A1/A4 drives and 8X scanning

This has baffled me ever since A1 appeared end 2006 :doh:, especially the apparent discrepancy between individual drives and even scans on the same drive. On the forum, I see examples of both types of 8X scans, those starting with abnormal error counts and those consistent with 4X scans. Things seem to have improved with A4 (and also newer A1) drives, but the issue is still apparent, even if to a lesser extent.

When I got my current 18A1H in January 2007 (after having dismissed two “raucous” 18A1P’s :rolleyes:), I was initially pretty happy with it being quieter and smoother than those first two units. Over time however I still found it to be an unrefined drive, although this is something very difficult to pin down. At times it’s been very noisy and resonant, in recent months however it seems to be quieter again. Just for the record, I don’t think I’ve burned more than 50 DVD and done more than 100 scans with it, although it’s been flashed and crossflashed dozens of times, had its suspension stiffened with washers (removed again) and lens cleaned a couple of times. My initial satisfaction with the drive was paired also with some nice 8X speed scans it returned on the first burns, but very soon (and that’s in a matter of weeks) disappointment set in with bad 8X scans and poor mechanical refinement.

That’s a bit of background. What I’ve done now is take out some of those very first burns (which have been sat in the original spindle all this time) and re-scanned them in the same configuration (8X speed, firmware and CDSpeed version). Now what can make a brand new drive return those perfect 8X scans and make it lose that ability very soon afterwards :confused:? My EEPROM backup dates from the time of the “good” scans and has been restored on several occasions, but never giving back the drive its clean 8X scanning ability. This is so frustrating :bigsmile:.

Scans coming up :slight_smile: (original scan first, re-scan second):

Verbatim DataLifePlus 2004 by CMC. Not too bad, but still significantly more PIF at the start of the scan. Looks like there could be some deterioration with those spikes at the edge :eek:.


I would trust the 8x scan for this example.
LiteOn 4x scans are very forgiving. I’m sure a BenQ scan would show those PIF spikes too.

If you look at the write speed graph, it fits with the 8x scan.

Verbatim by Taiyo Yuden 2005. A clearer example of the difference here.


Verbatim by Taiyo Yuden 2005, after crossflash to 20A1H. Another clear example of the scanning discrepancy. Getting more PIE at the edge as well, deterioration?


That’s it for now. If anyone’s got a grip on this, please let me know. :flower:

I’ve decided to always scan @4X with my 20A4P and 20A1P before it

It is very strange to see that with TYG03 (for me) the huge PIFs in the beginning of a scan are always there , but it also can show spikes and it needs worming the drive many times first to get rid of spikes but clumbs are always there :bigsmile:

MCC004 doesn’t show the PIF mountain but it clearsly shows higher PIF count also at the beginning

Benq 1640 @8X agrees with LiteOn @4X so I always use 4X with my 20A4P

[QUOTE=Cressida;2033388]That’s it for now. If anyone’s got a grip on this, please let me know. :flower:[/QUOTE]

Thanks for the interesting thread Cressida, as you know I’ve also experienced some strangeness scanning at 8x (with my 20A1H). :slight_smile:

[QUOTE=Womi;2033380]I would trust the 8x scan for this example.
LiteOn 4x scans are very forgiving. I’m sure a BenQ scan would show those PIF spikes too.

If you look at the write speed graph, it fits with the 8x scan.[/QUOTE]A Benq scans @ 8ecc of course it’s gonna show spikes, but an 8PIF spike on a Litey and Benq are 2 totally different things. My 20A1P is a marginal scanner in my opinion just like my 1620s, nothing to do with being too forgiving.

Here’s my 3 cents on the issue along with a 6s scan. The last scan at 6x seems to fall in between the 8x and 4x results.





[QUOTE=chaosoffar2k;2033476]A Benq scans @ 8ecc of course it’s gonna show spikes, but an 8PIF spike on a Litey and Benq are 2 totally different things. [/QUOTE]

I know :), I meant higher PIFs on those spots just like the litey 8x scan.

Thanks all :flower:, here’s one more I found. Nothing dramatic, but still several hundreds of PIF more in the first 0.5 GB.

I know I should move on and just scan @4X, but every time I see an 8X scan of an A1/A4 I’m reminded of this (whatever it is). My best guess is that it has to be something mechanical, and I’m ready to believe it’s down to the type of grease used on the laser-sled rails :bigsmile:. The A4 in my new EZ-DUB (December 2007) isn’t clearing things up for me either, putting the shakes through my desk. Doesn’t bother me much as 1. I don’t have my hopes up for A1/A4 anymore and 2. I have plenty of better LiteOn’s to take its place in that magnificent enclosure :).



First burn with the new EZ-DUB, all stock, on Panasonic TYG03.

I’ll have to withdraw my previous statement about the newer drives having improved 8X scanning :bigsmile:.



Hello !

