How to copy Philips 3575 / Magnavox 513 DVD recordings to PC?
The IsoBuster developer and I are looking at the Philips 3575 and by association the Funai/Magnavox MDR513 DVD/HDD recorders for video extraction. They seem to have the same filesystems. If development and testing go like the Toshiba's did I'll post an update in this thread.
Please excuse the 'click-bate' title, but it serves a useful purpose. Lordsmurf stated his preferences for which recorders to look at next. I looked at what recorders I had for testing and what I could get my hands on. As a direct result of that.. it drove the decision of what to look at next. Followup suggestions: RCA DRC8030 LiteOn LVW5045 Polaroid 2001G JVC MX/MH/DVM |
1 Attachment(s)
I finished looking at the 3575 for a while, it was HDDfs 07.0 format which meant it had a volume, volume table, and a static set of files. The files held a single .IFO and .MPG file set and the .IFO held recording content location, the .MPG held actual recording content. Hopefully the developer can figure out recovery from there.
Just started looking at the DRC8030N, it has a Zoran Activa100 chipset, with 3D Y/C comb filter and frame sync TBC with MPEG2 encoder. I've never seen one until tonight and the image quality with a good signal is quite good. Its filesystem says it is using TinyUDF. The software tool almost appears to read it natively. By that I mean it does not startup and immediately list the recordings, but after a forced scan identifies the TinyUDF volume as a UDF volume and files in that volume, extracting the first .mp3 reveals a watchable MPEG-PS file. TinyUDF seems to be a subcategory of a true UDF volume that might be missing some of the tell tale markers of a true Fully compliant UDF volume.. but after a manual scan the tool reconciles it by assuming its a damaged version of a Fully compliant UDF volume. The problem with a manual scan is it is time consuming, it has to read every sector of the entire disk to make sure it has found all recordings. It would be faster and more convenient to simply acknowledge it has found a correct UDF, list the recordings and extract the recordings based on the file allocation data at the beginning of the HDD. The file set names also appear to be separate .IFO and .MPG files per recording, so it may be simpler to list them rather than read them all from one .IFO file. Will have to see about that. Usually a .VRO title set has one .IFO per recording, assuming the redundancy and independency makes it more resistant to damage on a DVD-recording disc... which is probably true on a HDD disc, but less important since its better protected. That is to say.. unlike several other brands where there is one "single" .IFO file for the entire HDD and one "single" .MPG file for the entire HDD. This recorder has multiple .IFO and .MPG files, one set for each recording. (I cannot say that for sure and with confident certainty, at this point it simply "appears" to be the case.) The Zoran Activa line had at least three generations of its DVD chipset, the 100, 150 and 200. But the DRC8030 used the earliest version from 2005/6. It is possible the later versions used the next gen version, however its very hard to tell from the user manuals online whether a DRC had a HDD or did not have a HDD. The only "hint" seems to be if the manual mentions a feature called the "Library". |
Yep, I've said for years how stunning Zoran is quality-wise, and probably the best chipset in terms of legal NTSC values. If only it removed chroma, it'd beat the LSI for tape work. The Toshiba was in between Zoran and LSI, in terms of absolute chipset. But for tape work, the chroma noise removal of the LSI simply cannot be beat.
- I've used the RCA (Zoran) for off-air for 10+ years now. - Philips as backup, or for 16x9 sources (recorded to 16x9, but misflagged as 4x3 when ripped, needs manual correction). - JVC (LSI) for tapes. The RCA usually does not care about copy protection flags. Fox Sports is the only one I've seen where it may happen. The Philips picks up anti-copy even when none exist, so you have to use a Grex to record. Ironic, seeing as how the Grex normally screws up sources. (That leads me to believe the Grex was designed for perfect signals with obvious protection markers, not chaotic tapes.) I hate having to re-record to the RCA, with Grex in middle -- because the output will halt, cause RCA to lose signal. It's be better to plop HDD into enclosure, bypass that nonsense. JVC LSI doesn't care, especially the Professional line. None really correct input like Panasonic. DVD recorders were all a really beta-quality technology that never matured. We never got a great machine that did everything, worked perfectly. And a reason it was abandoned. But it's what we have, hence why I poured in so much research in the 2000s to identity the best units. |
Lordsmurf.. I've wonder about that "Chroma Noise Reduction" problem.
