TF4060 Beta Program

[terriblefire]

@terriblefire As there are perfectly working solutions (A4000T, A4000 w/o Z3 DMA cards, A3000x), I would offer a Jumper or SW solution (maybe with 1x reboot?) to disable the upper 128MB for these cases.
A new HW with fixed drill holes for the stands could introduce the Jumper to do this trick.

I mean by now there are too many cards in the wild to make a hardware change. must be close to 100 sold now.

[saft6luck]

I mean by now there are too many cards in the wild to make a hardware change. must be close to 100 sold now.

Oh, so Beta is done?

[terriblefire]

I mean by now there are too many cards in the wild to make a hardware change. must be close to 100 sold now.

Oh, so Beta is done?

Its gradually expanded over time. I guess its its no longer beta. I'm still taking bug reports

[saft6luck]

Its gradually expanded over time. I guess its its no longer beta. I'm still taking bug reports

• There might still be bugs, and there will be firmware updates

• Please report any bugs you may find

• Be patient

• Although very unlikely at this point, there MIGHT be hw changes needed

• NOT to be resold for a period of 6 months or until the general availability happens, whichever comes first

• You have 30 days to return the board for a full refund

• If posting or reviewing the board on any website/YouTube/article etc, please state that the card is an Open Beta, and things can change at any time, including performance and behaviour

To be frank, this doesn't sound like beta is over plus HW changes are to be expected.

Do you plan to fix at least the hole of the standoff?

[saft6luck]

Essentially all i am doing is blocking burst to SDRAM when DMA is in flight. It cannot burst when DMA is happening anyway as the clock speeds are 4x different at best case.

Then nothing from Amiga side can burst, regardless of DMA? And on the CPU board there is no DMA anyways?

tf4060r5_18533_2024_08_21_121f718_alpha.zip

EDIT: Tested on A4000CR with Fastlane Z3. Stable AF

Tested on an A4000T and A4000D+: much better. What happend to the cpuspeed command? And, is this change available in an official release, too?

[saft6luck]

Attached is an updated version of the original U714 which excludes the upper 128MB TF4060 RAM area from Buster Z3 space.

I tested DMA to a Zorro III 256MB RAM expansion from my onboard NCR which still works fine.
Of course onboard NCR DMA still works to the upper 128MB RAM area.

Maybe someone can test with an A4091 (devtest with option -m 16000000 should do the trick)?

EDIT: Tested the A4091 and the onboard SCSI. The results look promising.

Screenshot 2025-01-26 181517.jpg

[terriblefire]

Attached is an updated version of the original U714 which excludes the upper 128MB TF4060 RAM area from Buster Z3 space.

I tested DMA to a Zorro III 256MB RAM expansion from my onboard NCR which still works fine.
Of course onboard NCR DMA still works to the upper 128MB RAM area.

Maybe someone can test with an A4091 (devtest with option -m 16000000 should do the trick)?

EDIT: Tested the A4091 and the onboard SCSI. The results look promising.

Screenshot 2025-01-26 181517.jpg

Nice work.

[misha]

Hi everyone,

I just wanted to share some feedback on the TF4060 — what an amazing piece of hardware! I'm really glad I joined the beta program and I want to congratulate Stephen on this brilliant design.

The card arrived with older firmware, but the update process using a rPi was super smooth - really impressive!

The TF4060 has now replaced my Warp Engine 4040 (with 68060) in my A4000CR, and so far, I’ve tested it with:

• FastATA4000 mkVII

• BigRamPlus

• ZZ9000 + AX

• X-Surf-100

• RapidRoad USB

To make things even more fun, I switched from OS3.9 BB4 to OS3.2.2.1 during the process.

I managed to overclock my 060r.6 to a stable 94 MHz — it even hits 100 and 101 MHz, but things get a bit unstable there, with occasional freezes.

Now, a quick question:

While testing with Amiga Test Kit 1.21, I noticed a CIAB TOD fail (showing 10,149,055 ticks instead of 5,008) and a CIAB TOD IRQ failure. I double-checked using another tool that tests interrupts, and it also pointed to irregular behavior with the CIA at U300.

Interestingly, despite these errors, the system runs fine — no noticeable issues in regular use. But I’m wondering if anyone has come across something similar or has any ideas on what could be causing this?

Thanks again for this amazing card — loving every minute with it!

[terriblefire]

Hi everyone,

I just wanted to share some feedback on the TF4060 — what an amazing piece of hardware! I'm really glad I joined the beta program and I want to congratulate Stephen on this brilliant design.

The card arrived with older firmware, but the update process using a rPi was super smooth - really impressive!

The TF4060 has now replaced my Warp Engine 4040 (with 68060) in my A4000CR, and so far, I’ve tested it with:

• FastATA4000 mkVII

• BigRamPlus

• ZZ9000 + AX

• X-Surf-100

• RapidRoad USB

To make things even more fun, I switched from OS3.9 BB4 to OS3.2.2.1 during the process.

I managed to overclock my 060r.6 to a stable 94 MHz — it even hits 100 and 101 MHz, but things get a bit unstable there, with occasional freezes.

Now, a quick question:

While testing with Amiga Test Kit 1.21, I noticed a CIAB TOD fail (showing 10,149,055 ticks instead of 5,008) and a CIAB TOD IRQ failure. I double-checked using another tool that tests interrupts, and it also pointed to irregular behavior with the CIA at U300.

Interestingly, despite these errors, the system runs fine — no noticeable issues in regular use. But I’m wondering if anyone has come across something similar or has any ideas on what could be causing this?

Thanks again for this amazing card — loving every minute with it!

I'll add it to my list of things to check.

[misha]

I'll add it to my list of things to check.

I appreciate your response! I've now tested with other accelerator cards, and it turns out this is just how my A4000CR rev. D behaves, so there is no need to trouble yourself with it. Thanks for your time and interest, but the TF4060 is not the cause. By the way, I swapped the CPU from my Falcon with CT60e, and now the TF4060 runs fine even at 100MHz! 😃

PS. It kept bothering me, so I went ahead and replaced the Even CIA—I soldered in a brand-new chip, and the problem is completely gone. Amiga Test Kit now shows correct readings!