Turbo 1230 MK III - MMU problem

Hi guys,

Just got myself a Turbo MK III 68030 accelerator for my A1200, coupled with 32MB of fast RAM. According to the manual the 68030 chip does have MMU capability, and by all means when I use the CPU FASTROM option to manage the Kickstart ROM in fast memory, things seem to work okay - but trying to run WHDLoad with MMU to do more in-depth testing on installers just causes the whole system to freeze.

Any ideas?

[RetroMan]

Just use some system tool like SysInfo to check if your MMU Unit works correctly .... or maybe you got an EC030 CPU, that one came without a MMU Unit

Okay, I tried SysInfo, and it reported MMU status as "68030 (Not In Use)". However, according to Codetapper, that came up on his old board, which had a 68EC030 processor.

I've now tried both WHDLoad and Enforcer. WHDLoad still freezes the system completely, and Enforcer causes the power light to go dim and bright and dim a few times, but doesn't report hits as it should. I'm now led to the opinion that I've been duped by Analogic. A real shame, too Looks like I need to negotiate a refund with them, and try to find a proper 68030 board elsewhere. Anyone who can help me out either with a board they want to sell, or with any links, please by all means contact me :-) If such help is available within the UK, even better.

[RetroMan]

Hmmm, WHDLoad should also work on a no-MMU system !

Sorry, didn't make myself clear enough :-S

WHDLoad runs perfectly fine normally - I meant when you switch the MMU option on (which WHDL by default disables for 68030 processors, precisely because of this problem I'm experiencing).

I'm contacting Analogic to arrange a return for a refund, since a non-MMU board is of no use to me.

This problem is ticking me off no end - Enforcer runs, and if you cause a hit the power led does flash on and off and on - but no error is displayed. WHDLoad still just freezes with the MMU option turned on. I was advised to run a program called "CheckMMU" by Analogic, which reported that the MMU was working 100%. I'm at the end of my rope as to what's wrong here...

[Miggy2TheMax]

Originally posted by djmiketjg
This problem is ticking me off no end - Enforcer runs, and if you cause a hit the power led does flash on and off and on - but no error is displayed. WHDLoad still just freezes with the MMU option turned on. I was advised to run a program called "CheckMMU" by Analogic, which reported that the MMU was working 100%. I'm at the end of my rope as to what's wrong here...

Hi, could you look at the card and quote us the
model number you have on the CPU. That way we can determine if it is a MMU CPU or not. Another question... is the CPU on the accelerator socketed on soldered on board... if its socketed you should be able to find another 030 CPU realtively cheap and you could test it that way.

[Toni Wilen]

I have heard some 68EC030 CPUs are relabeled 68030's (MMU didn't pass Motorola's tests)

I had A500 accelerator card with 68EC030 CPU. Enforcer worked perfectly but all virtual memory programs failed to work correctly.

[jmmijo]

@Toni, isn't that due to the abence of the MMU since it's an EC030 CPU?!?

Also, can't you just replace the EC with a real 030 as I've seen these for sale for a pretty good price

[Toni Wilen]

isn't that due to the abence of the MMU since it's an EC030 CPU?!?

That can't be true because Enforcer needs MMU to work

It turns out the answer is something I totally didn't expect - the board runs perfectly fine - WHDLoad is at fault.

Enforcer, it turns out, was working perfectly - I didn't activate the STDIO switch to view the output in CLI.

And the reason WHDLoad crashes? I quote from the bugs page of WHDLoad..

'There are some rare hardware configurations which don't support the MMU related features of WHDload. The only systems affected are 68030 based. The option MMU/S must not enabled on these systems! If the option is set, WHDLoad will freeze the whole machine. Currently the following hardware is known to cause problems: accelerator boards "Apollo A630", "Apollo 1230" and "Magnum MK2".'

And my board... The Apollo 1230 - dang :-(

Oh, in reponse to Miggy - the code printed on the chip was 68030RP33B (I can't remember the code below that one, and I don't feel like taking the card out of my A1200 for the fiftieth today :-/)

As for the chip, unfortunately it's soldered to the board - no socket - so changing the chip would be a problem.

[Miggy2TheMax]

Originally posted by djmiketjg
Oh, in reponse to Miggy - the code printed on the chip was 68030RP33B (I can't remember the code below that one, and I don't feel like taking the card out of my A1200 for the fiftieth today :-/)

As for the chip, unfortunately it's soldered to the board - no socket - so changing the chip would be a problem.

The serial number on the chip would indicate that its a full 030. Shame its soldered. Toni, about motorola.. I heard the other way around... There was an EC030 shortage, so motorola relabled full 030s as EC030 and thats why you get ECs sometimes working as full 030... but I'm not sure where I heard it from so don't know if its right.