Speed Cart 64: Slow Down Your Commodore 64, Sometimes
Vložit
- čas přidán 31. 07. 2024
- If you've ever found your Commodore 64 runs too fast, maybe you need a Speed Cart 64 which provides a way of slowing down the C64. Sometimes. Sometimes it actually speeds things up, and sometimes it glitches programs in bizarre ways. And sometimes it just crashes.
To support 8-Bit Show And Tell:
Become a patron: / 8bitshowandtell
One-time donation: paypal.me/8BitShowAndTell
2nd channel: / @8-bitshowandtell247
Bremse 64 cart: retroport.de/rex-datentechnik-1/
Extra thanks to:
Chuck Hutchins: / chuckhutchins
Jim 64: / @jim_64s8-bitprojects5
Jim Brain: store.go4retro.com/
Rod @ TFW8b: www.thefuturewas8bit.com/
Index:
0:00 The Speed Cart 64
2:05 Slowing down 10 PRINT
4:14 Testing some games: GORF, Jupiter Lander
6:36 Jumpman Junior, Mountie Mick's Death Ride
11:20 Bruce Lee II, Wizard of Wor
13:53 Examining effect on the TI (jiffies) variable
17:25 Bremse 64 (Brake/Break 64)
18:08 I attempt to explain the circuit...
24:17 Potential improvements, clarifications
26:05 Thanks to my patrons - Věda a technologie
"This is my favorite game about a Canadian Mountie."
Such a competitive genre. How can you choose?
He could chose if he made a game "Moose in the Caboose". :D
“My favourite game about a Canadian Mountie.”
Uh…. right…. Much better than all those OTHER games about Canadian Mounties ;)
at 4:00 the line "Very irregular flash" gave me visions of Barry Allen making lots of super-speed trips to the bathroom ...
If I've learned anything from CZcams, that potentiometer might need some deoxit
(I see what you did there with the credits)
Yeah, I noticed too. Pretty sly ;-)
If a cartridge could make the 64 slower, Commodore would have developed it in the '80s.
I thought it was built in!
@@immortalsofar5314 Look out, we got a Speccy fan here!
Yay! I got a greetz (I get so few...) I *think* I know why they used a TTL IC. Well, we all know they used this design because it was cheaper than other options, but it is cheaper in a slightly tricky way. The goal is to provide a way to optionally slow down the 64, for whatever reason someone wants to do that (I won't judge). The sensitivity of the design allows the same circuit to do three things. Create a constant IRQ (when turned counter-clockwise, the input is ~0V, and this means the IRQ is always held low, which is not useful), create a variable cycling rate for the IRQ (the good part), and create a constant lack of IRQ (in essence, the unit is off). Without that 3rd behavior, the design would also be creating IRQs at some frequency, no matter how slow, and a switch would be needed to turn the unit off. Thus, the temperamental nature of the circuit and the fact there is a region where it will not create an IRQ was used to cheaply create an "off" mode.
Thanks for the extra insights! Good analysis.
I absolutely loved Jumpman Jr.! Many a pleasant hour spent with that title.
I'm just now learning about TTL chips so this was extremely useful.
Whoever designed that board was apparently just learning, too. That was the way things where in the 80's. Modding and doing all kinds of fun, crazy things was the norm. 8)
Love your videos!!
Funny! That's exactly what I did with my VIC-20 15 years ago. A 74HCT161 connected to the RDY line if I recall correctly.
Nice, thanks for the video.
They were so busy wondering if they could do it, they never stopped to ask if they should.
Never heard of this cart before, very interesting!
Even if they didn't want to pay for a 555, they could have built a much better voltage controlled oscillator using all 4 NAND gates on that chip. Alternatively, this simple circuit will work better if a resistor is added between the output of the first gate and ground.
I love how they named 1/60th of a second to a Jiffy haha
@8:50 "This is my favorite game about a Canadian mountie". Wait, how many mountie games are there?
