A6 and A7 are setup for true ADC and do not have digital pins. Nice video the graphic presentations worked very well and I found the information to be easy to follow and understand.
@@ezContents Your project is Great!!! I did build and it works great!!! Please! Is possible to use this device to test 4116 ones? I know how to generate -5v, +5v and +12v to power that chip. There is a way to use your device to test 4116 too? Very thanks once more!
@@ezContents Very thanks for replying friend!!! But and about the A7 line ???? What must I do about A7 line?? The 4116 does not have the A7 pin. The 4116 uses only till A6 address. What to do about Arduino line that should be connected to A7 ??? Thanks once more!
Very nice explanation! And the animations raised the bar quite a bit. Excellent! As for the pins, I would have guessed that you may run in problems with A4 and A5, as they are often related to I2C. I just learned from DJ Legion's comment that they are analog inputs only. Good to know!
A great first video -- I really enjoyed it -- very well shot and edited, great educational information and the animations are fantastic. I look forward to more! 👍
Glad to hear that you liked it. Looking at your videos they look very professional with amazing shots and sound. I feel that I have a long way to go to that level. :-)
@@ezContents Yes, it was a very good video, indeed. Thanks for your comments on my stuff. It is funny how self-critical I am -- I think my videos do not look as good as they could, especially when I compare to other RUclipsrs.
@@MindFlareRetro It's always good to be critical of yourself, otherwise you wouldn't improve. But I think that you shouldn't be too hard on yourself as your videos are great!
Thanks for the video and the article, I just built it for myself on a prototype board with soldered wires and a socket for the DRAM chip. If I rebuild it some time, I will use a zif socket. I modified the code, I added a second test round with all zeros for detecting stuck 1s.
This is a great video - I've always stayed away from using DRAM because it is more difficult to use. I'd be curious if you'd go further and explain refresh requirements (I dont know how that works) and maybe even show how you can put 8 DRAM in parallel and read/write entire bytes.
hmm, you only talked about interfacing though, not about refresh and how to keep the data for longer than a single write/read test. also can't you use Ports on the nano too, so simply the code a bit?
Thanks for sharing your tester. I built it and it seems to be working to spot bad chips, good job. However, by the way I noticed that sometimes it tests "ok" when there is no 4164 chip present, is it regular? And if so, do you know if there is a way to test for "present" chip? A notice: It would be better to apply an external 5v power supply because sometimes those 4164 chips get bad 'cause they short and can draw more than 0.2 amps from the arduino; this would fry the arduino itself.
Thanks for your remarks. I did not test it that extensively myself. It was actually never my intention to create a 4164 tester, but merely to show the operation of the chip.
What you are observing is the result of how simple this test is. The code simply writes 1’s (high) into the memory cells, and verifies that it reads back 1’s. So, if the DRAM has the fault of a stuck high output pin, the DRAM will incorrectly test okay. Likewise, if no DRAM chip is present and the unconnected input pin floats high, you’ll also register an incorrect okay test. Ideally, for testing a 1 bit wide memory chip, you should test storing 1’s as well as 0’s, and also alternate 1/0 & 0/1 bit tests (so adjacent cell bits differ). But, as ezContents said, it wasn’t his intention to actually create a 4164 tester. 🤓
This isn’t a problem for this example code, because he’s only writing and reading back a single 256 bit row at a time. So, the required refresh time is unlikely to be exceeded by the Arduino. But if he’d tried writing the entire chip (all rows) first, before reading back, it’s likely the resulting read delay would exceed the max refresh time (and the test would likely fail on at least some degraded / leaky bits). 🤓
I'm trying this out on some MK4560-20 chips and I'm noticing something strange. Maybe it is my lack of understanding of dram and its failures or perhaps a compatibility issue with this variation but if I run the test say 10 times, 9 times the ram goes through just fine but the 10th time (give or take I'm just ballparking) results in the following: Starting test... Testing row: 0 ERROR: row 0 number of bits was: 255 This is the only failure I'm getting on multiple chips which made me think something else is going on here. If anyone has any guesses or suggestions to try out I'd be grateful. Thanks!
Hello! a question. yesterday I added a memory 4164 to my Ibanez Digital delay guitar pedal and it fired. but when you touch the chips they were a little hot. is this normal? Thanks!!!
@ezContents Your project is Great!!! I did build and it works great!!! Please! Could anybody help me? Let me know if Is possible to use this device to test 4116 ones? I know how to generate -5v, +5v and +12v to power that chip. There is a way to use your device to test 4116 too? Very thanks once more!
A6 and A7 are setup for true ADC and do not have digital pins. Nice video the graphic presentations worked very well and I found the information to be easy to follow and understand.
Ah yes.. thank you for the answer. That explains there is no digital circuitry behind them.
@@ezContents Your project is Great!!!
I did build and it works great!!! Please! Is possible to use this device to test 4116 ones? I know how to generate -5v, +5v and +12v to power that chip. There is a way to use your device to
test 4116 too? Very thanks once more!
@@pcrockpc6080 If you connect all the voltages correctly, then I think it should work.
@@ezContents Very thanks for replying friend!!! But and about the A7 line ???? What must I do about A7 line?? The 4116 does not have the A7 pin. The 4116 uses only till A6 address. What to do about Arduino line that should be connected to A7 ???
Thanks once more!
@@pcrockpc6080 Just don't connect it.
Very nice explanation! And the animations raised the bar quite a bit.
Excellent!
As for the pins, I would have guessed that you may run in problems with A4 and A5, as they are often related to I2C.
I just learned from DJ Legion's comment that they are analog inputs only. Good to know!