Is there or (was there) a dedicated box for "CNR" or was what was available really bad? I've just started looking at gear Churches have been throwing out in droves in the last few years as they went "digital". There are a lot of "Scalers" for big LCD or Projection systems.. and they have what appear to be massive signal processing systems built-in.. in particular very fine control CNR filters. My guess would be long cables to the panels or projection systems made this mandatory. This stuff isn't tubes or transistors.. its dedicated custom, pre-FPGA silicon.. so I'm thinking its very recent tech, circa 2011 or 2014 at the latest. If it were used on the VCR output before the Zoran encoder.. would this be a winning combination? |
You'll have to test it. I've seen many claims of chroma NR, all have failed to do. Same for dropout compensation. Pretty words on paper the read impressive, but do nothing in practice. Or "is that it?", pathetic processing.
I'm all for finding a magic box to do external chroma noise removal. I think the key phrase is "reduction" or "removal". The LSI tends to entirely remove the noise. TBCs and other devices that "reduce" it are laughable quaint, performing even worse than the rudimentary filters we had 10-20 years ago. |
1 Attachment(s)
Quote:
I caution this is with "overrides" and needs to be "fixed" in the mainstream auto detection code to make it easier for people to use. But I've been able to play all of the recordings in VLC without any issue. Attachment 9668 |
Don't know if it's any help but the philips recorders apparently have a hidden RS-232 port (with a non-standard connector) inside that gives pretty wide access to the internals via the service mode, much more than your average service remote would.
|
Quote:
-- merged -- not much to report tonight. confirmed 3575, H2160MW, 513 all share the same filesystem HDDfs |
1 Attachment(s)
Ok.. lots happened really fast tonight. All recordings, fragmented or not, are extracting completely.
Only tested on 3575 for now. I'll be moving on to test with the 2160 and 513 maybe later tonight... may have something for other 3575 people to test by this weekend. The RCA DRV8030 will not be part of this pre-release. Support for that looks very good.. but it should be tested better first. |
ugh.. testing.. we found several edge cases with the 3575 that were easily handled
but.. the mag-2160 and mag-513 seem to form a second variant of the HDDFs.. they do not decode properly with a straight version of the philips-3575 code.. looks supportable.. explains some massive confusion people had years ago trying to decipher the format. Its not that they were encrypted.. its that they had the same internal labels but were actually "different".. duh... never assume conspiracy, where ignorance will suffice.. we didn't know we were looking at two different things. my technical perspective is the mags "switched" from one type of microprocessor to another and had to change operating systems.. they "tried" to maintain compatibility for their developers.. but slimmed down the feature set.. and introduced a "mistake or correction" in the HDDFs format that propagated forwards into later models.. switching entirely to a different format in the 865, 867 but maintaining the same/similar user features.. gosh.. it was a moving format circus that outsiders couldn't figure out.. again I'm not the developer.. but he is brilliant.. I'm just a spectator testing on the sidelines.. observing |
There was also a Philips 3576, so perhaps the Magnavox are more closely aligned with it? I'd bet that's the case, as the Magnavox clones came 6-18 months later than the 3575 release. People often insisted the Philips and Magnavox machines were "the same", but I was never convinced of it. Very similar, yes. Same chipsets, probably. But "the same", unlikely, given the time lapse. Not a straight rebadge.