@26:15 hahah, nice (interrupted roll call)
Apparently "WWF Superstars 2" for Game Boy has The Mountie as a playable character! But I wasn't thinking of that when I made my statement :)
And it was made by... Ariolasoft! (insert nipple joke here)
I'd like to hear more SID's slowed down. Like Agent X II
Dear Robin thanks for introducing this amazing cartridge.back in time I saw the simular thing on zx spectrum , but I didn't know that's exist for C64 too. Apparently games which use Raster does not work property. 🙏🌹
Looks like a great way to bork things, and I like it
In addition to being useful for some types of games, I could see it being useful as a debugging tool, potentially, either by slowing something down so that you can catch it more easily or by pairing it with a special IRQ handler that aided in debugging by checking a memory location or some such more quickly than the IRQ would normally be able to process it.
Suddenly riolu reinvents himself as a C64 speedrunner...
If Jupiter Lander were realistic the lander would simply descend thru clouds until it was crushed by atmospheric overpressure.
Wow, really weirdly interesting! I had no idea that you could control something digital by throwing an elementary analog stimulus into the mix.
Wow... This is like circuit bending :D
There's a lot of people using Raspberry Pis as various Amiga accelerators, flickerfixers and SCSI boards - I wouldn't be surprised to see someone use a Pi Pico in the place of a 555. I'm sure it could read a potentiometer and output appropriate number of pulses - scaled better than a raw pot too...
I remember having a cartridge for the Commodore Amiga, Action Replay revision 3 if i remember correctly and it had a pot on there and you could slow the machine down. But it has been so long I cant remember how well it worked.
What happens when you hold down a key on the keyboard while being set slow? Faster key repeats?
A decelerator! Clock your 2 MHz 128 back down to 1 like God intended!
The unused circuit is for a led(left)+ resistor(bottom) and a reset button (right of the led).
That's the most compact version of 10 PRINT you ever wrote on this channel. ;-)
I'm wondering if it's 10 PRINT's final form, or if there's an even more compact version :)
@@8_Bit Perhaps "9 PRINT"?
You don't seem *nearly* as flaky as most Bass players I have the pleasure of knowing....
Bassists are the coolest people I know.
@@ericanderson85 I didn't say they weren't also cool...
It could also be useful for debugging too. Seeing exactly where on the screen a glitch occurs, abit like stepping through now with VICE using ALT/+ (or SHIFT/ALT/P on newer builds)
Great Vid! I wonder if an audio-taper pot would yield more linear results since the charge-curve of a capacitor is logarithmic?
Complemented perfectly by the SloDOS ROM replacement for the 1541 for when you're just not done with the six-course gaming dinner yet
Interesting device. Yes, I'd imagine it was originally cased. It would have indeed cost less doing it that way as a 555 circuit would need more components. Maybe the value of the pot is too high or low, that's why it's so sensitive? I seem to remember there was also a similar device that worked via a joystick port.
I wonder if a logarthimic or inverse logarithmic pot would actually make it less touchy and would allow you to find the "sweet spot" of resistance easier.
Slower, faster, and the occasional crash?
Call me when you get it to run things backwards.
Oh god if I had this back in the day I could have finished Butcher Hill
I think you could cut the 6510's speed in half by cutting the trace leading to its suspend input and inserting a circuit to suspend the chip every other clock cycle.
I love my U2+ cartridge since it's a cartridge and a disk drive and an REU and a fast load all in one. It does tape too, but I've never used a tape image before.
I'm just amazed and baffled how the game cart and slomo cart work simultaneously.
C64 port expanders like the CMD EX3 I'm using in this video are a lot of fun. Not all cartridges get along with each other, but the slomo cart just shares the +5V, GND, and IRQ lines and doesn't electrically interfere with the operation of the other carts.
@@8_Bit Thanks for the response.
It's a very weird design, to be honest. Two inverters connected like that make an oscillator. Without the capacitor there, it would oscillate at an extremely high frequency, close to the limit of what LS logic can do, which can be anywhere up to 350MHz. So adding the RC part adds some control to the oscillator, but as you noticed, it's still not very controllable. I'm surprised they didn't use something like a 555 timer which would give them much more accurate control. It would take some experimenting to find the range of frequencies that worked best, but once you figure that out you can easily design a 555 circuit to be adjustable over that particular range. Add a switch to turn it off maybe, or have multiple frequency ranges. Even back in 1989, the 555 was a cheap IC, though the LS00 probably was cheaper. It just seems like a not well-thought-out design, hacked together. The boards are homemade, so it was probably just produced in small batches.
