Hi J�rg! * Joerg Schilling writes: >>From: Markus Plail <[EMAIL PROTECTED]> >>* Joerg Schilling writes: >>>Note that the encoding speed is content dependant. Also note that due >>>to a kernel design bug, Linux does not enable DMA for ATAPI drives if >>>the sector size is != 2048 (as with RAW sector size 2448). This results >>>in a high system load and may cause a buffer underrun even before the >>>estimated maximum speed of the edc library has been reached.
>>This is not true, at least not as a general statement. As I reported in >>a newsgroup, you are reading, I was able to burn a data CD with -raw96r >>at 20x speed with my ATAPI CD writer with less than 5% CPU usage with >>your closed source version. AUDIO CDs can be burnt with the same >>options with less than 3% CPU usage even with the open source >>version. So it's rather a bad setup than a general Linux problem. I >>tested this with 2.4.18 and 2.4.20-pre8-ac3, i.e. with the new and the >>old IDE subsystem. >The fact that Linux does not use DMA with ATAPI has ben reported by may >people! Having 3% load is a hint that there is no DMA.... You must be joking. If I disable DMA I can't even burn at 20x speed. And actually it's not been many people. I'd rather say a handful and considering that almost everybody has a CD writer these days, that's really few. It was also said, that doing c2 scans can't be done with DMA enabled. I can do them at 40x speed with only a few percent CPU usage. Also a sign for PIO? >In addition, it may be that Linux does not report the load correctly >when doing PIO to the drive. Why wouldn't it? It's all system time when PIO is disabled. Maybe this is a bug, but they load is showing at least correct numbers. >If yout load was less than 5%, then you must have a PC with more than 2 >Ghz. It the oad you reported is really correct, then try >1.11a37.... With the new free cdrecord, you should be able to write in >24x with less than 15% load. I have about 18% load at 16x with the new version. It says libedc is able to do 147x speed. The closed source version does 422x and a36 was about 125x IIRC. I am using a Athlon XP1800+ with CL2 DDRAM. So I am *really* sure that I am using DMA with 'strange' sector sizes. One oddity also seems to be that the CPU usage seems to jump a bit. At 12x I have almost no load and then it's about 15% per 4x speed increase. That may explain my low CPU usage with the closed source version. I would expect it to make a jump, if I'd burn faster than 20x, which I can't unfortunately. regards Markus PS: You don't have to CC me. I am on the list. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
