can get in-game with many other games and resolves dynarec, etc.?). Just taking a WIILLLD guess here, not trying to create a debate.
Posts: 2,485
Threads: 77
Joined: Dec 2013
Reputation:
32
Literally impossible to answer. 20 über skilled people could join tomorrow and we would be done in a year. Or not.
Asus N55SF, i7-2670QM (~2,8 ghz under typical load), GeForce GT 555M (only OpenGL)
Posts: 14
Threads: 3
Joined: Apr 2014
05-07-2014, 09:37 PM
(This post was last modified: 05-07-2014, 09:46 PM by haagens.)
That's why I said estimate ... given the factor of commit averages, percent needed to work on, and down and up times, it's very possible to estimate a span of time from which rpcs3 could indeed improve enough to play a greater deal of ps3 games. 20 uber skilled people joining is not realistic, but an estimation based on plenty of observable factors is, namely:
1.Commit averages (Git measures).
2.Work needed to be done (Roadmap).
3.Contributor averages to gain perspective (check Git history).
4.Estimated workload frequency (factoring in overall potential by averages).
I could analyze and try to get a decent guess, and of course not even that is perfect, but estimation is the best tool for the unknown.
I could make a ballpark guess and say that since it took 3 years to reach playable games (of the lowest stock), it will have to take at least three more years, but that would be a wild goose guess with no statistics behind it. I don't know enough of ps3 but I have estimated timeframes for some other projects and they were all wrong, but close enough to give a fair idea. I believe estimations of some sort give a plausible, sensical, determinant, realistic view of this project's future.
If nobody estimated, everything would be a wild guess ... and wild guesses can hurt. Wouldn't you guys be more comfortable going by a measure of "X point in time to see real progress," rather than blindly, aimlessly adding code over years and being enthralled when the next game improves a pinch? To add more "too long; didn't read" to this post, estimating a time can also set a goal that's worth working for, not just coding up in air.
Posts: 149
Threads: 14
Joined: Aug 2010
Reputation:
0
05-07-2014, 10:08 PM
(This post was last modified: 05-07-2014, 10:09 PM by Gundark.)
Two things that can make a difference is recompiler and spurs. We don't know the state of this things, it could be done in a few days or a year. When these are commited, especially spurs, you should see the next breaktrough.
Posts: 107
Threads: 0
Joined: Feb 2014
Reputation:
5
There's literally no one who can tell you. People work on it in their free time, this is not some software project where there's employees to give you a rough estimate, we could literally have a break for months with no progress.
I would really like a rule in the forum that an ETA question are disallowed, they contribute nothing and only lead to baseless speculation. While some might find that entertaining, I do not.
