Code Guys' Burn-Adjusted Firmwares

vbimport

#1

<HR>

Thread closed

Prepatched “burn-adjusted” firmwares are no longer available because an autopatcher is now available to easily make create them. Please use our new [thread=97590]OmniPatcher[/thread] tool to create your own custom “burn-adjusted” firmwares. :)<HR>

Burn-Adjusted Firmwares

C0deKing has been hard at work for the past week or so working on improvements to the burning process of 851S and 812S drives. During this time, there have been peeks at what was being worked on in the forms of the Ricoh swap and shift-fix firmwares that have been put out. We are now releasing all of this in a package that I have dubbed “burn-adjusting.” The final product is a firmware that burns smoother and faster at 8x! :slight_smile:

So what exactly is a burn-adjusted firmware?
1/ For some (not all) +R (does not affect -R) media types on some (not all) burners, the burner will sometimes fail to shift up from 4x to 6x at the proper shift point. After this upshift fails, the 4x region is extended. In many cases, this failed upshift affects laser calibration adversely, and results in higher errors in the extended 4x region. This has been dealt with by fixing the problem in the firmware that was causing this failed upshift. The result is a faster burn and lower errors.
2/ For an improved +R (does not affect -R) burning speed, the 6x region has been shortened. The upshift to 8x now occurs at an earlier point. The result is a faster burn.
3/ For RICOHJPNR01 media, the fallback to 4x at the end of the burn has been fixed. This fallback has been known to cause problems in the form of high error rates, so we did away with it by fixing the writing strategy. And of course, this media type has also been bumped up to 8x.
4/ For the few other +R/-R media types that also suffer the fallback at the end of a burn, the fallback has now been changed to 6x (instead of a more dramatic 4x fallback). It has been found that a 6x fallback at the end is not as harmful as a 4x fallback. This fallback only affects a few media types, and as noted above, RICOHJPNR01 (a special case) will no longer suffer a fallback at all.
5/ The Ricoh R01/R02 strategy swap has been abandoned. First of all, the quality improvement was negligible. Most of the quality improvement came from the removal of the high errors caused by the various shifting problems. It was decided to keep using the R01 strategy because we have started to see a few RICOHJPNR01 discs that have problems with the R02 strategy. So instead of replacing the R01 strategy with the R02 strategy, we decided to just fix up the shifting problems in the R01 strategy instead. And besides, the Ricoh swap was just for Ricoh media. Fixing the shifting problem in general affects many more media types!

What are the side-effects of this?
At the moment, we don’t know of any negative side effects. In our testing, this firmware has results in both better error rates and faster burn times. Of course, we won’t mind seeing how this firmware performs on other drives (since C0deKing and I represent just a small sample set), so please, post your results! :wink:

CD/DVD Speed and KProbe Scans
Attached at the end of this post are before and after pictures to illustrate what burn-adjusting does. Notice the removed fallback at the end, the absence of a failed 4x shift, and the shortened 6x region. In addition, I also attached a couple of KProbe scans of 8x burns done on a crossflashed 832S using the new burn-adjusted firmware and a firmware without burn-adjusting (using, instead, the Ricoh swap). The error levels seen in these two scan are nearly identical, and are consistent with the error rates that I typically get with discs from this particular batch of discs. This is to show that getting rid of the swap and using more aggressive shifting does not seem to have any undesirable side effects in our testing.

What firmwares have been patched? Will there be an autopatcher?
So far, all the patching has been manually done. We have patched the GS0H (851S), US0J (812S), and VS01 (832S) firmwares. Unfortunately, we have not had a chance to patch the 811S drive yet. The reason is that the 811S firmware isn’t as similar, and we also lack a _11S drive that we can use for testing. There currently does not exist an auto-patching tool for all this.

What else (aside from burn-adjusting) has been changed in this firmware?
1/ (Optional) Autobitsetting has been applied to the firmware.
2/ (851S version only) Reading speed of recordable/rewritable discs has been increased to 8x.
3/ (851S & 832S versions only) Crossflashing has been enabled.

<HR>Latest version is Release #2A
Download from: http://codeguys.rpc1.org/firmwares.html<HR>

Changelog
• Release #2A - Autobitsetting is now optional (2 different firmwares per package, one with and one without ABS), and name-change from burn-fix to burn-adjust (the firmware themselves are unchanged)
• Release #2 - Fixes 6x->8x upshift problem on some drives (first reported by pchilson)
• Release #1 - Initial release (not labeled as “Release #N”)

Enjoy! :slight_smile:


#2

I don’t think we tell you guys this enough but…

YOU ROCK!!! :bow: :bow: :bow:

I will start burning some tonight and post my results!!

Thanks again!!


