The bug résolut.

PFS2/3 OR SFS 1.193

What ? ? ?


As all possessors of FAST ATA bought this beast to pass the rod of 6 Mo per second, but it been not the case.

As knowing was sold for maximum 16,6 Mo (impossible ! ! !), me can ever pass the 5.5 Mo per second PFS3 coins, in short already not pain because some same not to pass the 3 Mo, me therefore not too much to complain.

But I always wanted better, and when j´ai something in the head me l´ai not elsewhere this not ;o).

During all my tests has this arrived while knowing that 5,5 Mo per second at home the PIO 3 PFS3 coins:

- 1er test :

to pass my FAST ATA in fashion PIO 4 and there all starts well except boils of 5 minutes a request that kills appears: " PFS error in line 45662 etc " and when I clicked on CANCEL he/it continued to function but he/it didn't drag to put back a second request in the stride, result,: unusable ;o(.

- 2eme test :

history to see the filesystem how was going to react pushed vice PIO 5 and this time l´AMIGA more wanted to start impossibility therefore total user plots it ;o(.

- 3eme test :

Tried to modify the Maxes transfer to decrease debits, believing brushes with too many debits and well it was also a failure, same requests ;o(.

- 4eme test :

I have simply changed of filesystem and put SFS 1.193, the last updating, here small change to pass in PIO 4 with better debits (6.5 Mo second) and even stronger, the PIO 5 is also admitted what makes me spend to 8 Mo second the top ;o) while letting my Max transfer to the maxi.

- 5eme test :

I begin to make thick transfers of partition data to partitions and CDROMS to partition to see if all behaves marvellously, the screen freezes, more possible access to the Workbench, obligated of reboot, me the same test and re - crash ;o(. Flute this always not steady.

On this do I bring down again PIO, do I get in PIO4 redo the test and guess what? ?, and very same thing but in color ;o(, all fact as to the first tests history to see if me missed not a stage but nothing to make, very unsteady in copies.

All these tests have nevertheless lasted more than 4 days without having given satisfactory results.

While regrouping all elements positive of these 5 tests one that PFS 2/3 don't make for the FAST ATA 4000 if one wants to pass the PIO 3 one must fall back therefore on the SFS that is as effective as PFS with the last version, then one doesn't arrive therefore to boot in PIO 4/5 SFS coins this the filesystem that is in reason.

After walls up reflection I fall back on the hardware (my small cute sin in makes), and there I cogitate one moment, an idea passes me by the head, why not to change the quartz of the card, to reduce the frequency it would decrease odds of crash.

I look for in my cave d´ALIBAQuartz and find the best compromise 32,000 Mhz in replacement of the 33,333 Mhz present on the FAST ATA in standard.

- 6eme test :

after soldering do I start again my MIGA adored, all functions normally same as having modified the frequency of work of the card of 1,333 Mhz, do I restart tests pushed in PIO 4, is the debit 6,5 Mos (roughly) and guess that stroke ? ? ?, and very nothing crashed, I have just copied 700 Mo without having no crashing, magnificent no ? ? ?.

Good push the plug again farther and let's try the PIO 5 with the new quartz, I will confess you that after having launched the copy test me didn't believe and yet it functioned and as me didn't believemakes make image of my system (600 Mo) as well as the engraving without reboot, all well unwound, not of crash.

-Conclusion :

The FAST ATA is a bomb but unexploitable on certain machines (overclocking etc.), it is necessary can be to come some to the change of quartz, to know that my machine is overclocked and arrange drank G-REX, the overclocking of the 68K modifies the frequency of the ZORRO bus because the jumper EXT clock is activated therefore on the motherboard of the A4000 the bus takes as a basis on the frequency of accélérator card and it can embarrass the working of certain Z2/3S cards (appreciable).

The change of quartz must be makes by a no one that knows a minimum it tries because a stupidity of made and the card would not start anymore.

I would not be on no account person responsible of all deteriorations committed on your AMIGA, to your own risks

-The word of the end:

Is PFS 2/3 very effective on the IDES of basis (buddha, IDE quadrupleur etc..) but PIO 3 to the delà it is not necessary to count more on him on the other hand, surely because the development has been stopped a few years ago, can be will take an updating in months to come, who knows? ? ?

The SFS remains means everywhere, what allows him an exceptional stability and to be able to make pass our AMIGAS in PIO5 he is made therefore for the FAST ATA 4000, he passed tests with success.


Thank you to my POLYMERIC friend (Frédéric Rignault) of motivated to try this terrific Filesystem (that simply adopted) and as had of the patience with my A4000 that didn't want to recognize my partitions more because SFS doesn't write correctly in the RDB of my hard, (this not POLY that of memories ;o).

Promised, juror, dice that I find a new trick for the hardware I keep you the primeur of it.