CompuServe Messages

JS >> Seems the Amiga is finicky about what kind of IDE HD you hook up to it. I can put any drive into my Intel or Mac PC's and those silly machines just let you do it – but not the ever-so-sophisticated and packed-with-great-features Amiga.<< Good point. The reason the Amiga has problems with some IDE drives, is that the IDE interfaces in the A4000/1200 expect IDE drives, NOT IDE subsets. Some of the IDE drives on the market don't uses the FULL IDE standard, but a common sub-set that works with most IBM/clone interfaces. One of the missing features of these IDE drives is the "MultiRead". The IBM/clones don't care about the extra interrupts, the Amigas do. — JS >>No ability to cold boot from the HD, of course, because that doesn't seem to be an Amiga feature.<< Gee, I've been cold booting from a Seagate in an A1200 for about a month now. The Amiga has a default path for booting. It checks for devices during powerup, then boots from the one with the highest "boot priority". A normal default may be: – CC0: (PCMCIA on A600/1200, if card is set to boot), – RAD: (Recovable rAm-Disk, if present and set to boot) – DF0: floppy disk in drive #0 – HD0: Harddrive if present and marked bootable There could be more devices on boot-up with boot status and higher bootpri. The Amiga gives controllers & interfaces so long to report their existance, and what device are connected to them. If a controller doesn't detect a drive, it assumes one isn't connected. The system then selects the available boot device with the highest priority. If a HD takes too long to initialize and answer its controller (or interface) from a power-up, then its controller will assume the HD was dis-abled or not connected, and will skip reporting it to the system (verses waiting forever.) If your HD won't boot from a cold-boot, but will boot for warm-boot, then most likely the HD takes too long to readily & report itself from a complete power down. As for your HD problem, I'm not sure about the A4000's interface, but some controllers can be set to wait longer for HD to report. Also, you could try powering the HD's separately and powering them up first. — JS >> (According to this source, there are precisely four drives "recommended" by Commodore, none larger than 120Meg. – and the list is not public.) My dealer told me what drives were CBM compatable, and a list of drives came with my CBM HD installer kit for my A1200, and also go a list of several user tested drives off of USENET. — JS >>The NewTek folks gave me a new string to insert in the "MaxTransfer" box under advanced options. […] It didn't work. And the Amiga manual, […] MaxTransfer: "Displays the hex number that determines the maximum number of bytes to moved during each DMA transfer". Why that explains everything!<< Ok, MaxTransfer depends on the HD itself. Hard drives have a built-in cache for buffering data to & from the drive. The Amiga is aware of this cache. The MaxTransfer rate setting, tells the Amiga the maximum amount of data the can attempt to send the HD with causing overflow problems. The cache can be different sizes for different drives. Most 3.5 inch drives have a 256K cache, but large drives may have caches of 512K or more. Many of the 2.5 inch IDE drive have only a 64K or 128K cache. CBM's HD installer software defaults to a MaxTransfer rate of 256K (displayed in HEX), smaller sized caches will cause problems if the MaxTransfer isn't reset to 64K or 128K.