forums.marpirc.net

Rule 2.g) reads:

"Games must be played at a MINIMUM of 90% of their full speed. Use a framerate counter if you are unsure."

It has been suggested that 90% is too lenient.

Also, if the 90% minumum is for the average of the total recording, or even the average for the actual gameplay, then temporary slowdowns below that threshold could still give a player unfair advantage in certain parts of the game, while maintaining a higher speed elsewhere to keep the overall average above the threshold.

Do we want to raise the threshold to 95%, or some other number?

Do we want to require that the recorded speed never drop below some threshold, or not drop below some threshold for more than a certain length of time anywhere in the recording?

Discussion please.

When you watch a recording made with a computer fast enough to emulate the game at full speed you will see three numbers: 99, 100 and 101%. Nothing else should be acceptable. A couple of random dips below 99% and slowdowns outside of the gameplay are no problem of course.

If the speed goes below 99% all the time the computer is obviously not fast enough for full speed emulation.

MARP's goal is to emulate the experience of "watching a virtual master play an arcade game", true to the arcade experience, as closely as possible.

That's why autofire and pausing is not allowed. MARP should stay true to the arcade experience in terms of speed as closely as possible as well.

Weehawk wrote:Do we want to raise the threshold to 95%, or some other number?

90% is fine by me for general play.

Weehawk wrote:Do we want to require that the recorded speed never drop below some threshold,

No. There is never any guarantee of consistent, regardless of system speed. You never know when the OS, or some other program may pull something that causes MAME to go down to 50% for a split second.

Weehawk wrote:or not drop below some threshold for more than a certain length of time anywhere in the recording?

Depends where slowdowns occur, and for how long. Screen transitions can often slowdown a fair bit, either by fancy video effects, or just shifting a large amount of RAM around.

95% average would be ok

I think 90% is fine. Below that is too slow, anything above that is splitting hairs. In my opinion the slowdown between 100% and 90% is so minimal it's not spotted without using a framerate counter, so is any gaming advantage really gained?

90% is fine with me, for reasons already stated by Barry.

I also must add that I agree with Rick here (posted in another thread and I don´t remember where right now). Sometimes it really doesn´t matter if the speed drops down bellow 90% (between stages etc) Sometimes it does (like the last boss on dodonpahci!). I think they will have to be judged in a case to case basis.

I agree with 95%. But don't count slowdowns due by animations and etc. between stages and don't count random slowdown due by OS.

90% is too slow for me and unfair.

destructor wrote:I agree with 95%. But don't count slowdowns due by animations and etc. between stages and don't count random slowdown due by OS.

90% is too slow for me and unfair.

Hmm I thought 90% was pretty fast for you

QRS wrote:Hmm I thought 90% was pretty fast for you

Sometimes. After 2 or more days of drink

I think 90% is ok. I also don't mind 95%. I worry though about having a minimum drop down point. Some games slow down dramatically when loading up. Other games slow down when there is an intermission scene in between levels.

I also think editors should be able to zero an inp even if the average speed is 95% in the case where a player blatantly slows down the recording during a critical part of the game.

for me only 99% is ok

90% but if its in the lower 90s then it should be added into the inp desc if anyone wants to do that.

DRN wrote:In my opinion the slowdown between 100% and 90% is so minimal it's not spotted without using a framerate counter, so is any gaming advantage really gained?

It depends on the game. For games like mahjong games and tile matching games etc. that have a timer to complete each stage, running at even 95% will give you extra time that might be the difference between completing a later level versus not being able to complete it.

There are some tile matching game scores from before alphamame and wolfmame came along that I strongly suspect were using slowdown cuz to play the game at 100% speed the timer clicks so quickly for a couple starting at the 3rd and 4th rounds that even if you INSTANTLY know all the moves to make you can barely complete the level if you make no control mistakes in just selecting the tiles to remove.

For other games like even mspac or pacman type games, I really don't think there really is a difference playing at 90% vs 110% playing in regular mode(not turbo). The game already moves slowly enough....don't need or really benefit from additional slowdown. it just makes the game take even longer than it already takes......boring.

For me, playing at anything other than 100% screws me up more than helps me on anything. When you are playing at less than 100% the speed will fluctuate some instead of being steady. That variance can screw up control input timings of your moves.

90% is good enough (95% reserved for tournaments, of course). Not all of us have supercomputers, and as MAME grows, the requirements for full speed grow higher as well.

90% is ok for me, but if any inp with over 90% on average speed slow down under 90% in critical phasis of gameplay, it will be removed. Beside an inp under 90% can be accepted if there is slow down only on no game play phasis.