Luckily did the other analog ports not cause more problems... I knew for sure to stay away from pin D1 and D2 as they are for serial RX and TX.
nice video. so this is only a retro thing right, they dont get produced anymore?
Thanks. No, I don't think that they are produced anymore.
A great first video -- I really enjoyed it -- very well shot and edited, great educational information and the animations are fantastic. I look forward to more! 👍
Glad to hear that you liked it. Looking at your videos they look very professional with amazing shots and sound. I feel that I have a long way to go to that level. :-)
@@ezContents Yes, it was a very good video, indeed. Thanks for your comments on my stuff. It is funny how self-critical I am -- I think my videos do not look as good as they could, especially when I compare to other RUclipsrs.
@@MindFlareRetro It's always good to be critical of yourself, otherwise you wouldn't improve. But I think that you shouldn't be too hard on yourself as your videos are great!
@@ezContents Thank you very much for saying so. But yes, I am always looking to improve. I look forward to your next videos.
@@MindFlareRetro Thank you as well for nice comments. I am going to start working on my next video, before I loose the momentum.
Cool video. Very educational. I really like the animation. Did you make that yourself?
Thanks! Yes, I made the Blender animation myself. Just used the existing 3D DIP package shape from KiCad.
Yet another way to utilize breadboard or plugboard - very educational 👍
Thanks! No breadboard shall be left unused!
Great project, also great article, you just saved me a ton of time!!! Thank you so much!
Thanks for the video and the article, I just built it for myself on a prototype board with soldered wires and a socket for the DRAM chip. If I rebuild it some time, I will use a zif socket. I modified the code, I added a second test round with all zeros for detecting stuck 1s.
Nice to hear that you have built it on a breadboard and continued upgrading the code.
@@ezContents I added a green LED on pin D13, for when the chip passes.
This is a great video - I've always stayed away from using DRAM because it is more difficult to use. I'd be curious if you'd go further and explain refresh requirements (I dont know how that works) and maybe even show how you can put 8 DRAM in parallel and read/write entire bytes.
Thank you. A good idea. This would mean to dive a bit deeper into the datasheet. 🙂
Great Video! I was looking at something to explain the RAM in my C64. Its been having RAM issues. Thanks!
hmm, you only talked about interfacing though, not about refresh and how to keep the data for longer than a single write/read test.
also can't you use Ports on the nano too, so simply the code a bit?
Thanks for sharing your tester. I built it and it seems to be working to spot bad chips, good job.
However, by the way I noticed that sometimes it tests "ok" when there is no 4164 chip present, is it regular? And if so, do you know if there is a way to test for "present" chip?
A notice: It would be better to apply an external 5v power supply because sometimes those 4164 chips get bad 'cause they short and can draw more than 0.2 amps from the arduino; this would fry the arduino itself.
Thanks for your remarks. I did not test it that extensively myself. It was actually never my intention to create a 4164 tester, but merely to show the operation of the chip.
What you are observing is the result of how simple this test is. The code simply writes 1’s (high) into the memory cells, and verifies that it reads back 1’s.
So, if the DRAM has the fault of a stuck high output pin, the DRAM will incorrectly test okay. Likewise, if no DRAM chip is present and the unconnected input pin floats high, you’ll also register an incorrect okay test.
Ideally, for testing a 1 bit wide memory chip, you should test storing 1’s as well as 0’s, and also alternate 1/0 & 0/1 bit tests (so adjacent cell bits differ).
But, as ezContents said, it wasn’t his intention to actually create a 4164 tester. 🤓
Goog job Man....
Thank you so much, now I can finally build my z80 computer
No problem. Let me know when you have something working. I am curious.
Cool video
danke gut gemacht!!!
und jetzt entwickeln eine pcb fur mega...hahaha, alles gute+bisbald !!!
A dram must be refreshed of course, once every few msec
This isn’t a problem for this example code, because he’s only writing and reading back a single 256 bit row at a time. So, the required refresh time is unlikely to be exceeded by the Arduino.
But if he’d tried writing the entire chip (all rows) first, before reading back, it’s likely the resulting read delay would exceed the max refresh time (and the test would likely fail on at least some degraded / leaky bits). 🤓
nice job good video just wanted to ask if the tester works also with mt4264-20 chips thx.
I did replace a broken 4264 chip with a 4164 once. But I am not sure how the chip will behave on this Arduino circuit.
@@ezContents ok thx I'll give it a try and if works I'll let you know
@@ezContents ok i tested it and it works with 4264 chips as well
Nice! Thanks for testing it out.
Nice video, but you might want to fix the typo at 0:27 - Electonics should be Electronics.
Well spotted. Thank you. I'll fix that in the intro in the next video.
I'm trying this out on some MK4560-20 chips and I'm noticing something strange. Maybe it is my lack of understanding of dram and its failures or perhaps a compatibility issue with this variation but if I run the test say 10 times, 9 times the ram goes through just fine but the 10th time (give or take I'm just ballparking) results in the following:
Starting test...
Testing row: 0
ERROR: row 0 number of bits was: 255
This is the only failure I'm getting on multiple chips which made me think something else is going on here. If anyone has any guesses or suggestions to try out I'd be grateful. Thanks!
Hello! a question. yesterday I added a memory 4164 to my Ibanez Digital delay guitar pedal and it fired. but when you touch the chips they were a little hot. is this normal? Thanks!!!
Not very normal. But as long it stays under the max. operating temperature which is 70 deg. C you should be ok.
Don't forget about 41256...
@ezContents Your project is Great!!!
I did build and it works great!!! Please! Could anybody help me? Let me know if Is possible to use this device to test 4116 ones? I know how to generate -5v, +5v and +12v to power that chip. There is a way to use your device to
test 4116 too? Very thanks once more!