Where can i get BB2?

[JetRacer]

Warning! Add modify your startup-sequence like this:

C:SetPatch QUIET SKIPROMUPDATES=exec.library

The exec in the BB2 was udated with one written in the C language (slooow). The original was written in assembler (snappy). The performance with BB2 exec on a 060/50MHz is like a AOS3.1 w. 020/14MHz.

Also, try doing something about the datatypes.library task. It got priority 15 (or something) and locks the computer whenever in use by out-prioriting the rest of the system. I havn't found any good utility to adress this problem  without messing with the rest of the system (Execute), so I can't recommend anything.

[JetRacer]

SilvrDrgn: ...Just try the line I recommended will you. It appears as if it is still running ok, until you put some load on it (ie. multiple tasks). Then it performs like a slug.

[JetRacer]

SilvrDrgn: Oh, sorry; can't check the BB1upd exec version w/o alot of hassle. Using 40.68 atm though.

Everyone: you find current exec version in your WB menu "Workbench/About System".

[JetRacer]

If you want speed on a 060 I also recommend the following patches found at Aminet:

CyberPatcher (aminet?)
CMQ060
PatchMult64
BlazeWCP
FBlit
FText

Also, switching from FFS to SFS vastly increases HDD performance. SFS is so fast (asm) that it outperforms the RAM disk in every aspect but transfer rates. You also don't have to worry about disk validation no more.

Speed things up even further by adding buffers with the AddBuffers command. A few megs doesn't hurt.

Cleaning up the startup-sequence and user-startup also help. The intepreting is dead slow. Remove comments, join all "Assign libs: whatever:libs add" to one line, etc.

[JetRacer]

And, yes, I have tweaked my own startup-sequence  beyond reason. The env: contense is unarchived with lha during boot (faster than lzx (w/o compression)). I have no user startup. All application assigns are collected in a separate script and sorted by partition and directory location. I "cd" to every directory used more than twice so less time is spent processing paths. All "Assign XXX to YYY/XXX add" are merged.

As I said: waaay beyond reason.