As the title of this post suggests, if you are not as sick as I am when it comes to sound quality you should stop reading now. What I'm describing here is not clearly audibile (or even audible at all) on an average audio system and with untrained ears. However, for those of us willing to spend the money on high quality audio gear and interested in CD-R recording, not all the burners do their job just as well - and I'm not talking about speed here.
On short, I own two internal IDE burners, a cheap (~ $60) Philips 4816 and a Premium, and here is the heresy: THE PHILIPS READS AND WRITES BETTER THAN THE PLEXTOR (even better overall than the 0.6 Gigarec recordings). No, the Philips drive is not an undiscovered audio genius - I only compared it directly against a Teac 52x, and the Teac was more detailed and coherent and had better transients, although it was brighter (farther away from the tonal balance of the original CD). I posted before about this on Head-Fi, source forum - by the way, excellent source of audio information and number one in everything about headphones.
I am reading with Plextools Pro (with Plextor) and with EAC (with both drives), both programms set for best accuracy; the writting is done normally at 4x with Plextools Pro (for Plextor) and with Nero (for Philips). The listening has been done on some very good $150 AKG K501 and $220 Sennheiser HD600 headphones out of a $300 Technics SL-PS770 and a very cheap Clatronic CD players. Yes, I know, the source could be better, but I don't belive this would change my conclusions. My wife, who is not so interested in sound quality, and an audiophile friend have heard the same things during blind tests, so it's not just me.
Here are my findings:
- first of all, I want to make something clear. ABSOLUTELY ALL copies I have ever heard, and I've heard a lot, including 1x - 52x recordings made on Imation, TDK, Verbatim, Plextor, Memorex, Traxdata, Sony and other brands of CD-Rs (including a black one) with burners of various ages, be it Teac, Philips, Plextor, Yamaha, Ricoh or Sony, sound obviously worse than the originals: colder, harsher and obviously less detailed - subjectively less transparent and less musical. This is true also for Premium - made low density recordings (using Gigarec set at 0.8, 07 or 0.6), and I suspect for Yamaha F1's similar function too (hope to find out these days);
- the Philips writes better then the Plextor: the sound is warmer, livelier, with apparently better frequency extention both upwards and downwards, losing less of the punch, dynamics and color of the original CD. Also, it's more coherent - it's easier to hear the relations between the melodic lines carried by various instruments and to perceive this way a unitary musical panorama. On the positive side, the Plextor copies seem to have slighty tigher bass (but also clearly diminished in quantity). With one word, the Plextor copies are lacklustre. This seems to be the characteristic sound of the Plextor Premium, no matter if it's reading or recording (at any speed) and regardless of the version of Plextor's firmware or of Plextools;
- EAC ripping produces slightly better sounding copies than Plextools ripping (using the Plextor in both cases and with both programs set for best accuracy) in the same way I described the writing with Philips, but the difference EAC-Plextools is small. This is true even when no errors are reported and even when it is never required a set of supplemental readings of a supposedly incorrect read sector. Why is it so is a total mystery to me!
- EAC ripping is almost always slightly safer (better correction of errors) with Philips than with Plextor, and the sound is slightly better (in the same way as writing, but the difference is about 2-3 times smaller) even when "no errors" is reported with both drives and even when it is never required a set of supplemental readings of a supposedly incorrect read sector. The C2 error retrieval of the Premium is flawed - and this was also confirmed in some reviews I'm sure you are aware of. Also, the Philips is better at reading damaged data CDs;
- the Philips made copies have less irregular pattern of the jitter graphic and worse beta (it starts to decrease slightly after min. 40 or 50, while it's said it should be a straight horizontal line). Premium copies generally seem to have the best beta and perhaps the most irregular jitter of all the burner brands. Especially after upgrading from firmware 1.00 to 1.02 (this applies also to 1.03 and 1.04) I noticed the jitter graph looks rather weird, with some great amplitude, sharp dips every now and then. I saw this pattern as well in the graphs that appeared this month in a post about Varirec, so I suppose my made in China drive is not defective. I've been told that a German magazine said the 1.01 firmware produced too much jitter because of burning slightly too long pits, but now it's v1.04 and I couldn't hear a change (at least not a significant one)! And don't tell me I should try to change the media, the TY made Plextor I received with the drive measured clearly worse than my usual Ritek made 80 min. 48x TDK Metallic and 74 min. 24x TDK Reflex Ultra (both of which, by the way, measure notably better than some Mitsubishi Chemical made Verbatim Datalife I tried);
- I only tried Varirec +1 and -1. +1 sounds slightly worse (less detailed, etc.), while -1 sounds identical or almost identical to 0;
- I was very excited when I found out about the Gigarec 0.8 - 0.6 options. On short, yes, they improve the sound the more the lower a density you choose: greater detail, frequency extention, space and dynamics, bass is more present and a lot tighter (on 0.6 setting it's tighter than on Philips copies by a serious margin). Also, the beta is even better than on normal Plextor copies. It "sings" more and invites you more to move into the rythm. But most of the characteristics of the "Premium sound" are maintained: it's still just as cold and "decolored", only with more weight on the bottom end and with better detail - but not clearly better detail than Philips. Overall, Philips is still preferable...
- generally, I couldn't hear a difference between a low quality and a better quality blank, after burning them in identical conditions they sounded the same to my ears. However, I didn't repeat the experiment, so I'm not 100% sure. But after upgrading the firmware from 1.03 to 1.04 and moving from a cheaper CD to a TDK Metallic, at least the bass improved a bit, both quantitatively and qualitatively. I don't know if it's because of the firmware or the media;
- finally, I compared two CD-Rs writen from the same image with Philips with Nero, one at 4x and the other at 24x. The two CD-Rs were the same brand and bought at the same time. To my surprise, the 24x copy sounds slightly better (more dynamic and detailed) than the 4x one. This is even more surprising to me as I tested both discs with the Plextools and found the 24x written one to have obviously worse beta and seemingly worse (much more irregular pattern of the graphic) jitter than the 4x one. So beta doesn't seem to affect directly the sound quality (the better sounding Philips copies also have slightly worse beta than Plextor ones), probably it affects the durability of the recording and the compatibility with various players. Jitter, on the other hand, SHOULD affect the sound quality, so it probably shows more the difficulties the Plextor has with reading the disc than the actual jitter of the recording?! Maybe the 24x written disc was better built form the beginning than the 4x written one??! Yes, I know, 40x discs or so could theoretically be more suitable for writting at high speed, but I've always found 1x written copies (on high speed CD-Rs) to sound better than higher speed recordings like 4x or 8x, and Plextools measurements, if they really have any meaning, seem to confirm that 4x > 8x >or= 16x > 24x.
Honestly, I expected much more from Plextor. Apart from the supposed durability and the Q-Check functions, what does it do better than the less-than-half-priced Philips?