Would you disqualify this score for slow speed?

greenspun.com : LUSENET : MARP Editors : One Thread

Ok, here's a theoretical recording (done by me) on the Game Senjyo.

Checking the game with F11 turned on while playing it seemed to be mostly staying at 100% of original game speed. However, obviously at the times where it was frantic and I was too busy playing to watch it was slowing down for a number of frames.

Here's the analinp speed checking information (game used m35tg3b):

Number of frames in this recording: 13706

-------- InpSpeed Report -------- Good frames : 10693 (average speed: 100%) Slow frames : 2953 (average speed: 78%) Fast frames : 60 (average speed: 112%)

Total frames: 13706 (average speed: 96%) -------- End Of InpSpeed Report --------

Although the overall game average was 96%, given that there were a large number of frames with an average of 78% what would your various opinions be on DQing this game for having too many slow frames?

Cheers, BeeJay.

PS: I can easily understand this part, but man does the histogram information confuse this newbie currently.

-- Anonymous, April 24, 2001

Answers

I would examine this recording further.

First of all: I think a large number (relatively or absolutely) of slow frames is not necessarily bad in general. Lots of times, you see the game "catch up", having slow frames followed by fast frames. If so, the number of fast frames would be in the same order of magnitude as the number of slow frames, usually half of the number of slow frames or more.

In this recording, that is obviously not the case. A lot of slow frames, but no fast frames to make up for that. I would look at the pattern of slow frames then. If you use "-S" (capital S), then, at the start of the output, you'll see more detailed speed information. I'd say that if there are significant periods of time that the game ran too slow (say a few seconds or more; divide number of frames by the game's framerate to get the number of seconds), I'd disqualify it.

Hmm... I thought that there were one or two posts here on the board that explained the three columns for the InpHisto information, but I guess I was wrong. There is some useful information in http://www.greenspun.com/bboard/q-and-a-fetch-msg.tcl? msg_id=003kfZ, but I guess I'll have to check my IRC logs. I might have done most of the explaining on InpHisto during the IRC meetings. I'll reply to your e-mail soon.

Cheers, Ben Jos.

-- Anonymous, April 24, 2001


Moderation questions? read the FAQ