(Topic ID: 168721)

How "powerful" Were SS Games?

By mbaumle

7 years ago


Topic Heartbeat

Topic Stats

  • 21 posts
  • 9 Pinsiders participating
  • Latest reply 7 years ago by thedefog
  • Topic is favorited by 4 Pinsiders

You

Linked Games

No games have been linked to this topic.

    #1 7 years ago

    Maybe a moot point, but I'm hoping will make an interesting discussion topic; I went to goodwill the other day and picked up an old Ti-1200 Calculator for $2. Originally introduced in 1975, and originally costing $24.99 ($111 in today's money) I can see how impressive that calculator was for it's time, honestly, even looking at it today, I can appreciate it's technological abilities.

    Anyway, I'd imagine that Solid State games of pinball machines, of the late 70s to the late 80s, had to have been pretty powerful for their time, considering they had to not only perform calculations, but also provide light shows, play music, operate machinery, and do so without crashes or freezes. Would that be a fair assessment? How much processing power did these earlier games have compared to other computers at the time?

    Additional thought while typing this: A William's System 9 board set is essentially the same as a System 11 board set, yet there's a clear difference in what, say, Space Shuttle or Comet can do vs. Whirlwind. What bottlenecked the hardware of the system 9 boards from keeping William's from making complex games?

    #2 7 years ago

    The processing power didn't change a whole lot. For example, classic bally/stern used a 6800 8-bit CPU which ran at 1MHz. WPC games ran on a 6809 16-bit 2MHz CPU.

    Computers during the WPC era were generally running at 60MHz in 1993, slowly increased up to 800MHz throughout the span of the 90s, and eventually broke the 1GHz barrier in March 2000.

    One thing that did change rapidly was sound technology. Which is why sound chips & amps always seem to be hard to obtain today--they were only made for a short time before being discontinued and moving on to the next latest and greatest thing.

    Display technology also leaped ahead in fits and spurts, but not nearly as quickly as sound tech.

    But generally, the CPU in pinball machines didn't need to be too powerful. It followed a few simple if-then instructions and did some simple addition and multiplication, and that was about it.

    #3 7 years ago
    Quoted from mbaumle:

    there's a clear difference in what, say, Space Shuttle or Comet can do vs. Whirlwind. What bottlenecked the hardware of the system 9 boards from keeping William's from making complex games?

    1. Sound ability (whole separate CPU, often running faster than the main CPU)
    2. Playfield design
    3. PROM space: Sys3 had maybe 2KB of space for all its code, WPC had up to ~2MB iirc. Also cheaper ROMs meant more sounds
    4. Competition with their older games (if your games aren't getting any 'better' it looks bad, so you need to keep outdoing yourself and your competitors). Technically, besides from a fancier sound, Black Knight and Hot Tip run on HW with the same capabilities, but they didn't make Black Knight then, because they didn't need to

    #5 7 years ago

    Mad powerful. Many gigaflops and terahoozits.

    #6 7 years ago
    Quoted from zacaj:

    3. PROM space: Sys3 had maybe 2KB of space for all its code, WPC had up to ~2MB iirc. Also cheaper ROMs meant more sounds
    4. Competition with their older games (if your games aren't getting any 'better' it looks bad, so you need to keep outdoing yourself and your competitors). Technically, besides from a fancier sound, Black Knight and Hot Tip run on HW with the same capabilities, but they didn't make Black Knight then, because they didn't need to

    That's sorta what I was thinking--There really weren't any technical bottlenecks with older hardware, other than ROM space. If they could theoretically make a Black Knight in 1977, then I'd have to think that their original hardware specifications were pretty top notch for their time, at least when they were being developed in the early to mid '70s.

    Thanks for the comments, guys. I've always had a fondness for computer hardware--especially with older technologies, and what they were able to accomplish with what they had at the time.

    #7 7 years ago

    Comparing with computers of the late 70's and 80's is a tough comparison, because it wasn't until the mid to later 80's that people were really buying home computers. Obviously the servers of the time were much more powerful. And code applications and compilers were different than what was done with pinball.

    Pinball MPUs really were using relatively inexpensive off the shelf processors & RAM for the most part, so they weren't overly powerful in comparison to even some of the home systems that were available. x86 & 680x0 processor based machines of the time were much more powerful.

