forums.marpirc.net

JoustGod wrote: And sorry...had to chuckle when you of all people bring mention to someone else that they "...didn't need to write a novel...". ;-)

HAHAHAHAHAHAHAHA

I'm sorry Rick, but it's true :D

JoustGod wrote:And sorry...had to chuckle when you of all people bring mention to someone else that they "...didn't need to write a novel...".

I can understand that... you too BBH.

He totally explained the game as if I had never even seen it before though. Given I have just tried to assist him in playing Centipede based on my own centipede and other anlog game experiences...I think it goes without saying that shows I understand the game.

I only asked what the "blob technique" was cuz I wanted to be positive that by that he didn't mean the mushroom grouping that keeps the feed-out/through stuck on the edges of the board...so you are then free to leech off spiders. I didn't want to assume he just meant preventing the fleas from dropping. I know that's a crucial technique especially for the later stages in a loop where the centipede is very divided even initially.

Anyway, I hope he finally got his sensitivity issue figured out.

LN2 wrote:Can you get specific on what you tried? It sounds like you just tried a few extremes...instead of trying a range of realistic sensitivities for centipede. I had suggested trying in the 25-60% range.

I have tried every setting 0-255, one by one. None of them has any affect on the fact that the shooter will not move fast enough for me to play the game in any fashion similar to the way I play it on the arcade machine.

LN2 wrote:Note you might need to reset the game after changing that...not sure.

No, the change is made "on the fly".

LN2 wrote:Have you turned off your OS mouse acceleration as suggested above?

My mouse on my computer at home does not have acceleration, but on computers that do, I have turned it off.

LN2 wrote:I play on a mac so I'm not that intimate with pc-mames.

Could it be the problem doesn't exist on Mac mames?

LN2 wrote:Well, do you admit that you can move your mouse or in your terms...your OS cursor a lot faster than you can move that Centipede man?

From one side of the screen to the other, sure. As I mentioned before, there was a physical limit to how fast the trackball could be free-spun. As a matter of fact this varied greatly from one machine to another, depending on the condition of the trackball. Within the range of motion you could achieve while keeping contact with the trackball however, the shooter could move just as fast as your hand could. If there was a ceiling to it, it was irrelevant because my hands could not move fast enough to detect it, and my hands can move pretty fast.

LN2 wrote:If you make the same flaw or mistake on different PCs...of course you will get the same results.

LOL, I wish you would at least open your mind to the possibility that this doesn't have anything to do with the settings on my machine, or the others I have tried. I'm 99% sure that it doesn't at this point, but I'm still willing to consider the alternative...

LN2 wrote:Ok, you didn't need to write a novel about the game. I'm very familiar with the game

Ahh...then you must be aware that the scores posted on MARP are not particularly strong by arcade machine standards. To what do you attribute this?

You can see for yourself how decent the gameplay is with a properly tuned sensitivity.

People thought the gameplay in Donkey Kong was decent, but no matter how hard you pull on that joystick, that little guy is going to move at his own pace. The beauty of Centipede was that it could move as fast as you could. That's why some of us loved it. That's what I want now. Does that make me a bad guy?

Does anyone know "negative1"? He seems to "own" the high scores on Centipede and its clones on MARP. Perhaps if he is also familiar with the arcade machine he could comment comparatively on the control response between the two.

Thanks again,

John

Weehawk wrote:I have tried every setting 0-255, one by one. None of them has any affect on the fact that the shooter will not move fast enough for me to play the game in any fashion similar to the way I play it on the arcade machine.

Well, the problem there must be something with your mouse drivers or your OS settings(did you try a lower OS setting as suggested by another above?) I just know that using my mouse for trackball games works quite well.

I think it's likely working for you also because of what you said in your original message....but for some reason your mouse driver or OS settings has it where even at "0%" sensitivity that's too high for you...so try lowering your OS sensitivity. It must be something that very few others if anyone have to do though.

In your original message you said barely moving the mouse makes the player zoom across the screen. Well, there you go. You need to lower sensitivity so to get that same motion it requires a faster and larger mouse movement than what you said in your original post above.

If you are getting that behavior even at 0%...then set your OS mouse drive sensitivity really low and try it again...see if any differences. You might be using an extraordinarily high OS-mouse sensitivity that most others don't use.

Could it be the problem doesn't exist on Mac mames?

Nope cuz I do remember helping a friend get started with mame on his PC where I got the mouse going for him and demonstrated how to tune it for an anolog game. I used Crystal Castles as a sample...and it will worked just fine after a few trial and errors.

