Since the first retrobright experiment has lead into a big disaster (#111), we cannot really stop here. So let's do it right this time!
This is what I've got: A transparent box, chemicals, the Youtube video from the 8-bit guy, and yes, two new cute puppies:

I really feel sorry for number 1 and 2 who I made to look like a marble for their rest of their life. Sorry guys, I didn't mean it π. Really.
BTW, the second one came with a mysterious second box... π€

I never dreamed that this story would continue. So cool!
More more more .... π€€
... and may the force be with you this time.
Wow we are in season 2 now. Do we also get new pictures from the mainboards too ? Boardrevision, Revisions on chiplevel and so on ???
Let's start with the box. Fasten your seat-belts for a time-travel...
π€€ An external floppy, real disks, and an original mouse (not unpacked yet). Now that I have a workbench disk, I can finally test the various floppy drives I own now.

Hot literature, too. Oh look, one of them is even "special" π€©.

And there is this (never seen it before) π―:

If we plug this in, we could probably run UAE 0.1 on the AT board π€.
Uuuuhhhhh. Scary π³.

From outside, both Amigas look quite well, better than number 1 and 2. The first one had some kind of colour accident though. Poor thing. Maybe with ink?

Any cleaning experts out there? A free ride in the washing machine might be too harsh, I guess.
The best way might be warm water, washing powder and a (tooth) brush. Avoid to spread the color on surrounding surface of the amiga, which are currently still clean and not affected by the printer color devastation .
ATonce (seems to be rev1) is listed in Amiga Hardware Database http://amiga.resource.cx/exp/atonce as an expansion->emulator expansion.
Printer color... yes, makes sense.
Commodore put in alien screws from outer space in here. So mean π. Need to check my magic tool box...

ATonce is listed in Amiga Hardware Database
80286 @ 7.2 MHz
I've got a performance beast here π.
The interstellar alien screw looks like a descendant of a terranian torx screw.
Alien screws removed successfully, but the keyboard sticks like hell. Can hardly be removed from the motherboard.
Finally, I've got the keyboard removed. Let's see what we've got here:

At first glance, the board seems to be in really good condition π.
BTW, removing the colour from the case was a piece of cake. I did not even had to apply washing powder. Plain water did all the job. I'm gonna put the case in the bathtub for a couple of hours now...
The board is working. First I thought, it's broken, but when I pushed the power coord again, it started booting as usual (from the Gotek, I didn't test the original drive yet). I'll apply some contact cleaner to the port outlets. This should solve the power plug issue.
This is a code RED. I repeat. This a code RED. π³

We'd better get to the lifeboats.
Start preparations for separation sequence of the stardrive section immediatly. All men into the saucer section please....
Commodore Amiga is speaking: This is no exercise. I repeat. This is no exercise. π¨π»ββοΈ
That was close, I was able to isolate it in the last second. No more danger of an imminent core meltdown (I think).
Yummy π:

But there is another issue here π€’:

This is all dirt from the 80s. I am not sure if an immune system from 2019 can handle that. Maybe I've just set free hordes of ancient life-forms π¬.
Amazing. Even the floppy drive still works, after such a long time. And I did not even clean it π. Just plugged it in and inserted a disk.
The left button of the Amiga mouse is broken though π’.
open the mouse and have a look on the click mechanism. Maybe the left button contact is corroded and can still be repaired/cleaned ...
There is not much inside here. The chip is OK, I guess, because the right button and the light sensor stuff is working.

