Actual vs Nominal burning speed (DVD)

[SaSi]

I have a NEC 3520 burner now and I made a few attempts on burning at "exotic" speeds with HQ media. I used Verbatim, TY and TDK +R media, rated between 8X and 16x. I used a full size DVD image for testing.

With all media, I get practically the same results:

Burning at 4x completes in 15 minutes (as it should).
Burning at 8x completes in 12 minutes (against the 8 minutes Nero - and logic - suggest).
Burning at 16x completes in 8 minutes (against the 3:30 Nero suggests).

All the resulting DVDRs verify succesfully and playback perfectly on several standalones - so I have to assume burning is succesful.

During burning, the Nero indicator on DVDRW buffer (the small one) fluctuates badly and appears empty very often - indicating a transfer rate issue between the IDE controller and the burner. On the other hand, the read buffer (large top one) remains full at all times - indicating the the PC can read the source data from the hard disk fast enough.

The burner is primary on an IDE channel with another HD NOT used during this test. Actually, the source disk is a brand new SATA on a separate controller with only the DVD contents on it, totally unfragmented - not that it matters in this particular case.

My question is: Does the above appear familiar to others that tried burning at 16x - or should I be looking for a problem.

Thanks

[alsyed]

my burner at 4x burns a fullsize dvd image in just over 14 minutes, at 8x just under 10 minutes.

My burner operates on zones, ie, begins burning at 4 speed, then accelerates to 6 speed, then at about 65% speeds up to 8 speed (max). I assume, the times you are getting are a result of this zoned constant angular velocity. or Z-CAV. Because of the nature of dvd, if it is spinning at a constant angular velocity, ie,the same amount of rev/sec, then the linear velocity is greater the further from the centre.

I use dvd decryptor to burn images, as it shows me the burn speed as it goes. YOu can see at certain zones the rate increases. Why not try that to see what is happening?


NOw the problem with your buffer, on the burner. Most are 2MB, not really enough in the first place to cope with high rate transfers (8x is approximately 10MB/s) as long as it doesn't empty or go dangerously low, then the burn speed won't be affected.

[steve2713]

Your burn speeds are a bit slow, although they don't increase in a linear manner as you suggested, especially past 4x. The buffer needs to be built up, as well as accounting for lead in and lead out, finalizing, etc. An 8x burn completes in about 8 1/2-9 minutes for me. I don't own a 16x burner, but burn times of around 5 1/2-6 minutes are more realistic even on a 16x burn.

P.S. - your problem is likely from one of the following: IDE controller/IDE cable/drivers for it. It's unlikely to be a fault of the drive itself.

[MOVIEGEEK]

All my 8x burns are 8-9min when using Nero or RecordNow,it sounds like the drives buffer is malfunctioning.
Do CD's behave the same way?

[koolaidwa]

4x around 14 mins
8x 8-9mins
16x 5-6mins

those are what i get, my liteon only take a lil xtra due to the lead in and lead out wait, but my other btc 16x drive doesnt have the lag

[SaSi]

Your burn speeds are a bit slow, although they don't increase in a linear manner as you suggested, especially past 4x. The buffer needs to be built up, as well as accounting for lead in and lead out, finalizing, etc. An 8x burn completes in about 8 1/2-9 minutes for me. I don't own a 16x burner, but burn times of around 5 1/2-6 minutes are more realistic even on a 16x burn.

P.S. - your problem is likely from one of the following: IDE controller/IDE cable/drivers for it. It's unlikely to be a fault of the drive itself.

I did a bit more "research" on the issue and did the following:

Re-arranged the on board controller devices so that the OS disk was Master on the primary channel and the DVDRW Master on the slave channel. Same results

Replaced the IDE cable to the DVDRW with an high quality shielded IDE cable certified for up to 133MBPS. Same results.

Upgraded to the latest NERO version. Same results

Used DVDDecrypter to burn the disks. Same results.

Took away the DVDRW and put it in another P4 system. Same results.

I guess it has to do with how the particular drive works.

[johns0]

The buffer emptying is normal with the newer nec,the only buffer that shouldnt empty is the record buffer.

[SaSi]

johns0,

your observation is correct.

Due to another problem I had with avi codecs, I had to revert to a backup of my system volume. Fresh installation but same tools and h/w.

The fresh OS installation solved the burning speed problem.

The DVDR buffer is again shown to drain, however the drops in fill factor are now smoother and they don't often get below 15%. Typically draining stops at about 25%.

More importantly, burning continues normally without the noticeable drop in rotation speed heard from the recorder.

A full DVD disk was completed at about 4 minutes and what's more interesting, verification started at about 5~7x (in contrast to the 3.3x I had) reaching 9.5x at about 25% of the disk (in contrast to less than 7x I had before).

Therefore, I think I can assume that the overhead in RW performance was OS related - don't know what exactly other than it is related to the age of the installation (6 months) and the application of patches and service packs.