Perhaps in changing this sensitivity you are changing the wrong ones?!?!?
Have you tested changing the key sensitivity and see how that affects the key press control? I know you can't really play it like that....but how you see that affect the key response is similar to how it should be affecting the mouse response for the track x and track y sensitivities. Perhaps your mouse is setup as 2nd player somehow so you need to also try setting the track x 2 and track y 2? That shouldn't be the case but given you say nothing seems to change it simple logic would say you aren't changing the right thing. Maybe you have been changing track x 2 and track y 2 for this...which is supposed to be for the 2nd player controller. Of course normal behavior is changing those won't affect your first controller.

Have you tried other analog games(take a simple racing game like pole position or even crystal castles or marble madness or something) and see if you get the same behavior you are getting for centipede or if it works.

the shooter could move just as fast as your hand could.

I played a lot of Centipede and never came across one where you could move it around as fast as you can move your hand.

Which rom set are you using? There are some rom sets for certain games that have issues.

What version of mame are you using?

I wish you would at least open your mind to the possibility that this doesn't have anything to do with the settings on my machine, or the others I have tried.

Given you have repeated this on several PCs as you claimed above, I was opening your mind that if you are doing it the same way on several machines, of course you likely are going to get the same result.

As I said there are tons of scores here for trackball games that all show they have excellent control using their mouse.

Ahh...then you must be aware that the scores posted on MARP are not particularly strong by arcade machine standards. To what do you attribute this?

I disagree. I think 700k is quite a decent score vs arcade scores.

Negative1 has excellent scores posted. He has 362k for the rev 3 of centipede which is the one most of us played in the arcades.

A 300+k score is an excellent score for Centipede. All those arcade scores you find in the millions are all using the spider leech trick which is now illegal for the arcade records by Twin Galaxies. TG is still in the process of setting up a new system and once setup tons of new scores will be added plus many older scores are being reviewed game by game with the result being many of them are removed as they used cheap techniques to get the high score where TG is now making all of those types of techniques illegal.

I am curious what the actual high score for Centipede is playing it regularly. You mentioned Eric Ginner. I can't quite remember but I thought his high score was around the 800k mark...although a 3 million score comes to mind also. Do you remember? TG seems to not list eric's scores so perhaps even he used that spider leeching technique so the record was removed.

As far as MARP....you can find many scores here that are perhaps only "so-so" scores. There are tons of games to play within MAME....only so many are playing games in MAME....so there are likely great players of each individual game that just aren't aware of MAME etc. or care to post a score for it.

The beauty of Centipede was that it could move as fast as you could.

Well, I disagree here. It might have seemed that way cuz the way that trackball was you couldn't really spin it like other games to get that ridiculously fast motion...at least I never came across one that allowed that. If you tried to spin it you more than often would get pinched.

Also, when you think about the actual controller, there is a MAX voltage those produce from the trackball movement. If you move the trackball even faster it is NOT possible to give a higher voltage than this max value.

Now some games in their in-game setup panels do have a setting to adjust the sensitivity of the control vs the voltage reading. I don't think centipede has this but would need to check. Even if it does it's not necessary for the above tuning to work properly.

The difference in control from that and mame is you wouldn't have to worry about moving it "too fast"....cuz too fast would still be moving at the max speed the game allows. In MAME you do have to worry about going too fast which will result in jittering in place or going the other direction. Tuning can help that some..but there is a compromise between that and wanting a sensitivity so typical movement is fine.

That's something that's subjective for each of us. I have it where the general sense is close to what the arcade one was....although moving a mouse around can't be compared to a trackball.

Just remember that Negative1 has had a lot of scores DQ´d due to slowdown problems. And yes, even on games as old as centipede. I just want to point that out. Of course his centipede scores are ok until proven bad. Also I´m not saying Negative1 is a cheater or something like that. He just had some problems with the speed and a too old pc I think

Just something to think about.

QRS, you bring up an interesting point...very interesting.

Negative1 was very active for about 4 months last fall and winter...but he suddenly stopped playing at the same time alphamame came along. He hasn't submitted 1 score using alphamame. Coincidence with his past of using slowdowns? maybe...maybe not.

LN2 wrote:In your original message you said barely moving the mouse makes the player zoom across the screen.

Well, I said that moving the mouse a smaller distance made the shooter move farther when i set the sensitivity to 255, but then that is to be expected.

