Philips DVP642 / DVP630 - The three firmware files explained

[Jefferson Ryan]

Hi,

We were wondering for the last weeks why some guys can´t do a downgrade to the 0531 version and why the 1109 firmware has three files.

It´s because there are at least three different hardware configs of the DVPxx in the market.

In the firmware screen you can see a code like this:

http://ryan.com.br/imagens/dvp642/dvp642_firmware.jpg

Ver0531 642/78 01 M 4

The letter "M" is the hardware config. I counted three configs in the wild: "M", "S" and "E".

The "M" uses only the file dvpxx.rom (tested - I have one unit)

The "S" uses only the file dvpxxDV.rom (tested - I have one unit)

The "E" requires the file dvpxxEW.rom , not present in the 0531 firmware (not tested, but seems obvious)

So, you can´t downgrade because you do not have the file for your hardware. Maybe it was not Philips intent to block the downgrades. I think that some parts changed in the production line and so the firmware was changed acordingly. Philips puts the three files in the same ZIP to simplify the process to the users point of view. Each unit picks the right file based in information stored in the 512Bytes EEPROM (a little chip).

When the firmware loading process begins, the player displays the name of the picked file for a moment in the top left of the screen.

But what parts changed? What difference we can expect in the unit behaviour?
I don´t know for sure, but we will need to take that letter in consideration for now on when trying to understand the DVP642 quirks.

The full reasoning for this was published in the last week in my site, but in Brazilian Portuguese. My English is not enough to explain everything here, sorry.

If you understand Portuguese:
http://ryan.com.br/prod_dvp642.htm#27022005

[]'s
Jefferson Ryan
http://ryan.com.br

[fewtch]

Finally, the secret emerges! Thanks much for that info, it finally clears things up.

[Jefferson Ryan]

Just to expand:

In
Ver0531 642/78 01 M 4

"01" were allways present in all firmwares, DVP630 or DVP642. Never changed for me.

"4" Is the default region of the player. Even if you apply the Region Hack and modify the player region, the firmware screen will show the factory default. It´s stored in the EEPROM and it´s due this that every time you reflash your player you need to reapply the Region Hack.

[Jefferson Ryan]

I compared two CPU boards of each "M" and "S" versions (four players in total) and found the following:

Board "S" (http://www.ryan.com.br/prod_dvp642_cpu_dvpxxDV_comentada.htm)
4Mbit RAM:Elpida EDS6416AHTA (S6416AHTA-6B-E) (http://www.elpida.com/pdfs/E0439E50.pdf)
Servo Driver: AMTEK AM5868S (http://www.amtek-semi.com/5868.htm)
Flash:MX26LV800BTC-55 (http://www.olimex.com/dev/pdf/MX26LV800T-B-1.2.pdf)
Vibratto: ES6698F G254 and ESS ES6603S
HC374 (http://www.pci-card.com/sn74hc374n.pdf)

Board "M" (http://ryan.com.br/IMAGENS/DVP642/dvp642k_cpu_comentada.jpg)
4Mbit RAM:SAMSUNG K4S64132H-75 (641632H-TC75) (http://www.samsung.com/Products/Semiconductor/DRAM/SDRAM/SDRAMcomponent/64Mbit/K4S641632H/K4S641632H.htm)
Servo Driver: AMTEK AM5868S (http://www.amtek-semi.com/5868.htm)
Flash:MX26LV800BTC-55 (http://www.olimex.com/dev/pdf/MX26LV800T-B-1.2.pdf)
Vibratto: ES6698F F164 (or F074) and ESS ES6603S
HC374 (http://www.pci-card.com/sn74hc374n.pdf)

As you can see, the only differences are in the RAM used and in that number in the main chip. I compared all the major chips in the boards. I don´t believe that a difference in a simple chip as a RAM could cause a firmware revision, so what´s the difference between the G254 and the F074/F164 Vibratto II CPUs?

I´m also suspecting of a possible difference in the loaders, because there IS a difference in the noise in some units, but I´m not sure.

--Edit: I edited the Vibratto chips that I marked incorrectly, sorry!