    So to answer your question, they weren't powerful for the time because they didn't really need to be. Technically they still don't need to be. You can build a happily functioning pin with an ATmega256 microprocessor. They only had to run fast enough to catch and register switch hits and run simple subroutines. And since they were programming these in straight-up assembly, they had the ability to maximize the order in which things occurred as to not clog up the CPU cycles with useless instructions and loops. In other words, these guys did amazing stuff with code.

    What really is interesting was the choice to use the 6800 processor by Williams when the 6502 was available at a much cheaper price with roughly identical specs. Gottlieb took advantage of this, but they were the only ones. I don't know the full history behind those choices though.

    #8 7 years ago

    The early bally computer runs at 0.6MHz, have 256x4bit + 128x8bit of RAM, and used 4kb of software. Its a pretty amazing they pulled off great games with so little.

    #9 7 years ago
    Quoted from barakandl:

    The early bally computer runs at 0.6MHz, have 256x4bit + 128x8bit of RAM, and used 4kb of software. Its a pretty amazing they pulled off great games with so little.

    Pure code. Really efficient ASM programmers that understood priority of instructions. It blows my mind when I watch the VPinMame Debug window and I see how quickly the CPU will get out of subroutines, with seemingly no time to spare, all while juggling all the other data being sent all over to the various peripherals. These guys were undoubtedly knee deep into the mechanics of the game layouts as well, not just code monkeys. I'd love to hear from some of the earlier game original programmers. They don't get as much recognition and credit as the layout designers and artists.

    #10 7 years ago
    Quoted from thedefog:

    Pure code. Really efficient ASM programmers that understood priority of instructions. It blows my mind when I watch the VPinMame Debug window and I see how quickly the CPU will get out of subroutines, with seemingly no time to spare, all while juggling all the other data being sent all over to the various peripherals. These guys were undoubtedly knee deep into the mechanics of the game layouts as well, not just code monkeys. I'd love to hear from some of the earlier game original programmers. They don't get as much recognition and credit as the layout designers and artists.

    Bally multiplexed to the extreme. The PIAs do amazing work in these games too toggling between updating displays, reading switches, etc.. Bally uses two PIAs, WMS was using 4 or 5 in the same era.

    Anyone know who was the main engineer when designing the Bally -17 electronic board set?

    #11 7 years ago
    Quoted from barakandl:

    Bally multiplexed to the extreme. The PIAs do amazing work in these games too toggling between updating displays, reading switches, etc.. Bally uses two PIAs, WMS was using 4 or 5 in the same era.
    Anyone know who was the main engineer when designing the Bally -17 electronic board set?

    I can only image how much of a PITA programming on the Bally board must have been...at least initially while they were figuring out how to juggle everything. Programming in Assembly is not fun unless you're making a drinking song in it.
    http://www.99-bottles-of-beer.net/language-6800-assembler-2423.html

    #12 7 years ago

    That's the thing - there's no layers of crap over top slowing stuff down. No OS, no graphical interfaces, nothing but direct hardware addressing.

    Can you imagine how fast modern computers would be without so much cruft?

    I also wonder at the choice of the 6809 vs. 6502.

    #13 7 years ago
    Quoted from jwilson:

    That's the thing - there's no layers of crap over top slowing stuff down. No OS, no graphical interfaces, nothing but direct hardware addressing.
    Can you imagine how fast modern computers would be without so much cruft?
    I also wonder at the choice of the 6809 vs. 6502.

    The 6809 was more advanced at the time, and from the article says, easier to write code for.

    https://en.wikipedia.org/wiki/Motorola_6809

    #14 7 years ago
    Quoted from barakandl:

    Bally multiplexed to the extreme. The PIAs do amazing work in these games too toggling between updating displays, reading switches, etc.. Bally uses two PIAs, WMS was using 4 or 5 in the same era.

    I never really thought of that before. And to think that, generally speaking, Bally games had much deeper coding, rules, and more "stuff" to do than their William's counterparts.

    And given that, I'd be curious to know, had William's software engineers multiplexed as much as Bally's engineers did, would their hardware have been able to produce even more complex games?

    Quoted from jwilson:

    That's the thing - there's no layers of crap over top slowing stuff down. No OS, no graphical interfaces, nothing but direct hardware addressing.
    Can you imagine how fast modern computers would be without so much cruft?

    I think the most modern example of this happening is with the Nintendo Wii. The OS is offloaded into a small ARM processor, whereas the actual game programming is run on bare metal, meaning that the main CPU and GPU can utilize 100% of it's resources to running the game. The main CPU and GPU of the Wii were/are dog slow, but they were able to accomplish pretty amazing visual effects due to this style of programming.