The potentiometer looks exactly like the ones I bought for various hobby projects ;)
Someone should tell Commodore that Jupiter is a gas giant.
So would this be useful if I were to run PAL region games on an NTSC system? I hear you can, but they run faster than they should
It may have been "touchy" because the pot was dirty. Use some pot cleaner (or electronics cleaner) and turn the knob a lot and it should smooth out.
I should have mentioned in the video that I did give it a good cleaning beforehand, and it seemed very smooth when I tested it with my multimeter afterwards. Pretty sure the preciseness required is a result of the design of the circuit.
I have one of these !!! Bought it from eBay.
9:43 Vanellope von Schweetz would be proud. :)
As far as improvement, adding one or two resistors and playing with the pot value should really improve the control resolution (ie make the pot much less finicky).
While higher resolution/smother range would help, it's not the best solution since some games would still crash based on their design. As suggested by Robin, NMI and/or DMI line manipulation would probably work better.
Hmmm the final cartridge III :-)
I wonder if the potentiometer needs a little contact cleaner.
I actually did give it a good clean while preparing to record the video; as far as I know, the cartridge is working as designed.
0:12 How does a Commodore Security badge make your computer faster?
2:30 Reminds me of Slow Mobius from "Rick And Morty": czcams.com/video/j-5wmWuJHMI/video.html .
2:59 You can accomplish a similar thing by POKEing the high byte of the IRQ timer.
4:30 Won't the game get very confused if it uses raster interrupts? This is probably one of the things that happens to your favorite Mountie game.
13:52 I was spoiled by the schematic diagram in your (original) thumbnail!
24:34 NMIs wouldn't be feasible. The thing about NMIs is that they're *Non-Maskable*! If they're generated too quickly, they interrupt each other until the stack wraps around and garbles itself. I ran into this problem with SwiftLib @ 115.2kbps. The advantage of IRQs is they don't produce nested interrupts. Places where the system appears to pause with the current device (from a new IRQ being initiated immediately after the preceding one completes) would see the NMIs overflow.
Yes, probably most games that use raster interrupts will get confused by this; I'm sure that's why Bruce Lee II was so crashy. But if the interrupt handler is coded a certain way (arguably, "correctly") then it will look for the correct interrupt source and handle the extra IRQs in a robust way. Of course, I wouldn't blame any C64 programmer if their code doesn't handle this strange situation well!
And yeah, for sure NMIs wouldn't be nearly as useful, but if the circuit was better behaved (for instance, if it used a 555 timer and had a wider range of usable interrupt frequency that didn't so easily cause continuous interrupts) then perhaps in some software it would be another usable option?
I don't think this was ever meant for games. I think it was meant for programmers to slow things down so they can see what is happening. I'm guessing that POT is probably dirty as well given how touchy it seems to be.,
@@8_Bit 6:42 Mine too I played it a lot had the disk version and owned both the 1541 and the 1541-II
@@8_Bit 8:59 that sound reminds me of one of Eagle Soft Incorporated crack intros
Does the slowdown on jumpman stop you from getting hit by the bullets? Could be useful then
uhhhh, slowing down the cpu by overloading it with interrupts
urrrk, I shiver inside
Neat! I wonder if you can pause a game and do a screen shot. Interesting. Btw, what are you using to do screen capture?
Yes, it can work as a pause feature on some games, so that's another benefit. I use a Cloner Box Pro for all my video capture.
This would be a little bit more useful if it actually had NMI / IRQ selector (if someone pointed that out already - I didn't read comments first)
Probably you should clean out the potentiometer with Deoxit.
Jupiter Lander uses the opening music and other songs from the coin-op "Moon Cresta".
How would someone with the cartridge port multiplexer use this cartridge with a game cartridge.
"Canadian Mounted, Baby. Police force that works. Red and black, its their color scheme. Get their man, in the end."
I would definitely like to see a 555 version. It might actually work better since it would be easier to fine tune the frequency of the interrupts.
I was thinking exactly the same!
I wanna design and build it!
Use a Precision potentiometer, 10 turns Poti for better setting ;)
That. But first try to clean the current pot with an electronics cleaner.
I was thinking a multitude pot as well, although it would probably take some pretty frantic twirls during the throes of a game! Also thought maybe hacking a super long sliding pot instead…
Multi-turn pot, not multitude
This reminded me of a SYS # or was it a (poke #) that changed the speed of the cursor but slowed down a print program. Does anyone remember this address #?