And to repeat: whatever changes I make to my OS mouse "sensitivity" (they call it "speed") does not appear to change anything in the game.

LN2 wrote:Have you tested changing the key sensitivity and see how that affects the key press control?

Yes, it works as expected.

LN2 wrote:Perhaps your mouse is setup as 2nd player somehow so you need to also try setting the track x 2 and track y 2?

No, changing the sensitivity in MAME has the expected effect, it just doesn't change the fact that the shooter's speed seems grossly limited.

LN2 wrote:Have you tried other analog games(take a simple racing game like pole position or even crystal castles or marble madness or something) and see if you get the same behavior you are getting for centipede or if it works.

Yes, as a matter of fact about a week after I joined MARP I captured the high score on Snake Pit. The mouse worked well enough there that I thought there was no problem. After I tried to play Centipede however, I went back to Snake Pit to verify that the problem did not occur there, but I found that it did. It's just that in Snake Pit you don't have to make the same lightning fast moves that you do in Centipede, as a matter of fact you generally are moving with much caution, and so the problem never was an issue.

LN2 wrote:I played a lot of Centipede and never came across one where you could move it around as fast as you can move your hand.

Maybe your hands are faster than mine, otherwise I can't explain this.

LN2 wrote:Which rom set are you using? There are some rom sets for certain games that have issues.

"centiped" - Revision 3

LN2 wrote:What version of mame are you using?

My tinkering on this has been using mame69.

LN2 wrote:I disagree. I think 700k is quite a decent score vs arcade scores.

But there's only one score in that range, and the gentleman that did that pointed out that he used the blob method, which does not require the speed that playing in the traditional fashion does - and now it's being pointed out that he may not have been playing the game at full speed anyway!

LN2 wrote:As far as MARP....you can find many scores here that are perhaps only "so-so" scores. There are tons of games to play within MAME....only so many are playing games in MAME....so there are likely great players of each individual game that just aren't aware of MAME etc. or care to post a score for it.

Well, that's true enough, but I still believe the scores on MARP are lower than they would be if MAME could offer the same type of control repsonse that the arcade machine did.

LN2 wrote:I am curious what the actual high score for Centipede is playing it regularly. You mentioned Eric Ginner. I can't quite remember but I thought his high score was around the 800k mark...although a 3 million score comes to mind also. Do you remember

My problem is that for any score I saw recorded, I would have had no way of knowing what method was used to achieve it. I watched a friend do over 800k once playing straight-up, and my own highest was probably just under 700k, so I assume there were those who were good enough to break a million that way. Multiple millions would be made extremely difficult however because......

(Deep breath again) (but this time to explain something that even someone "very familiar with the game" might not know)

....between 996,000 and one million points, you earn an extra "man" for everything you shoot. You can only store 6 extras though, so most will be wasted, and for each one you are "penalized" after you go over one million, in that you will not earn an extra for that same number of times after one million that you otherwise would have. So unless you are careful to go from 996,000 to one million by shooting just 4 1000 point scorpions, or 5 900 point spiders, you may be going a long time without earning any extra men. If you are fighting through the later stages of the sidefeed cycle at this point, this is rough.

LN2 wrote:If you tried to spin it you more than often would get pinched.

LOL, I think this only happened with trackballs that were a little out of whack, but it brings back memories.

LN2 wrote:Also, when you think about the actual controller, there is a MAX voltage those produce from the trackball movement. If you move the trackball even faster it is NOT possible to give a higher voltage than this max value.

Wouldn't the same hold true for a paddle controller in Pong? This didn't change the fact that the paddle could move faster than the eye could see.

LN2 wrote:That's something that's subjective for each of us. I have it where the general sense is close to what the arcade one was

Maybe that's the subjective question: How close is close? To me its like playing a completely different game.

Any other arcade Centipede players out there want to comment on how well the control response in MAME compares to the real machine?

John

Weehawk wrote:....between 996,000 and one million points, you earn an extra "man" for everything you shoot. You can only store 6 extras though, so most will be wasted, and for each one you are "penalized" after you go over one million, in that you will not earn an extra for that same number of times after one million that you otherwise would have.

That's very interesting... I didn't know that. It makes those multi-million scores even more suspect knowing that....cuz even if you do what you suggested, that means you are going the next 60 or 72k or more without earning an extra man. I guess if at 996k you only had 1-2 men then you aren't losing much really cuz you would earn 4 in that...but then not earning 5-6 so net loss of only about 2. hehe...although it could be a ton more if you happen to be hitting centipede segments when that extra man stuff starts up.

