Nec 2500A vs 3500A. Let's compare!

I have had the chance to test some media burnt with a Nec 2500A and a 3500.
The firmware for 2500 was Herrie 1.07 beta 5.
The one for 3500 was Herrie 2.17.
Scans by Kprobe 2.3.0 on a Lite-On 166S (so take it with a grain of salt).

Here we go with a Ridisc Ritek G04 burnt at 4x.
First comes the scan of the 2500 and then 3500.

And here there are 2 printable Ritek R02 burnt at 4x.
First 2500, second 3500.

I am not sure I would trust the results other to say that the pi failiures are the same on the two drives and the errors are higher on the 3500 anyway both those error rates are way to high for 4x burns on either drive.

Ahh… I was saying :rolleyes: OMG! :rolleyes: looking at these results, then I read more carefully and saw that you used a 166 as reader… :wink:



Yes, unfortunately it’s the only thing I have handy for testing.
I know it’s not like testing with a burner but at least it’s better than not to test at all…

Actually, it might be worse than no test at all, because it might be misleading. If your tests can be believed, then the results are not that different. I, too, own a Nec 3500 and a 2500. For me, the biggest difference so far is that I can write at 8x on several brands of 4x media using the Nec 2500, but not the 3500. Here is a scan of an OPTODISCR04 DVD+R 4x disc, burnt at 8x on the Nec 2500 with Herries 107V2B5 firmware. By the way, you can get 4 packs of 25 for $27.96 with free shipping from here:

I agree completely, its better with a questionable test than no test.

It is fair to discuss whether the test is trustable or not, but we have something to discuss from. Before we were only guessing, now we are discussing data.

If somebody with a Liteon writer could perhaps redo the test and come up with some more conclusive results?

I will post 3 scans in 3 separate posts. All 3 will be OPTODISCOR4 discs, which are DVD+R 4x rated discs. First, the disc burnt at 4x in the Nec 3500 with MadDog 2.F8 firmware. Then, one burnt at 4x in the Nec 2500 with Herrie’s 107V2B5 firmware. And then, one burnt at 8x with the same Nec 2500.

First, the 3500 at 4x:

Next, the Nec 2500 with Herrie’s 107V2B5, burnt at 4x

And last, the same OPTODISCOR4 burnt at 8x on the 2500 using Herrie’s 107V2B5. Here is a quick summary:

3500 at 4x, avg PI 0.93 avg PIF 0.01
2500 at 4x, avg PI 1.81 avg PIF 0.06
2500 at 8x, avg PI 6.13 avg PIF 0.13

The best burn is clearly the 3500 at 4x, but all 3 burns are good and read back perfectly in my settop player. I would choose to use the 2500 at 8x and save 5 minutes on every burn.

Icon, you need to download cdbremse and lock the 166s @ 4x cav before running kprobe of cdspeed quality tests. Setting kprobe read speed to 4x wont work on the 166s.

I’ll do and try this way.

I followed dicer’s advice and I locked DVD speed @ 4x CAV.

I started Kprobe and it showed me a speed of 2.2x at the beginning, it raised till 4x at about 50% of the disc and it reached 5.1x at the end of a full DVD.

By the way, at the bottom line, I see “Driver:SPTI”. Is that correct ?

I attach the new results from re-scanning the previous two Ritek R02 in the same order: differences seem to be less evident.

And here are the results of the previous two Ritek G04.
In this case, I see much more PI errors in ND3500 burnt disc.

Hey, this is really interesting! I traded my 2510 for a 3500, but still wonder if this was a good trade on all occasions. The best comparison however, is when using a burner with PI/PO testing capabilities, so I’m waiting too!

In my case I burn with a Nec 3500 with firmware 2C8_SE_Ritek_4X_D2L. I also used a Lite-On DVD-Rom Reader with CDBremse 1.37 to fix the scan speed to 4x.
Media: Ridata Ritek G04 burned at 4X
Scan by Nero CD Speed 3.42 on a Lite-On 166S (It seems to give a better scan with the 166S)
Burning software: RecordNow Max v.4.5 + pxEngine507 + pxdrv1[1].1.9.0+nec3500A

Nice scan for a G04, really.
I get PI max at around 50 on my 2500A.

Could you redo at 4X speed instead of max to confirm ?


The scan was done with the Lite-On DVD-Rom Reader speeed fixed to 4X. I used CDBremse 1.37 to fix it. I attach the scan of the same disc without CDBremse and you can seee that the final speed went to 8.65X instead of 5.01X of the previous scan.
For Lite-On readers you have to use this application since the sppeed selection in CD Spped does not work.


Thank you for the trouble.
I’ll post a G04 scan on 2500A soon.

On single layer, if I use very good media like Taiyo Yuden with my Nec 2500A, the Kprobe scans are comparable with the ones shown for 3500A burns.
On Ritek G04, and dual layer discs, I must admit that the 3500A scans are better. The problems is that those better scans were most of the time made with far better readers than mine (832s vs 811S)…

Here are a scan of a Verbatim Pastel TY G01 burnt 4X and that of a Ritek G04 also burnt 4X on Herrie’s 216 2500@2510.

I’m happy with both considering the lower price I paid for the G04. They give me no reason for desiring an upgrade to 3500A.