I am just starting to get back into the hobby. I want to buy a logic anayzer.

At the top of my list is this one.

http://www.linkinstruments.com/mso19s.htm#price

A bit pricey for me but it is so full of features, it is hard to beat.

And then there is there are these two:

http://www.usbee.com/sx.html

http://www.saleae.com/logic/


These are all only 8 channels. Should this be fine?

Any other suggestions for a data analyzer that are sub $250?

We rented a Hewlett-Packard logic analyzer for a 6502 project in the late 1980's but ended up not having any use for the analyzer, which I think will normally be the case if you watch the timing diagrams in the data sheets and are careful about your work. If you go by the "Tips" column with troubleshooting tips here, you'll be fine with just an oscilloscope.

I do have an HP 1610 or something like that here I'd be glad to give someone. I think it topped out at 10MHz which is why it's considered obsolete. A man said he wanted it several years ago, so I boxed it up with lots of styrofoam packing material and then found out it was going to cost $50 at the time to ship it from So.Cal. to the Sacramento area, and when I tried to get back to him with that info, he had disappeared. It's still sitting in the garage, boxed up, with the manual. AFAIK, it works, but I won't guarantee it. If it's worth the shipping to you (which will undoubtedly be more by now) to try it, it's yours.

I tend to concur with Garth. I can't recall ever having a use for a logic analyzer, but sure get plenty of use from my trusty logic probe/pulser and, of course, 'scope. You'd be better off, in my opinion, investing in a reasonably good 'scope, especially if you intend to get elaborate with your 65xx projects.

Keep in mind that for best results the 'scope should have a bandwidth at least 10 times that of the highest clock frequency you plan to use. Since WDC's 65Cxxx MPU's can be reliably clocked at 20 MHz, you could find yourself wishing that 'scope really did go up to 200 MHz or beyond. Lower bandwidth 'scopes will tell you if you have signal activity but if you run into, for example, a problem with Ø2, only a 'scope with high bandwidth will tell you if Ø2 is a sharp square wave or not (the CMOS MPUs tend to be fussy about Ø2 waveform).

While I'm busy spending your money for you, don't forget to use a quality probe. The high bandwidth 'scope won't do much for you if you use an el cheapo probe.

Last but not least, I use an old PC-AT power supply to power my projects (they are still available from several sources). Doing so provides all the common voltages needed to run the types of contraptions I build, plus gives me a safe power source that will crowbar if a massive fault occurs in my latest invention (yes, it's happened now and then). As most PC switching power supplies require a minimum load at all times in order to stay up, I equipped mine with panel mount indicators with incandescent lamps. The lamps provides a nice visual indication that power is on, plus adequately load the supply. One indicator is wired to the 5 VDC output and the other to 12 VDC. If something goes kaput in the power supply, it'll be patent. Obviously, you don't need to get that fancy, but that's why it's called a hobby.

Enjoy making the old 6502 do new tricks. After some 30 years of fiddling with the thing and writing endless amounts of code, I never tire of it. In fact, I have a little project started involving the W65C816S...

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

The piece of equipment that gets used most on my workbench is the Global Spcialties 1310 triple power supply which has a 5V 1A supply and two variable supplies that are supposedly 20V 1/4A but at around 12V they seem to be happy to provide more than twice that much current if you need it. It has current limiting and SOA protection so shorting it won't hurt it, and can display the voltage and current on the two variable outputs. It looks like

but has an older styling that the one in the picture. The 1300 which you might find on eBay is the same thing from the 1980's but with an analog meter instead of digital. These things have an outstanding reliability record. I've hardly used the 5V supply. My workbench computer uses ±12V (actually about 8-14V-- it's not critical) and has its own regulator from the +V power input to +5V, and uses the unregulated power inputs for the op amps for the A/D and D/A converters, the RS-232 line drivers and receivers, and the speaker amplifier (only +V for that one), and feeds these power supplies to the 65SIB (the multi-purpose synchronous-serial bus we devised on this forum) for 65SIB devices that don't have their own power supply, and feeds the +V to my home-made PIC programmer which needs 13V for the programming voltage (Vpp) for most PICs. The Vdd of the PICs is set by the D/A converter so you can program at 5V and test at anywhere from 2V to 6V, under software control. Some LCDs (especialy the STN) need a negative supply for the backplane and viewing angle setting, but this one doesn't.