Yes, it's similar to doing e.g. POKE 56325,0 which will cause the main system IRQ to fire very rapidly, increasing the cursor flash and slowing down BASIC.
Of course you bought this. I was thinking of buying it myself when I saw it on eBay cause it looked neat, but when I decided to, it was gone. Glad it went to a good home though.
Yeah, it was too weird to pass up!
I promise not to buy the next one that shows up on eBay :)
@@8_Bit Nah, it's cool, I just found it funny that the neat thing I saw on eBay and my uncertainty caused me to loose would pop up not to much later on one of the channels I sub to. Appreciate you at least sharing this little gizmo on your channel.
It's an odd one definitely, I think I saw something similar advertised in the past - and there were joysticks with "slow motion" buttons that were very ineffective. The 1989 date for the IC is intriguing too - it makes the cart much later than you would expect.
I have memories of seeing adverts for something similar too, but couldn't manage to find anything after a few hours of searching both my magazine collection and archive.org. If you ever run across an ad, let me know please! :)
@@8_Bit Lots of third-party controllers for consoles in the early 90s had "slow-motion" buttons -- all they did was press the START button really fast over and over. If the START button happened to pause the game, it *sort* of worked. But if it did something else then it was useless.
ah one kind of the IRQ "jammer" or how you call it today?
Was in the "magic 64" disc mag, build it inside my c64
afair sid was running faster and system could also crash.
yes like you show in the rail road game. glitching. untill it crashes haha
the circuit I have build back in the good old days is with a NE555, a 10kOhm pot,, 150 Ohm resistor, 10µF cap
Pin 1,2,4 on ex-port.
There were examples it was possible to speed up things in the c64
that is the dependency between the chips and their tasks (or how to say in english)
interesting they used a 7400 ...
yes there was also a circuit for the DMA ...
Do I remember there being a POKE that has a similar effect? The cursor blinks fast and the computer lags. One of those documented registers the Kernal uses.
You're probably thinking of POKE 56325,0 which causes the system IRQ to fire much more often than usual.
Why do I remember the 68000 family having a trap state that was "spurious interrupt" and getting worried by all of this even if it is a different platform? I would be especially worried if you had a disk drive attached. I know the C64 has that weird semiautonomous drive, but something about messing with with interrupts while (e.g.) writing to disk sounds like a recipe for head crashes or something.
"Speeding up faster"? As opposed to... "speeding up... slower"? Haha! :-p
That card needs a log pot instead of a linear one.
I'm trying to decide if this cart is useful or just a weird way to DDoS your 6510 ;-) Nice plywood.
Why not both? And yeah, this is a full 4' x 8' sheet :)
I'd imagine this was an excellent debugging tool for the time, and probably cheap too, but we have emulators and affordable better solutions as well.
Robin, would you like me to print some cases for your bare boards? It makes me nervous to see naked boards like that. Also, some better supports for your cartridge slot expander would be good.
Any diference with pal or ntse ? Great channel by the way :)
Thanks. I didn't try it on my PAL computer but I'd expect it would in pretty much the same way.
Great video, Robin. And great concept with the cartridge. The 65xx never really had the CPU step facility that other CPUs had, because its internal registers needed a constant refresh, but this is an interesting way to achieve something similar. I had wondered if it held down the bus request line, that I think the CP/M cart used, to halt bus access (rather than CPU speed). Wonder how the Datel Action Replay acheived similar on the Amiga 500 with its 68000?
The 6502 _does_ have a CPU step facility. You can't do it by stopping the clock because, as you point out, the CPU core is not static, but you can pull the RDY line low and the CPU will stop executing instructions until you let it go high again. (The clock continues to run at the same rate during this.) This is the exact mechanism used by the Commodore 64 for DMA (such as with the REU).
The MOS January 1976 _MCS6500 Microcomputer Family Hardware Manual_ gives a description of "static test" circuit for single-cycle and single instruction execution in section 3.1.2 on page 124, with a schematic in diagram 3.1 on page 125. (Sorry about not giving URLs for this, but CZcams tends to send comments with URLs to /dev/null, so you'll have to search.)