If I push the button with the finger, there is kind of a click (not as strong as with the right button though).
I'll spray it with contact spray, maybe it's just the contacts.
Oh, Mitsumi again. Seemed to be one of Commodore's main suppliers.
Contact spray is my new best friend π₯³ (but I won't apply it in the living room again π·).
So our mouse works now ? This is like bringing the millennium falcon back to live. π
So our mouse works now ?
Works perfectly now π.
So you got 2 new machines? Impressive :)
So you got 2 new machines?
Well, yes, the first two have... how shall I put it... hmmm... small retrobright issues π.
BTW, besides cleaning the new Amigas, I am trying to understand the magic in function decide_diw(hpos) in playfields.js. Is there some documentation somewhere in the wild about the DIW details? I think I am pretty close to the solution, because all my DIW tests pass in vAmiga and I can explain why thing are alike. However, the Sword of Sodan intro doesn't show up in vAmiga, and I can explain (in my theoretical model) why that is. But since it does appear on a real Machine (and in xAE), I must be missing a tiny piece of the puzzle.
Sorry, but can't help you with DIW yet. I don't know of any other docs, except the usual. But maybe the problem is related to something else, like DMA. Perhaps you could disassemble the SOS-intro to see what's going on.
BTW you infected me :) I'm currently in my summer-holiday (3 weeks :) and started to read the HRM again :)
and started to read the HRM again :)
That virus is spreading fast π³. Alessandro has started to refurbish his old Amiga 2000, and I am sure Mithrendal will soon have a Gotek drive in his A1000 (which is really cool BTW).
Regarding the DIW issue, I think I got it right now. All my test 11 test cases as well as the Sword of Sodan intro work correctly when I extend my theoretical model (the way I think DIW works) by a simple rule: "Set the horizontal DIW flipflop at the beginning of each frame."
Hence, I only needed to change a tiny portion of the code to get it working.
Old code (Agnus::vSyncHandler()):
// Reset the vertical DIW flipflop
vFlop = false;
New code:
// Initialize the DIW flipflops
vFlop = false;
hFlop = true;
The hard part was to write suitable test-cases to back up my theory.
Regarding practical things, I have another lessons learned for today, which is:
OK, applying the contact spray to the drive also had it's positive side. It's clean and shiny now. The downside is that it doesn't work any more π¬.
Wether forecast is predicting a lot of sun for today. It's showtime π.
This is what I understood from the 8-bit-guy video π€:
I'm so scared π. The "new" Amigas are in pretty good shape even without bleaching. Besides, if I screw this up again, the retro community will hate me, because I'm destroying all the lovely old machines. One after another.
π This is getting a nightmare.

And now undercover... I mean under cover:

Furthermore, because I was not sure if it is save to touch the liquid with bare hands as the 8-bit-guy did, I decided to wear the full package (safety eyewear, gloves etc). Now my neighbours think I'm conducting a mysterious bio-hazard experiment. But I only want to turn π into π.
For the second keyset, I'll try the "traditional" zipper bag approach.

For the first set of keys, I turned them all upside down so that they can float on the liquid. Maybe they get enough sunlight reflected from the bottom π€.
What about the temperature of the liquid ? 8 Bit guy used warm βliquidβ 58 celsius.
I have an idea for the bag of keys: place the bag on top of a small mirror. Then they will get hopefully UV from all sides.
What about the temperature of the liquid ?
I don't have a thermometer that can be applied from outside like the 8-bit-guy had. I would have to put my thermometer inside which means that it'll came out bleached. But I boiled the water before filling it in. Unfortunately it is not as sunny as my weather app predicted. It has gotten a little cloudy.
of a small mirror
Good idea. Need to look for a mirror... on the other hand, a mirror could generate hot spots π€.
Are the clouds static ? I mean no wind ? Then the clouds make cloud shapes on the case... We dont want an Amiga CE (cloud edition) do we ? In that case put all the stuff into your car and drive to a place where it has more sun ... No?
hopefully you have a windy sky which will bloat the clouds out of the way for our experiment.
Amiga CE (cloud edition)
I am expecting the worst π¬.
I already have an
Regarding the mirror idea, I've put the zipper bag on a board wrapped in aluminum foil. The foil is more diffuse than a mirror.

Before:

After:

Unfortunately, it's hardly impossible to shoot comparison pictures easily. On the pictures, the result looks much more yellowish that the original, but this is only because the image is more yellow in general (have a look at the floor tiles).
Judging the preliminary result, I think it's really good π₯³. But it has to be taken into account that this case was in really good condition right from the beginning and not very yellow at all.
The question of success will be decided by the keys (still taking a bath). These were clearly more yellowed than the case.
Did it already bath 6 hours in the sun?π€
Or were you afraid of getting it too white?
No stripes. Very even result. π
Final result for today:
Keyset bleached together with the case π:

Keyset bleached in the zipper bag π₯³:

The zipper bag keys have gotten really nice. They have no stripes at all and look even better in reality than they do on the picture (on the picture they still appear to be a little yellowish).
The keyset that was put into the container (together with the case) has to be redone. It's still π€’.
Since I used pure 12% H2O2 in the zipper bag, I think that my 1:2 H2O2:H2O ration in the container was not optimal. A higher concentration of H2O2 will likely do a better job. However, to bleach the case, a large container is needed (it's hard to find a transparent container with perfect size, so mine is a little too large).Hence, a lot of H2O2 is required. I needed 12 litres fluid in total to get the case covered with liquid (8 litres H2O and 4 litres H2O2).
On Tuesday, I am going to redo the second keyset and the second case. Unfortunately, I am running low on H2O2, so I am unsure how far I can get.
The case looks good to me. It's really is a very even result. The cover has no stripes at all and I think I can use it as it is. Together with the zipper bag keys, this A500 will be a beauty again π.
Very nice !
Concerning the too big box I have an idea for a better ratio but without using just more of the H2O2. You could fill one transparent plastic bag with warm water and give that also into the box. Carefully tied up of course. This way you could raise the fill level of the box without to lower the concentration of H2O2. Place the plastic bag on the side. Maybe also put a stone in the plastic bag to keep it sinking to the ground.
You could fill one transparent plastic bag with warm water and give that also into the box.
That's it!!! Sometimes I'm blind to the simple solutions π.
The Retrobrite forecast says good conditions for tuesday between 14:00 and 20:00.
Please prepare for countdown .

Retrobrite Forecast, this is launchcontrol. I repeat time frame tuesday 14:00 - 20:00. Roger. Radio check over. Roger. Air pressure nominal. Roger, nominal. Temperature above high, nominal. Roger. Control, retrobrite box alignment complete. Roger. Waiting for Pilot voice check. Over.
Breaking news +STOP+ Mission aborted last minute +STOP+ Weather conditions unstable +STOP+ Clouds all over my place +STOP+ All existing H2O2 evaporated (or drunken by cat) +STOP+ Need to restock supplies +STOP+ Next mission window: Wednesday +STOP+
Stocks are up, the eagle got refuelled.

Winds are bad, though. Have to stay in the dock until further notice. Weather forecast predicts a suitable starting window for tomorrow...
This is houston ground control: Roger - weather conditions verified - shifting eagles countdown to tomorrows timeframe -- whish you all the best -- and over
Set sail four hours ago. Weather conditions are so-so.
What I can already tell is that bleaching the keys does not work well. They simply refuse to loose their yellow colour π.
In the meantime, I found another approach in this vid:
https://www.youtube.com/watch?v=c7rIgt6jXzg
The guy used creme to prepare a yummy keyboard salad π€€ (at 1:35). He said it worked out well for him.

The problem is that I swore to myself I wouldn't use the cream anymore π.
π

Good, not an overkill unnatural bleach. For me it looks also not yellow at all. It looks quite original. π€πΏ
Is that the space key there in the middle? It looks very good.
Considering today's weather and the frightening forecast for next week, it's D-day πͺ.
Because I am running out of possibilities, I have decided to break my promise and prepare the key cream salat π. Say goodbye to a once-good-looking key set... π’

Don't worry about the blue coloration of the image. This is because I live next to a graveyardπ¦...
I understand in such a desparate situation you donβt think twice whether to give them the kiss of death or not or not not or not not not ...
Here is the final result π. Both puppies got their π drives built in and are ready for duty:


To install the π drive inside the newer model, I had to cut off a small plastic part from the top case with a bread knife. I know the retro community hates people who do that. You won't tell anybody, will you? π¬
They are looking so beautiful. Honestly like new in 1989. Well done. You can connect two of them with a null modem cable to play networkgames. The cat is still okay?
The cat is still okay?
Don't wanna talk about it. Maybe the neighbours won't notice π.

Most helpful comment
π This is getting a nightmare.
And now undercover... I mean under cover:
Furthermore, because I was not sure if it is save to touch the liquid with bare hands as the 8-bit-guy did, I decided to wear the full package (safety eyewear, gloves etc). Now my neighbours think I'm conducting a mysterious bio-hazard experiment. But I only want to turn π into π.