[fewtch]

Originally posted by Jefferson Ryan
As you can see, the only differences are in the RAM used and in that number in the main chip. I compared all the major chips in the boards. I don´t believe that a difference in a simple chip as a RAM could cause a firmware revision, so what´s the difference between the G254 and the F074/F164 Vibratto II CPUs?
I looked at the product brief here (http://www.esstech.com/products/AVplayer/avplayer.shtm) for the ESS6698 chipset, but it doesn't go into enough detail. There's a tech support link at that URL, maybe they could explain (I didn't bother to try it).

[Jefferson Ryan]

Originally posted by fewtch
I looked at the product brief here (http://www.esstech.com/products/AVplayer/avplayer.shtm) for the ESS6698 chipset, but it doesn't go into enough detail. There's a tech support link at that URL, maybe they could explain (I didn't bother to try it).

I don´t think that they will bother to answer any question from me about this.

[fewtch]

OK, I decided to try it... here's what I sent to them:

Hello,

I own a Philips DVP642-37 DVD player containing the ESS6698-F chipset, and am trying to determine the difference between mine and others of the same model. Can you tell me the technical differences between the following revisions?:

* ES6698F G254
* ES6698F F074/F164

Is there anything significantly different between these chips that could affect the process of upgrading firmware for the Philips DVP642 player?

Email response preferred (if you have the time) -- thanks very much.
I'll post any (potential) response in this thread. Chances are probably better with them than Philips, anyway.

[SeeMoreDigital]

Thanks for posting such detailed information JR.

I've just added a link to your thread from my own Std-Def: ESS chip-set based - Player Information (http://forum.doom9.org/showthread.php?s=&threadid=82765) thread.


Cheers

[Jefferson Ryan]

Today, someone at the Philips_DVP642 Yahoo Group answered my message about this topic (that I posted there):

I discovered the difference between the xx and the xxDV versions at
least. I flashed the xxDV version to my xx player (by renaming the
file) and the firmware acted properly EXCEPT for the loader. The
loader code is different between the two, apparently. Now no discs are
readable (and the loader makes strange attempts at seeking).

So basically I killed my unit, but hopefully this will at least
enhance the understanding of this player.

As I suspected, the difference seems to be in the loader!

If this is true, the player is not really dead. It´s possible to fix this by swapping loaders with a model "S" model and reflashing.

[]´s
Jefferson Ryan
http://ryan.com.br
Philips DVP642 Brazilian Portuguese Review at
http://ryan.com.br/prod_dvp642.htm

[fewtch]

Originally posted by Jefferson Ryan
Today, someone at the Philips_DVP642 Yahoo Group answered my message about this topic (that I posted there):

I discovered the difference between the xx and the xxDV versions at
least. I flashed the xxDV version to my xx player (by renaming the
file) and the firmware acted properly EXCEPT for the loader. The
loader code is different between the two, apparently. Now no discs are
readable (and the loader makes strange attempts at seeking).

So basically I killed my unit, but hopefully this will at least
enhance the understanding of this player.
OUCH... I don't think it was worth it... :(

As I suspected, the difference seems to be in the loader!

If this is true, the player is not really dead. It´s possible to fix this by swapping loaders with a model "S" model an reflashing.
How can one go about swapping a loader (and what exactly is a loader anyway, is it a ROM chip?).

P.S. I received this irritating response from ESS today (you were right):

If you are experiencing difficulties with using your player, please
contact the original manufacturer or marketer of the hardware directly
for the assistance. ESS does not supply end-user support services to the general public. Rather, ESS works with the OEM manufacturers to ensure that they have the required software solutions for their respective products.
Regards,
ESS Tech Support


I was annoyed, and answered like this:

Hello,

I was requesting technical details on your ES6698 chipset, not help with my DVD player. I'm sorry your company doesn't employ competent people who know about your products.

Cheers,

:D

[Jefferson Ryan]

Originally posted by fewtch
[B]OUCH... I don't think it was worth it... :(


It depends on your point of view :)

I risked my own players several times doing tests for my Brazilian review of the DVP642 and trying to discover the role of the three files.
I'm not rich and here in Brazil a DVP642 costs at least US$100, if bought in the black market. In the "official" market, costs US$170

BUT... I'm very curious and the pleasure to obtain this kind of info worths the risk :)

Nevertheless, I would never rename the firmware files, because I was almost sure that would kill the player. As I said, I'm not rich :)


How can one go about swapping a loader (and what exactly is a loader anyway, is it a ROM chip?).


