User avatar
intric8
Seattle, WA, USA

Posted Sun Apr 03, 2022 10:00 am

Folks,

Thanks to recent troubleshooting work by EzdineG and jdryyz they discovered a bug with the clone Rejuvenator boards.

J100

There is a jumper option on the Rejuvenator that allows one to switch between using the Kickstart ROM on the Rejuve back down to a stock floppy-based KS setup. If you short J100 (i.e. put a jumper on the 2 prongs) this will cause the Amiga to black-screen.

Removing the jumper, thereby taking the machine back to using the Kickstart ROM (which is what 99.9% of folks with Rejuve's do, so I'm not surprised this bug got missed) works perfectly.

This is being investigated.

FYI

User avatar
EzdineG
Springfield, MO

Posted Sun Apr 03, 2022 10:56 am

intric8 wrote:
Sun Apr 03, 2022 10:00 am
J100

There is a jumper option on the Rejuvenator that allows one to switch between using the Kickstart ROM on the Rejuve back down to a stock floppy-based KS setup. If you short J100 (i.e. put a jumper on the 2 prongs) this will cause the Amiga to black-screen.
Thus far we have confirmation of this bug on three Rejuvenator boards, covering :
  • 1MB and 2MB Agnus
  • 22V10 Atmel U30,U31
  • 22V8 Lattice U30,U31
If anyone else can confirm (or refute) this behavior, it would go a long way toward finding the issue. Boards should display a blue screen if you jumper J100 and remove the front 256K RAM expansion, and request the Kickstart disk with it attached.

The currently observed behavior is that you always get a black screen.

Thanks!

User avatar
Dragonheart

Posted Sun Apr 03, 2022 12:07 pm

A friend and I are selling the clones on Amibay and so far no issues have been reported with the j100 option. On my own one it works as intended, someone on the German a1k forums who bought one of my boards also confirms that it is working.

I’ve messaged a few people who bought Rejs we made but I’ve not heard back yet.

User avatar
EzdineG
Springfield, MO

Posted Sun Apr 03, 2022 12:38 pm

Dragonheart wrote:
Sun Apr 03, 2022 12:07 pm
A friend and I are selling the clones on Amibay and so far no issues have been reported with the j100 option. On my own one it works as intended, someone on the German a1k forums who bought one of my boards also confirms that it is working.

I’ve messaged a few people who bought Rejs we made but I’ve not heard back yet.
Is it possible to get a close up photo of your Rejuvenator with working J100? I'm interested in the PAL chips, revision of the board and perhaps a good look at the parts you've used. Have you deviated at all from the published BOM?

What revision is your Amiga 1000 motherboard?

Thanks for your feedback!

User avatar
Dragonheart

Posted Sun Apr 03, 2022 12:54 pm

I’m out of town for a couple of days, unfortunately I cannot help you with pictures. I think the photos I have attached to the sales thread on amibay are full resolution though, maybe you can take a look at them.

Edit: I just received feedback from two buyers that the KS option works. One of them doesn’t even use a ROM chip at all, the other dude has a switch installed. Seems to work fine.

User avatar
Acill

Posted Sun Apr 03, 2022 4:50 pm

Hmm, I have built a few of these from the original boards and have one from a very early batch of boards I made from the first public files released. I havent thought to try reverting back to booting from disk though. (Why would you even want to do that?) I have mine out of the 1000 currently and am testing some side car things out. I can try and see what mine does, but i never got any complaints from the others I made. I would suspect that its not something many people thought to try.

User avatar
EzdineG
Springfield, MO

Posted Sun Apr 03, 2022 5:30 pm

Dragonheart wrote:
Sun Apr 03, 2022 12:54 pm
I’m out of town for a couple of days, unfortunately I cannot help you with pictures. I think the photos I have attached to the sales thread on amibay are full resolution though, maybe you can take a look at them.

Edit: I just received feedback from two buyers that the KS option works. One of them doesn’t even use a ROM chip at all, the other dude has a switch installed. Seems to work fine.
Thanks! Your boards look identical to the two I've built, down to the board revision. Perhaps the issue could be down to the Amiga motherboard revision; I assume your tests were all done on PAL machines?

User avatar
EzdineG
Springfield, MO

Posted Sun Apr 03, 2022 5:37 pm

Acill wrote:
Sun Apr 03, 2022 4:50 pm
I havent thought to try reverting back to booting from disk though. (Why would you even want to do that?)
Agreed, I wasn't even going to solder the header for this feature.

That said, we stumbled upon this when trying to diagnose why one board wouldn't boot when paired with a Parceiro, but works fine otherwise. Others don't have this issue, so there seem to be some subtle "differences" I'd like to document for future builders.
Acill wrote:
Sun Apr 03, 2022 4:50 pm
Hmm, I have built a few of these from the original boards and have one from a very early batch of boards I made from the first public files released.I have mine out of the 1000 currently and am testing some side car things out. I can try and see what mine does, but i never got any complaints from the others I made. I would suspect that its not something many people thought to try.
I'd appreciate if you could find out which (if any) of your boards behave this way so we can find a common denominator.

Thanks!

User avatar
jdryyz

Posted Sun Apr 03, 2022 6:46 pm

I know I'm in the minority here but I see the A1000's ability to load kickstarts into RAM as a bonus. I also know one of the primary selling points of the Rejuvenator was to allow users to install real ROMs and do away with pesky kickstart floppies. I totally get that.

My belief is that if the original Rejuvenator allowed for this fallback method to work, then the new Rejuvenator should be able to handle it as well.

Consider also that this ability to softload kickstarts could be incorporated into an expansion device so that more than a single ROM can be used. You need not stick with one OS version.

User avatar
intric8
Seattle, WA, USA

Posted Sun Apr 03, 2022 9:24 pm

Today I learned of one additional bug with the Rejuvenator.

Most folks here will never, ever care about this. But I was working with a friend today *(Crispy) with a new program he'd written that allows A1K's and StarDrives to soft-reboot without the need of a floppy disk. Really, really cool.

And this is regarding the RAD:, the recoverable RAM disk, which should be persistent until you manually remove it with the RemRAD command. The RAD command should make the program in question resident, and allow it to survive reboots.

I went through a process tonight to see if the clone Rejuve would honor the RAD.

It did not.

On soft-reboot, it cleared it out.

That being said, there does not appear to be any mention online anywhere of the Rejuvenator not honoring the RAD:. However, I'm not currently in the mood to pull my clone board to put in the OG board to re-test this.

Either way, it is something the clone board (and possibly original) suffers from. Again - this is an edge case, but it is something that affected a project I've been working on for weeks and surprised me.

Crispy:
It should survive any number of reboots.

For some reason it's messing with the chip ram memory, and overwriting or corrupting code that was in there before the reboot.

There's no way to do [my program] autoboot, or any other resident program, on an A1000 with a Rejuvenator.
Or, at least a cloned Rejuvenator. But possibly all of them.





Return to “Hardware”