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.
Please understand that you need to create an account to be able to post, guest posting was disabled as an anti spam measure.
Will Lyra v3 fit with the Indivision mk 3?
yes.
Does it make any sense to wait for a v4 with full USB support ?
No, because USB will only report a maximum of 6 keys pressed at the same time, which kinda breaks a key feature of Lyra: It supports actual n-key-rollover with any number of keys pressed at the same time. This lets you play games that require pressing several keys at the same time. We wouldn't be able to offer that with USB keyboards.
Jens
Question: Is the disparity in read performance simply a matter of how different CF cards behave with resepct to the interface (luck of the draw), or does the ACA check some things and determine safe parameters?
No, we don't check for the device's capabilities. Long-term, it might be an idea to load the IDE device's identification block, as there are hints at what speed they work at. However, that might give a false feeling of safety, as not all cards really tell the truth here. So should we ever implement that, we'll also add an override for those automatically-found parameters. Since you still need to make sure that everything works in the hardware combination that you're running (CF card adapters, possible buffers and the card itself), it may be pointless to attempt automatic speed settings.
Sidenote, but there's a noticeable lack of initialization lag when turning on the computer, or after resets - the card starts instantly!
The delay is caused by setpatch, which in turn activates the MMU library. This one requires some time to initialize.
Jens
Didn't watch the video, but interlace/no-interlace glitching sounds like the SuperPlus screenmode driver missing a VBlank IRQ. Indivision AGA MK3 is actually doing what it should: Switch between interlace and non-interlace handling.
So there appears to be a software conflict between the Amiblaster software and the SuperPlus screenmode driver. They may be sharing timers and not playing nice with each other - or it's an MMU setup problem (we're currently investigating another issue where an additional NOP before reading a register fixes things, which should not be the case for properly-serialized IO space).
Jens
Since Thor didn't change anything on the driver, it might have been a side effect of identify.library not playing nice. Please do test that again and report back - we don't have that hardware.
Jens
Yes, that's correct.
That's possible, but might clash if you press keys on both keyboards at the same time. If you strictly use "one or the other", there should be no conflicts.
Jens
Before we try to debug this on an OS that is supported by a third entity (Hyperion), please use a plain 3.1 system for all tests. I use the AUX CF card regularly, and it works, so it's surely not a general problem.
The 68040/68060 libraries are isntalled on that system?
Jens
Thanks for your patience - your observation of "works with a clean install, doesn't with your existing install" might lead us to the root cause - which is of course software, but we need to find out what part. I'll discuss this with Thor.
Jens
Will a simple RTC Clock also disturb ?
No, the Debrick interface only uses a data line that is not used by the RTC, and it also only does read operations - it never actually writes to the RTC registers.
I see that AmiBlaster is a community project. Maybe they can shed some light on what they're doing for timing. The flicker fixer must be getting *some* trigger. Note that Indivision makes use of unused (empty) Chip registers that will have unwanted effects if hit by any other software than our config tool.
Jens
I should probably analyze that Gayle adapter. If you could send me high-resolution pictures via eMail (I have no limit on incoming attachment size), that might give me enough information.
Jens
Jens, you mentioned that XC68040RC40M out of stock. But I have one.
Would it be possible to send to you cpu to place it to the card?
No, we don't work with customer's CPUs. Please also check the the FAQ for the ACA1240/1260.
BTW. customers who have purchased our products from resellers can still be added to the ACA1260 group if you send me an eMail with the warranty IDs of the products you have.
Jens
I'd have to consult the Keyrah manual myself - please check that to learn what key codes these keys *should* send, which actually depends on the position of the power switch.
The left Amiga key doing nothing indicates a bad contact. Note that all qualifier keys (shift, CTRL, ALT, Amiga) of the Amiga keyboard have their own dedicated signal, so they can be used conflict-free.
Jens
Order ID = 156226
eMail with invoice and latest version left on February 17th, 12:00 CET.
I have your power supply!, you aren't getting out of that one that easily
Good! Are you running menu V0.148?
(oh, and I officially don't work on weekends...)
Jens
However, on AmiBlaster activity, the Indivision "restarts" every 5s (small flickering, the OSD appears).
If software hits the HStrobe or VStrobe registers, Indivision will count that as "coming from Alice" - please ask the programmer if they hit these registers. We've had that with some mod player as well - it's really just bad programming, not defective hardware.
Is this the computer that you're trying to get the ACA1240 in Debrick mode with? Note that Debrick mode uses RTC registers, which might be occupied by the clock port device. For an attempt, you could try to remove that. Same thing for that buffered CF card adapter: It may have an influence on clock port operation. There's also some board on the PCMCIA slot that might cause trouble.
Last not least, Debrick modes means that your computer appears like "without accelerator", so you may need to unpack all data first, then start without startup-sequence and navigate to ACA1260tool manually, on the early shell.
Jens
I did not tick any CPU box at all, because it was not asked actually.
Maybe it only asks that in expert mode.
3.x from Cloanto is unsupported from our side. 3.2.2 is supported by Hyperion. We develop for 3.1 and lower, so your 3.0 ROMs should be fine.
Jens
That's unusual. In Debrick mode, the accelerator is completely tri-stated, it's like it's not installed.
I don't know what Kick 3.2.2 does, though. You could try to revert back to 3.1 and see if that fixes it.
Jens
This is something you have to purchase in the IComp store
Buy this and get your download link from your order history in the store
We still have database problems with the shop. So no, you don't get a download link, as what you're downloading is still V3.4.0. You'll get an e-Mail with V3.5.0 and the invoice - let me know what order ID you've purchased with, and I can give you a hint on where to look for that e-Mail.
The older V3.4.0 does NOT contain the Indivision ECS V4 driver. Also, you're not expecting to see a Picasso IV mode. It's really called Indivision ECS V4.
Also, this is all still very experimental. Expect tinkering while we haven't even made a blitter for Indivision ECS V4. The installation procedure will most likely be updated to check mmu library versions and the P96 version, so the current installation pain will be mitigated, if not eliminated.
Side note, the Aux CF flash slot just crashes my computer and it reboots any time I put in a CF card once the OS is booted up. (Any OS, even plain 3.1) If I remove the ACA1260 and put on my ACA1234 it works perfectly fine, so not sure if that is some indicator that maybe my ACA1260 is bad somehow? I sure hope its just bad hardware, because that would explain alot with everything I've been struggling with...
Might be a bad/weak power supply.
Jens
If this is an ACA500plus, you need to change the global settings for RTC to "A1200 accel", otherwise the Debrick mode won't work.
Jens
Try Auto resolution in combination with VSync - that's a sure shot with HDMI.
Jens
With Kickstart 3.x from Cloanto
We don't support that. Please use Kickstart 3.1, or turn to the support of Cloanto.
-> So I would say the hardware is not the issue, but some weird software stuff.
Right, and you should report that to Cloanto. Meanwhile, the workaround is to stop using the Cloanto Kickstart.
By the way, the installer script did NOT installed any 68060.library, only 68040. Is it normal ?
It is normal if you only ticked the box for 68040. The script actually lets you install both, if you choose to do so (at least in expert mode). That said, if you have the ACA1240 (not ACA1260), it appears that the install script did the right thing.
Any tips to get OS3.1 running ?
Really use a clean OS3.1 - don't mix with the wrong Kickstart version.
Jens
