exxos's Atari, Amiga & retro forum

I couldn't make it any faster than 48Mhz for some reason. Annoyingly don't understand why.
By trial and error I managed to get this : https://github.com/arkadiuszmakarenko/T ... v2F722.git

I found another bug in USB HS HID Clock initialization. It does two init one by one, second overrides some settings from first one, as a consequence USB HS stops working at all. Commenting out one line fixes the issue.

Anyway If this would be working stable for you then we have base project.

I managed to register a mouse which did not work on f105. So there is a hope for my knock off SNES crappy controller.
I will try to port my library to this hopefully over next few days.

arkadiusz.makarenko wrote: 20 Apr 2020 08:35 I managed to register a mouse which did not work on f105. So there is a hope for my knock off SNES crappy controller.
I will try to port my library to this hopefully over next few days.

I think i have one of those controllers too. Certainly sounds the same. I'll try and fire this up later.

is 48Mhz fast enough? The Chinese may have put on crappy linear regulators on my board... is that repeatable between the HID2AMI and TF?

terriblefire wrote: 20 Apr 2020 11:41 is 48Mhz fast enough? The Chinese may have put on crappy linear regulators on my board... is that repeatable between the HID2AMI and TF?

Don't know, I should have one decent regulator left, will swap it and check.

Even with 48mhz I can start working on firmware, if I get to bottom of this issue , I can update clock setting and speed it up.

YEY! Got it.

I think it was power configuration issue. I need to read more about those features in this chip.

In RCC->RCC Parameter Settings->Power Parameters
If you change to Power Regulator Voltage Scale 1 then my chip is stable after this at 96Mhz.

If this was not enough then we would need 25Mhz crystal and we could squeeze 216Mhz from it. ( am not certain about it 100%)

Arek

arkadiusz.makarenko wrote: 20 Apr 2020 19:15 I think it was power configuration issue. I need to read more about those features in this chip.

In RCC->RCC Parameter Settings->Power Parameters
If you change to Power Regulator Voltage Scale 1 then my chip is stable after this at 96Mhz.

Does this fix everything by itself or do you also need the clock config changes.

terriblefire wrote: 20 Apr 2020 19:28

arkadiusz.makarenko wrote: 20 Apr 2020 19:15 I think it was power configuration issue. I need to read more about those features in this chip.

In RCC->RCC Parameter Settings->Power Parameters
If you change to Power Regulator Voltage Scale 1 then my chip is stable after this at 96Mhz.

Does this fix everything by itself or do you also need the clock config changes.

By itself.

terriblefire wrote: 20 Apr 2020 08:51

arkadiusz.makarenko wrote: 20 Apr 2020 08:35 I managed to register a mouse which did not work on f105. So there is a hope for my knock off SNES crappy controller.
I will try to port my library to this hopefully over next few days.

I think i have one of those controllers too. Certainly sounds the same. I'll try and fire this up later.

Yup.
I managed to start it. This chip definitely has improved USB circuit compared to f105, as either my mouse and this controller couldn't communicate reliably.

I am waiting for dual controllers, it may take a while to get to me.

OK in the meantime i'll try and get the integration with the Amiga working.

Please check with me before soldering on the DB23 connector. There are resistors to sort out that are epic to deal with if the DB23 is already on.