it feels so good to watch this series from the beginning and understand what's happening in little incremental steps i love learning stuff like this, because everything feels much more intuitive, now a day's people want stuff as soon as possible without knowing it's inner workings, but going over every single step makes ot SO MUCH MORE ENJOYABLE love your work, I'll try to get sone money to support you, because content like this doesn't deserve to be abandoned because of financial reasons
This took me back to college learning about CANBUS systems you find in newer cars. This channel never gets boring, I love electrical engineering (even if it was only motor vehicle specific)
Dear Ben, What a nice video series. This is what I was looking for. I do not have any engineering background. I have a medical background. Your explanations are so simple to understand everything effortless. You are a great teacher. Thank you very much. Please upload more videos in this topic.
First person I've ever pledged too on patreon. You have taught me so much that it's the only way I can thank you. Just took my first computer architecture class, we did a lot of similar stuff to this series and I breezed through the class up until the stuff this series hasn't covered yet. You're the man Ben, I can't wait to see what's coming next.
Hey Ben, It seems like a youtuber with the youtube name/ID called "Suryaprakash105" is reuploading all your videos. I don't know if you gave him permission for this or not, otherwise you can report him for stolen content.
A great video :) Just a small improvement I could think of at 5:15 "Synchronous means it counts using a clock" could have been put in a better way as "Synchronous means that all Flip flops get the same clock unlike asynchronous where successive FFs get divided clock. Both of them count using a clock input though". We additionally need Next State Decoder in synchronous counter. 1000th like 🤘
It's all coming together. I joined late but I have a lot to do to catch up with the rest of you. Just got the Clock put together. I found out that Jameco is a great place to get parts. They have everything in stock and ship fast.
Awesome! I suggest if you could connect an actual keyboard to it so it would detect scancodes from the keyboard and then make some kind of assembly language for it that you can type with the keyboard :)
Yeah man, I celebrate whenever you put out a video! This one's great again - as always - but it ends right when it's getting exciting :( Please, PLEASE, don't have us wait too long until you're getting to guts with the PC and the control logic.
In this diagram I see a "MAR" element connected to RAM and te BUS, What is used for? I saw the ram videos, but it doesn't seem be explained in that videos. By the way GREAT EDUCATIONAL CONTENT please keep it up.
So wait, do we not need the JK flip flop chip? It was on the buy list 😂 Also it's interesting how in the Counter chip, the flip flops don't circle back into one another via clock input. Would have been interesting to hear a few words on that! I suppose all of them having the same clock signal synchronizes the increments better, at the cost of some extra round about wiring.
9:02: "This is an eight-bit thing; we're only going to use four of them, but that's fine." I wonder if it would be possible to build some sort of parasitical circuit using all the extra pins and gates that are not used by the main circuit. It might be too difficult to include these pins because their state is influenced by the main project's circuit logic, but some of the other ICs used in the wider project had spare gates on them that were not connected to anything else and that arguably could be hooked up to some separate thing. That would be quite a challenge to make something interesting out of those leftovers. Obviously not the thing you'd want to include in a "simply teach the fundamentals" design, but it might be an interesting challenge for anyone out there who's building a copy of Ben's breadboard computer. You could invent something extra using what you already have. If you do it, please send me a reply and tell me about it.
I can't see why you couldn't change this 8 bit addresses. It would mean that all instructions conserning memory addresses would be two bytes long eg. need to fetch two 8 bit words into the instruction decode unit, so that would need major overhaul. The bus is 8bit wide so you only need to widen certain registers to 8 bits and of course buy more memory.
Thanks Ben, Your series has inspired me to restart a project I worked on years ago to make a simple 8 bit computer entirely out of NAND gates (along with an EEPROM chip and a few 555 chips). Thankfully I already have most of what I need, do you happen to know where I could get a few hundred dip 14 wire wrap sockets? Wire wrap seems to be a lost or at least dying method and most of the sockets I've been able to find online are $1.20+ per socket.
What kind of studies should I opt for to acquire this knowledge deeply? Computer Science, Electronic- connected studies (if yes - which one?) or Automatics and Robotics?