#3

WoW! That’s amazing! Can’t wait for my 812S next week, so I can try fixed firmware. :slight_smile:

Maybe some hacking too… :wink:


#4

Very Nice.


#5

please, post your results!

US0J_burn_fix fw
Maxell 1x-4x 15piece cakebox

00000000 08 02 00 00 01 0F 01 00 00 03 00 00 00 26 05 3F …&.?
00000010 00 00 00 00 00 00 01 52 49 43 4F 48 4A 50 4E 52 …RICOHJPNR
00000020 30 31 02 38 23 54 37 09 00 3C 67 00 AC 62 16 18 01.8#T7…
<g…b…
00000030 0B 0B 0A 0B 01 19 1B 0B 0B 0E 0F 01 00 00 00 00 …

This drive never had an upshift problem before. hmmm



#6

@pchilson

Oh my. Hmm. We didn’t have that happen with our tests, but I guess this is why we’re having people test it so that we have a larger sample set. :wink:

I guess there’s a bit more that needs to be done in the way of adjustments. We’ll get back to you in a few hours. :slight_smile:

Out of curiosity, how many other people are encountering similar problems?


#7

What are the affects of bypassing the 4x area completely?


#8

As I’ve mentioned before, great work c0deking and code65536 :wink:


#9

We’ve considered that. But we haven’t tried it because it’ll require the motor to spin pretty fast, and we were not quite ready to take the risk of the motor spinning above design specs. As you may notice from the angular velocity graph that I posted, we shortened the 6x region to the point where the angular velocity matched that of the 4x->6x jump (which, BTW, if you look at OC’s reviews, is the official 4x->6x jump point; we didn’t touch that jump point; we just prevented a failed upshift at that jump point) so that the angular speed with our firmware will never exceed the angular speed acheived with an official firmware.


#10

Are there any plans in the works for 811s I hope there are I need all the help I can get.

After futher reading of the first post I got my answer somewhat. If you guys need a test drive I’ll test on mine if you guys can fix up the firmware the way you like.

Thanks
Shawn


#11

Hi Guys,
Can’t try it as my 812S died this morning (bios can’t see the drive) :sad:
Good job I have a 411S as a backup :bigsmile:
dare2001


#12

I can confirm the same thing on my 812@832.
It fails to shift from 6x to 8x then it does shift later in the burn. Same media.


#13

451S@851S via the 851S burn-fixed firmware. This is a RicohJPNR01 4x rated DVD+R burned at 8x. :cool:

I have not tested this on stand-alones yet to see if it works. Will be doing that shortly. At 8x it skipped like I normally get from this drive. Just burned one at 6x and attached the kprobe results. I have to run to the store but will test after that.




#14

Firstly, I’d like to thank the codeguys for the work they’ve been doing

451s@851s burn-fixed f/w. The result is similar to the one posted by Braxas (at 8x). Looks like the drive doesn’t do well in the initial phase, but after shifting to 6x everything is perfect. My scan doesn’t show the spikes at the end, but this is not a full DVD - 3.7 GB movie. Media branded as Maxell 4x, burnt at 8x.

00000000 08 02 00 00 01 0F 01 00 00 03 00 00 00 20 86 3F ............. .?
00000010 00 00 00 00 00 00 01 52 49 43 4F 48 4A 50 4E 52 .......RICOHJPNR
00000020 30 31 02 38 23 54 37 09 00 3C 67 00 AC 62 16 18 01.8#T7..<g..b..
00000030 0B 0B 0A 0B 01 19 1B 0B 0B 0E 0F 01 00 00 00 00 ................

This is much better than I used to get with the shift-fixed f/w. And the original f/w… uhm… since joining cdfreaks I forgot what it is ;-))) Seriously, I had the same high PIs in the beginning with the original speedhacked GSOH.


#15

LITEON 812 @ 832 - Retail
Production date: March 2004
FW: VS01-CodeBurnFix-ROM-R1
RecordNow Max v4.5 - (PX Engine=507)

[i]Notes:

  • EEPROM ZMOD
  • case temp: 25.1C/77.18F (sensor at hottest corner of case)[/i]

*************************************
*** Memorex x4 +RW - (Ricoh W11) @ 4x ***
*************************************

[i]Notes:

  • over 30 R/W cycles on these RW disks
  • full-erase before RW tests[/i]

DISK 1 - (4.2gig data)
PI Max: 17 Average: 4.220
PO Max: 9 Average: 0.175
http://premium.uploadit.org/bichi007/MemxW11-812832-VS01-CodeBF-ROM-001.png

DISK 2 - (4.3gig data)
PI Max: 18 Average: 4.001
PO Max: 9 Average: 0.245
http://premium.uploadit.org/bichi007/MemxW11-812832-VS01-CodeBF-ROM-002.png