:wink:

I’ll probably say a (very :D) stupid thing, but i have an idea :wink: that can fit with these strange results.

At least it has worked for me.

When my 20A1H and my external aluminium enclosure arrived to me 2 months ago, the first impression was"very nice, but too much vibrations !".

The thing was probably emphatized by the fact i had not the unit in a comuter case but in a little aluminium box.

However, first i tried some slow burning and scans, and i was not satisfied. I expected some result like the one in your review, but i had worst result.

I only used MMMC004 verbatim for these test, and these supports worked great with my old Nec 2500.

The first thing i tried was to reduce the vibration with the brown paper rings commonly used to secure HD : with one ring for each monting screw vibrations had been reduced, burns and scans improved (at that time i only made 8x scans, but since only recently i stared saving them i can not post these early results).

With 2 paper rings for each screw things improved greatly, but was hard to make them fit in the right place.

At the end i tried to reduce again vibration with a single 1cmx5cmx2mm neoprene adhesive strip below the unit. In that case things were getting worse, probably because it was too rigid and all che chassis was too tigh.

I think there was resonance instead of attenuation, since neoprene was too compressed.

My scans before the paper rings were like the one that Cressida posted.

As I told you before, I never saved them, since i tought they were bad result and i wanted to improve things.

In the next days i’ll try to use some softer element (hard disk sponges) below the unit to improve again vibration reduction, but in the meanwhile i’m quite satisfied.

Hope this can help in finding a good solution for you :flower: and for all of them :D.

:wink:

Gyxx

P.S. i enclose a 4x and 8x scan of a TY03 16x Verbatim DVD+R burned at 12x with the same unit, new :bow:C0deKing LLOD firmware, all on.

I don’t know if the 8x scans were good or not, you are the real experts, for me they are quite fine ;).



Hello Gyxx,

That’s a fresh approach, for you to attack the 8X scanning problem with external damping measures. It made me think about whether I had my A1 drive mounted internally (inside the PC tower) for those first scans (probably, but I’m not 100% sure now). I mostly use my PATA drives in external enclosures (something to do with Linux, and convenience), which will expose vibration more easily like you said.

When A1 drives were new, end of 2006, we had several threads about noise/vibration and we even had some examples of modifying the drives internally. The link with 8X scanning is only now emerging however.

But, even if the damping measures you used have improved 8X scanning, I think we must consider the following points:

  1. There is a significant difference between individual drives, as far as their levels of noise/vibration are concerned. As mentioned in my opening post, I didn’t keep my first two A1 drives, because they were unacceptable to me in this respect. Meanwhile I have acquired two A4 drives (more by accident than for actually wanting them) and one of them is good, the other is bad again in this respect. This does indeed translate into good and bad 8X scans! I’m attaching 8X scans of the same disc I posted in #12, now scanned in my second (AOpen labeled) A4 and A1 drives. All drives are mounted in an external enclosure.

  2. Most of the “good” 8X scans that I’ve seen so far on the forum have been with very good media and you can understand that the less PIE/PIF-errors there “are” on the disc, the less chance there will be to “exaggerate” them. I now seriously doubt there’s any A1/A4 drive out there, which will consistently turn out “good” 8X scans with all media and burns (“good” in this case meaning “consistent with a 4X scan”).

  3. A 4X Disc Quality scan starts a little below 4.0X speed, whereas an 8X scan starts a little below 3.5X speed. It is my feeling that in this interval (between 3.5X and 4.0X) the spindle drive doesn’t run as “round” as at other (higher) speeds, contributing to the reported errors on the disc. Now as the scanning speed gradually increases, we can see the error levels gradually drop to a normal level (as detected at other scanning speeds) as well. This doesn’t explain to me however, why my first scans with the new drive were significantly better than later and current scans.

So as I definitely accept your explanation, I don’t believe that reducing vibration can ever fully eliminate the problem these drives have when scanning at a speed between 3.5 and 4.0. As far as I can tell at this point, damping vibrations will alleviate the symptoms, but can not eliminate the root cause of the problem.

Thank you very much for your idea, Gyxx!

P.S. What I’ve also noticed is that the “good” drives will report lower Jitter in general throughout the scans (my Jitter scan speed is set at 4X for all scans with C0deKing’s registry patch tool).



I completly agree with you Cressida :wink: , but you know that if we write here we like to play with our optical units :smiley: !

In this way we can arrange to do some more experimet !

:smiley:

Ok, I’m just joking, I’m very happy to hear that I may have told a good idea to this community.

It’s a long time I’ve been reading discussion here at cdfreaks, only recently i decided to regiter, and I’ve found a lot of solutions thanks to all of you in this time ;).

And many times thanks to your posts.

By the way, regarding the grease used in the rails of optical untis, i think is pure vaseline. I use it to replace the original grease after the periodic “cleaning sessions” of my units, all the times it had worked vey well.

I put it on the rails with the ease of a cotton fioc : just one drop for a single rail.