|
Quote:
I'm finishing off the testing tonight and he is hunting edge cases.. but they are coming fewer and fewer, he's starting to anticipate their every move. We are so far down the rabbit hole I don't think anyone will ever run into these edge cases. In summary, to be clear.. due to my more limited collection or access to philips and magnavox models, these are now working correctly: 3575, 2160, 513 (not) the 865, 867.. we both agreed the 865 and 867 will have to be a project to themselves.. if there is ever interest shown.. if no one speaks up.. they may never be looked at. Right now we're looking specifically at the models you've mentioned Lordsmurf. Next up after this alpha version for general feedback.. will be the RCA DRC8030.. but that looks basically done, it works.. we're just not trying to throw everything out at once. The 5045 and 2001G I "guess" will be next.. but the JVC MX/MH series looks intriguing.. possibly doable.. maybe be more of a challenge. The DVMs looks like they may be harder. I'd truly like to get to the Pioneers and Panasonics.. but zero interest has shown up so far.. could be a long time down the road. Either I or the developer may end up getting bored or distracted by other things. Its not a priority for either. I will have to look on 'Bay to see if someone is tossing a 3576.. I didn't know they exist. Spending money isn't my preference.. if it were a 515 would be nice to try. Have access to an 865, 867 but I think we're lumping those into the Panasonic camp due to their disk labels that "seem" to say internally they are (were?) actually Panasonic Blu-Ray recorders. |
Quote:
Quote:
Quote:
I sold my LiteOns some years ago, so those aren't as interesting to me anymore. But the method is probably similar to the Polaroid. BTW, all of those have LSI chipsets. Quote:
However, something else we've not really discussed is a disaster recovery plan. If the drive goes sideways once removed, we need some sort of reimaging ability. That aspect has less to do with ISOBuster, but it's something we can provide downloads of those images (and tools?) here on this site. In fact, I'd like to have a whole series of guides (images/text) on this site, per recorder, showing how to remove the HDD (tips, things to avoid), safely plug it in (Linux? Windows?), and then how to use ISOBuster to extract. Would you mind doing that? I'd get that published on the main site. In PM, I can show you an early beta of the new site, so you'll have an idea of how it's look. It's quite nice. Quote:
Quote:
Quote:
We're going to start sending out monthly emails to members soon (by summer), something we had to shy away from due to technical (and time) limits in more recent years. I've been prepping some important items to go in the first send-out, and I could add a short blurb about us seeking out old HDD DVD recorders. Odds are that we have a few members here that have recorders collecting dust in a closet, that would donate those for R&D purposes. I'll PM you as that date approaches, so we can plan some materials for it. |
I found a 3576, thanks I didn't know it existed.
I'll test that model when it arrives. Your welcome to the Alpha version of IsoBuster for testing, the 3575 works well with it now. The support for the 2160 and 513 was just cleaned up early this morning.. I weakly tested it on the way out the door to work.. it needs better testing. The developer is concerned the segment map is larger than the philips map and he might be missing a few millisecs before or off the end of a recording. That said, it looked perfect.. all recordings recovered and played back. Consider the Polaroid 2001G next after the RCA DRC8030. "Disaster Recovery" - IsoBuster has an image mode built into its right click toolkit when browsing the drive contents.. you could back a drive up to an image "before" doing anything if you like.. I don't know about restore though.. I'd have to ask him. - One of the reasons I tried to contact this developer is IsoBuster does not use windows to try and (mount) the hard drive.. its "touchless".. so it doesn't mess around with trying to "write" to the drive. After a dozen or so models and drives.. I'm overconfident it won't invalidate a drive. "Guides" - I rant and write a lot while working on problems.. but I'm not sure other people always understand me. For now I think its better to attempt as many brands and models as we can, where people show interest.. before the developer looses interest. We both have other things to do and this is "fun" for the moment. It may not always be like that. He's also suggested I do more videos.. to