04:00 But the loading of a new content should still be synchronous with the clock, right? Or could we load it asynchronously as well? I'm asking this question because there's not many counter chips with synchronous load, especially reversible counters (those that can also count down towards 0), but there are quite some reversible counters that can be loaded asynchronously. So I'm wondering if they could be used here as a replacement of the 74LS161 one? Counting down might not be very useful for the program counter, true. But one may still want to use it if one doesn't have the 74LS161 chip, or if one wants to build a stack pointer register that counts down instead of up (because the stack usually grows down from the top of the memory space). Also a reversible counter could perhaps also be used as a replacement for the D registers, since a counter is basically a register too, and being able to increase it and decrease without the need of passing its value through the ALU could be a cool thing to have as well ;>
isn't switching from falling edge triggered ff to rising edge triggered ff simply just changing the position of clock inverter from slave to the master latch?
I am surprised you are using the 74LS161 binary counter when you already had the 74LS76 flipflops in place. I understand that the 74LS161 has the option to accept data with an enable pin, but your JK flipflops have a preset (PR) and a clear (CLR) pin that could have been used as data input signals.
Updated Program Counter Schematic, Sorry I had the 74LS245 DIR pin 1 tied high +5 like Bens other build Schematics where A, B, IR Register, and Memory. Now I see in this Program Counter build Ben tied DIR pin 1 to ground instead.
Whoa! This is a great explanation. I gotta see where you started this topic. Do you have any favorite books on this topic, and also sysadmin, bash, linux driver programming? Thank you for good teaching here.
Just finished my program counter but if the clock is too slow, like under 6 Hz for example, the clock cycles after hitting 2 or maybe even 3 (two lowest LEDs lit) and then back to 0000 for starting over. If I increase the clock, to 7 Hz then the counter cycles thru all values 0 to 15 as it should. Not sure what is causing this, any advice or comments?
Bill Bowlio: I have been asked the same questions recently. As I have all but completed my SAP-1. I am using a 5VDC 3A wall charger. It appears that the smaller 5V 1A have a problem powering a fully assembled SAP-1. If you are having problems with power, I would suggested trying a 5VDC 3A as that seems to work for me, but even that has some slight limitations. I can not use blue LEDs... They seem to require more power. drive.google.com/open?id=0B6LIgQzDkH8fQlFZU01VajZzdmM drive.google.com/open?id=0B6LIgQzDkH8fUTlCYWpJV3B2dDA
@rabius sany Apu They are not really compatible, but you can mix them if you respect a few things. - *First off: there's also HCT, which you actually CAN mix with LS unworried* - supply voltage: LS and HCT require 5V pretty much spot on, whereas HC can be used within 2V-6V - gate propagation delay is about the same, ~10ns - both can go well above 20 MHz, although this is of no relevance for this kind of project - LS uses more power than HC and HCT (mWs rather than µWs at low frequencies) - LS is pretty resilient to static discharge, while with HC and HCT you have to be careful - LS inputs have "auto-pull-up" (Ben takes advantage of this several times), ie if not connected they default to HI. HC/HCT inputs must *always* be tied to either 0V or +Vs (even if that part of the IC is not even used). - *you cannot reliably drive HC inputs from LS outputs* (HC*T* inputs from LS outputs is fine) - you can drive LS inputs from HC/HCT outputs - LS has smaller fan-out, that is: 1 LS output can drive only up to 10 LS inputs (many more HCT inputs), whereas 1 HC(T) output can drive at least 50 HC(T) inputs - *but only up to 10 LS inputs* Finally: the above is in no way comprehensive. There are more details/pitfalls. For example there are "open-collector" output versions, or "Schmitt-trigger" input versions in both families - things are different for these... So please do your own research, eg via Google: "74LS vs 74HC"
rabius sany Apu Besides not leaving open any inputs (as said), it really depends on the actual circuit. Again, *avoid driving HC inputs from LS outputs*, but vice versa it's fine. The problem is that LS considers anything above ~2V a logic 1, and a 1 output may be as low as 2.4V. However, HC (not: HCT) expects at least 2.5V for a logic 1, so it doesn't quite match. You may get along with it though, possibly with additional pull-up. The other thing is that LS can't source a lot of current (it can sink much more though). That means in particular that you should not adopt Ben's practice of putting bare LEDs on HC outputs, but rather use low-current ones (1 mA or less), and always put a current-limiting resistor on them (good practice anyways IMHO, even for LS). You have to try on the actual circuit, use your multimeter.
