How (NOT) to Diagnose TF536 IDE Speed issues.

[terriblefire]

I can see at least 5 bridges on the shoulders of the CPLD. Even with the poor quality photo.

No photo of the bottom of the board.

EDIT: Everyone else who has had bad PCBs has proven it by finding a short or open connection. If you've actually measured it then you'd find the problem.

EDIT2: I can see the issue.. its a short of IDEINT... its marked in Yellow... Other shorts in red.

TF536.jpg

The pin its shorted to is GND. So yeah.. no brainer why its not working.

[exxos]

@Mario_PL Maybe post close up images of the pins themselves (side view of pins) as a top view isn't really that clear.

[8 Bit Dreams]

@Mario_PL
Ok what definitely helps to boot faster - is to add this line in startup sequence:

2020-07-13 22.15.31.jpg

[Mario_PL]

Maybe you can see better now

IMG-20200713-223356544.jpg

IMG-20200713-223421180.jpg

IMG-20200713-223444328.jpg

IMG-20200713-223335967.jpg

[PhilC]

@Mario_PL yes those photos do help but as Terriblefire has said, it looks like there are some bridges between pins where the pins come out if the chip.

Best advice I can give you is to use a microscope to view the pins as I'd said earlier.

An example of my first TF536 where visibly it looked fine, under the microscope it wasn't.

20200608_210834.jpg

[terriblefire]

OK.. first point.. attach photos to your posts don't externally link. This is against the forum rules. But i've updated the post for you.

I still see shorts on the shoulders but not on IDEINT this time. IDECS1... Which is a possible master/slave issue.

You need to actually work now. We cannot teach you diagnostic skills. By taking on one of these projects you have to start using your own brain. Its a possibility that the PCB has an issue but its very very unlikely. Because you've already said you've measured the things that could cause an issue (even if they were on the PCB you would still see the issue if you looked at with a scope or multimeter). My feeling is you havent measured properly or you would find the issue.

Either that or its a software issue. Are you on Kickstart 3.1?

Start measuring all the lines mentioned above and write down where they are connected to. Something is wrong. Waving your hand and saying the PCB is bad is the lazy way of saying i cant be bothered to figure out whats wrong. The PCB isnt some magic thing... its just a bunch of connections and there are only handful associated with this issue.

Still no pictures of the bottom of the board.. THE BUS BUFFERS ARE CRITICAL TO IDE speed. thats why they are there.

[terriblefire]

Also.. if you can get WB to start run CLOCK. Check the second hand moves at the expected speed.

Again i ask you to do lots of things but i never get confirmation that you did them. So i'm basically going to stop helping you at this point.

[8 Bit Dreams]

@Mario_PL
Any news about Your progress?

Have also one question: how big is is block size on Your card? Please try to increase it to 4096 and report how it changes boot time, if all remains same - You definitely have a hardware issue on your card..

[terriblefire]

Not sure why i'm bothering given the attitude from this guy but it might help others...

Test Firmware (It does not actually work and is only meant to work with the adf disk below)

tf536r2_main_top_AMIGA_TEST.zip

This does not work normally but allows this test disk to test the various parts of the hardware...

tfdisk.zip

TestIDEINT - works and checks IDEINT works in by checking that Gayle reflects the state of IDEINT. I Could do a little more here to check the pull ups later.

TestINT2 - Verifies the man in the middle code is working it asserts the internal INT2 line and checks that the correct bit is set in INTREQR. (PORTS).

TestIPL - Forces the IPL lines to go into 101 pattern from Gayle emulation and causes an interrupt. Measures that it was int2 and not another interrupt that happened.

All three shut down all other interrupt sources before starting.

Honestly given the attitude this guy has given i do not know why i've written this for him. Its rare i've seen someone so unwilling/incapable of basic diagnostic skills.

[terriblefire]

Also.. tell us what the disk access speed is from inside SysInfo.