65816 build -- completely stuck -- please help :)

[akohlbecker]

Awesome! Great to see it working

[Dr Jefyll]

Happy to hear about your success, rehsd. (Nice catch, Adrien!)

-- Jeff

[rehsd]

akohlbecker wrote:

Awesome! Great to see it working

I chuckled today. I went to connect a VIA to my 65816, and I pulled up the code I had originally written a few weeks ago for it. Guess what? It had the reset vector in it! So, somewhere along the way, I accidentally dropped the reset vector. Oops.

If you're interested, I have a VIA running now.

https://www.rehsdonline.com/post/65816-via1-working

https://youtu.be/kHPb11BXB1A

[Sheep64]

Your six(?) chip decode is clever because it has no wasted sections. However, you might have under-estimated the official latency by not counting the address latch. The result is still likely to require less than 100ns in good conditions and attain 4MHz with 150ns ROM. However, it is definitely possible to do more with less.

I'm alarmed by the size of your board which is almost four times the size of the smallest EuroCard - and that's above average on the 6502 Forum. Some of your problems might be caused by board area. In particular, you might have power distribution problems.

Dr Jefyll on Sat 26 Feb 2022 wrote:

You mentioned you've been following Adrien Kohlbecker's video series, and he's facing some of the same issues.

I quite like akohlbecker's "warts and all" approach due to its honesty. It is a very good compliment to numerous failures found within the 500 messages of POC Computer Version One. We learn more from the mistakes of others than their success. However, it is highly inadvisable to copy unresolved problems and I usually have to lurk on the The Plagiarist’s section of Reddit to find this level of unintentional comedy.

Actually, I've quite enjoyed your numerous messages on Reddit and I'm surprised about your second system effect.

[BigDumbDinosaur]

Sheep64 wrote:

I quite like akohlbecker's "warts and all" approach due to its honesty. It is a very good compliment to numerous failures found within the 500 messages of POC Computer Version One.

Indeed, failure, in the words of one of my instructors way back when, has a “divine” purpose: it tends to eliminate that which doesn't work. Over the years, I have found that to be eminently true and have never hesitated to report my failures, as well as my successes. At first, the former greatly outpaced the latter, but thanks to ever-increasing knowledge and experience—as well as a desire to avoid wasting money on stuff that doesn’t work, I’ve managed to partially reverse the failure:success ratio.

Quote:

We learn more from the mistakes of others than their success. However, it is highly inadvisable to copy unresolved problems...

That’s the crux of the aphorism (attributed to Albert Einstein) that says the definition of insanity is repeatedly doing something and expecting a different result each time. Which brings to mind something...

There are those around here who do not want to read anything negative about someone’s project, even though it may be rife with technical errors that will all-but-guarantee failure. All-too-often, said people will say we must be encouraging, no matter what. Dunno about you, but I’m not in the habit of encouraging failure—it will naturally occur, even absent encouragement. Encouraging someone to repeat the mistakes made in the past is insanity; the same mistakes will result in the same failures. Better to be encouraging by rationally describing the problems seen in a design and offering suggestions on how to correct them. If the designer adopts the suggestions, failure is less likely. If not, the designer will merely prove Einstein's aphorism to be true.

[sburrow]

BigDumbDinosaur wrote:

There are those around here who do not want to read anything negative about someone’s project, even though it may be rife with technical errors that will all-but-guarantee failure. All-too-often, said people will say we must be encouraging, no matter what. Dunno about you, but I’m not in the habit of encouraging failure—it will naturally occur, even absent encouragement. Encouraging someone to repeat the mistakes made in the past is insanity; the same mistakes will result in the same failures. Better to be encouraging by rationally describing the problems seen in a design and offering suggestions on how to correct them. If the designer adopts the suggestions, failure is less likely. If not, the designer will merely prove Einstein's aphorism to be true.

100% behind you BDD. Being a math teacher, I know the power of failure: To learn what NOT to do, and try again. The hardest thing for any student of any topic is to try again. And again. And again if necessary. It is even hard for me. If pointing out problems comes across as 'gruff', perhaps the receiving person is being too sensitive. I am a sensitive person myself, but I value the words of experts more than my feelings.

Your post was very encouraging to me BDD, especially above where your success/failure ratio as flipped. I hope to one day say the same. One day.

Everyone here has helped encourage me as well, and spur me on to better things. And that includes you BDD. So, thank you.

Lastly, to rehsd, I'm glad your board is working now! It is a good feeling indeed. Sorry for jumping in unexpectedly but BDD's post was too good to pass up.

Chad

[BigEd]

Everyone here, I'm sure, is in favour of helpful criticism and input. It would be wrong to suggest otherwise.

[Sheep64]

Sheep64 on Sun 6 Mar 2022 wrote:

Your six(?) chip decode is clever because it has no wasted sections.

Apologies. I was looking at akohlbecker's design. You might be using the same design software but I am often amazed by other people's ingenuity.

Anyhow, now that you've switched from 6502 to 65816 and CPLD, you have enough dormant resources to make a UART. You might need an inverter or want a buffer but it is possible to use 65816 EMU pin for UART transmit. plasmo recommends using 65816 M/X pin for UART transmit and CPLD for framing of 8N1 input.

EMU and M/X pins can also be used for servo control or digital LED control. Meanwhile, unused CPLD cells can be used for potentiometer reading or mouse positioning. See EEVblog #1054 - How an Analog PC Joystick Works for potentiometer. The POKEY chip in an Atari 2600 game console may have similar functionality.

[akohlbecker]

Sheep64 wrote:

Sheep64 on Sun 6 Mar 2022 wrote:

Your six(?) chip decode is clever because it has no wasted sections.

Apologies. I was looking at akohlbecker's design.

Thanks, it is quite satisfying to optimize a circuit to fill only a limited number of chips, but use every gate available, even though wasted gates don't actually matter that much. A fun challenge!