What's new
By:

What would be the correct 12mhz crystal oscillator to upgrade an A board

I have a spare SF:WW board that I could use for a conversion, if anyone ends out offering this service. Please PM me.
 
I don’t remember if this is the only thread on this subject, but it’s at the top of my watched list, so thought I would post it here.


Captured gameplay from a B+C SF2HF running on each of the following ‘A’ boards:


1) Stock 10 mhz 'A' board

2) Modified 10 mhz 'A' board, with a 12mhz oscillator replacing the stock oscillator

3) Stock CPS Dash 'A' board


Then, took the best “apples-to-apples” samples, for side-by-side comparison of the frame data.I’m sure there are much more empirical ways to do this analysis/comparison, but I think this is adequately representative.

While it's a small sampling of comparisons here, I think it gives a good snapshot of the hardware behavior. From these comparisons, it is hard to say that there is any difference in gameplay speed, from one configuration to the next.


https://www.youtube.com/watch?v=b7ZK8SwFitY
 
YesAffinity said:
While it's a small sampling of comparisons here, I think it gives a good snapshot of the hardware behavior. From these comparisons, it is hard to say that there is any difference in gameplay speed, from one configuration to the next.
Click to expand...
So in your opinion is it an advantage or a disadvantage playing a 12Mhz game on a 10Mhz board or vice versa?

I'm not too sure. If what you are saying is correct then a 12Mhz game on a 10Mhz board runs at the same speed but could suffer from slowdown where it wouldn't on a 12Mhz board?

I think its questionable if the slowdown is an advantage or a disadvantage. For example say Area 88 on a 10Mhz board you kind of rely on the slowdown to dodge bullets on a busy screen, same with Final Fight, the slowdown gives you a little bit of time to make decisions on a busy screen. If you play these on a 12Mhz board and it reduces that slowdown it could be argued that this makes it harder? You read about people who play Cave shmups crying about the Xbox360 ports not suffering from slowdown like the PCBs do.

People kind of assume that slowing down hyper fighting is an advantage because "the game runs slower" but I'm not so sure it slows the game down as much as its just a matter of the system having less cycles to work with. I think if there are moments of slowdown at all they could mess you up just as much as they might benefit you.

I wonder if any of the MAME gurus have a way to tell if the CPU is ever a bottleneck during a playthrough of SF2' at all running at 10Mhz vs 12Mhz.
 
Yeah, I think that becomes for of a question of practicality, and not necessarily a cut-and-dry answer.

From my brief tests, there was no difference. Taking a wider sampling, looking at more games, etc. could produce a clearer understanding of what the benefits or changes would be depending on the title and the A board version it is paired with.

One thing I am somewhat convinced of but haven't had a chance to empirically test yet, is the speed at which chainable attacks can be executed, comparing 10mhz to 12mhz. SF2HF on 10mhz does not have ability to chain (for instance, Ryu or Ken's crouching short kick) as rapidly as can be done on a 12mhz board. If it can be proven, that would be a clear disadvantage if not running SF2HF on a 12mhz board - less damage potential, less dizzy potential, in those situations.
 
Thinking about this, I wonder if MAME could be configured to run SF2:CE at 10Mhz?

If this is possible we could record an .inp with the CPU at 10Mhz then play it back with the CPU at 12Mhz and see if it de-synchronizes. Then do the same test in reverse. That might provide some insight.

Maybe @l_oliveira or some other MAME guru would know if it's possible.
 
YesAffinity said:
I don’t remember if this is the only thread on this subject, but it’s at the top of my watched list, so thought I would post it here.


Captured gameplay from a B+C SF2HF running on each of the following ‘A’ boards:


1) Stock 10 mhz 'A' board

2) Modified 10 mhz 'A' board, with a 12mhz oscillator replacing the stock oscillator

3) Stock CPS Dash 'A' board


Then, took the best “apples-to-apples” samples, for side-by-side comparison of the frame data.I’m sure there are much more empirical ways to do this analysis/comparison, but I think this is adequately representative.

While it's a small sampling of comparisons here, I think it gives a good snapshot of the hardware behavior. From these comparisons, it is hard to say that there is any difference in gameplay speed, from one configuration to the next.


https://www.youtube.com/watch?v=b7ZK8SwFitY
Click to expand...
Sorry for reviving an old thread. It is expected that an overclocked @12Mhz A board will behave as an original 12Mhz, the processor which is overclocked may heat a bit more but will behave the same. The good part of that test was that it showed that the 10Mhz version works basically like the 12Mhz version. What one would expect running a 12Mhz game in a 10Mhz board would be more slowdowns when many things go on the screen.
 
I have experienced extra slowdowns in Pang 3 using a 10MHz oscillator instead of 12MHz when there are many bubbles in the screen.

I tried a 12MHz oscillator from Farnell and it didn't work, I used this one recommended by other user and it worked perfect: https://www.ebay.es/itm/401396958941

I also had success with a 'Sunny' oscilator from a scrap board.
 
Last edited:
rsignal said:
I have experienced extra slowdowns in Pang 3 using a 10MHz oscillator instead of 12MHz when there are many bubbles in the screen.
Click to expand...
Yes, that's exactly what I meant. I would expect the same on other fighting or Beat'em up games when multiple enemies show on the screen at the same time.
 
I've spent a considerable amount of time playing Final Fight lately.

This sequence with two lots of fire men brings the game to it's knees, I wait out the lag while standing on the bins.


If you want to compare something in Final Fight, this is the only bit where you really notice the lag except maybe the fire bridge at the start of stage 4.
 
BladeZX said:
The 12mhz upgrade still working ok on the 10mhz A boards?
Any links to the proper 12mhz xtal we need?
Thanks
Click to expand...
I can send u a part number when i get home from work, but it seems we won't be needing it since there's the option to put a DCS on the multi. I have been running it on my mercs board without issue for a couple months now. Pretty sure i got mine from digikey
 
i'm surprised nobody has tried 16MHz yet :)

the 68000 itself can be overclocked way more than you would think,
in the amiga days we where running 10MHz chips at nearly 16MHz continously.
the first thing to be a problem when you overclock is usually the ram or eproms.
 
if you get a crystal that doesnt work, check the datasheets.
some have an "enable" pin that needs to be connected, others dont.
i dont know what cps boards need, but given the age - i doubt they use an enable pin.
 
You must log in or register to reply here.
Back
Top