With reference to the discussion at th-cam.com/video/-arYx_oVIj8/w-d-xo.html, I was quite happy with the no-clock-gating approach but now I can see that the 74LS163 is inverting the clock, namely using a not gate. At this point, I'm a bit confused. Any thoughts from anyone here? Thank you in advance for your comments. And thank you, Ben, for this amazing video series!
This series is the best thing that has ever appeared on youtube. Or appeared anywhere. Thanks for continuing it!
Agreed. It's very addictive, I can't stop!
it feels so good to watch this series from the beginning and understand what's happening in little incremental steps
i love learning stuff like this, because everything feels much more intuitive, now a day's people want stuff as soon as possible without knowing it's inner workings, but going over every single step makes ot SO MUCH MORE ENJOYABLE
love your work, I'll try to get sone money to support you, because content like this doesn't deserve to be abandoned because of financial reasons
Please keep up these awesome breadboard computer videos!! As someone who is trying to learn how computers work, these videos are teaching me so much.
I LOVE your 8Bit Computer series. So glad you are continuing it.
Thank you Ben for continuing the 8 bit series! this is just great.
Alert notifications always on for Ben
lol ya me also lol
lukas russell same! 😃
same
This took me back to college learning about CANBUS systems you find in newer cars. This channel never gets boring, I love electrical engineering (even if it was only motor vehicle specific)
I want a computer called "The BONG system"
You have a gift. You are an excellent teacher and I for one appreciate you sharing your gift!
Dear Ben, What a nice video series. This is what I was looking for. I do not have any engineering background. I have a medical background. Your explanations are so simple to understand everything effortless. You are a great teacher. Thank you very much. Please upload more videos in this topic.
First person I've ever pledged too on patreon. You have taught me so much that it's the only way I can thank you. Just took my first computer architecture class, we did a lot of similar stuff to this series and I breezed through the class up until the stuff this series hasn't covered yet. You're the man Ben, I can't wait to see what's coming next.
What do/did you study?
@@vicvic553 computer engineering probably
Looks like Ben's Breadboard CPU is getting traction again. Well done Ben; I love it. Please keep it going.
Yes. This is indeed one of the best engineering/tech series *ever*
Hey Ben, It seems like a youtuber with the youtube name/ID called "Suryaprakash105" is reuploading all your videos. I don't know if you gave him permission for this or not, otherwise you can report him for stolen content.
Perfectly presented. I almost feel like this channel is too good to be true because of how great it is, what's the catch Ben?!
A great video :)
Just a small improvement I could think of at 5:15
"Synchronous means it counts using a clock" could have been put in a better way as "Synchronous means that all Flip flops get the same clock unlike asynchronous where successive FFs get divided clock. Both of them count using a clock input though". We additionally need Next State Decoder in synchronous counter.
1000th like 🤘
Thank you Ben,i feel lot confident in understanding internals now
It's all coming together. I joined late but I have a lot to do to catch up with the rest of you. Just got the Clock put together. I found out that Jameco is a great place to get parts. They have everything in stock and ship fast.
This info is very helpful for learning about computer science
Synchronous means there's no ripple effect, Asynchronous counters also count with a clock.
Awesome! I suggest if you could connect an actual keyboard to it so it would detect scancodes from the keyboard and then make some kind of assembly language for it that you can type with the keyboard :)
Outstanding
thank you for continuing
Brilliant! Explained so nicely, it’s easy to follow. Thanks a lot.
Yeah man, I celebrate whenever you put out a video! This one's great again - as always - but it ends right when it's getting exciting :(
Please, PLEASE, don't have us wait too long until you're getting to guts with the PC and the control logic.
Thanks from Texas.
In this diagram I see a "MAR" element connected to RAM and te BUS, What is used for? I saw the ram videos, but it doesn't seem be explained in that videos. By the way GREAT EDUCATIONAL CONTENT please keep it up.
YES!!! I've been waiting for this! Thank you Ben!
I major computer electronics at school, but I study with Ben :)
I jist love u man... It just amazing to get a feeling of how this brilliant machine (the computer) works ❤️
Thanks for share
thanks this video was very helpful
will you be doing a demo on a shift register like you did for the flip-flops?
Good stuff. Do you do hand modeling on the side?