"Loader" is, in the dvd player jargon, the same as "drive", in the computer jargon. So, swap the loaders is a matter of loose some screws and exchange a couple of cables :)


P.S. I received this iritating response from ESS today (you were right):


Yeah. I was sure that I would receive this kind of answer or no answer at all.

Excuse me for my bad English.

[]´s
Jefferson Ryan
http://ryan.com.br
Philips DVP642 Brazilian Portuguese Review at
http://ryan.com.br/prod_dvp642.htm

[drf666]

Nevertheless, I would never rename the firmware files, because I was almost sure that would kill the player.

There something else I might try, but not sure how the player would end.

Latest FW for DVP642: 1109 (2004)
Latest FW for DVP630: 0112 (2005)

Both contain the same .ROM filenames... I don't plan on renaming anything... Only burning a DVP630 Upgrade disc and trying it in my DVP642. Is it obvious that I will screw things up trying this?

[Jefferson Ryan]

Originally posted by drf666
There something else I might try, but not sure how the player would end.

Latest FW for DVP642: 1109 (2004)
Latest FW for DVP630: 0112 (2005)

Both contain the same .ROM filenames... I don't plan on renaming anything... Only burning a DVP630 Upgrade disc and trying it in my DVP642. Is it obvious that I will screw things up trying this?

No. The DVP630 and DVP642 are the same, with some minor differences in the default setup. For example, the DVP642 default is NTSC while DVP630 default is PAL.

[]´s
Jefferson Ryan
http://ryan.com.br
Philips DVP642 Brazilian Portuguese Review at
http://ryan.com.br/prod_dvp642.htm

[BxWrapper]

Does any body know if ESS-6698 supports JTAG ?

I got a "dormant" DVP-642. I found this player is very prone to fail during firmware update especially during writing phase. Last error message I got during flashing was "FAILED".
I am very sure Philips did not design the flashing procedure robust enough. Actually this was the second time I failed flashing this model. I have never failed doing it for other player like KISS and Samsung.

If somebody can pointing me please to DVP-642 schematic or ESS-6698 datasheet, I like to revive this player. I am very familiar with JTAG and been using it with plenty devices particularly sat receiver :) like sammy, fortecstar and other receivers. The flash rom MX26LV800BTC-55 is common to be found in sat receiver. I am very sure I got it in my jKeys.def file.

TIA, Bxwrapper

[Jefferson Ryan]

Originally posted by BxWrapper
[B]Does any body know if ESS-6698 supports JTAG ?



I have NO experience with JTAG, but as I understood reading some pages, the JTAG standard requires some dedicated pins (labeled like TRST*,TCK, TDI, TDO and TMS) but the Vibratto II Product Brief shows NO such pins.

So, I guess that Vibratto II has no support to JTAG, unfortunately.

Vibratto II Product Brief:
http://www.esstech.com/products/AVplayer/Prod_Briefs/pb6698.pdf

[]´s
Jefferson Ryan
http://ryan.com.br
Philips DVP642 Brazilian Portuguese Review at
http://ryan.com.br/prod_dvp642.htm

[BxWrapper]

Thanks Jeff,

It is pity that ESS makes full chipset datasheet not public.

Then I wonder how would Philips service center re-loaded the firmware to flash ROM. I do not believe they un-solder and solder it again since it requires too much work. I have yet to see an "in-situ" clamping socket for the flash ROM. I guess, there must be some service points on the PCB for re-writing the ROM. But the question is then which point and what interface are required. Hope in the future, this information will be unsurfaced and become public knowledge.

[acidsex]

According to that pdf, it says AAC is supported but for some reason, none of movies with AAC work on the DVP642. Is there a specific fimrware that I need to enable ACC playback? Also, if so,is it only AAC-LC or does it playback HE-AAC as well?

[SeeMoreDigital]

Originally posted by acidsex
According to that pdf, it says AAC is supported but for some reason, none of movies with AAC work on the DVP642. Is there a specific fimrware that I need to enable ACC playback? Also, if so,is it only AAC-LC or does it playback HE-AAC as well? As I've mentioned over on some other forum threads... If you Philips player owners want AAC support... you could do no worse than to bombard Philips with requests for it!