In my little experience, the lowest the grease level, the better the results.

Best wishes

;).

Gyxx

Thank you, Gyxx :flower:. For the rails, I like to use a little bottle of silicone oil for bicycle chains that I have :p.

Another thing that confuses me about the vibration theory is my 16A7S, which is easily one of my quietest and smoothest drives, yet at the same time probably the worst 8X scanner. This model uses the same hardware (also the Sanyo DS10HLF OPU) as the 20A1S.

As an example, the same Panasonic disc from posts #12 & #14 above:



And just to show that it’s nothing to do with this drive’s old firmware, here it is again, masquerading as a 20A1S (strange spike as well at 0.5GB with 8X). It’s really a weird phenomenon, it doesn’t even recover when the speed goes over 4X.



Cressida, I’m start thinking that all i have supposed is a fake :frowning: .

I’ve tried to use some soter material to remove more vibration, but since i dismounted my unit and tried all the combination all the scan were worst.

The softer the material, the worse the scan.

Now I’m trying to return to the condition of the two scans above, but is difficult.

I have doubled the PIF errors (same disk to scan) on both speeds, even returning at the working configuration that has produced the results above.

Now i have reassembled the unit without any vibration stopper, and things seems better but still far from what i first posted in this discussion.

Probably there’s some other factor that comes to play, still related to assemply of the unit in the case.

Since my first play has worked i supposed that vibration stopping was the key, now scientific analysis of new result seems to say i had missed.

I enclose some of the scan i made, first (8x) was produced with a soft sopnge below the unit and with the aid of the paper rounds, second one (still 8x) was made after mounting the unit with parafilm spacers ( for details on parafilm see here : http://en.wikipedia.org/wiki/Parafilm ), third one was 8x scan of today’s assembling with paper rounds, fourth one has been made right now with the naked unit (no spacer), fifth same as fourth but made at 4x.






Yeah, it’s crazy Gyxx. The differences between your 8X scans aren’t very big though. A couple hundred PIF is within the tolerances of the scanning setup/method we use. A tiny dust particle or scratch on the disc can add an even larger amount. Your sample disc is a very good burn however, so the difference between 4X and 8X isn’t dramatic in any event.

I think so far we can say that:

  1. It’s not caused by the OPU (A1 and A4 use different OPU’s and both suffer from the same problem). :slight_smile:

  2. It’s not caused by the firmware (unless this has been deliberately sabotaged maybe?). :stuck_out_tongue:

  3. It may be a chipset bug? Are there any Samsung or Asus drives with the same MediaTek MT1898/1899 chipset that don’t have this problem? :confused:

  4. It may be a hardware issue? Vibration/resonance and/or component choice? I don’t feel like opening up my two A4 drives yet to compare them, maybe later on. :bigsmile:

Meanwhile I’ve installed the 20A1H in my PC tower again, exactly the same way as when it was new, and still can’t get the good 8X scans back. Example with same disc as used in post #5 (PIF have gone down by ~200 though, similar to your tests :)):


[QUOTE=Cressida;2036448]Yeah, it’s crazy Gyxx. The differences between your 8X scans aren’t very big though. A couple hundred PIF is within the tolerances of the scanning setup/method we use. A tiny dust particle or scratch on the disc can add an even larger amount. Your sample disc is a very good burn however, so the difference between 4X and 8X isn’t dramatic in any event.

I think so far we can say that:

  1. It’s not caused by the OPU (A1 and A4 use different OPU’s and both suffer from the same problem). :slight_smile:

  2. It’s not caused by the firmware (unless this has been deliberately sabotaged maybe?). :stuck_out_tongue:

  3. It may be a chipset bug? Are there any Samsung or Asus drives with the same MediaTek MT1898/1899 chipset that don’t have this problem? :confused:

  4. It may be a hardware issue? Vibration/resonance and/or component choice? I don’t feel like opening up my two A4 drives yet to compare them, maybe later on. :bigsmile:

Meanwhile I’ve installed the 20A1H in my PC tower again, exactly the same way as when it was new, and still can’t get the good 8X scans back. Example with same disc as used in post #5 (PIF have gone down by ~200 though, similar to your tests :)):[/QUOTE]
If it was the firmware it should be all drives, but obviously some people’s drives seem not to be affected. I agree with your earlier statement about higher jitter readers having the problems with 8x scanning. The drives that scan well are spitting out 7 or 8 average jitter frequently, whereas drives that can’t scan well at 8x seem to almost always scan jitter over 9 average.

As far as other drives my 182D is best at 4x(still higher errors than other scanners though) while my 203B only provides scans close to my other scanners at 18x, all other speeds show much higher errors. If you look through the 203B scan threads you’ll see that they’re a mixed bag scanner wise, some on par with other scanners but alot significantly pickier or less picky then other scanners at 8x. Not sure if there are other drives with the same chipset.