LOL, I think this only happened with trackballs that were a little out of whack, but it brings back memories.

Perhaps I never played one as pure as you got to play then. It also sounds like you move around much more and more quickly than I did even when I played in the arcade. My arcade personal high for that game was around 280k.

Maybe that's the subjective question: How close is close? To me its like playing a completely different game.

Maybe...cuz perhaps you are getting good response but exaggerating how it behaves so much I am getting the wrong picture. Even so, how much you have to move the mouse corresponding to movement and speed of movement should be drastically changing as you vary that analog control sensitivity. You said it doesn't change a thing going all the way from 0 to 255. That makes no sense to me. If I set the sensitivity to something low like 10%, then I have to move my mouse quite a few feet to get it from one end to the other in centipede. I have it set at 50% and still have to move it a foot to get it from 1 side to the other. Yes, if I go spastic in my mouse movement I get the overload also. I explained what causes that. How fast you need to move the mouse before you reach that overload changes with sensitivity changes as well. You seemed to indicate it's the same regardless of the settings.

You don't really want to set that where you can move the mouse as fast as possible to be that max...cuz then you have to move it a long distance to get from 1 end to the other. There is a compromise there.

I agreed with you to an extent that a new feature to mame could be added where the max is handled properly...if not to change the rons by adding a file with that info...perhaps in the mame game settings a max input value that is adjustable..so for each game you could tune a max that matches the game's max and doesn't overload it....yet use whatever sensitivity you want.

That's the only aspect I am focusing on right now...cuz until you can properly tune that you can't really judge how the max speed you can move around relates to the arcade game.

Try a game like Marble Madness..or even millipede(which I always liked better anyway). That snake game you tried doesn't sound like a good test anyway.

Given you have only tried the 2 games....try a few others. What speed is the game running at? Have you hit F11 to see what speed the game is running at? Are you using automatic frameskip?

I'm guessing if you have a slow PC where some frameskip is being used that very easily could influence the response and thus speed of being able to move around cuz the device isn't being polled for game input at the full rate. I'm guessing you have a more than adequate PC though...but figured I would ask.

when you say you are using mame69..which one? Have you tried other builds or versions?

I guess I should say something here...

All those arcade scores you find in the millions are all using the spider leech trick which is now illegal for the arcade records by Twin Galaxies.

Well, not *all* of them

I am curious what the actual high score for Centipede is playing it regularly.

November 5, 2000: 7,111,111 (9 hours)
The game was videotaped and approved by Twin Galaxies to be "shoot-em-up" style of play (no blob, no side-feed trap).

between 996,000 and one million points, you earn an extra "man" for everything you shoot

Almost, but not completely accurate. The "extra man" you mention is for everything that you score points for which is not the same as everything you shoot. The case where this happens is when you shoot and damage a mushroom...that does NOT count as an extra man. If you happen to lose your shooter then the damaged mushrooms count up and then those actually count as extra men.

That's very interesting... I didn't know that. It makes those multi-million scores even more suspect knowing that....cuz even if you do what you suggested, that means you are going the next 60 or 72k or more without earning an extra man.

Well, not really. First, 60-72K is actually not hard to do with a full complement of free men. My personal first-man best is over 200K (shoot-em-up style, of course). Second, that's not how I do the rollover anyways.

I guess if at 996k you only had 1-2 men then you aren't losing much really cuz you would earn 4 in that...but then not earning 5-6 so net loss of only about 2. hehe...although it could be a ton more if you happen to be hitting centipede segments when that extra man stuff starts up.

Having only 1-2 men at 996K would not help you. The counter that the machine keeps for when to give you the next free man starts immediately at 996K regardless of how many men you have built up. So, if you hit 6 things to pass 1M, you're next free man would be at 68K (note that it's not 72K...brownie points for anyone who can deduce why)

(deeper breath...)
Ok, the "trick" to doing the rollover properly is to think in the opposite direction. Instead of hitting as few things as possible, what happens when you hit lots of things? This is similar to what happens in Defender, I believe. So, if you were to hit (which really means "score points for") 10 things, the next real free man is delayed 120K. 20 things, it's delayed 240K. Continuing on to 80 things, that's 960K. Well, what about 84 things?? That's 1008K. But the machine doesn't keep track of the millions, so this becomes 8K which would be very manageable as a "delay" for getting your next free man (it's actually sooner than you would normally get it!) So, that's the basic idea of how to do it.