So wait, do we not need the JK flip flop chip? It was on the buy list 😂
Also it's interesting how in the Counter chip, the flip flops don't circle back into one another via clock input. Would have been interesting to hear a few words on that! I suppose all of them having the same clock signal synchronizes the increments better, at the cost of some extra round about wiring.
9:02: "This is an eight-bit thing; we're only going to use four of them, but that's fine."
I wonder if it would be possible to build some sort of parasitical circuit using all the extra pins and gates that are not used by the main circuit. It might be too difficult to include these pins because their state is influenced by the main project's circuit logic, but some of the other ICs used in the wider project had spare gates on them that were not connected to anything else and that arguably could be hooked up to some separate thing. That would be quite a challenge to make something interesting out of those leftovers. Obviously not the thing you'd want to include in a "simply teach the fundamentals" design, but it might be an interesting challenge for anyone out there who's building a copy of Ben's breadboard computer. You could invent something extra using what you already have. If you do it, please send me a reply and tell me about it.
I can't see why you couldn't change this 8 bit addresses. It would mean that all instructions conserning memory addresses would be two bytes long eg. need to fetch two 8 bit words into the instruction decode unit, so that would need major overhaul. The bus is 8bit wide so you only need to widen certain registers to 8 bits and of course buy more memory.
Great video, thanks!
Thanks Ben,
Your series has inspired me to restart a project I worked on years ago to make a simple 8 bit computer entirely out of NAND gates (along with an EEPROM chip and a few 555 chips). Thankfully I already have most of what I need, do you happen to know where I could get a few hundred dip 14 wire wrap sockets? Wire wrap seems to be a lost or at least dying method and most of the sockets I've been able to find online are $1.20+ per socket.
What kind of studies should I opt for to acquire this knowledge deeply? Computer Science, Electronic- connected studies (if yes - which one?) or Automatics and Robotics?
04:00 But the loading of a new content should still be synchronous with the clock, right? Or could we load it asynchronously as well?
I'm asking this question because there's not many counter chips with synchronous load, especially reversible counters (those that can also count down towards 0), but there are quite some reversible counters that can be loaded asynchronously. So I'm wondering if they could be used here as a replacement of the 74LS161 one?
Counting down might not be very useful for the program counter, true. But one may still want to use it if one doesn't have the 74LS161 chip, or if one wants to build a stack pointer register that counts down instead of up (because the stack usually grows down from the top of the memory space).
Also a reversible counter could perhaps also be used as a replacement for the D registers, since a counter is basically a register too, and being able to increase it and decrease without the need of passing its value through the ALU could be a cool thing to have as well ;>
You didn't really explain why all the LS163's clocks are tied together, and how it's able to load in new data to the flip-flops :(
A stack would be interesting.
Thanks!
i tried building a computer and started from the point of the program counter :)
isn't switching from falling edge triggered ff to rising edge triggered ff simply just changing the position of clock inverter from slave to the master latch?
Well explaination of program counter could have tell about stack tho
Struggling here. W/O knowledge of decoder, how do you plan to sort out all these high/low enables? You need a truth table.
Nice scheme :)
I am surprised you are using the 74LS161 binary counter when you already had the 74LS76 flipflops in place. I understand that the 74LS161 has the option to accept data with an enable pin, but your JK flipflops have a preset (PR) and a clear (CLR) pin that could have been used as data input signals.
Updated Program Counter Schematic, Sorry I had the 74LS245 DIR pin 1 tied high +5 like Bens other build Schematics where A, B, IR Register, and Memory. Now I see in this Program Counter build Ben tied DIR pin 1 to ground instead.
Whoa! This is a great explanation. I gotta see where you started this topic. Do you have any favorite books on this topic, and also sysadmin, bash, linux driver programming? Thank you for good teaching here.
it is a little old but i recommend "cmos cookbook" by don lancaster
Thanks! In this instance, older is better. Love that meme.
now repeat the same, using 74HC.... series chips and build a 16 bit computer.
Is it possible to make a PC using the 74LS175 Quad D flip flop instead of using the JK or the 4 bit counter chip?
Why there's tow different enable of they doing the same thing?
Can you tell me how to build a device that will send out Manchester code fm addresses? Any idea?