@@Curt_Sampson had no idea, thank you!
Why do some games get so messed up by it?
The cartridge is generating many extra IRQs which cause the CPU to execute the IRQ handler code which is often custom code written for each game. That code has often been designed to be executed only 50 or 60 times a second, not e.g. 1000 times per second, and may have other dependencies in the main game loop that get out of sync with each other. So in short, this cartridge may force a game to get out of sync with itself.
Jumpman JR looks a lot like Wizzard
The original Jumpman was actually released in early 1983, well before Wizard sometime in 1984. They're very similar looking!
I guess Mountie On The Run is your 2nd favorite game isn't it?
Why don't you just try to get your hands on that German cartridge?
Maybe it is Less effective on cartridge games that may also manipulate the IRQ line.
But then I would of expected the easy flash Games to be more like disk games.
This reminds me of something I did, circa 1990-ish. I was in my electronics class in high school and found a crystal that was just above 1.1MHz and wondered if I could every so slightly boost the performance of my C64 if I installed it. I had enough sense that I didn't think I could boost it up to 4.77MHz like an IBM PC, but just and extra 100+KHz maybe would work??? I remember someone telling me in a message on a BBS that they didn't think it would work because the timing between the C64's components were too tight. He was correct. However, it DID boot up but the screen was really distorted with a many horizontal bars skewed into like a saw-tooth pattern. I think there was also a screechy sound coming out of my speaker. I don't remember if it responded when I typed on the keyboard. I don't think I tried, I was too afraid to allow it to continue in that state for fear of breaking it. So, that was the end of that idea. Glad I tried, though. :-)
I'd have loved to have seen that myself! Yeah, it's especially the integration with the video chip that needs to generate a valid NTSC or PAL video signal; even a very small change in speed would be enough to make the video signal invalid, though I'd think the CPU would still be running okay. I've thought about it before but have never seen it demonstrated; that would make a really cool CZcams video, especially if there was a way of smoothly adjusting the speed and watching what happens to the video output.
@@8_Bit I would think in this age there's a variable oscillator somewhere one could wire into the timing circuit?? Its just a measly 1MHz. I wholeheartedly agree that would be a fun thing to try and share with the world. One caveat - the crystal MAY have been something like 1.08MHz. Even if it wasn't 1.1+ , it was close. I'm struggling to remember after all these years, but the rest I'm sure of. I still have that ol' 64. Tried to resuscitate it a couple years go. I re-capped it completely, replaced a couple of chips while troubleshooting it until it appears the VIC chip died. Will revisit the effort later.
I don't see how it could have booted at all: the C64 uses a 14.31838 MHz crystal resonator, nowhere near 1 Mhz. (The 14 MHz clock is divided down into several slower clocks used for various parts of the system, particularly for video, as well as the 1 MHz clock used by the CPU. See page 6 of the _Service Manual_ for all the details.)
Seems to me a cart like this could be useful for being able to step through code slowly during the creation process of a program and also be able to run that program at full speed easily. But it is a rather kludgy way of doing it!
Yes, it could potentially be useful for debugging if it had a better system of control.
this would be much better with a switch to turn the oscillator off and on and the pot only changing the speed.
Bill Herd on his live show a few days ago mentioned problems slowing down the C64 CPU due to its design. Here's a link to the section. czcams.com/video/Ou7iRrHhLrE/video.html
Yeah, most of these 8-bits that could connect directly to a television have a very tight integration between their CPU and video displays, and if you try to speed up or slow down even a small amount, the video signal goes out of spec and won't be displayed anymore. Among other problems!
Hi Robin! Listen my friend, I desperately need your help. I just bought a c64, it's an old gray case model and I connected it up and all the games are sped up, especially the music. I have no bloody clue what's going on, but I'm not a happy camper! Could you please help me? Thank you so much for all your videos..
Hi! What are you comparing the speed to? Do you have another C64 that runs slower? Or have you been using an emulator or other recreation? Maybe you're used to PAL speeds and real NTSC speeds seem fast now. Can you post video of a few games running on your channel? Just handheld from your smart phone should be fine; I'm curious to see this.
@@8_Bit I compared the speed to some videos of the same games I found on youtube, they all run at normal speed, aswell as my ccs64 emulator. And yes, I will post a couple videos of the gameplay. Thanks, my friend.