Any other arcade Centipede players out there want to comment on how well the control response in MAME compares to the real machine?

I bought a $20 trackball mouse at CompUSA to use for playing Centipede on MAME. I found that the analog settings in MAME were useful in making the "feel" of the trackball closer to the arcade machine but it never got quite good enough for my taste. I have my own Centipede machine at home anyways so it hasn't been a priority for me to get it working in MAME. Part of the problem is also the trackball itself...it just wasn't a high-quality piece of equipment and it just wouldn't respond to the full beating I normally give to the arcade machine's trackball

An interesting thread ! Have any of you tried a 'marble mouse' ?
I have one , but haven't had time to or reason to use it for Centipede . Could be worth checking out...

As for those 'Novel-type threads'......well Ben Jos would be proud

Cheers ,

AL

dbh wrote:Having only 1-2 men at 996K would not help you

You didn't get my logic there....if you only had 2 men at the time you hit the 996k mark...then 4 of the extra men you get on your way to turning it over at 1 million actually are stored as more extra men for you...so those aren't "lost" men. So...if you only had 2 left when you hit that versus a whole compliment of 6 it might be good for you to get those to go from 2 back up to 6. if that happened then you really only "lost" 2 men....cuz if you gained "6" in going from 996k to "0"...and got to store 4....then you really only lost 2. If you already had 6 built-up...then you are losing all 6.

It actually gives a player some strategy to think about...cuz if you are at 950k and have 6 men....you can take it a little easy and perhaps plan on things so when you hit the 996k point it can be at/near the start of a fresh loop....where you can more easily just pig out on a few spiders to quickly go from 996k to >1 million versus having to shoot individual centipede segments where you could end up having 10+ "earned" then so wouldn't get another man until like 120+k.

Continuing on to 80 things, that's 960K. Well, what about 84 things?? That's 1008K. But the machine doesn't keep track of the millions, so this becomes 8K which would be very manageable as a "delay" for getting your next free man (it's actually sooner than you would normally get it!) So, that's the basic idea of how to do it.

ahh, that's even a better strategy than what I said above...as long as you keep an accurate count.

Playing that for 9 hours is very good...I would think timing and coordination would start to break down after a few hours of that game....although I guess if you have your "channels" setup well...then the game gets a little easier versus even the first parts of the game where you don't have those channels setup.

You didn't get my logic there....if you only had 2 men at the time you hit the 996k mark...then 4 of the extra men you get on your way to turning it over at 1 million actually are stored as more extra men for you...so those aren't "lost" men. So...if you only had 2 left when you hit that versus a whole compliment of 6 it might be good for you to get those to go from 2 back up to 6. if that happened then you really only "lost" 2 men....cuz if you gained "6" in going from 996k to "0"...and got to store 4....then you really only lost 2. If you already had 6 built-up...then you are losing all 6.

Actually, yes, I did get your "logic". What I am saying is that it does not matter whether you have 1 or 2 or even 6 men when you hit 996K. You will still not gain free men after passing 1M based on how many things you hit that score you points. If you had 2 men at 996K and hit 6 things before passing 1M, your next free man will be at 68K (you will have 6 men in the bank, though). If you had 6 men at 996K and hit 6 things, your next free man will still be at 68K. There is no difference between the two.

It actually gives a player some strategy to think about...cuz if you are at 950k and have 6 men....you can take it a little easy and perhaps plan on things so when you hit the 996k point it can be at/near the start of a fresh loop....where you can more easily just pig out on a few spiders to quickly go from 996k to >1 million versus having to shoot individual centipede segments where you could end up having 10+ "earned" then so wouldn't get another man until like 120+k.

Yes, there is strategy involved with doing the rollover properly, but it has nothing to do with the number of men you have before reaching 996K.

ahh, that's even a better strategy than what I said above...as long as you keep an accurate count.

Yes, you need to be able to count. It's a very controlled process for me so it is now second-nature.

Playing that for 9 hours is very good...I would think timing and coordination would start to break down after a few hours of that game....

Not really...once the game really got going I was pretty much in the zone for the whole time. I only had one "scare" at 2.7M (about 3.5 hours in) where I was down to my last man. The rest of the time I don't think I hardly ever dropped below 3 men in reserve.

although I guess if you have your "channels" setup well...then the game gets a little easier versus even the first parts of the game where you don't have those channels setup.

I don't intentionally shoot out any "channels" when I play. They increase the risk of having a stray shot going all the way up the screen which leaves you defenseless.