Booting Flip and Flop

[+bob1200xl]

I am trying to boot an original Flip and Flop (First Star) disk, but it will not complete. It reads a couple of sectors, seeks someplace else and reads some more... never gets to an opening screen. It does load a background graphic after a few reads.

 

I thought maybe it didn't like the XF551 so I tried a Happy 1050... no. Tried an 810... no. Tried an 800... no.

 

You can tell when a drive gets a read error. Not doing that. The diskette surface is very clean and unblemished.

 

Does anyone have F&F? Is there a trick to it?

 

Bob

[+mytek]

This ATR works: Flip and Flop.atr

 

[_The Doctor__]

I'd get a couple kryo or scp images of it for analysis. Perhaps it wants a different os version. I don't remember having the issue myself.

[+Nezgar]

Try with Stock 1050?

 

A stock 810 can seek to other tracks faster than the 1050 if its protection is timing based.

 

Happy 1050 does track buffering which will also affect timing, but unhappy mode may be close enough to stock behaviour.

[Mclaneinc]

Flip and Flop has from the sound of it loading NUMEROUS protection checks, it works XL or 800 mode and as Nezgar says really needs a true stock drive, any non turned off Happy or other drives will stop loading.

Edited July 26, 2019 by Mclaneinc
spelling...as always

[+bob1200xl]

Yes, it does seem to wander about the disk as it loads. A stock 1050 works OK.

 

I may have drive problems with my 810s, which isn't helpful at all. It looks like the 810 can crash a disk even if it is write-protected. Not oxide damage or anything like that, just stepping on the track data.

 

Thanks for the help, guys.

 

Bob

 

[+Nezgar]

5 minutes ago, bob1200xl said:

I may have drive problems with my 810s, which isn't helpful at all. It looks like the 810 can crash a disk even if it is write-protected. Not oxide damage or anything like that, just stepping on the track data.

I've read that the 810 has no protection from a surge going out the write head on poweron/off which can cause it to corrupt data on the disk wherever the head happens to be (usually track 39, since it always seeks back there on idle)  This was accounted for and is not an issue with 1050's.

 

So hopefully not from you turning the drive on/off with a disk inserted and latch closed...

[+DrVenkman]

Never heard of this title but it must have some really heavy-duty copy protection. The ATX file from the A8 Software Preservation Initiative *will* boot from my SDrive-MAX on the v1.1 firmware but it's a really slow, choppy boot process. Watching the SIO status messages during boot shows a bunch of SIO errors, pauses, hitches, then a sectors, more errors and pauses, then a few more sectors very slowly, etc. But eventually it does boot up.

[+Nezgar]

Another thought... If the protection is failing because the 810 seeks track to track faster than the 1050, try a made in Hong Kong 1050 which has a World Storage mech and "WSTR5" stock ROM. Those drives stock ROM also seek the head faster than stock Tandon (made in Singappre) 1050's.

 

It would be interesting to see if the protection fails with that variant of 1050 as well.

[Mclaneinc]

@Nezgar, yes I also read (possibly the same place) about the 810 power on / off corruption issue a LONG time ago (I'd forgotten about it tbh), I also seem to remember one of the guys who did the Computer House Controller card talking about it....Amazing screw up by Atari...

[+Nezgar]

@Mclaneinc On the 810 power on/off corruption issue, I recall trying to reproduce this problem and was unable to get it to corrupt data on a disk. Every sector on track 39 remained readable after a dozen on/off cycles. This was in an 810 with the updated power, analog, and data separator boards.

 

This makes me wonder now if this was more a problem with the early 810's that didnt have those revised components. I recently acquired two non-working 810 drives that appear not to have the upgrades. Maybe theres a chance I can test this theory if I can revive one.

 

I DO remember from personal experience that a disk latched in an ATR-8000 drive would get blasted at poweron. One of the many reasons leading up to me eventually regretfully electronics recycling the thing...

[Mclaneinc]

I don't remember what version of the 810's it was, the latch or the slide down lid?

[+Nezgar]

1 hour ago, Mclaneinc said:

I don't remember what version of the 810's it was, the latch or the slide down lid?

The "slide down lid" or "garage door" mechanism was MPI, and the later "latch" mechanism was Tandon.

 

The upgrades I'm referring are independent of the drive mechanism used. The MPI mechs were used throughout the time Atari released the first 810's with the original rear board, no data separator, and Rev B ROM, until after they released the replacement power, analog, data separator board, and rev C ROM. All of those would have been included in a warranty upgrade, but I hope anyone that opted for the changes out of warranty would have opted for all of the changes.

 

I believe they were all sorted out by the time the Tandon mechs were used, so I highly doubt there is such thing as a Tandon 810 without the upgrades.

 

I haven't personally owned a Tandon 810, but at least with the MPI 810's, one can easily peer inside the front opening to see if the top analog board is present above the drive mech. If present, it's a quick visual indicator the other upgraded components should be present as well.

[_The Doctor__]

little to do with the mech installed, all about the boards, Nezgar  has the right idea...

[Mclaneinc]

Oh well at least we know more....I just wondered if it was linked to the style as all would have had some board in there, I presumed the earlier version, naturally...The point being it was a cock up by Atari either way...

[+DjayBee]

16 hours ago, Nezgar said:

Another thought... If the protection is failing because the 810 seeks track to track faster than the 1050, try a made in Hong Kong 1050 which has a World Storage mech and "WSTR5" stock ROM. Those drives stock ROM also seek the head faster than stock Tandon (made in Singappre) 1050's.

 

It would be interesting to see if the protection fails with that variant of 1050 as well.

Flip and Flop has no fancy protection, only several duplicate/quadruple sectors and lots of bad ones with state "deleted"and one with abad CRC. Some tracks have 19 sectors. There is nothing timing related.

 

Reading occurs mostly sequential (sectors 1-3, 717, 62-125 (=loading screen), 128-130, 59-60, 271-432). The deleted sectors are scattered throughout the last block of sectors and make loading very "unrhythmic".

 

Sectors 128-129 are quadruple and the algorithm reads both sectors until it encounteres all four instaces of each sector. Depending on the rotational delay this can take quite some time.

My quick test in Altirra with standard (not full disk) 1050-emulation took some 50 reads until sector 128 had passed and some 15 for sector 129. YMMV

 

AFAIR only missing sectors result in a head recalibration to track 0, so there should not be much head scratching due to the deleted sectors.

 

BUT: I just tested in Altirra with full drive emulation and there it loads completely different sectors right from the beginning.

Set the drive to unhappy and give it another try.

[+Nezgar]

3 hours ago, DjayBee said:

BUT: I just tested in Altirra with full drive emulation and there it loads completely different sectors right from the beginning.

Set the drive to unhappy and give it another try.

Good stuff @DjayBee! What did you test with Altirra full drive emulation, a stock 1050 or a stock 810?

 

I believe the stock 810 has a longer delay before it can read the next sector due to a couple reasons from a discussion in a previous thread, including doing checksum after the full sector is read as opposed to as each as it is read in like the 1050, and the 810 runs a 500khz CPU vs 1mhz in the 1050.

 

So yes its most likely due to not reading consecutive sectors fast enough, not because of seeking between tracks "too fast" .