Just finished my program counter but if the clock is too slow, like under 6 Hz for example, the clock cycles after hitting 2 or maybe even 3 (two lowest LEDs lit) and then back to 0000 for starting over. If I increase the clock, to 7 Hz then the counter cycles thru all values 0 to 15 as it should. Not sure what is causing this, any advice or comments?
sir u r great
Wooot!
Thanks for the amazing videos!
Where do you get your electronics stuff?
Did it mess up at 4:15?
Looks like a jump cut from when he was editing the video
love u bro
I LOVE YOU
I'm having what appears to be low power problems. How much current is your power supply?
Bill Bowlio 0.5A is standard in an USB charger, which he mentioned in an earlier video that he was using. iirc.
Thanks!
Bill Bowlio:
I have been asked the same questions recently. As I have all but completed my SAP-1.
I am using a 5VDC 3A wall charger. It appears that the smaller 5V 1A
have a problem powering a fully assembled SAP-1.
If you are having problems with power, I would suggested trying
a 5VDC 3A as that seems to work for me, but even that has some slight limitations.
I can not use blue LEDs... They seem to require more power.
drive.google.com/open?id=0B6LIgQzDkH8fQlFZU01VajZzdmM
drive.google.com/open?id=0B6LIgQzDkH8fUTlCYWpJV3B2dDA
I want to use 74lsxx and 74hcxx both together.. what should I do?
@rabius sany Apu They are not really compatible, but you can mix them if you respect a few things.
- *First off: there's also HCT, which you actually CAN mix with LS unworried*
- supply voltage: LS and HCT require 5V pretty much spot on, whereas HC can be used within 2V-6V
- gate propagation delay is about the same, ~10ns
- both can go well above 20 MHz, although this is of no relevance for this kind of project
- LS uses more power than HC and HCT (mWs rather than µWs at low frequencies)
- LS is pretty resilient to static discharge, while with HC and HCT you have to be careful
- LS inputs have "auto-pull-up" (Ben takes advantage of this several times), ie if not connected they default to HI. HC/HCT inputs must *always* be tied to either 0V or +Vs (even if that part of the IC is not even used).
- *you cannot reliably drive HC inputs from LS outputs* (HC*T* inputs from LS outputs is fine)
- you can drive LS inputs from HC/HCT outputs
- LS has smaller fan-out, that is: 1 LS output can drive only up to 10 LS inputs (many more HCT inputs), whereas 1 HC(T) output can drive at least 50 HC(T) inputs - *but only up to 10 LS inputs*
Finally: the above is in no way comprehensive. There are more details/pitfalls. For example there are "open-collector" output versions, or "Schmitt-trigger" input versions in both families - things are different for these...
So please do your own research, eg via Google: "74LS vs 74HC"
TheRealMeisl Thank you very much. If I want to use 74hcxx series to build my 8 bit computer, what necessary change should I apply?
rabius sany Apu Besides not leaving open any inputs (as said), it really depends on the actual circuit. Again, *avoid driving HC inputs from LS outputs*, but vice versa it's fine.
The problem is that LS considers anything above ~2V a logic 1, and a 1 output may be as low as 2.4V. However, HC (not: HCT) expects at least 2.5V for a logic 1, so it doesn't quite match. You may get along with it though, possibly with additional pull-up.
The other thing is that LS can't source a lot of current (it can sink much more though). That means in particular that you should not adopt Ben's practice of putting bare LEDs on HC outputs, but rather use low-current ones (1 mA or less), and always put a current-limiting resistor on them (good practice anyways IMHO, even for LS).
You have to try on the actual circuit, use your multimeter.
Finally :D
the program is stored in ROM
But that's not compulsory. I guess Embedded systems generally use ROM to store programs but General purpose computers use RAM
I guess explaining the load logic is left as an exercise to the viewer 🤣
l love you.....
With reference to the discussion at th-cam.com/video/-arYx_oVIj8/w-d-xo.html, I was quite happy with the no-clock-gating approach but now I can see that the 74LS163 is inverting the clock, namely using a not gate. At this point, I'm a bit confused. Any thoughts from anyone here? Thank you in advance for your comments. And thank you, Ben, for this amazing video series!
wr, no such thing as useflx or not usefx, can use anyx no matterx wx
sir i want like u from pakistan
Binary. Three syllables. Not bi a nar y.