Posts by Jens

Caution: Non registered users only see threads and messages in the currently selected language, which is determined by their browser. Please create an account and log in to see all content by default. This is a limitation of the forum software.


Also users that are not logged in can not create new threads. This is a, unfortunately needed, counter measure against spam. Please create an account and log in to start new threads.

Don't Panic. Please wash hands.

  • Compatible expansion for A500plus

    A pure memory expansion for the A1200 is assuming that it's talking to a 68020 processor. That is not the case with the ACA500plus. It has nothing to do with "more money for iComp" - we do support properly-designed A1200 accelerators, as you can see on the Blizzard cards, which are supported, but won't change anything on our bottom line.


    Jens

  • Non-functional ACA1233n (Hot CPLD normal?) query

    Didn't check the pinout, but as you've found out, the A1200 bus is separate from the local CPU bus. There are two AS signals, one from the 68030, and another one generated by the CPLD, mimicing the 68020 side for the A1200 bus.


    FPU select is happening based on addresses and function code. Termination of the access is purely done by the FPU with DSACKn. If you see FPU select happening, you already know that the FPU has been detected, as without FPUsense, you'd see a bus error instead of a chip select.


    Jens

  • Holidays from October 8th to October 15th, 2023

    After a very successful Amiga 38 show in Mönchengladbach, Germany, we're taking a week off. We'll be back in the office on October 16th to ship orders that were placed in the meantime. Orders that have been paid until today (Sunday) 11:30am have been prepared for shipping and will be picked up by UPS on Monday, October 9th.

  • Cropped image on overscan

    I know the MK2 tool is a bit techy, but you can just spend some time to play with the numbers and try them out. It's not as comfortable as the MK3 tool (which is instand and fully graphical), but that's just called "progress" - MK2 was just more complicated to configure.


    Jens

  • Indivision aga mk3 a1200

    Quote from Tj1978

    Does the card auto initialise itself now from a cold boot ? I saw a utube video a from a year ago suggesting you have to run the config tool from a floppy each time to get the card to be recognised , which i dont want to have to do ..

    Indivisino AGA MK3 (as well as it's predecessor) has a flash ROM that stores the configuration. This is loaded automatically at power-up, no extra software needed.


    What you've seen in that video is most likely one of the very few cases where the unit doesn't start at all from a cold boot, and the customer is running a "reset" tool which was published in this forum. The computers where this problem occurs are extremely rare - so far, the product has been available for over three years, and not a single one has surfaced in "mainland Europe" - until this week! One unit is in eastern Germany, and the customer agreed to help reproducing the error, so we can finally get to the bottom of it.


    That said, the probability of the same thing happening at your place is very small. So far, we have less than five report of this problem, and there's an intermediate solution for it, so it appears like people are still OK with it, as it's just a tool you need to insert into your startup sequence (using the rescue disk is also an option, but not required).


    Jens

  • Indivision AGA MK3 green screen

    I believe your parcel is the one that could not be delivered here yesterday. Did you ship from Germany, very close to the Polish border? I may be picking up the parcel when I drive to Mönchengladbach later today.


    Jens

  • Is there anyway of making the Keyrah v3 act as a single-interface HID?

    Quote from rockztah

    Does that mean the v2 was implemented in another way?

    Yes, fundamentally different. It had the six-key-limit, and there was quite some cross-effests with joysticks, especially if you use diagonal directions on two-player games. Back when I developed the first Keyrah (17 years ago), I just never thought that this would be used for anything more than an office prank. Then came the RPi, and Keyrah sales exploded :-)


    V2 isn't that much different from V1 - it only introduced the ACPI switch and added jumpers for the different keyboards. This reduced stock-holding complexity (only 1 version instead of 3). A few updated connector components made it scale better at higher quantities. V2 and V3 use mostly the same connectors.


    V3 addresses all requests that we've ever had with Keyrah. I did prepare for quite some extensions with the DB25 connector being very flexible, but wouldn't have expected that an up-to-date-product fails on implementing something that's been standard for many years in PC Bios routines. You might want to drop that bit of information (the comparison to a PC Bios) with some product support/manager at Digilent. They should really update the PIC formware, as Keyrah V3 is not the only thing that will have trouble on their "USB host".


