CD-R's

[soco]

there is a rather simple way to test this. assuming there were no special commands needed to enable this, if you have any pce programmer friends out there, they could do sequential reads from the cd and time them. with the magic engine kit it's a rather simple program. you shouldn't even need to care about the ram as you would repeatedly do disc reads from an increasing position back and ignore what was read in.

from what i've seen of the system card disassemblies, it's possible that the drive could've been 2x as they never seem to guarantee a specific seek or read time for any operation. they just do it until it succeeds or timesout.

perhaps if no one has done it in a few months, i'll do it so you can end this

[nodtveidt]

Fragmare suggested a similar thing, but the only real way to test this would be to use the arcade card, since that's the only way you're going to get a sustained rate long enough for any noticeable differences that could prove its speed. Since there's no acd api developed yet, it's gonna be awhile before this can be tested. Just using the standard functions isn't going to work because of the very small amount of RAM the normal setup has. A track jump test would only further prove the seek time difference that is already well-known, and doing a spinup time test isn't likely to be possible through code.

[soco]

unless the hardware access method changed between the drives the amount of ram is irrelevant, simply because the cdrom drive has an internal sector buffer, that the system cards read out byte by byte. there no DMA from the drive to system ram (only to ADPCM RAM).  thus the amount of ram is irrelevant.

you just keep sending commands to read one sector, wait until it's returned success, and then read the next, without transferring the data back to system ram. this is enough to keep the rate up to what it should be.

you shouldn't need the acd library, because you can access the hardware directly, inthe same way the card does.

if the hardware differs in some access way, then yes you would need to use the ACD. however, i thought the ACD only added extra memory and the special shifting registers to the system card.

a spinup test wouldn't really help much, but i think it should be possible by simply stopping the drive and checking the hardware that it's spundown. then start a timer, play a track, and read the Q channel data or the PCM data.

[NEC Avenue]

I'm in the process of scanning old magazines that feature the Duo. Rest assured the CD ROM in the Duo is double speed and I will post the relevant clippings when I find them.

When I finally got the Duo I was expecting much faster load times due to the double speed CD ROM, and was pretty surprised not to get them in a terribly pronounced way.

Did you find the relevent ads?