CT63 only starts from 030 mode

[coonsgm]

Hey there Mikro, thank you for all you've shared. Extremely helpful.

[foft]

Hey there Mikro, thank you for all you've shared. Extremely helpful.

Seconded.

Sorry if my reply was a little defensive. Also sorry to confuse matters by posting multiple places.

[mikro]

No problem, actually I think my message also sounded a bit rude. It's just... with every new batch of CT6x, there's a wave of people on Facebook who basically expect to turn the computer on and have everything ready, without reading anything.

I'm a little sensitive on the topic as back in 2001, you'd send ~700 EUR in today's prices to France, wait for two years (!), then receive the unit as one of 300 people and ... start exploring. In the next two years there would be updates to firmware, software even hardware based on our feedback. So it required patience, systematic approach, precise bug reporting and definitely a lot of reading and experience with the system.

Today, people tend to treat new Atari hardware like iPhone's gadgets, everything must work immediately else they start screaming REFUND! SCAM! (not your case of course).

So while it's totally possible that you've got a damaged unit (I find Centuriontech's behaviour quite odd and unprofessional), it's also possible that you started from a wrong end and need to rewind, sit back and start from the beginning, with simplest setup possible, cross-checking on other machines etc (oh, that reminded me, that was also the reason why people tend to buy two of everything - two Falcons, two CT60s etc :D)

Frankly, I wouldn't recommend messing with CT60 to anyone who's new to Falcon and doesn't have at least a year of experience using the machine.

[foft]

I think I jumped in at the deep end of Falcon somewhat, I bought a broken unit and started connecting FPGAs to it :lol:

Back in the day I had an Amiga Blizzard 1260 so feel like I have experience with 68k 060 systems, but really don't since its a different platform. :roll: On that platform its just a case of installing the right 68060 libraries then it pretty much works, so a very different experience.

So I was looking at your website... You ported Starstruck to Falcon, nice! Also Mystic Bytes, that is the same group as Grey right? I went to Silly Venture one time a few years ago back in 2018.

Honestly I don't think Centuriontech are being that unprofessional with me (*), I was probably being super intense and annoying. Probably if I start again and ask a few sensible, clear, questions about the software he'll explain. I got off on the wrong foot by not being able to insert the 68060. The socket is self made which was I'm sure very hard, though the positions are slightly off vs an original socket. Then by spotting the solder bridges (which are _usually_ abnormal) and trying to fix them.

I am not so surprised about using different flash chips in the same batch since parts are hard to get these days with the chip crisis. My projects are completely on hold since I can't get any Cyclone V or Max 10 chips at all.

I can understand its annoying all these newbies jumping around in excitement and confusion - asking all the same old questions every 6 months. The vibe can be different when things are new too, with working things out with the creator. Nowadays Rudolphe has moved on (right?) and its about mining old dead websites and a bunch of outdated into to try to understand where the truth lies.

* Though one page of docs included or a link to a website by them would have been nice.

P.S. A few things I can't find in the manual and a gazillion websites I've searched:
i) How to enable the fan after connecting it
ii) If the boost cable can be used on the CT63 and, if so, which parts to populate. I'll check the schematic here of course too: https://github.com/smduncan/ct63

[foft]

The flash code is here: https://raw.githubusercontent.com/mikro ... 60/flash.s
Looks to check for:
" dc.l 0x00ff2258, xx29f800_bottom_boot_block - devices ;Most of known flashes are identified like this."

Now the datasheet mentions this for the FB, while this is an FT. Which apparently means bottom boot/top boot. Now I don't pretend to know what that means - yet. https://www.eevblog.com/forum/microcont ... oot-flash/
I guess we don't care, except it has a different device code.

op device codes
– M29F200FT: 0x2251
– M29F400FT: 0x2223
– M29F800FT: 0x22D6
– M29F160FT: 0x22D2
• Bottom device codes
– M29F200FB: 0x2257
– M29F400FB: 0x22AB
– M29F800FB: 0x2258
– M29F160FB: 0x22D8

I presume top boot/bottom boot doesn't make a difference, other than flash codes, and this isn't the cause of having to hit reset so many times?! edit: seems to just be some kind of page size thing for flashing.

[mikro]

If you take a look at what Willy had done for this CT60e:

https://github.com/mikrosk/ct60tos/comm ... d5f344d3c3
https://github.com/mikrosk/ct60tos/comm ... eee4fc6b75
https://github.com/mikrosk/ct60tos/comm ... 8737578952

he basically disabled the check for upper 16 bits of the flash id (at least that is how I understand it). So judging from your list (M29F800FT: 0x22D6 = your one, M29F800FB: 0x2258 = old one) it would seem that now even *that* is not enough and downloading https://github.com/mikrosk/ct60tos/releases/tag/1.03d is not going to help.

But just to be sure, try the tools from 1.03d if you still see the unknown flash error. If so, I'd seriously asked Pavel (Centuriontech) for releasing his patches.

EDIT: looking at the supported devices in 1.03c: https://github.com/mikrosk/ct60tos/blob ... ash.s#L329 and 1.04a: https://github.com/mikrosk/ct60tos/blob ... ash.s#L377 ... I see why the lower 16 bits were checked *originally* but in Willy's source: https://github.com/mikrosk/ct60tos/blob ... ash.s#L468 there's basically just the one device present so his patched tools are not really needed (we learn something new every day, I had no idea!) ... so yes, now I'm 99% sure that Pavel has patched the CT60 TOS and flash060 and hasn't released the changes according to GPL...

[foft]

Thanks.

I think the upper 16 bits are the manufacturer and the lower 16 bits the device. So ignoring the top 16 bits is a good change for multiple manufacturer support.

I downloaded/installed this version and can confirm its indeed the same error.

I'm not sure if the flashing needs changing to take account of the different sector layout. See page 10 of the data sheet (https://media.digikey.com/pdf/Data%20Sh ... 5,2018.pdf). I'm still not quite sure what the sizes are - different erase sizes perhaps?

[foft]

so yes, now I'm 99% sure that Pavel has patched the CT60 TOS and flash060 and hasn't released the changes according to GPL...

Perhaps flash060 is not patched and he flashed it via jtag?

I guess he might have installed this 1.05:
https://www.atari-forum.com/viewtopic.php?t=35594
edit: Shows 20190330, similar date.

[mikro]

I'm afraid only Pavel can answer this, I'm quite inexperienced in that area. :-/

But as I said, there's no way he didn't do some manual patching by himself. With this new knowledge I could perhaps create new Flash060 and CT60CONF.CPX from source (same applies to 1.05 "bios" which uses the same access code to flash) but I'd need to have this confirmed from the author.

I think he just binary-patched the CPX and PRG and/or 1.05 "bios" with new flash ID but who knows, maybe there are other changes, too.

EDIT: I'm not aware of a way to upload the image via JTAG (only via the flash060) but perhaps that is just for user convenience. So yes, it's possible that he didn't need to patch anything but then you're quite stuck with what he'd flashed for you.

[mikro]

Hm... now that I'm thinking about it... if a patched CT60CONF.CPX (let's forget any flashing or updating for now) hasn't been released... how on earth you, as a user, are supposed to store flash parameters if support for given flash id is not in the original package?

So you see, we have actually managed to find the culprit of your troubles, CT60CONF.CPX is basically ignoring your changes because it doesn't know your flash!