Thankfully Philips had the presence of mind to include .MP4 container support (with MP3 audio) so they should be able to offer AAC audio too.... unless there's a licensing issue preventing them from doing so!

And... full/correct decoding of AAC-HE audio is not included in this chip-sets design... You will be able to play AAC-HE but sadly not the SBR element.


Cheers

[acidsex]

Yeah, I sent them an email not too long ago and they responded that they had no plans to include any AAC decoding for this player but would considerate it for future players.

Just found that pdf to be contradiciting that Philips said that AAC would not work on that chipset. Just wish I could find one of the Nero players somewhere here in the states.

[SeeMoreDigital]

Hi acidsex,

Well it's clear the chip-set does support AAC decoding... which is nice! Maybe you (and others) could ask Philips to confirm why they have no plans to include AAC decoding in their current ESS Vibratto II chip-set player line-up?

In any event, the more "official" confirmations we get, the better!


Cheers

[Jefferson Ryan]

I just tried today swaping loaders between a model "M" and a model "S". The two players don´t work with the wrong loader. The message "no disc" is always displayed after a few seconds trying to focus the disc.

After putting the loaders in the correct players, everything works as usual.

Also, the loaders are phisically different. I published a page about this (many photos) here:

http://ryan.com.br/prod_dvp642_loaders.htm

The page is in Portuguese, but I tried to make the photos self-explanatory.

And, if your disc is locked in the player after a bad upgrade, you can eject the disc following the instructions (many photos) here:

http://ryan.com.br/prod_dvp642_ejection.htm

[]´s
Jefferson Ryan
http://ryan.com.br
Philips DVP642 Brazilian Portuguese Review at
http://ryan.com.br/prod_dvp642.htm

[davidh44]

Originally posted by Jefferson Ryan
so what´s the difference between the G254 and the F074/F164 Vibratto II CPUs?

The numbers referenced above are date codes. The last number is the year of production, and the two numbers before that are the week of production.

G254 = Made in the 25th week of 2004.
F074 = Made in the 7th week of 2004

I don't know, however, whether the preceding letter has any significance.

[davidh44]

I have 2 DVP642/37 (U.S.) units, both unfortunately of the "E" variety in which there is no older (macrovision disabled) firmware available.

The production dates are:
November 2004
February 2005

I believe Oct. or Nov. '04 is when they started making the "E" hardware version of the DVP642/37 with the 1109 firmware preloaded.

Has anyone in the U.S. been able to get an "M" or "S" DVP642 with a production date in 2005?

[BluIron]

Yes I have an S unit I purchased back in 02/21/2005 and works very well. I have been able to downgrade & upgrade successfuly I recently downloaded DVP630 EW firmware and so far everything is working great. I have had 0531,1109 and others on my unit with no failur's. I can't remember what firmware this unit last had LMAO, I flashed it too many times hehe.

Here's some info on my receipt if it helps.

Philips DIVX DVD ACCS
CPU: NMS
Purchased from: WalMart.Com
Purchase Date: 02/21/2005
Price: $69.84
Model: DVP642/37
Firmware used: v1109, v0531, v0603 (Also v0112 - Released 22/23 Feb 2005)

[Jefferson Ryan]

I have 2 DVP642/37 (U.S.) units, both unfortunately of the "E" variety in which there is no older (macrovision disabled) firmware available.

Actually, we managed to extract a 0531 "E" firmware directly from a flash chip of an 0531 "E" model. The chip was dessoldered and put in a EPROM Programmer.

After the reading, the chip was soldered in place and the extracted firmware was burned in a CD and tested in the same unit. The "upgrade" was OK.

But we not tested yet if the 0531 "E" firmware from Philips really is Macrovision free like the others. We compared the 0531 "E" with the 0531 "S" and our extracted firmware is newer (this is not a surprise, though).

[oayz]

The numbers referenced above are date codes. The last number is the year of production, and the two numbers before that are the week of production.

G254 = Made in the 25th week of 2004.
F074 = Made in the 7th week of 2004

I don't know, however, whether the preceding letter has any significance.

Usually it means revision of the chip. G is a newer revision, maybe some bugs are fixed (and some new are introduced :-)

[romgohan]

It is over a year when that question was asked, I wonder if now someone knows the answer... Does DVP-630 support the JTAG interface? Is there a way to revive the player?