I Have succeded in porting PocketQuake to CASIOS

[CARPEDIEM]

[Dan East]

[CARPEDIEM]

Yes, i imagined that since i've been waiting for it to start coding tonight and i didn't get it<br>please send it to carpediem@velocom.com.ar<br>don't forget the .ar, i live in argentina )<br>

[999]

Congrats Carpediem!<br>Last modification: 999 - 02/22/01 at 21:08:10

[CARPEDIEM]

Hey that's not fair!!<br>I don't know what TIMEREFRESH does but it sure doesn't give an accurate result.<br>I assure that i have about 4-8 fps on the initial level, but when i run timerefresh it get's pretty slow and i get a cracking sound when it's facing the big open area.<br><br>Sound is flawless by the way and i have an E-125 @ 180<br><br>So... i guess i'll just have to say that timerefresh says that i run PocketQuake at 1.9 fps<br><br>:o(<br>Not true.<br>

[CARPEDIEM]

here's the link by the way.<br>http://www.rojakpot.com/Files/Misc/mipsquake.zip<br><br>(Admin Edit)<br>There is a mirror available at http://www.pocketmatrix.com/pocketquake/download.html<br><br>Last modification: Chris Edwards - 02/22/01 at 22:01:15

[999]

Carpediem:<br><br>A Timerefresh is about as accurate a number as you can get for FPS comparisons.  This is all assuming you are doing the Timerefresh in the same area which Dan has performed his.  Judging from your Numbers, I'm assuming you are.<br><br>An FYI, the Game is much slower when you are Playing it, as opposed to watching the Demo.  The Demo runs nearly 2x faster, this could be the culprit for any FPS confusion.  Perhaps your Timerefresh numbers will reflect something a little more favorable once you've applied Jacco's source tweaks which have already led to a 24% increase in PQ's overall speed <br>

[entropy1980]

Carpe you rule ... you all do i am so happy i can't even see straight!!!! woohoo rejoice!

[Dan East]

Doing a TIMERERESH at the opening level represents an average FPS, including the areas which Quake requires that most processing; looking across a large open area. Of course higher FPS is obtained in other more confined areas, or when you can only see a short distance. At this point you're getting less than half the fps of the original iPaq version, which I personally would have trouble considering playable.<br><br>Dan East

[Mark (killzat)]

Yes,<br><br>CARPEDIEM you must run the Benchmark at the 'Entrance Level' where you first spawn. don't move forward, just do TIMEREFRESH in the console and let the cycle complete.<br><br>This is the only way to get a good standard framerate as you can really push the engine with all the polygons. The original iPAQ release got 4.1-4.9FPS (Not Clocked/Over Clocked) depending on apps, sound and yadda yadda.<br><br>I'm trying your Casio ver now and it gets under half the iPAQ, so please be more intelligent with your words in the future:<br><br>"I don't know what TIMEREFRESH does but it sure doesn't give an accurate result."<br><br>DOh... DOOOOH =)<br><br>Also remember that like you, my Casio is clocked to 180Mhz and so the standard 150Mhz is likely to be more like the 1.2FPS everybody originally predicted.<br><br>Can't wait to see what the new code changes do to quake.

[CARPEDIEM]

Ok, i'll take your word for it, but if you runned TIMEREFRESH on a casio you'd know what im talking about. <br>For starters when timerefresh is executed, the screen starts to rotate slooowwwly and when it does facing the large area the sound gets corrupted, and i figure that's because the CPU can't handle it anymore. But when you fool around the level manually it is faster and there's no sound corruption.<br>But who cares, im starting work on the new source right now.<br>

[Dr. Phat]

How can I overclock my E-125... Safely?

[Dan East]

Yeah, that's the same basic thing with the iPaq. Sound doesn't mess up, but it appears to be much slower, especially when looking out across the whole room. The reason TIMEREFRESH seems so slow is that it takes 128 steps to rotate around, which of course is far more exact movement than when mlooking around. Either way it still takes the same amount of time to render the frames, based on the direction you're looking and the surroundings. Using TIMEREFRESH like this provides a good standard for comparing versions / devices, but does not indicate the actual average FPS you get while playing.<br><br>Dan East

[CARPEDIEM]

Ok, im looking at a 19% faster PocketQuake for mips with Phantom's Optimizations. Hey, this is a lot of progress for such a short amount of time!<br>Anyways Dan, Jacco sent me the updated files and i updated the mips version, i didn't touch your version. Should i compile with your changes?, what do you think?<br>Now i get 2.26 fps on the timerefresh test and i think it's very playable on closed areas.

[CARPEDIEM]

Hey Dan, im looking at d_scan.c and i see a lot of castings, from ints to floats, from floats to ints etc etc. <br>The ipaq is A LOT slower in casting than other devices. Perhaps we can work a little there too to increase performance.<br>