Previous, burned with 812@832 VS01, R01/R02, bitset - (4.2gig DVD)
PI Max: 19 Average: 4.313
PO Max: 9 Average: 0.189
http://premium.uploadit.org/bichi007/MemxW11-812832-VS01-R01R02-ROM-001.png

Previous, burned with 812@832 VY02, PSAROM, bitset - (4.1gig data)
PI Max: 24 Average: 4.461
PO Max: 8 Average: 0.196
http://premium.uploadit.org/bichi007/MemxW11-812-DRU700-VY02-PSAROM-001.png

Previous, burned with 812 US0J Rswapped, bitset - (4.3gig)
PI Max: 36 Average: 6.873
PO Max: 15 Average: 0.797
http://www.uploadit.org/bichi007/Memx4-W11-812-US0J-Code-002.png

Comments:
Disk 1

  • Excellent burn!

Disk 2

  • Still doing good!

Conclusion:

  • just as good as US0J, DRU700 VY02, and VS01-R01/R02, bitset
  • “red” burn LED, as opposed to “green” with DRU700 FW

**************************************************
*** Memorex x4 +R R01 Printable - (Ricoh R01-02) @ 8x ***
**************************************************

Disk 1 - (4.2gig DVD)
PI Max: 11 Average: 0.435
PO Max: 10 Average: 0.017
http://premium.uploadit.org/bichi007/MemxR01-812832-VS01-CodeBF-ROM-001.png

Disk 1 - (4.2gig DVD) - KP 2.1.0
PI Max: 13 Average: 0.46
PO Max: 9 Average: 0.00
http://premium.uploadit.org/bichi007/MemxR01-812832-VS01-CodeBF-ROM-001-KP2.png

Nero v3.0 Scan Test - Disk 1
PI Max: 11 Average: 1.07
PO Max: 9 Average: 0.02
http://premium.uploadit.org/bichi007/MemxR01-812832-VS01-CodeBF-ROM-001-Nero30-scan.png

Nero v3.0 Speed Test - Disk 1
http://premium.uploadit.org/bichi007/MemxR01-812832-VS01-CodeBF-ROM-001-Nero30-xFer.png

Disk 2 - (4.2gig Data) - Maxell +R x4 (R01-02)
PI Max: 53 Average: 8.731
PO Max: 13 Average: 0.464
http://premium.uploadit.org/bichi007/MemxR01-812832-VS01-CodeBF-ROM-002-Maxell.png

Disk 3 - (4.2gig Data) - Maxell +R x4 (R01-02)
PI Max: 28 Average: 2.509
PO Max: 10 Average: 0.124
http://premium.uploadit.org/bichi007/MemxR01-812832-VS01-CodeBF-ROM-003-Maxell.png

Disk 4 - (4.0gig Data)
PI Max: 15 Average: 0.814
PO Max: 9 Average: 0.048
http://premium.uploadit.org/bichi007/MemxR01-812832-VS01-CodeBF-ROM-004.png

Disk 5 - (4.1gig Data) - Memorex 25pak/R01-02-(shiny-black, as opposed to dull-black bottom)
PI Max: 12 Average: 0.573
PO Max: 9 Average: 0.018
http://premium.uploadit.org/bichi007/MemxR01-812832-VS01-CodeBF-ROM-005.png

Nero v3.0 Quality Test - Disk 5
PI Max: 12 Average: 1.28
PO Max: 10 Average: 0.02
http://premium.uploadit.org/bichi007/MemxR01-812832-VS01-CodeBF-ROM-005-Nero300.png

Previous, burned with 812@832 VS01, R01/R02, bitset - (4.2gig DVD)
PI Max: 10 Average: 0.396
PO Max: 6 Average: 0.020
http://premium.uploadit.org/bichi007/MemxR01-812832-VS01-R01R02-ROM-001.png

Previous, burned with 812@832 VY02, PSAROM, bitset
PI Max: 9 Average: 0.409
PO Max: 5 Average: 0.017
http://premium.uploadit.org/bichi007/MemxR01-812-DRU700-VY02-PSAROM-001.png

Previous, burned with 812 US0J Rswapped, bitset
PI Max: 10 Average: 0.388
PO Max: 4 Average: 0.020
http://www.uploadit.org/bichi007/Memx4-R01-812-US0J-Code-001.png

Comments:
Disk 1

  • RecordNow Max starts at 12:35, then ~10secs, settled to 7:23 burn time
  • one more speed shift at about 1mByte, then settled at 6:16 burn time
  • one more speed shift at about 1.7mByte, then settled at 3.22 burn time
  • Excellent burn!
  • whats with 72% Quality in Nero v3.0 CD Quality test?

