Dvd Burning - 100% CPU utilization

[Druga Runda]

???

Anyone knows what could be the reason

DVD - NEC 8xRW+/8xRW- Dvd burner (will check the model number later)
AMD 2400+
654MB ram
Win 2000
NForce 1
GF44200

It did the same on my old (probably broken now) Nforce2 with 2500+ Barton

Windows 2000, sp4 in use, with latest Nforce drivers.

Ant it's 100% at 4x speed already.

Not only that - media is 4x, nero reports 4x, but the actual speed is 2x ~ takes 30 mins to burn , and CPU utilization is 100%

On the old Nforce2 it was utilizing 100% CPU but at least it burned @ 4x speed as reported...

any ideas on how to solve this problem would be appreciated.

 

[BRiT]

Very odd, I have the same NEC 8x burner. I don't see those symptoms at all. I am running on an nForce 2 system with a Barton 2400+. The drive is in an external FireWire enclosure. When I burn with normal 4x DVD-R media, cpu utilization is barely above 2% and average throughput is 4.3x with most burns in 12.5 minutes. I'm running WinXP SP1, with 512Meg DDR PC2100, and I'm burning most stuff off from a very slow 5200 rpm hard-drive.

 

[swaaye]

Try a new IDE cable. Also make sure it's in a DMA mode and not PIO. That would do it right there and sounds like what you're experiencing.

 

[Dio]

Is it sharing with another (possibly much older) drive? All the IDE drives on the same cable use the same interface mode afaik. It's why it's so important that old and new hard disks, or hard disks and CD's, aren't on the same cable.

 

[mczak]

Is it sharing with another (possibly much older) drive? All the IDE drives on the same cable use the same interface mode afaik. It's why it's so important that old and new hard disks, or hard disks and CD's, aren't on the same cable.

No, that's just not true. It was true pre-1994 or so, but since then all chipsets can use different pio/dma modes for devices on the same channel (otherwise you wouldn't think I'd use my PIO-mode-0 zip on the same channel as my primary udma-100 hd, would you ). The sole and ONLY reason to use different channels for your devices is because you can't read and write simultaneously on one channel with two devices (so when I'm actually using that PIO-mode-0 zip, hd has to wait...).

That said, this really looks like the drive is operating in PIO mode instead of DMA. Reasons can be bad (for instance too long) IDE cables, problems with chipset drivers or just not switched DMA on in device manager.
Windows XP (W2k might be affected too, I'm not sure) though introduces another reason why you might not use DMA: if errors happened on the ide bus due to some reason (can for instance happen with hardly readable DVDs coupled with slightly buggy firmware on the DVD drive), it will switch off DMA for that device WITHOUT TELLING the user, i.e. device manager will still say it's switched on, even if it's not! You can get dma only back on by removing and reinstalling the ide/device driver (some registry hackery would probably also do the trick).

 

[Tim Murray]

Actually, when the errors occur, it tells you fine when you go to the properties of the disk controller. Just remove the disk controller, let Windows redetect it, and you're gravy.

 

[Wixard]

Is it sharing with another (possibly much older) drive? All the IDE drives on the same cable use the same interface mode afaik. It's why it's so important that old and new hard disks, or hard disks and CD's, aren't on the same cable.

No, that's just not true. It was true pre-1994 or so, but since then all chipsets can use different pio/dma modes for devices on the same channel (otherwise you wouldn't think I'd use my PIO-mode-0 zip on the same channel as my primary udma-100 hd, would you ). The sole and ONLY reason to use different channels for your devices is because you can't read and write simultaneously on one channel with two devices (so when I'm actually using that PIO-mode-0 zip, hd has to wait...).

That said, this really looks like the drive is operating in PIO mode instead of DMA. Reasons can be bad (for instance too long) IDE cables, problems with chipset drivers or just not switched DMA on in device manager.
Windows XP (W2k might be affected too, I'm not sure) though introduces another reason why you might not use DMA: if errors happened on the ide bus due to some reason (can for instance happen with hardly readable DVDs coupled with slightly buggy firmware on the DVD drive), it will switch off DMA for that device WITHOUT TELLING the user, i.e. device manager will still say it's switched on, even if it's not! You can get dma only back on by removing and reinstalling the ide/device driver (some registry hackery would probably also do the trick).

Just wanted to back him up on the ide.

Newer HDD standards can re-send data thats been corrupted. If you have rounded ide cables, maybe try moving them away from the case. Or try a differant cable.

 

[Druga Runda]

Tx guys,

I checked and indeed my secondary ide channel was in PIO mode?!? No idea why, but well it was, and device manager - win2K has a nice drop down box to choose between PIO only, and DMA if available mode.

It was in PIO only, (the gray box above says "automatic detection" )

still it let me choose "DMA if available" setting in the second drop down box... I was asked to reboot, and voila all OK.

Which is, CPU usage 1-2% and DVD's being burned in 14.5 mins - which is OK I guess @ 4x... with this media...

Thanks all

 

[Dio]

I stand corrected!

 

[Tahir2]

Dio you are right to an extent. Some DVDRW manufacturers explicitly state that their drives need to be on the secondary IDE channel as master.

 

[mczak]

Dio you are right to an extent. Some DVDRW manufacturers explicitly state that their drives need to be on the secondary IDE channel as master.

Are you sure? AFAIK this is only needed if you want to flash the firmware of some CDRW/DVDRW drives, IIRC some firmware flashers are too stupid to use anything else than secondary master as a target...
Some very old CD drives also might have problems with other devices on the bus, but that would be pre-1994 hardware too.

 

[Dio]

Righto. I stand confused.