@@8_Bit Lets just do this the easy way for now....watch this video of cybernoid. This is how it is supposed to sound. mine is alot faster
czcams.com/video/k6xhfDLaVtU/video.html
P.S. I will post my version on my channel a bit later today.
5:25 That is from another game. I think it's a shooting game, but I just cannot think of what game it is. Maybe "Eagle" also known as Moon Cresta. (also at 5;45) These are not original to this game.
Good catch, yes, Moon Cresta released in 1980 has this same melody. HAL Laboratories created Jupiter Lander for the VIC-20 and C-64 in 1981 and 1982, and those young programmers were known to "borrow" almost all the elements that went into their games. But the code was all original, as far as I can tell! They had lots of programming skill, but lacked original ideas.
That module must allow you to cheat at certain games.
I was sure about the place for a reset button. It is the law of the universe that every C64 cartridge MUST have a reset button or the Earth will be in great danger. It is our way of overcompensating for the lack of an official built in reset button :)
делал такое на zx-spectrum-е - просто через кнопку соединяя один из делителей частоты с сигналом WAIT процессора
чтоб точно проходить игры
There surely isn't much in it for DM 89,-
Heh, I've always had the feeling that the JJr. intro. song is a rip-off of the Ms. Pac-Man song. Ya think there's any factual basis for that thought?
I can hear some similarities, but a note-by-note analysis would show they're really quite different. Ms. Pac-Man's melody is mostly a continuous climb upwards, with only small steps downward, while Jumpman Junior's melody climbs up and down quite a bit more. They do have a similar feel to each other though.
You're a bass player!?
Yeah, that's my playing (bass, guitar, and often vocals) on the end credits music in most of my videos. My daughter plays the keys and drums for those songs. Here's one of our songs: czcams.com/video/KcI68ww76ds/video.html
@@8_Bit thank you!
These seem to be CIA interrupts. So, if a game turns off CIA interrupts (which many do, I guess), the cart will not affect the gameplay.
From the CPU's perspective, anything that pulls the IRQ line low will cause the CPU to interrupt what it's doing, and execute the code that the IRQ vector is pointed at. Normally the CIA generates the IRQ that drives the system interrupt that flashes the cursor etc. but other sources can trigger the IRQ as well such as the VIC-II raster interrupt, or anything hooked up to the IRQ line on the expansion port, such as this slow-mo cartridge. So even if the CIA timer is disabled (you're right, many games do) this cartridge can still cause extra IRQs; it just depends on what code the particular game has the IRQ vector pointed at, and whether the handler bothers to check the source of the interrupt.
@@8_Bit Oh, I see. You’re right. So, if a game has only one raster interrupt, it will be fired lots of times with this cartridge. And the line will not be checked, so the code is processed and the game glitches away. If a game has 2 raster lines where it wants notification, it might ask the VIC if it’s on eather of the two lines. So, the extra interrupts have no effect. Reasonable?
Yes, children. There was once a time in history when people would actually go "oh no, my computer is TOO FAST".
That's what the turbo button on PCs was for (or go slower if this button isn't pressed) - when 12MHz was too fast for 8MHz games.
@@immortalsofar5314 yup, I remember trying to play a PC port of Joust that had used the CPU clock for timing. On a faster machine in Turbo mode it was humanly impossible to press the flap button fast enough to take off...
DeoxIt.
ah. so 'deathride' is like. gta-sa. :P
💪😁🔥🔥🔥
I was wondering how I could make my c64 bitcoin mining rig slower to leave bitcoins for others. Now I have a way!
I think it's clone of Rex Bremse cart. I've got another (with cover and sticker) where one sided PCB is used. See: czcams.com/video/kNTNDnKJadQ/video.html
after the tests you made that indicates its probably affecting the graphics units refresh rate as it doesnt affect the main thread that counts but it stopped refresh rate that without causing trouble to counting progra. that indicates it directly changes refresh program/logic somehow. its mostlikely changing graphics chips refresh rate by adjusting dropped frame count. mostlikely changing soma settings on driver software/thread.
Besides Space Taxi, Jupiter Lander was one of my favourite games in the 80s. Nowadays i play Star Citizen as their successor 😀
4:40 I don't think this is for games. I think this is for programmers.