What piece of equipment I use second to most is a toss-up between the mid-range DMM and my basic 20MHz dual-trace triggered oscilloscope. 20MHz is the minimum guaranteed 3dB bandwidth (probably with 20MHz probes), not an absolute limit. Obviously it won't do what BigDumbDinosaur is talking about (in spite of my 50MHz probes), but just following what we nebulously call "good engineering practice," I haven't needed anything more for getting my 6502 projects-- even commercial-- going at up to 7MHz, and, I expect, higher in the future. I use the workbench computer with its A/D and D/A converters to substitute for an audio digital oscilloscope and spectrum analyzer.

Of the other commercial pieces of test equipment on my workbench, the 512MHz frequency counter is the only one that is somewhat related, and I usually don't even have it plugged in, because it has a crystal oven to get .01ppm accuracy which takes power even when it's off. I really got it for radio work, but although I've kept my amateur radio license current, I haven't been "radio-active" in 25 years.

I make myself other small pieces of equipment in a couple of hours as the need arises. One was a variable-frequency oscillator with a 5V square-wave output to see how fast the workbench computer would run. Another is the PIC programmer mentioned earlier, which is controlled by the workbench computer.

Other than that, I use the old DOS PC as a host for software development (using the very full-featured programmer's text editor and hi-res monitor). Since I have compilation and assembly capability on the workbench computer itself, I only do those on the PC if I want to put it in EPROM; and the EPROM programmer is in the PC with a ribbon cable that comes out to the ZIF socket. The old dot-matrix impact printer with fanfold paper has a switch to select its data input between the workbench computer and the DOS PC. Fanfold is nice for long program listings.

The only simulator I have ever used-- or should I say tried-- was Microchip's for the PICs, but it's so slow and cumbersome it's not worth it, so I just use debugging techniques I put in the "Tips" column.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

I don't think it is a question of what I need. More of a question of what I want.

I feel that a logic analyzer will help me to understand what is going on inside my project.

So maybe I should restate my question. If I was to buy a logic analyzer then which one?

I am leaning toward this one.

http://www.linkinstruments.com/mso19s.htm#price

But tt is only 8 channels. Is that enough? I have 16 address lines, control etc. That adds up to a log more than 8.

It as a 2GSa/s Oscope. How does that translate to mega hertz?

If you can afford it you should really look at this one: http://www.pctestinstruments.com . I bought one about two years ago and it's one of the best value for money items I have in my lab.

/Pontus

I'm try to imagine what you would use the Link Instruments product for. It doesn't look like it will catch glitches or small timing errors (setup time, hold time, etc.) that would keep actual hardware from working. (An oscilloscope would usually be more useful for that.) It also definitely does not have enough inputs (like the Intronix unit does) to watch the buses so that you could for example say, "when the address bus has $F14A (during phase-2 high), record the address, data, and R/W\ for the next 2000 clocks, or until the address reaches $F160 for the 64th time," and scroll through a display of the info gathered, along with instruction mnemonics beside the op codes. That would be almost the same function as an emulator, and real emulators are very expensive.

Simulators OTOH are software only, and there are quite a few you can download free. The simulator can run your code and tell you what should happen on the buses; but it doesn't have the connection to your real-world I/O. It may use stimulus files and have other ways to simulate I/O, but it's not as good in that regard. It's also much slower. Between the lack of real-world I/O and the slowness, it's not suitable for any real-time work. In one of the projects I mentioned earlier, if the computer did not run at full speed, the hardware it was connected to would not work at all. IOW, the computer could not be slowed down at all for debugging. Unfortunately, especially the word "emulator" has been misused so much that its original definition and the lines between "emulator" and "simulator" have been blurred.

