TF1260 Testing FW

[terriblefire]

No worries.

What is he verdict on iteration 4?

[matt020]

No worries.

What is he verdict on iteration 4?

Interation 4 / 68090
1D4 motherboard, 060 Rev 5
3.2 ROM, Workbench 3.2.2.1
Using IDE on A1200 motherboard only
No problems, no changes, stable at 67mhz (maybe not, see update)
Only got to test for 20 mins or so, played some Quake, did a mass file copy from HDD to RAM, all good. AIBB did hang after a dhrystone test, I've not seen it do that before nor have I seen it do it again. Appears to be a one off. After a reboot it ran the dhrystone test fine.
Edit: I'll spend some more time testing tonight (AUS time) to ensure that hang was a one off.

2PM update: Quake at 67mhz seems to crash after 15-20 mins. That’s about the same amount of time it took for AIBB to hang

4pm update: for me, the 67mhz hangs are repeatable. After 20 mins, exit Quake and it locks. Or after 20 mins, run AIBB and it locks up.

[terriblefire]

No worries.

What is he verdict on iteration 4?

Interation 4 / 68090
1D4 motherboard, 060 Rev 5
3.2 ROM, Workbench 3.2.2.1
Using IDE on A1200 motherboard only
No problems, no changes, stable at 67mhz (maybe not, see update)
Only got to test for 20 mins or so, played some Quake, did a mass file copy from HDD to RAM, all good. AIBB did hang after a dhrystone test, I've not seen it do that before nor have I seen it do it again. Appears to be a one off. After a reboot it ran the dhrystone test fine.
Edit: I'll spend some more time testing tonight (AUS time) to ensure that hang was a one off.

2PM update: Quake at 67mhz seems to crash after 15-20 mins. That’s about the same amount of time it took for AIBB to hang

4pm update: for me, the 67mhz hangs are repeatable. After 20 mins, exit Quake and it locks. Or after 20 mins, run AIBB and it locks up.

ok that might be phase related. i'll see if i can reproduce in the week

[Miniworks787]

Iteration 4 appears to of resolved an issue I've had with my rev 1 hanging at black screen intermittently at power on with dim power LED. This was after updating from the 999 firmware to the alpha.

@supaduper Kindly helped me diagnose the issue and it appears to be specific to my rev 1 cpu.

Since updating from iteration 3 to iteration 4 it's working perfectly, not a single black screen at power on.

1B motherboard.
Kickstart 3.1.
Booting from onboard IDE
Rev 1 CPU running at 50mhz

Thank you very much for the update :D

[idc]

Iteration 4 appears to of resolved an issue I've had with my rev 1 hanging at black screen intermittently at power on with dim power LED. This was after updating from the 999 firmware to the alpha.

@supaduper Kindly helped me diagnose the issue and it appears to be specific to my rev 1 cpu.

Now you mention it, I experienced this on the 1000 firmware when testing a specific MC68LC060RC66 (rev. 4) processor on a 1B board. It was always directly after power on. Turning the power off and on again once always got it to boot. The same CPU with a 1D.4 board was fine. I swapped it out with an XC68060RC50A (rev. 5) and the issue with the 1B board went away, and I didn’t give the issue any further thought.

I could pop the LC CPU back on to test it again with iteration 4 (‘68090’) firmware, if it helps in any way.

[matt020]

So iteration 4 was giving grief after 15-20 mins when overclocked at 67mhz.

I went back to iteration 3 as it was last reported to be stable. Well, with some extensive testing, it too is also unstable at 67mhz if left long enough and more than just a demo or game is run.

With iteration 3, it locks up and freezes when doing a bulk read from the hdd. I can reproduce it constantly by copying a HDD partition to RAM: or another folder on the HDD. I have conducted long running RAM tests and it passes so I don't believe the RAM is faulty.

If I can get it to copy 70, 80, or 90mb from HDD to RAM or another HDD folder, and then run AIBB, it'll lock up. Or it'll run a couple of benchmark tests and then lock up. Or if I do a mass file xfer, then run sysinfo, it'll lock up at *some* point. In fact, even 62mhz was giving grief.

"But other people report no issues at 67mhz"
I was one of them and running a demo or game is fine. It's when the system is sorta loaded up, multiple apps running. Thats how I can reproduce it. Dopus and AIBB together, after a heavy load of data xfer in Dopus... lock ups.

All of the above runs fine at 50 and 60mhz.

[matt020]

Evening update (evening in West Australia, anyway).

I think I've sussed out what is going wrong here. I now have a stable system, iteration 4, at 67mhz with a Rev 5 060.

I have been using the 6ns RAM jed because I have a 6ns CPLD on the TF1260. However tonight I reflashed using the standard RAM jed, and she's as solid as a rock. No lock up's, no HDD block errors, no AIBB hang, no nothing. I have multitasked, I have copied the contents of DH1: to RAM:, I have run a stack of tests in AIBB and it hasn't locked up once.

I'm confident that you can disregard my post above, because since re-flashing with the standard RAM jed, none of that has occurred. I hope this feedback helps and thank you Stephen for your patience with my updates, trials and tribulations.

[terriblefire]

Evening update (evening in West Australia, anyway).

I think I've sussed out what is going wrong here. I now have a stable system, iteration 4, at 67mhz with a Rev 5 060.

I have been using the 6ns RAM jed because I have a 6ns CPLD on the TF1260. However tonight I reflashed using the standard RAM jed, and she's as solid as a rock. No lock up's, no HDD block errors, no AIBB hang, no nothing. I have multitasked, I have copied the contents of DH1: to RAM:, I have run a stack of tests in AIBB and it hasn't locked up once.

I'm confident that you can disregard my post above, because since re-flashing with the standard RAM jed, none of that has occurred. I hope this feedback helps and thank you Stephen for your patience with my updates, trials and tribulations.

Thanks for that. If everyone else has all their issues cleared up i think we can call this a stable release.

[PainDictator]

Hi,

nice work, but for me the EHIDE does not boot. I have an 3.2.1 ROM with integrated ehide.device, but the screen stays black when I connect my CF to that one. If I however use the internal connector, the machine boots just finde.

Also, I still have a maximum of 4,4MB/s chip read and write speed (PAL Hires noninterlaced, 4 planes) according to bustest (and it feels like that). That problem however also existed with the previous alpha.

Best,
Pain

[terriblefire]

Hi,

nice work, but for me the EHIDE does not boot. I have an 3.2.1 ROM with integrated ehide.device, but the screen stays black when I connect my CF to that one. If I however use the internal connector, the machine boots just finde.

Also, I still have a maximum of 4,4MB/s chip read and write speed (PAL Hires noninterlaced, 4 planes) according to bustest (and it feels like that). That problem however also existed with the previous alpha.

Best,
Pain

Think we covered this. try putting a blank cf card in the internal IDE slot and see if that helps.