Three developers affiliated with the Pacific Northwest National Laboratory provide an in depth look for Dr. Dobb's Journal at the options for optimizing a breadth first search (written in C) on the Cell Processor.
For the non-technical the results can be summarized as "On a Pentium 4 HT running at 3.4 GHz, this algorithm is able to check 24-million edges per second. On the Cell, at the end of our optimization, we achieved a performance of 538-million edges per second. This is an impressive result, but came at the price of an explosion in code complexity. While the algorithm in Listing One fits in 60 lines of source code, our final algorithm on the Cell measures 1200 lines of code"
Definitely one for the programmers amongst us.
Salman from Tech4Gamers writes "Mortal Kombat 9 revived the series from a low point after bringing it back to 2D combat. It marked a new high-point for the franchise due to its incredible roster, exciting cinematic story mode, and high-octane combat."
That game was actually goated. It was the first time ever that I actually tried to get good at a fighting game. Unfortunately the online connection was so dogshit it made it hard to enjoy and eventually I gave up. Haven't really played much fighting games since.
Taking a trip back to Housemarque's forgotten platformer.
The Simpsons: Hit and Run, The Legend of Dragoon, and Chrono Trigger would benefit substantially from a modern remake.
The Xenosaga trilogy, the Wild Arms games, Xenogears, Parasite Eve, Dino Crisis, Suikoden. The Warriors game I know nothing about but it looks interesting. Anybody can do a list of games they want to see get modernized I would lose my shit if Namco brought the Xenosaga trilogy to modern consoles.
This reminds me of my MS paint project.
Make MS Paint in VB and C++.
Took me weeks to do it in C++.
Took only 2 days to do it in VB.
**Breaking news**
**Programming the cell is hard** hehe. Well the cell is capable of so much, but there is a lot of work that needs to be put into it. Guess the trade off is how much time an effort is worth the results.
prove the power with the games.
do you now see why there are so many cell and rsx threads.
GAMES.
The end of the article reads:
"In short, the Cell offers an impressive potential for performance. However, due to its architecture and limited support offered by the compiler, you can't expect to exploit this potential by just recompiling your current applications. Applications must be radically redesigned in terms of computation and data transfers. Computational bottlenecks must often be analyzed and addressed manually, and data transfers must be properly orchestrated in order to hide their latency completely under the computational delays."
Basically, unless something is built from the ground up for the Cell to process, don't expect drastic improvements over a system that has a different processor running at the same speed.
SONY went for the "Cutting Edge" factor when they decided on the Cell for the PS3, but with very little in the way of tools to efficiently program with the cell, they've left many developers hanging out to dry.
they did it with the ps2, toy story like renderings. And they did it with the ps3, 4d, killzone, etc. had they just said it was a little bit more powerful, and let the games speak for themselves then there wouldn't be all this rioting.