If the slowness of single-cycling is acceptable (which is most often the case when you're trying to debug problems that occur soon after reset), the CMOS 6502 can be stopped indefinitely while you probe the all the pins you want to. I've done this a couple of times over the decades, breadboarding an external clock source that advances by one clock cycle every time I push a button.

If you really want a logic analyzer, go for the 34-bit one. Otherwise just use a decent oscilloscope and maybe a free simulator.

It is common for beginners to want all the fancy tools too soon, and these tools insulate the beginner too much from really learning what goes on at the heart of the machine, giving him a disadvantage that takes much longer to overcome. A sure result down the road is more bugs, some of which they won't be aware of until after the bugs have been a secret cause of a lot of inconvenience, equipment damage, or worse. It's kind of like giving young kids a calculator to multiply and divide with before they really have an understanding of what these procedures even are. The appropriate thing is to get them well acquainted with, and practiced at, doing it with paper and pencil, before moving on. A very smart engineer I work with marvels at how I can quickly calculate logarithms (for decibels) in my head in an instant. He's just slightly too young to have used a slide rule, having started with calculators.

I use old PC power supplies as well. They put out loads of current at needed voltages and also have the short protection built in. But instead of loading it to keep it running, I use a jumper between the green and black wires on the main motherboard connector.

Here is a quick easy guide I found: http://www.overclock.net/faqs/96712-how ... y-psu.html

BTW, 1 more vote for a scope only. Very "cheap" on ebay. I recently bought a 500MHz Techtronix dual trace for 1/50th of what they cost new. I've seen the high speed logic analyzers on there too. Complicated looking pieces of equipment they are.

_________________
65Org16:https://github.com/ElEctric-EyE/verilog-6502

You should carefully re-read my comment about the use of an AT power supply. There's no "jump start" (as the referenced page author calls it—the correct designation for this signal is PS_ON#) on the PC-AT power supply motherboard connector. See here for more info about ATX power supplies.


Affirmed here. Again, I've never run into a situation with any of the digital stuff I've built where a logic analyzer was of any particular help. When a circuit didn't behave as expected, I've always been able to sort it out with my trusty logic probe/pulser and a 'scope. For now, a logic analyzer, in my professional opinion, is a needless extravagance.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

OK I will take all your sage advice.

I am back to an oscope I guess.


How fast? Yeah I know as fast as you can afford. Well what about 200MHZ?

This distinction seems quite antiquated. Today, you'll find "emulators" that are pure software (e.g., a Commodore 64 emulator, Apple II emulator, and MAME: Multi-Arcade Machine Emulator).

It's more precise to use "in-circuit emulator" (ICE) when you really wish to refer to any device which behaves the same as, and substitutes for, the isolated chip.

Regarding the usefulness of logic analyzers, they are invaluable when you:

* have more than 32 channels to play with (64 preferred if you're developing contemporary ASICs or FPGAs),
* are debugging a DMA circuit of some kind,
* can set "watch points" (as Garth described).

This lets you monitor, for the most part, the entirety of the I/O signals on whatever bus or custom logic you're attempting to debug.

As far as I'm aware, the original purpose of a logic analyzer was, in fact, to discover bus contention faults between multiple bus masters. Since the 6502 doesn't have many bus masters (typically only one other bus master, running during phase-1), the need for the logic analyzer isn't that great.

My memory on this could be faulty though.

Glad you listen to your "elders." Of course, I really am an elder and enjoying my meager social security check, as well government health care rationing. I may be on a death-list somewhere, but given the general inefficiency of anything run by the government, it'll be years before they finally get to me.


200 MHz should do you well. I managed for years with a 20 MHz Beckman unit salvaged from scrap (an employer from long ago left it for dead). You got a particular unit in mind?

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Well I am right behind you. But I guess I am a baby when it comes to the hobby. Or at least born again.


I just purchased a Tektronix 475 Oscilloscope 2 Channel 200Mhz. I paid $159 + $37 shipping. It did not come with probes. I got a pait of 100mhz for $15.

An eBay special by chance?

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?