Can you shoot a short video? Maybe some artefacts are visible? And do you still have a monitor that displays a picture on the native DB23 monitor output?
Jens
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.
-
-
First of all, Indivision AGA MK3 cannot drive each and every of those pixels, as the maximum output resolution is full-HD, often referred to as 1080p or 1920x1080 pixels.
Games mostly use PAL or NTSC screens, so you should adjust these. Note that you can have multiple output modes for a single Amiga-mode, which can be switched with the live config feature (keyboard - see manual).
In order not to trigger the monitor's scaler (possibly washing out pixels and adding movie-"enhancements"), you should attempt to output a resolution that is a whole-number division of the actual resolution. In your case, I'd try creating a 1720x720 mode and adjust the picture with padding-pixels left and right, so the monitor thinks that the black bars are actually part of the picture you want to see.
Jens
-
Thanks for the feedback. Yes, surprisingly, reading the manual does help in many cases
Jens
-
The CPUs of the ACA500plus and ACA1221lc both only have a total of 16MBytes of address space. From that, you need to deduct Kickstart, IO registers (chipset/CIAs, Zorro space) and Chipram, then you have the total addressable memory.
Since the address space of the ACA1221lc fully overlaps the ACA500plus, the memory on ACA500plus can't be used.
The ACA1221lc already does it's best to fill all the available gaps with memory. It actually has 16Mbytes physical memory, but you can't see it all . The card does support switching to the "invisible" areas in hardware, but to date, there is no sowtware that makes use of this feature. Probably because the only sane thing you can do with it is a RAM disk. You will never be able to make that other memory usable as fastmem.
The only way to get more memory is to use a CPU with full address space, like a 68030, 68040 or 68060 CPU. The ACA1234 comes with 128MBytes of memory, and the ACA1240/1260 will come with 256MBytes of memory.
What's your (memory-eating) application?
Jens
-
Please be more specific: Which key doesn't work?
Does that same key work with a normal Windows program (not Vice)?
Jens
-
Sounds like a software thing. Any special patches installed?
Jens
-
What's the warranty ID of the card? Did you buy directly from us (if so, order ID?), or was this a used card?
Jens
-
The technical data appears correct, but you never know if the manufacturer really fulfills it's own specification. That's why we put all supplier parts through extensive testing, send local persons to the China factory and do further tests that our product liability insurance requires. You might think that this is an unnecessary amount of red tape, but your experience shows that it's totally worth it: Even with the point-of-load regulator on the C64 Reloaded board, quality of input power still plays a huge role.
Jens
-
I am interested, but I fear that this will cost more in shipping and customs hassle than the price of a Lyra-3. If you're willing to spend this money, I'm all for it, but can't promise any date that I can turn to this project.
Jens
-
We're only catching up with work - when the sick leave craze started here in september 2023, a lot of work had to be postponed, and this return/issue is one of them. Sorry about that - this case is not forgotten, and it's moving higher up the priority scale.
Jens
-
I hear a volunteer. What will your contribution be?
Jens
-
Our network install disk contains a newer AmiTCP version than the free 3.0 version you're quoting. You probably have two versions mixed up, which always causes trouble.
Did you read my SCSI issue
I've read it, but I don't see a relation to any iComp product.
Jens
-
I measured it a few weeks ago it appeared fine then but I am now in doubt.
What measurement steps did you take?
Jens
-
Ordered and tried an old mechanical ide drive - this was found and I could partition it. When formatting computer lock-up. After reboot it ”works” but Starting any software from the drive crash (black screen - reboot)
What kind of drive is it? Manufacturer, type, capacity?
I have seem multiple youtube videos working but with the previos Buddha version.
There should be no difference in timing between the old and the new version. Like I wrote before, I believe you have a faulty adapter there.
Jens
-
Yes, we do repairs. I'm personally all for repairing over buying new, but I'm not going to get into details of why I'm a fan of sustainability
In any case, we MUST keep in mind that your A1200 board is questionable.
Jens
-
Glad to hear that the 50MHz license has lead to a success. However, if you see a red screen again, it may still be that the 28.37516MHz crystal in your A1200 is off. You also reported that you've had a TF accelerator that didn't work on your machine, but supposedly did on the seller's computer, so there is something to investigate - if you have the drive to do that.
Putting screws back in is a good idea, as the screws make contact between the mainboard and the shield. This will also improve stability of your system
Jens
-
Please provide more context. I don't see where iComp hardware is involved with that.
Jens
-
That's not a problem, I can see when you've reported things here, and I won't bail out of my obligation to provide a working product.
Jens
-
no idea where those dupont connectors will go. doesnt appear to be any pins on the board apart from the eeprom reset jumper and the clock port.
Like I wrote earlier: The jumper needs to be removed, as that's where the debug-out is being sent. Dual-use of headers, no components wasted
The pin on the outer edge is GND, but I don't remember which of the other pins is the Tx pin. You can just try that out - set your terminal program to 57600 Baud, 8N1, no handshake, and see which pin sends data. You can't break anything by choosing the wrong one. Just make sure you're connecting to the Rx pin of the adapter.
Jens
(yes, license codes are checked manually, as the shop doesn't know which card has which speed licensed already - check your eMail, I have sent that out a minute ago)
-
but if the Commodore one is not good enough, at least it did work with my first MK3.
It's on the edge - I have one A1200 board that I had put aside for years with a label "solder new Alice on", and at some point I started digging into what the Vbb voltage is all about. Then tried to increase the capacitance and blitter errors were gone. Here's a picture of the location on the bottom side of the baord, taken from a German thread of this forum:
https://forum.icomp.de/core/in…ttachment/3221-board-jpg/
The D520 does not give any flickering with the same tests, AmiQuake especially, i have tested them yesterday back to back while i was doing the videos.
Was this while the MK3 was also installed and set to the output resolution that triggers the problem?
Jens