show the procedure.. I may do that. He doesn't have access to any of this hardware and he is on the opposite side of the planet from me. I'm making suggestions and doing testing. Its fantastic he seems to understand me most times and seems to get so many impossible puzzles "right"... its been like a game.. each time I think we're toast.. no way he can solve that.. he pops up in my email and says (Its Ready). Think Sherlock Holmes and Watson.. I'm Watson.. he's Sherlock... lol.. I kind of look like M.F. too. "JVC DVMs" well.. I've looked at them.. ahh.. I hestitate to say.. (I have been able to extract recordings from them) so they are not enrypted.. but the dataset (might be encrypted).. I think they would take longer than 1 day to figure out.. I haven't tossed one the developers way yet. (I'm probably over thinking them) they are probably super easy a puzzle.. but I've been picking the popular.. seemingly in the past.. most reverered recorders to look at first. "JVC MX/MH" these look a lot easier to decipher.. and might be stepping stones to the DVM. But no one has shown any interest in them except you and I in a small way. "Panasonic EH" i've seen other people worked on these.. and produced code.. thats a good sign.. but we've never looked at a Panasonic yet.. if there is real interest we might get to this. The real danger to ending the cases of Sherlock Holmes.. is lack of interest for the moment ps. On second thought let me do a video tonight or tomorrow night.. lets not rush things. You probably need confidence you aren't going to loose any recordings and want a carefully described procedure... as I said, I'm not that confident my writings will be legible to anyone.. so a video might be the best thing. ( I'm also not much of a script writer.. but I'll try not to ramble.. :) ) |
Quote:
Quote:
Quote:
2. The I can/will edit anything you write, images you take, into a more streamlines guide. So don't worry about rambling. Quote:
Quote:
|
Enable or Disable Automatic Mounting of New Disks
HDD Raw Copy Tool - Has been my "go to" tool for raw image backups and restores. I did bring up Linux a couple times.. he doesn't do Linux. No interest in it. "Guides" - I'll send you a draft "Videos" - squelched "20 years old" - yeah |
Found the filesystem on the Polaroid 2001G
Its something called TFS2 and was part of the LSI Logic 2004 EVB package. Apparently it went into a lot of LG, Tevion, Salora and (unconfirmed) possibly the JVC MH series DVRs (maybe even the JVC DVM series) I can't be sure but the tantalizing clues are it follows a fairly regular pattern and video clips extract clean and playback from my recorder okay.. so its mostly about sorting out the directory structure, and the VOB segments. It will (not) be part of the next Alpha release.. but it looks doable at some point. |
Tested this alpha with the Phlips 3575, and now its been tested with a Philips 3576.
Recordings are copyable from the HDD to PC as before. However the 3576 seems to have adopted a run-out leader similar to all recordings on the 2160 and 513. I don't think it will be fixed since it does not effect the recording, its simply a nuisance at this time. The RCA DRC8030 will be the next focus.. then the Polaroid 2001G. The next alpha will only support Toshiba, Philips and Magnavox, but that is two more brands than the previous alpha. |
Quote:
HDD DVD recorders were essentially this: - JVC LSI - other LSIs - Panasonic - Philips and Magnavox - Pioneer - Toshiba - RCA - Sony - random brands with chipset kits from Cirrus, and .... I forget. (Mediatek was another?) I need to pull out my research. Finish my unfinished list (but at this site): https://forum.videohelp.com/threads/...-LIST%21/page2 - LSI was best for tape conversions, removed chroma noise, but IRE a bit hot on some units/models - Zoran (RCA) was best for off-air = clean - then Toshiba = some filtering, pretty clean - then Philips/Magnavox = 16x9, pretty clean - then Pioneer = pretty clean, some softening and maybe +2 IRE - then quality tanked, and you got noisy AC3-problematic Panasonic - then Sony managed to make decks that made Panasonic look good - then all the other decks, mostly crap I'm sure there are model variations, but I'd assume those would all be fairly similar. So you you have 8 basic recorder types here, excluding random brands like Funai |
The RCA DRC8030 is now working, it will be in the next public Alpha shortly.
Today we'll start focusing on the Polaroid DRM2001G - B |
Site design, images and content © 2002-2024 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2024 Jelsoft Enterprises Ltd.