    #15 7 years ago
    Quoted from mbaumle:

    And to think that, generally speaking, Bally games had much deeper coding, rules, and more "stuff" to do than their William's counterparts.

    I'm not sure if they had more complex code or just better designed rules. Starting with system 7 Williams actually had a sophisticated, multithreaded byte code operating system allowing them to fit more rules in less ROM space. I've also noted Williams games tend to have a lot of sophisticated lighting effects (the pulsing bonus counter or the animated displays during multiball on black Knight, for instance). Stern and Gottlieb also supposedly switched to a state based architecture, while bally didn't make any changes, according to some things I've read, which supposedly caused them problems

    #16 7 years ago
    Quoted from ForceFlow:

    The 6809 was more advanced at the time, and from the article says, easier to write code for.
    https://en.wikipedia.org/wiki/Motorola_6809

    There are advantages to both processors. The 6809 was apparently easier to program on and had more hardware registers (I think 6, vs 3 on the 6502). That right there could have been the crucial decision point. The 6502 was pipelined, which allows it to process more stuff in a single cycle.

    #17 7 years ago
    Quoted from ForceFlow:

    The 6809 was more advanced at the time, and from the article says, easier to write code for.
    https://en.wikipedia.org/wiki/Motorola_6809

    I actually meant the 6800, which was out before the 6502.

    I asked an industry veteran and he pointed out an important fact - Motorola was only 15 miles away from Williams on California Ave. compared to MOS, who were down in SV. So, proximity probably played a big role, as well as timing.

    #18 7 years ago
    Quoted from thedefog:

    The 6502 was pipelined, which allows it to process more stuff in a single cycle.

    How about the 6803? It seems like Bally, keeping in their tradition of multiplexing things in their -17 board set, they moved to even more multiplexing with their 6803 board set. I was reading some old repair guides where it explains that Bally was able to double their outputs by multiplexing by using zero crossing on the AC power line. I thought that was amazing. Other than it's built in ability to multiplex, I can't imagine it being too different from other processors that were being used at the time.

    I'd imagine that from a theoretical standpoint, that Bally's 6803 hardware was more capable than William's System 11 hardware?

    #19 7 years ago
    Quoted from mbaumle:

    I think the most modern example of this happening is with the Nintendo Wii. The OS is offloaded into a small ARM processor, whereas the actual game programming is run on bare metal,

    I think the xbox and PS dedicate a single core to the os if I'm not mistaken.

    #20 7 years ago

    It is amazing to think about early SS games when they were new. I remember when calculators were new and expensive.Digital watches were LED and you mashed a button to see the time. Really cutting edge technology for the time. Cash registers and gas pumps I remember the old ones. I recall seeing my 1st SS Mata Hari and Power Play back in 78. And look at us now.

    #21 7 years ago
    Quoted from mbaumle:

    How about the 6803? It seems like Bally, keeping in their tradition of multiplexing things in their -17 board set, they moved to even more multiplexing with their 6803 board set. I was reading some old repair guides where it explains that Bally was able to double their outputs by multiplexing by using zero crossing on the AC power line. I thought that was amazing. Other than it's built in ability to multiplex, I can't imagine it being too different from other processors that were being used at the time.
    I'd imagine that from a theoretical standpoint, that Bally's 6803 hardware was more capable than William's System 11 hardware?

    The 6803 is a rock solid design, definitely the best of the era. A shame it was the end of Bally on its own. They really managed to get huge mileage out of very little, considering the sounds deluxe near the end on some of those games was much more powerful than the MPU, using a 68000 CPU. For comparison, the Capcom SYS-1 CPU is entirely driven off of a 68000 CPU, a Z-80 (gameboy), & a Raster Display IC. CPS-1 was Street Fighter II, Strider, Ghouls & Ghosts, Magic Sword, etc.

    Reply

    Wanna join the discussion? Please sign in to reply to this topic.

    Hey there! Welcome to Pinside!

    Donate to Pinside

    Great to see you're enjoying Pinside! Did you know Pinside is able to run without any 3rd-party banners or ads, thanks to the support from our visitors? Please consider a donation to Pinside and get anext to your username to show for it! Or better yet, subscribe to Pinside+!


    This page was printed from https://pinside.com/pinball/forum/topic/how-powerful-were-ss-games and we tried optimising it for printing. Some page elements may have been deliberately hidden.

    Scan the QR code on the left to jump to the URL this document was printed from.