    Jens

  • Is there anyway of making the Keyrah v3 act as a single-interface HID?

    Reporting more than one keyboard is required in order to support more than six keys pressed at a time, which is the limit for a single USB keyboard. So no, this would not be possible without work on the USB code. Please understand that what you're asking for is a reduction of functionality (removal of n-key rollover), causing more work for testing, in order to support a system that isn't fully USB-compliant.


    My guess is that the cost for this code change is in the 2000-2500 EUR range. If you find enough people who would be willing to chip in some money, we could of course do that. However, if you're paying over 400,- EUR for an FPGA board that has a plain description of "USB supports HIDs", and it turns out later that keyboards with n-key rollover don't work, it feels more like false advertising than the fault of the keyboard. If you provide a USB port, call it "USB Host" and hide the fact that it does not have hub support on page 10 of the manual instead of mentioning it with the online product description, there's something wrong. Many keyboards that support n-key rollover use this trick of pretending to be a USB hub with two keyboards connected.


    The good news is that the Nexus A7 100T USB interface is driven by a flash-based microcontroller, so the hardware does support an upgrade path (just like Keyrah V3 does). Granted, implementing hub support (even for just two ports, knowing that there's two identical keyboards connected and no further hub) is more complex than the Boot HID interface protocol only, but it's not prohibitive, as current and not-so-new PC BIOS implementations show.


    You are right that "one of the two sides" needs to move. Both involve code changes. My guess is that Digilent has a much higher margin on their boards to cover such code changes. Keyrah V3 is a low-cost product with a lot of functionality, but low margins. I just have no budget to put someone on such a code change that effectively reduces functionality.


    Jens

  • Jumpy mouse with Indivision ECS v3

    Quote from dwight01

    I have measured VCC from pin 14 of the 68000 CPU, it measured a steady 5.6V,

    Switch off the machine IMMEDIATELY. This is beyond the 5.5V "absolute maximum rating" for many 5V parts. If any of your components has a Z-Diode for clamping, you are burning excessive electricity and putting your components at risk. That's not worth it.


    Quote from dwight01

    Finally, I have measured the E-Clk from pin 20 of the 68000 CPU, again no missing cycles or hiccups, and healthy Pk-Pk voltage of 5.76V, running at 709.2KHz, which I think is normal for a PAL Amiga?

    You don't look for missing cycles on a divided clock such as E. You always look for anomalies on the source side, the 28.37516MHz. However, if you are pushing 5.6V into a 5V oscillator, it is likely to misbehave. Solve your power supply problem first, then you will most likely see that many other problems are gone with it.


    Jens

  • Jumpy mouse with Indivision ECS v3

    If a "busy screen" makes a difference, this may not be due to lots of screen updates, but because of HD activity and CPU power usage.


    The Phoenix board was developed by hobbyists back in the days, and lacks quite a few of the otherwise-often-used EMI filters, bulk caps and other goodies to make the board stable. The phoenix board never passed any CE or FCC testing, so the smallest interference may make the board glitch, which is then caught by Indivision ECS V3.


    You should check your power supply for 5V drops, and all clocks for missing cycles. Before you jump to any conclusions, please test the Indivision ECS V3 unit in a real Commodore-made Amiga, so you can exclude an effect induced by a board that was not mentioned to be a compatible target in the first place.


    Jens

  • Indivision ECS V2 ghosting

    Please register in this forum, so you can take a picture of what you desribe as "smear" - the V3 version has a vastly updated DA converter that is a lot sharper than the V2 version discussed in this thread, so it's either a typo, or a defect that you may be talking about here.


    Jens

  • Product Improvement Suggestion Indivision AGA

    Mouse lag is often introduced by USB mouse adapters. The problem with USB is that the protocol is slow, no competition to the speed and reaction time of PS2. This, added to the comb-effect of the mouse on interlaced screeens might make the effect that mahen is describing.


    There's a reason why we're only making PS2 input devices: Micromys has unparalleled low-lag, and Lyra supports pressing any number of keys at the same time that the PS2 keyboard supports - both are perfectly suited for games. But hey, that's just me, I'm biased :-)


    Jens