Posts by Jens

Popping out won't be a problem, but "half-contact" will be. If the power-up ramp is not controlled, at least the Xilinx chip may suffer a latch-up (i.e. "dies"), essentially causing a short on the 3.3V rail. So you should really make sure that it does not pop out. I've watched your video, and it does indeed clip down nicely, so I would not expect what I see next in that video.

One more thing you could try is to clean the Lisa and the socket with alcohol and an old toothbrush. This "easy popping back" does seem like there's some kind of grease on the pins (did you use a contact spray on this chip?), making it way too easy for the socket to be moved. With pure alcohol and a toothbrush, you will remove any such layer, and it may hold good enough for stable operation.

I don't see anything special - neither on the socket, nor the Lisa. THe best explanation I have is that you're not pushing hard enough, and that not all four sides of the socket are all the way down. It does require quite a bit of pressure, but once in place, the unit will hold very stiff.

Please take a picture of both the Lisa and the socket. This is the first time in over six years of product availability that someone is reporting this - I can only assume that there's something wrong with the socket or there's something special with your A1200 that does not let the socket come down all the way.

Hi,

most devices will work out of the box, but some will require the new software, like the IDE to CF card adapters. Download the upgrade disk from the Wiki and execute the flash update (not the CPLD upgrade), this will make most CF cards work, and it may even make the Pata2Sata adapter work.

Yes, the Buddha OS Installer is primarily for installing devices on the Buddha IDE bus, and product support ends where you want to extend it's use. However, it should be easy to find the right commandline options to call the installer with if you want to install a different device name.

No need to be sorry - you've bought a product and deserve the required service to make it work.

All I can see is that "something" with the voltages is not right, and the values you've posted really point to a hardware defect - either on the board or with the power supply. I fear that there's not much left for a diagnosis from a distance. Next step would be to pack it all up and send it to us - if there's a fault, we'll find and fix it. Please include chips and the PSU, so we can test the exact same setup that you're having trouble with.

You can find the company address at the top of your invoice, or in the "Impressum".

acamaprom needs the library in libs: - the program won't work all by itself. The latest archive contains an installer if I remember right.

ACAtune is an old piece of software for older cards, so please stick with the tools specifically for the ACA1233n. I know for a fact that some betatesters have been using the ACA1233n while OS3.1.4 was under development, so it's a known-good combination.

That's really on the far-low side. Before I jump to conclusions, please remove the VIC-II chip and measure on pin#28 again. Does the voltage come up into the area where we'd expect it? (i.e. 8.5V)?

As you have already pointed out, this is an issue of 3.1.4, which should be reported to the vendor of that. Caches MUST NOT be enabled on anything else but memory, and the Z3 autoconfig clearly does not add the space to the freemem list.

Are you a betatester, or did sales of 3.1.4 already start?

Please remember that yesterday was a holiday in Germany.

Next step would be to measure the PSU under load. Connect the PSU to your board, switch it on and use a mount hole as GND pin. Then measure on a few points:

- at the power input

- at pin#28 of the SID (that's "top right")

One thing I didn't mention yet is that some people use the ZIF sockets the wrong way round, and I've seen it even in a Youtube video: Closing the sockets means that the lever is "down". Before inserting the chip, the lever must be "up", then you drop the chip into the socket and push the lever down. For best contact, you can slide the chips sideways while the lever is down (this does require a little force). This will scratch the pins on the contacts, removing a possible oxide layer that may have built up over the years.

Still on the list, not forgotten. Pretty much all family members but me were ill in september, and now I'm recovering, trying to catch up with work.

MacOS drivers from Prolific are truly bad, and you see frequent rants about that on the internet. THere's a terminal program that can talk to the USB chip directly, the name is just "serial". While I do not recommend to buy it due to it's non-working XModem support, the free version will at least be able to show you the remote control menu:

https://www.decisivetactics.com/products/serial/

The mere fact that the two LEDs do blink shows that the MCU on your board is working, so that part is already known-good. We do have a fairly long list of things that are checked during QC, and every single board must pass this list before it gets it's warranty ID. Without checking the database for details, I am very sure that your board has passed QC and is to be rated "known-good".

One thing that you haven't mentioned is the source of your USB cable: Could it be that this cable is a power-only cable? There are cables that are meant for charging only, and they do not have data lines. Hard to see from the outside, and due to the small connectors, even harder to measure. A simple change of cables could already do the trick.

Please also mention the type of PSU that you're using. The power supply does have an influence on the chip recognition part.

The 68030.library is not our product, but it's a free piece of software, part of the MMU libraries from Thomas Richter. This is available on Aminet:

http://aminet.net/package/util/libs/MMULib

Thanks for your hint - we will see if we can get the rights to host this on the iComp network drive, which will be launched later this month.

Both are in production, with Indivision ECS V2 arriving first (probably late october).

The A500+ has it's own RTC which can be used, hence you don't need an extra RTC module. The ACA500plus has quite a few options for RTC, and the location on the A1200 accelerator is just one of them.

We will put a few ACA1221ec through QC later this week. Please monitor the shop site, or subscribe to product-related news by ticking the box at the top of the product page (you need to be logged in to see that box).

The manual asks a very simple question that's not answered here: Is the LED on?

Since I don't have access to the current Betas, I would like to see if it works on a plain 3.1 install. At this point, you have introduced quite a few variables that each can be the cause.

Jens

First of all, installing an FPU is not covered by product warranty or official support. With the high amount of fake chips out there, we strongly recommend to not do this modification.

Burst mode is switched on/of with the "CPU" command of the OS. To enabe full caches and burst, type:

To switch off burst, but enable cache, type:

You may want to check if your FPU has clock: Either close the solder-jumper labelled "FPU clk" (this will clock the FPU at 40MHz), or install a canned oscillator at place U13. Using the canned oscillator, you can run the FPU at the speed that it's rated for, independent of the CPU frequency.

Once again, this modification voids warranty of the product.

Jens

The RR cable MUST NOT be any longer than the one included - this is required to ensure signal quality. I know that one reseller offers longer cables, but those are "use at your own risk" modifications. The clock port interface is really delicate, requires a short cable and the added ground cable to be mounted.

I'll see if I can snap a picture and put these cables into the shop as spare parts.

The clock port on the A512 is really only for the RTC module; there is nothing else that works on that port. Besides, the Poseidon USB stack requries a 68020 (or higher) processor and more memory than the average A500 has. Adding that requries the ACA500plus and an A1200 accelerator attached to that. The ACA500plus has a clock port on it's own where you can connect the RR.

Jens

P.S: Product added to the shop: https://icomp.de/shop-icomp/en…uct/clock-port-cable.html

No need for an e-mail, I'll post here when a solution is found.

Jens

Just remove the "Maprom" jumper.

Jens