Disk 2

  • Maxell +R x4 (R01)
  • appears to be slightly less quality than Mem Printable R01

Disk 3

  • Maxell +R x4 (R01)
  • better than Disk 2
  • next disk, back to Mem +R x4 Printable (R01)

Disk 4

  • typical Memorex +R x4 excellent burn…

Disk 5

  • Memorex 25pak/R01-02 - (shiny-black, as opposed to dull-black bottom)
  • on sale, Fry’s 25pak, $24.99 - May 23, 2004
  • good burn!

Conclusion:

  • just as good as US0J, DRU700 VY02, and VS01-R01/R02, bitset
  • Maxell +R x4, (R01 mc-02), same as Memorex +R x4 Printable, burns slighly differently.
  • “red” burn LED, as opposed to “green” with DRU700 FW
  • might have to try Code’s BurnFix R2 to see effect on Maxell +R x4 media

#16

@all / @pchilson

Release #2 is available now. This should fix the upshift problem that you brought up. The changes made in release #2 should effect only the upshift problem. If you didn’t encounter this upshift problem in the original release, then your results with release #2 should be identical.

@VAD

Interesting that you got a high error range in the (very short) 4x region. This was something that went away for us with the Ricoh swap, and seemed to stay away after we returned to the R01 strategy. Hmm… the thing that’s keeping us away from going back to using the R02 strategy is that both C0deKing and I got some new RICOHJPNR01 discs recently, and for both of us, these “new” R01 discs seem to be faring much worse with the R02 strategy than with the R01 strategy, which is why we switched things back to the R01.

@bichi

Nice results! :slight_smile:

@all

Anyone testing this on something other than Ricoh media? :wink: :stuck_out_tongue:


#17

So far on RitekJPN01’s my results have been good. I’ll drop in a CMC Magnetics F01 here shortly to test with. So far knocking on wood it appears to be playing in my stand-alone burners very well as long as I keep the burn speed to 6x. Haven’t tried at 4x as of yet.

I have a question. Does this firmware adjust the laser write strategy of the drives at all? It seems to be much improved over my standard 451S Firmware (GSB7 with auto-bitset). This firmware has led me to the decision of keeping this drive in the computer as primary slave when my HP 420i goes in for Primary Master. Very nice firmware and a great job done by the codeguys indeed. :slight_smile:

I had my worst result so far at the very beginning of a 6x burn with the RicohJPNR01 discs but can’t seem to be able to upload attachments for some reason at this point in time. :Z I don’t know what’s up with that but basically had a high 160ish spike at the beginning and then went back into a normal burn. Looked good at the end and it should be a very passable disc.


#18

Nice job once again, Codeguys! This is the best results I’ve had on the first time out with a CMC Magnetics F01 (TDK 4x DVD+R) disc in this 451S running on any firmware revision out there. You guys should definitely go and teach Lite-On a thing or two about making firmware for their drives for compatibility and such.

Here is the TDK 4x DVD+R burned at 4x. It defaulted to 2.4x but burned fine at 4x from what I can tell by Kprobe. I’m anxious to test it out in the pickiest of the stand-alones I have.

The Ricohjpn-03.png attachment is the one I was talking about the high spike in the very beginning. The only problem viewing wise on this one so far was a couple stutters in the very beginning of the movie. Otherwise it has been flawless.

So far so good viewing the CMC Magnetics disc. No stutters to be seen. :slight_smile:




#19

scratches his head Well, I just finished up all of my Ritek R01’s and started a 50 pack spindle of Ritek R03’s. Kprobe scans are through the roof. Not sure what to think. Here is the info on the media with Kprobe results. Hopefully the Codeguys can come up with something. This is with Release 2 of the burn-fixed firmware. They are 4x DVD+R’s burned at 4x. Burned at 6x the results were pretty similar.

I just noticed that the media code speed only lists 8x… Hrm! Interesting. hehe

                        Media Information                            

Disc Regions are 1,2,3,4,5,6,7,8
Media code/Manufacturer ID RITEK R03
Media Product Revision Number 02h
Volume Name KILL_BILL_VOL_1
Application id
Implementation id AHEAD Nero
Recording Date/Time (mm/dd/yyyy) 5/22/2004 7:16:53
Format Capacity 3.37GB(3.62GB)
Book Type DVD-ROM
Media Type DVD+R
Media Id Code Speed 8.0x 11080KBps
Data area starting sector 30000h
Data area end sector 1AFE8Fh
Linear Density 0.267um/bit
Track Density 0.74um/track
Number of Layers 1



#20

Give them a run at 8X burn speed! :cool:

This is 8X rated media, to my knowledge.

Edit; I see you just noticed this…